Targeted intracerebral delivery of the anti-inflammatory cytokine IL13 promotes alternative activation of both microglia and macrophages after stroke

Targeted intracerebral delivery of the anti-inflammatory cytokine IL13 promotes alternative... Background: Subtle adjustment of the activation status of CNS resident microglia and peripheral macrophages, to promote their neuroprotective and neuroregenerative functions, may facilitate research towards curing neurodegenerative disorders. In the present study, we investigated whether targeted intracerebral delivery of the anti-inflammatory cytokine interleukin (IL)13, by means of transplanting IL13-expressing mesenchymal stem cells (IL13-MSCs), can promote a phenotypic switch in both microglia and macrophages during the pro-inflammatory phase in a mouse model of ischemic stroke. eGFP/+ RFP/+ Methods: We used the CX CR1 CCR2 transgenic mouse model to separately recognize brain-resident microglia from infiltrated macrophages. Quantitative immunohistochemical analyses were applied to characterize polarization phenotypes of both cell types. Results: Distinct behaviors of both cell populations were noted dependent on the anatomical site of the lesion. Immunohistochemistry revealed that mice grafted with IL13-MSCs, in contrast to non-grafted and MSC-grafted control mice, were able to drive recruited microglia and macrophages into an alternative activation state, as visualized by a significant increase of Arg-1 and a noticeable decrease of MHC-II expression at day 14 after ischemic stroke. Interestingly, both Arg-1 and MHC-II were expressed more abundantly in macrophages than in microglia, further confirming the distinct behavior of both cell populations. Conclusions: The current data highlight the importance of controlled and localized delivery of the anti- inflammatory cytokine IL13 for modulation of both microglia and macrophage responses after ischemic stroke, thereby providing pre-clinical rationale for the application of L13-MSCs in future investigations of neurodegenerative disorders. Keywords: Stroke, Microglia/macrophage polarization, Interleukin 13, Mesenchymal stem cells, Neuroinflammation * Correspondence: mathias@sf.mpg.de Somayyeh Hamzei Taj and Debbie Le Blon contributed equally to this work. Peter Ponsaerts and Mathias Hoehn shared senior authorship. In-vivo-NMR Laboratory, Max Planck Institute for Metabolism Research, Gleuelerstrasse 50, D-50931 Köln, Germany Department of Radiology, Leiden University Medical Center, Leiden, Netherlands Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 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. Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 2 of 17 Background can be considered as competent carrier cells, since they Ischemic stroke following cerebral artery occlusion is a (i) are conveniently engineered with viral vectors to pro- major cause of chronic disability worldwide and an effect- vide long-term gene expression [12, 13], (ii) have re- ive therapy to improve functional recovery after stroke is duced immunogenicity because of a low MHC class I missing in the clinical settings [1]. A time-dependent re- expression and the absence of MHC class II molecules cruitment and activation of immune cells, including and co-stimulatory factors [14], (iii) display a robust sur- brain-resident microglia, monocytes/macrophages, granu- vival rate upon transplantation in CNS tissue [15, 16], locytes, and T cells are a hallmark of the secondary dam- (iv) have a potential immunomodulatory effect on natural age after ischemia [2]. Under inflammatory condition killer cell function and B and T cell proliferation [17, 18], caused by ischemia, there is a complex interaction be- and (v) are able to produce immunomodulatory, neuro- tween brain-resident microglia and infiltrated macro- trophic, and angiogenic factors [19–21]. Therefore, the phages. These two phagocytic cell populations share many therapeutic relevance of MSCs in stroke models and other antigenic markers, although rapidly growing literature on neurodegenerative disorders can be applied upon intrace- this subject indicates that both cell types have distinct spe- rebral or intraventricular administration. cialized functions [3, 4]. In this study, we performed stroke The first aim of this study was to investigate the eGFP/+ RFP/+ experiments in CX CR1 CCR2 knock-in fluor- spatial distribution and the behavior of recruited resi- escent protein reporter mice, in order to distinguish dent microglia and infiltrating, peripheral macrophages brain-resident microglia from infiltrated macrophages in a transient focal cerebral ischemia model, to better after ischemic stroke. In these transgenic mice, the che- understand the contribution of these cell populations to mokine receptor CX CR1 is mainly expressed by the inflammatory process after stroke. By using the eGFP/+ RFP/+ brain-resident microglia, while CCR2 is overexpressed CX CR1 CCR2 transgenic mouse model, our in infiltrating activated macrophages [5]. A profound results have provided new insights into the trafficking and characterization of these cell populations in damaged distinct functions of resident microglia and infiltrating CNS may provide the opportunity to manipulate spe- macrophages in the post-ischemic brain. The second aim cific subsets for therapeutic benefits. of this study was to discover whether MSC-based delivery In the early stage of ischemic stroke, these immune of IL13 can modulate the spontaneous polarization shift cells can acquire a protective function known as of anti-inflammatory (M2) to pro-inflammatory (M1) anti-inflammatory (M2) phenotype, i.e., presenting char- phenotypes in microglia and macrophages during the acteristic features comparable to in vitro IL4 and/or second week after ischemic stroke. We agreed on ap- IL13 polarized cells. In the late stage of ischemic stroke, plying the simplified classification of pro-inflammatory they can polarize into a classically activated state known and anti-inflammatory states as M1 and M2 pheno- as pro-inflammatory (M1) phenotype, i.e., presenting types, respectively, in order to best describe our intention characteristic features comparable to in vitro LPS and/or of modulating the polarization of microglia and macro- interferon gamma (IFN-ϒ) activated cells [6]. Emerging phages. Our results have provided confirmatory evidence knowledge in targeting cerebral inflammation argues that transplantation of IL13-MSCs, which continuously se- that therapeutic approaches should move from extensive crete IL13, is able to polarize both microglia and macro- and forcible suppression of immune cells towards careful phages to a neuroprotective M2 phenotype during the adjustment of the balance between their different pheno- pro-inflammatory status in ischemic stroke. These data types [7]. Therefore, to drive microglia and macrophages strengthen the idea that IL13-MSCs could be considered as towards a more protective anti-inflammatory phenotype potent modulators of the cellular and molecular responses after stroke, we selected the immunomodulatory cyto- in neuroinflammation, with strong anti-inflammatory po- kine interleukin 13 (IL13), which is a well-known modu- tential, for further explorationinanimalmodelsofstroke lator of immune responses in vitro and in vivo [8, 9]. and other neurodegenerative disorders. The neuroprotective importance of the cytokine IL13 has been demonstrated in several experimental models of Methods neurodegenerative disorders. It can strongly decrease the Animals and experimental groups pro-inflammatory cytokine secretion, reduce inflammatory Wild type C57BL/6J mice were obtained via Charles cell infiltration, and suppress axonal loss [10, 11]. River Laboratories (strain code 027) and were used Effective delivery of therapeutic proteins, such as IL13 for in vivo detection of IL13 mRNA produced by in the present study, is a main challenge in neuroinflam- grafted IL13-MSCs. In total, 24 mice were included mation research. To efficiently deliver IL13 to the in the experiment, divided over the following three ischemic brain, we used autologous mesenchymal stem/ groups: (i) a non-injected control mice (n = 8), (ii) stromal cells (MSCs) genetically engineered to secrete MSC-grafted mice (n = 8), and (iii) IL13-MSC-grafted IL13. As previously reported by us and others, MSCs mice (n =8). Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 3 of 17 eGFP/eGFP Transgenic CX CR1 mice (strain code medium (IMDM; Lonza) supplemented with 8% fetal bo- RFP/RFP 005582) and CCR2 mice (strain code 017586) vine serum, 8% horse serum, 200 U/mL + 100 μg/mL peni- eGFP/+ were obtained via Jackson Laboratories. CX CR1 cillin/streptomycin, and 1 μg/mL amphotericin B (all RFP/+ CCR2 mice were obtained by breeding products from Thermo Fisher Scientific). Culture medium eGFP/eGFP RFP/RFP CX CR1 mice with CCR2 mice. During for IL13-MSCs was further supplemented with 5 μg/mL the entire study, mice were kept in the animalarium of puromycin (InvivoGen). MSC cultures were split 1:5 the University of Antwerp (UA) under normal day-night twice a week using 0.05% trypsin-EDTA (Thermo cycle (12/12) with free access to food and water. All Fisher Scientific) for cell detachment. animal experimental procedures were approved by the Ethics Committee for Animal Experiments of the UA IL13 protein secretion assay (Approval No 2015-84 and 2018-36). Wild type MSCs and IL13-MSCs were plated at a con- eGFP/+ 5 Ischemic stroke was induced in 32 male CX CR1 centration of 2 × 10 /well in a six-well plate and allowed RFP/+ CCR2 mice (11–13 weeks, 22–26 g) by middle cere- to adhere during overnight incubation. After the following bral artery occlusion (MCAO), as described below. 24 h of culture, supernatant was harvested and analyzed Stroke lesions were checked by magnetic resonance im- for the presence of IL13 protein by means of ELISA, ac- aging (MRI) scans 48 h after MCAO, before transplant- cording to the manufacturer’s instructions (Peprotech). ation of mesenchymal stem cells (MSCs). Two mice died after stroke induction (stroke-only group), one mouse Cell transplantation in healthy mouse brain before stroke during anesthesia. Four mice were excluded All surgical experiments were performed under sterile due to too small or absent stroke lesions. Eight mice died conditions, as previously described [26–28]. Briefly, mice after cell implantation (n = 3 for MSC group; n =5 for were anesthetized by an intraperitoneal injection of a IL13-MSC group). In total, 17 mice were included in the ketamine (80 mg/kg, Pfizer) + xylazine (16 mg/kg, Bayer study, resulting in the following three groups: (i) control Health Care) mixture in 0.9% NaCl solution (Baxter) group subjected to stroke only (n = 6), (ii) MSC-treated and placed in a stereotactic frame (Stoelting). A midline group, which received MSC transplantation 2 days after scalp incision was made and a hole was drilled in the MCAO (n = 5), and (iii) IL13-MSC-treated group, which skull using a dental burr drill (Stoelting). Stereotactic co- received a transplantation of IL13 producing MSCs 2 days ordinates were as follows (relative to bregma): AP 0 mm, after MCAO (n = 6). All mice were perfused at day 14 Lat. 2.3 mm and − 2.3 mm, and DV − 2.3 mm. Next, an after stroke induction. An overview of the experimental automatic microinjector pump (kdScientific) with a protocol is presented in Fig. 1. 10 μl Hamilton Syringe was positioned above the ex- posed dura. A 30-gauge needle (Hamilton), attached to Genetic engineering and culture of MSCs the syringe, was lowered through the intact dura and In this study, we used a previously established and charac- positioned at the respective depth. After 2 min of tissue terized C57BL/6 mouse bone marrow-derived MSC line pressure equilibration, a suspension of 5 × 10 MSCs in [22] and a derivative thereof, genetically engineered to ex- a volume of 0.4 μl was injected. Following 4 min to allow press murine interleukin-13 (further named as IL13-MSCs) pressure equilibration and to prevent backflow of the [23]. The latter MSC line was generated by transduction of injected cell suspension, the needle was fully retracted. parental MSCs with the pCHMWS-IL13-IRES-Pac lenti- The exact same procedure was performed at both left viral vector, according to previously optimized procedures and right side of the brain. The skin was sutured (Vicryl, [24, 25]. For expansion, both MSC lines were cultured in Ethicon), and 100 μl of a 0.9% NaCl solution was adminis- standard cell culture plastic ware in “complete expansion tered subcutaneously in order to prevent dehydration medium” [24] consisting of Iscove’s modified Dulbecco’s while mice were placed under a heating lamp to recover. Fig. 1 A time line of the experimental protocol. Stroke was induced in all mice via the middle cerebral artery occlusion (MCAO). All mice were scanned by MRI at day 2 after MCAO. MSCs or IL13-MSCs were transplanted into the right hemisphere, ipsilateral to the lesion, after performing MRI. Motor and sensory performance were tested and ranked according to the modified neurological deficit scores (mNDS), once before MCAO (baseline) and every 2 days after MCAO until the day of perfusion. All animals were sacrificed at day 14 after MCAO induction. Abbreviations: B = behavior, MCAO = middle cerebral artery occlusion, MRI = magnetic resonance imaging, IHC = immunohistochemistry Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 4 of 17 qRT-PCR for transgenic IL13 mRNA and received a subcutaneous injection of 4.0 mg/kg Car- Mice were perfused with ice-cold 0.9% NaCl solution, profen (Pfizer, Karlsruhe, Germany) for analgesia after directly followed by removal of the brain and dissection the surgical interventions. With a neck incision, the of the transplantation areas (left and right). The ex- common carotid artery (CCA) and its proximal branches tracted tissue sections were snap-frozen in liquid nitro- were exposed. The internal carotid artery (ICA) was oc- gen and kept at − 80 °C until further processing. As cluded for a short time with a metal microvessel clip. A negative and positive control for qRT-PCR analysis, cul- silicon rubber-coated filament with a tip diameter of tured MSCs and IL13-MSCs were harvested, washed, 170 μm (7017PK5Re, Doccol Corporation, Sharon, MA, and resuspended in RNAlater (Qiagen) solution at 4 °C USA) was inserted into the ICA. The filament was ad- for further processing the next day. Total RNA was ex- vanced through the ICA until it blocked the blood flow tracted using the Purelink RNA kit (Invitrogen). RNA to the middle cerebral artery. Animals were allowed to quantity and purity were determined using an ND-1000 recover during the 30 min occlusion in a temperature micro-spectrophotometer (NanoDrop Technologies). stable box (MediHeat, Peco Services Ltd., Brough, UK). Two micrograms of total RNA was reverse-transcribed Afterwards, the animals were re-anesthetized, the fila- using Omniscript RT kit (Qiagen). PCR primers were ment was carefully removed to initiate reperfusion and designed using online available primer design software the CCA was permanently ligated. During 1 week after from Thermo Fisher Scientific and were purchased from the surgery, body weight was daily monitored in all Thermo Fisher Scientific. The forward primer (5′ to 3′) animals. Animals were randomly assigned to the three GAAGCCGCTTGGAATAAGGC and the reverse primer different experimental groups. (5′ to 3′) ACCTTGCATTCCTTTGGCGA cover part of the IRES sequence within the pCHMWS-IL13-IRES-Pac IL13-MSC injection following MCAO lentiviral construct, thereby allowing to detect only trans- All surgical experiments were performed as described genic IL13 mRNA and not endogenous IL13 mRNA. earlier under section cell transplantation in healthy Real-time quantitative RT-PCR analysis was carried out mouse brain. The needle was positioned at a depth of using Power SYBR Green PCR Master Mix (Applied 2.5 mm into the striatum close to the ischemic lesion (as Biosystems) detection, using the StepOnePlus Real-Time determined following MRI analysis). First, a suspension PCR System (Thermo Fisher Scientific). Thermal cycling of 2 × 10 MSCs in a volume of 0.4 μl was injected. Fol- conditions were 10 min at 95 °C and 40 cycles of 15 s at lowing 4 min of pressure equilibration, the needle was 95 °C and 1 min at 60 °C. Melt curves were performed retracted to a depth of 1.0 mm and a second cell infusion upon completion of the cycles to ensure specificity of (2 × 10 MSCs in a volume of 0.4 μl) was performed. the product amplification. Housekeeping genes for normalization were peptidylprolyl isomerase A (ppiA) Behavioral tests for detection of transgenic IL13 mRNA in control and To observe the different aspects of neurological func- MSC- and IL13-MSC-grafted mouse brains, using for- tions, a modified neurological deficit score (mNDS) was ward primer (5′ to 3′) CAGACGCCACTGTCGCTTT performed before and every 2 days after MCAO, based and reverse primer (5′ to 3′) TGTCTTTGGAACTT on a modification of a previous report [31]. In brief, the TGTCTGCAA and GAPDH for detection of transgenic modified NDS consists of a set of motor tests (muscle IL13 mRNA in in vitro cultured MSC and IL13-MSC, status and abnormal movement), sensory tests (tactile, using forward primer (5′ to 3′) AGGTCGGTGTGAAC and proprioceptive), and reflex tests on a scale of 0–18. GGATTTG and reverse primer (5′ to 3′)GGGG Increasing score indicates the severity of the stroke TCGTTGATGGCAACA. Data were analyzed with damage [32]. qbase+ analysis software. For comparison purposes, ob- tained values for MSC-grafted and IL13-MSC-grafted MRI acquisition mice are displayed as fold expression versus the mean At 2 days post induction of stroke, mice were subjected value of control non-injected mice. Similarly, expres- to MRI using a 7T Pharmascan MR scanner with a sion of transgenic IL13 mRNA for IL13-MSC is dis- 16-cm-diameter horizontal bore (Bruker, Ettlingen, played as fold expression versus control MSC. Germany). This system is equipped with a standard Bruker cross-coil setup, using a quadrature volume coil Middle cerebral artery occlusion for excitation and an array mouse surface coil for signal Focal cerebral ischemia was induced by transient occlu- detection. The system was interfaced to a Linux PC sion of the right middle cerebral artery (MCA) in all ani- running Topspin 2.0 and Paravision 5.1 software (Bruker). mals, using the intraluminal filament model as described Mice anesthesia was induced using 2% isoflurane (Forane previously [29, 30]. Briefly, mice were anesthetized with Abbott, UK) in a gas mixture of 30% O and 70% N at a 2 2 1–2% isoflurane in a gas mixture of O /N O (30:70%) flow rate of 600 ml/min. During MRI acquisition, 2 2 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 5 of 17 isoflurane concentration was set at 2%, and the respiration Fisher Scientific, A21245). Before mounting with Prolong rate was continuously monitored using a pressure sensi- Gold antifade reagent, nuclear staining was performed tive pad. In addition, body temperature was monitored via using TOPRO-3 (1/200, Thermo Fisher Scientific). Images a rectal probe and was held constant between 37.0 and were acquired using a BX51 fluorescence microscope 37.3 °C using warm air coupled to a feedback unit (SA equipped with an Olympus DP71 digital camera. Olympus instruments, NY, USA). Both respiration and body cellSens dimension software was used for image process- temperature control systems were controlled by PC-Sam ing. In each brain section, five different regions of interest monitoring software (SA instruments, NY, USA). Following (ROIs) were selected with the constant exposure time: the MR image acquisition, mice were left to recover separately border and the core region of the ischemic hemisphere in under a heating lamp before returning to their respective the striatum, the core region of the ischemic hemisphere cages. Following scout scans, we acquired an axial proton in the cortex, and two ROIs in the cortex and striatum of density weighted Rapid Acquisition and Relaxation En- the intact contralateral hemisphere. hancement (RARE) image using the following parameters: repetition time (TR) = 3000 ms, echo time (TE) = 13.3 ms, Quantitative immunohistochemical analysis matrix size (256 × 256), field of view (FOV) = (17.5 × 17.5) For quantitative phenotypic analysis of microglia and 2 2 mm , resolution = (0.068 × 0.068) mm , 10 coronal slices, macrophage activities after stroke, immunofluorescence slice thickness = 0.8 mm, RARE factor = 8, and aver- images were analyzed using NIH ImageJ analysis soft- ages = 2. Next, for determination of the T2 values, we ware (ImageJ) and TissueQuest 4.0 (TissueGnostics, applied a multi-slice, multi-echo sequence based on the Vienna, Austria). An entire picture of all ROIs, taken at Carr-Purcell-Meiboom-Gill sequence. The following × 20 magnification, was used for quantification. Cells parameters were used: TR = 3000 ms, number of were recognized based on the nuclei staining (nuclei echoes = 10, echo spacing = 11 ms, matrix size (128 × 128), size, staining intensity, and discrimination by area was field of view (FOV) = (17.5 × 17.5) mm ,resolution= optimized manually), followed by the analysis of specific 2 + − (0.137 × 0.137) mm , 10 slices, and slice thickness = staining. The cellular densities of eGFP RFP microglia eGFP/+ − + RFP/+ 0.8 mm. (CX CR1 ), eGFP RFP macrophages (CCR2 ), + + and eGFP RFP double-positive microglia/macrophages eGFP/+ RFP/+ Immunofluorescent staining (CX CR1 CCR2 ) were quantified. To achieve All immunofluorescence analyses were performed ac- optimal cell detection, the background threshold was de- cording to previously described procedures [26, 28]. fined. To discriminate false signals due to the coverage of eGFP/+ Briefly, mice were transcardially perfused with 0.9% CX CR1 cell ramification with total nuclei intensity, NaCl solution followed by 4% paraformaldehyde (PFA). the cut-off was defined for each ROI. Scattergrams were Next, brains were isolated and post-fixed with 4% PFA generated to visualize the corresponding positive cells in for 3 h, then dehydrated through a sucrose gradient of the source ROIs through the real-time back gating com- 5, 10, and 20%. Brains were snap-frozen in liquid nitro- ponent. Mean intensity and the relative number of the eGFP/+ RFP/+ gen and kept at − 80 °C until further processing. co-expressed CX CR1 or CCR2 with pro- or Ten-micrometer-thick cryosections were made using a anti-inflammatory markers F4/80, Arg-1, and MHC-II microm HM500 cryostat. Immunofluorescent staining were obtained. Afterwards, the mean values were deter- was performed on brain slides using the following anti- mined from analyses of at least three brain sections per body combinations: a primary rabbit anti-RFP antibody mouse. Based on the above cell density analysis, the pro- (2.5 μg/ml, Abcam, ab62341) with a secondary donkey portion of microglia or macrophages expressing F4/80, anti-rabbit Alexa Fluor 555 antibody (2 μg/ml, Thermo Arg-1, or MHC-II was calculated. To observe anti- and Fisher Scientific, A31572), a primary rat anti-F4/80 pro-inflammatory representative markers in double stain- + + antibody (4 μg/ml, Bio-Rad, MCA497R) with a second- ing with eGFP or RFP , the above-defined five different ary goat anti-rat Cy5 antibody (10 μg/ml, Thermo ROIs provided the following number of cells/mm :(i) Fisher Scientific, A10525), a primary rat anti-MHC-II total cell density according to the nuclei staining of + + antibody (2.5 μg/ml, eBioscience, 14–5321-82) with a TOPRO-3, (ii) eGFP microglia cell density, (iii) RFP + + secondary goat anti-rat Alexa Fluor 350 antibody macrophage cell density, (iv) eGFP RFP double-positive + + (10 μg/ml, Thermo Fisher Scientific, A21093), and a microglia/macrophages cell density, (v) F4/80 /eGFP or + + + + primary goat anti-Arg1 antibody (4 μg/ml, Santa Cruz, RFP cell density, (vi) Arg-1 /eGFP or RFP cell density, + + + sc-18354) with a secondary donkey anti-goat Alexa and (vii) MHC-II /eGFP or RFP cell density. Fluor 350 (10 μg/ml, Thermo Fisher Scientific, A21081) and a primary rabbit anti-Ym1 antibody (1:50 dilution, Statistics StemCell Technologies 01404) with a secondary goat For statistical analyses of qRT-PCR and ELISA data, anti-rabbit Alexa Fluor 647 antibody (10 g/ml, Thermo GraphPad Prism software was used. For all other Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 6 of 17 experiments, statistical analyses were performed with post-implantation, as compared to control non-injected SPSS version 22 (IBMSPSS statistics, Ehningen, Germany). (p = 0.0216) and MSC-injected mice (p = 0.0214) (Fig. 2c). The Normality test and homogeneity of variances were Overall, these data demonstrate that IL13-MSC can effect- assessed for all data. The nonparametric analysis ap- ively produce transgenic IL13 mRNA in situ following proach, Kruskal-Wallis H, was performed for analysis of grafting in the CNS. behavioral scores (mNDS). Immunohistochemistry (IHC) and qRT-PCR data from in vivo experiments were ana- Characterization of the ischemic lesion lyzed for significant changes between the three groups Occlusion of the right middle cerebral artery with a sili- using one-way analysis of variance (ANOVA) with cone rubber-coated filament resulted in cerebral ische- Bonferroni-corrected post hoc comparisons. qRT-PCR mia. Lesion size and location were assessed by MRI data and ELISA data for comparison of in vitro cul- using quantitative T2 maps. Ischemic territory shows up tured MSCs and IL13-MSCs unpaired t tests were per- as area of increased T2 values on T2 maps, representing formed. Differences were considered statistically significant the area with augmented tissue water content. In the at a p value < 0.05. Data shown represent mean value per hemisphere ipsilateral to the lesion, vasogenic edema in- group ± standard deviation. creased T2 values and a clear lesion was detectable in the right ischemic hemisphere on T2 maps. In the intact Results contralateral hemisphere, T2 maps showed no signs of In vivo transgenic IL13 mRNA production by grafted lesion. Two days after MCAO, two characteristic lesion IL13-MSC types were detected: (1) in 5 mice, lesions were restricted In our previous studies related to in vivo grafting of to the striatum and (2) in 12 mice, lesions involved both IL13-MSCs, we assumed transgenic IL13 mRNA (and striatum and cortex. Subsequently, mice were divided protein) to be effectively produced as determined by the over three experimental groups: (1) control mice with appearance of alternatively activated microglia and mac- untreated stroke, (2) mice with stroke receiving a control rophages [23, 33, 34]. In this first part of our study, we MSC graft, and (3) mice with stroke receiving an aimed to determine whether IL13-MSCs effectively pro- IL13-MSC graft. Representative lesions in each experi- duce transgenic IL13 mRNA after in vivo transplant- mental group are displayed in Fig. 3. ation. At first, results obtained from cultured MSCs and IL13-MSCs demonstrate that both mRNA and protein ex- Neurological deficit pression levels of IL13 are only detected in IL13-MSCs We determined whether the procedure of cell grafting and not in MSCs (respectively; p <0.0001 and p = 0.0003) (MSC or IL13-MSC) into the ischemic brain influences (Fig. 2a, b). Next, following transplantation of MSCs and the motor and sensory performance after stroke in mice. IL13-MSCs in healthy mouse brain, we were able to detect In all three experimental groups, the neurological deficit significant levels of transgenic IL13 mRNA expression scores increased at day 2 after MCAO (up to score 7 in in five out of eight IL13-MSC-injected mice at day 7 mNDS) compared to baseline (score 0 in mNDS), Fig. 2 In vitro and in vivo expression of transgenic IL13 mRNA by IL13-MSCs. a Graph showing the analysis of transgenic IL13 mRNA expression by cultured MSCs and IL13-MSCs in vitro. b Graph showing the analysis of IL13 protein expression by cultured MSCs and IL13-MSCs after 24 h in vitro. c Graph showing the analysis of transgenic IL13 mRNA expression in the non-injected control group, the MSC-injected group, and the IL13-MSC-injected group at 1 week post-injection. Statistical significances are indicated by *** for p ≤ 0.001 and * for p ≤ 0.05 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 7 of 17 Fig. 3 Representative T2 maps for lesion location and size of two representative animals from each experimental group 2 days after ischemic stroke. T2 maps are displayed as coronal brain section. Lesions show up as areas of elevated T2 values on T2 maps, representing the area with increased tissue water content following stroke induction. Two characteristic lesion types are visible in the right hemisphere 48 h after MCAO: a lesions involving both striatum and cortex (n = 12) and b lesions restricted to the striatum (n =5) indicating the impaired motor and sensory functions macrophages were diffusely distributed across the whole due to the stroke. No differences were observed between territory. The only population within the intact contra- eGFP/+ the three experimental groups within the two-week obser- lateral hemisphere contained resident CX CR1 vation period. Nevertheless, our findings clearly indicate microglia (Fig. 4c). The microglia density on the ische- that the additional surgical procedure, intracranial MSC mic hemisphere was significantly higher than on the transplantation at 2 days after MCAO, did not worsen contralateral hemisphere, in all groups and analysis the neurological deficits after stroke (Additional file 1: (Fig. 4d–f). No obvious difference in this distribution pat- Figure S1). tern was observed between the three experimental groups. Next, we assessed whether any differences in cell density Distinct spatial distribution of microglia and infiltrated existed between resident and infiltrating cells 2 weeks macrophages following ischemic stroke after stroke. When evaluating the whole lesion site (i.e., In order to histologically investigate brain trafficking of the cortical and/or striatal region), the vast majority of re- eGFP/+ recruited microglia and macrophages after cerebral is- cruited cells within the lesion are CX CR1 microglia eGFP/+ RFP/+ chemia, we took advantage of CX CR1 CCR2 in all three experimental groups. Statistical analysis con- eGFP/+ double transgenic mice. eGFP and RFP expression in the firmed that CX CR1 cell density, being of microglial RFP/+ brain sections of these mice enables us to simultaneously origin, is significantly higher than CCR2 cell density, + eGFP/+ track resident eGFP microglia (CX CR1 ) and infil- being of macrophage origin, in all three experimental + RFP/+ trating RFP monocytes/macrophages (CCR2 ). Using groups at day 14 after stroke (stroke-only group p =0.013, these transgenic mice, we observed differences in displace- MSC-treated group p = 0.038, IL13-MSC-treated group ment of migration between microglia and peripheral p = 0.017) (Fig. 4d). When analyzing the cell density of monocyte-derived macrophages. Histological analysis re- microglia and macrophages, separately for striatal and vealed a noticeable infiltration of both microglia and mac- for cortico-striatal lesions, the statistical significance rophages throughout the ischemic territory at day 14 after between both cell populations was preserved in all stroke induction, (Fig. 4a–c). From the representative im- cases, except for the two MSC groups in the cortical munofluorescence images in Fig. 4b, we noted that eGFP analysis (Fig. 4e), where only a trend was reached (MSC and RFP expression shows a spatially different distribution group p = 0.06; IL13-MSC group p = 0.063). Despite the in the cortex and the striatum. In the cortical part of the difference in spatial distribution of microglia and RFP/+ ischemic lesion, CCR2 macrophages were located in macrophages between the striatal and cortical lesion eGFP/+ the infarct core and CX CR1 microglia were mainly area, further quantitative histological analysis revealed accumulated at the infarct border. In contrast, in the stri- closely comparable numbers of infiltrating macro- eGFP/+ RFP/+ atal lesion area, CX CR1 microglia and CCR2 phages and of resident microglia between cortical 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 8 of 17 Fig. 4 Localization and cell density of microglia and macrophages 2 weeks after ischemic stroke. a Representative coronal T2-weighted MR images are displayed above the corresponding immunohistochemistry images of all experimental groups. The hyperintense regions in the coronal MR images show the ischemic lesion 2 days post stroke. b–c Fluorescent microscopic images of brain coronal sections of all experimental eGFP RFP + groups represent the distribution of CX CR1 microglia and CCR2 macrophages at day 14 after stroke; b localization of eGFP microglia and RFP macrophages in ipsilateral cortical and striatal ischemic lesions, and c in contralateral cortical and striatal intact regions. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages. d–f Representative stacked column graphs show the cell density quantification of all experimental groups at day 14 after stroke. d Evaluation of microglia/macrophage cell density in the whole hemisphere. No significant difference + + eGFP/+ was observed in the total number of eGFP microglia and RFP macrophages between the three experimental groups. CX CR1 cell density RFP/+ + is significantly higher than CCR2 cell density in all three experimental groups at day 14 after stroke. In intact contralateral hemisphere, eGFP microglia was the main population. e Further evaluation of microglia and macrophage cell density in cortical lesion areas, and f in striatal lesion areas. The cell density of recruited microglia and macrophages was higher in large cortical lesion areas compared to striatal lesion areas. Data represent mean ± SD. The data were compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’s eGFP/+ RFP/+ post hoc test. To compare the CX CR1 cell density to CCR2 cell density in each group, an independent one-tailed Student’s t test was used (Fig. 4e)and striatal (Fig. 4f) lesions over the three (Fig. 5, left column): (i) cells with long thin arborized experimental groups. processes and a small cell body, defined as ramified cells (top row), (ii) cells with swollen processes and elongated eGFP/+ Recruited CX CR1 microglia exhibit various amorphous larger cell body, classified as intermediate morphological phenotypes within the ischemic cells (middle row), and (iii) cells with round shape and hemisphere no plasmalemmal processes, specified as amoeboid cells eGFP/+ eGFP/+ CX CR1 microglia are visible in both the intact (bottom row). Ramified CX CR1 cells were mostly 3 3 and the ischemic hemisphere, but are found in different observed in the healthy periphery of the lesion and in numbers, distribution patterns (see above), and morpho- the contralateral hemisphere, but were not seen in the eGFP/+ logical appearance. In this study, we could observe at core of the lesion. Intermediate type CX CR1 cells least three distinct morphological appearances of were closely associated with peri-infarct regions. eGFP/+ eGFP/+ CX CR1 microglia at day 14 after stroke induction Amoeboid CX CR1 cells were mainly localized in 3 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 9 of 17 Fig. 5 Different morphology of microglia and macrophages at day 14 after ischemic stroke. Representative photomicrographs with subsequent magnified images represent different morphology of microglia/macrophages in a intact contralateral hemisphere, b border of the ischemic lesion of ipsilateral hemisphere, and c the core of the ischemic lesion of ipsilateral hemisphere. eGFP microglia display at least three distinct phenotypes at day 14 after stroke including ramified cells (top row), intermediate cells (middle row), and amoeboid cells (bottom row). RFP macrophages show oval or round shape without apparent processes. F4/80 cells, alternatively activated cells, display intermediate and amoeboid shape. Scale bar 50 μm RFP/+ the ischemic core. In contrast, recruited CCR2 mac- macrophage populations (central and bottom row) within rophages displayed a uniform oval, kidney form, or the lesion area, at least in part, express this F4/80 activa- round shape without apparent processes (Fig. 5, central tion marker. Notably, morphological inspection of F4/ column). Some CCR2 cells showed cellular processes, 80-expressing cells shows that these cells display an inter- suggesting that these cells may acquire an intermediate mediate or amoeboid shape. Next, we assessed whether phenotype in the damaged tissue. Examples in Fig. 5 microglia and/or macrophage activation, based on F4/ were taken from the stroke-only group, but no obvious 80 expression, was altered upon transplantation of differences between the appearance of microglia and IL13-producing MSC, as compared to the situation macrophage morphology was observed between the after transplantation of MSC alone or without any three experimental groups. transplantation after stroke induction (Fig. 6). As shown by the representative immunofluorescent images Transplantation of IL13-producing MSCs increases the in Fig. 6a, and further supported by the quantitative proportion of highly activated infiltrating macrophages analysis provided in Fig. 6b (relative proportion of F4/ Not only the recruitment of microglia and macrophages 80 expression, stacked bars) and Fig. 5c (absolute pro- is a key feature of neuroinflammation, but also the sub- portion of F4/80 expression, pie charts), F4/80 expres- sequent phenotypic alterations associated with their acti- sion was observed on both microglia and macrophages vation status. Here, we first investigated the appearance in all three experimental groups. Interestingly, we ob- of F4/80 expression, a general marker of activation, on served a significantly increased expression of F4/80 by RFP/+ microglia and macrophages following stroke (Fig. 5, right CCR2 macrophages upon IL13-MSC transplantation RFP/+ column). F4/80 expression was not detected contralateral as compared to F4/80 expression by CCR2 macro- to the lesion site (top row), while both microglia and phages following stroke or stroke + control MSC grafting Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 10 of 17 Fig. 6 Transplantation of IL13-producing MSCs results in alternative activation of microglia and macrophages at day 14 after ischemic stroke. a eGFP/+ RFP/+ Representative fluorescent microscopic images of ischemic cortical region display the expression of F4/80 on CX CR1 and CCR2 cells, acquired from the three experimental groups at day 14 after stroke. Illustrative ×20 close-up magnification represents the changes in F4/80 expression by microglia and macrophages. F4/80 biomarker is expressed on both cell populations in the hemisphere ipsilateral to the lesion. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages; white, F4/80. b Detailed analysis of the expression of biomarker F4/80 RFP/+ revealed a significant increase of F4/80 expression of CCR2 macrophages upon IL13-MSCs transplantation in comparison to other two control + + groups. c The corresponding exploded pie charts show the distribution of F4/80 expression by GFP and RFP cells in all three experimental groups. + + In the pie chart, GFP microglia and RFP macrophages are encircled in green and red, respectively. n =5–6 mice in each group. Data represent mean ± SD. The data were compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’s post hoc test (p = 0.008 and p = 0.007). The increased number of F4/80 immunofluorescent images in Fig. 7a, and further sup- microglia detected following IL13-MSC grafting was not ported by the quantitative analysis provided in Fig. 7b significantly different from other groups. (relative proportion of Arg1 expression, stacked bars) and Fig. 6c (absolute proportion of Arg1 expression, pie Transplantation of IL13-producing MSC following stroke charts), significant increase in Arg1 expression was promotes the induction of alternatively activated observed within and surrounding the lesion area on microglia and macrophages both microglia and macrophages following grafting of To further analyze the phenotype of recruited microglia IL13-MSC (for microglia, stroke + IL13-MSC vs. stroke + and macrophages within the ischemic stroke lesion, as MSC and stroke only, p = 0.002 and p = 0.005 respectively; well as the effect of IL13-MSC thereon, we first performed for macrophages, stroke + IL13-MSC vs. stroke + MSC and additional immunofluorescence staining for Arg-1. Not- stroke only, both p < 0.001). Furthermore, the induc- ably, we wanted to investigate whether IL13-producing tion of this alternative activation program, as charac- MSC were able to convert stroke-associated neuroinflam- terized by Arg1 expression, was significantly higher in matory immune responses into an alternatively activated macrophages than in microglia in the IL13-MSC inflammatory response. As shown by the representative group (p = 0.002). Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 11 of 17 Fig. 7 Transplantation of IL13-producing MSCs increases the number of recruited neuroprotective microglia and macrophages at day 14 after eGFP/+ ischemic stroke. a Representative fluorescent microscopic images of ischemic cortical region show the expression of Arg-1 on CX CR1 and RFP/+ CCR2 cells in all three experimental groups at day 14 after stroke. Illustrative 20× close-up magnification display the changes in Arg-1 expression by microglia and macrophages. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages; white, Arg-1. b Detailed eGFP/+ RFP/+ phenotypic quantitative analysis of CX CR1 and CCR2 cells expressing Arg-1 showed a significant increase in the number of + − + − + + GFP RFP Arg-1 and GFP RFP Arg-1 cells in IL13-MSC-treated group, in comparison to both control groups. c The corresponding exploded pie + + + charts show the distribution of Arg-1 expression by GFP and RFP cells in all three experimental groups. In the pie chart, GFP microglia and + + RFP macrophages are encircled in green and red, respectively. In the IL13-MSC group, the Arg-1 fraction of macrophages is significantly higher than that of microglia. n =5–6 mice in each group. Data represent Mean ± SD. The data were compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’s post hoc test Transplantation of IL13-producing MSC following stroke microglia (p = 0.008, p = 0.007 and p = 0.002, respectively, may reduce pro-inflammatory MHC-II expression on in each group). Following IL13-MSC grafting, no signifi- infiltrated macrophages cant difference in MHC-II expression was observed on Likewise, we investigated the degree of MHC-II expression microglia. Even so, infiltrating macrophages seem to dis- within and surrounding the lesion area over the different play less MHC-II expression following IL13-MSC grafting, experimental conditions. As shown by the representative albeit this decrease was not statistically significant. immunofluorescent images in Fig. 8a, MHC-II expression is observed mainly inside the ischemic lesion in all three Characterization of MSC and IL13-MSC grafts experimental groups at day 14 after stroke induction. Fur- Similar to our preceding studies [23, 33–35], MSC and ther quantitative analyses provided in Fig. 8b (relative pro- IL13-MSC grafts were able to survive in the pro-inflam- portion of MHC-II expression, stacked bars) and 7C matory stroke environment and displayed a similar re- RFP/+ (absolute proportion of MHC-II expression, pie charts), in- modeling pattern, MSC graft-infiltrating CCR2 dicated that MHC-II expression is significantly higher in monocytes/macrophages at the core of the MSC RFP/+ eGFP/+ eGFP/+ CCR2 macrophages, compared to CX CR1 grafts and brain-resident CX3CR1 microglia and 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 12 of 17 Fig. 8 Transplantation of IL13-producing MSCs switches the polarization of microglia and macrophages towards less pro-inflammatory condition at day 14 after ischemic stroke. a Representative fluorescent microscopic images of ischemic cortical region show the expression of MHC-II on eGFP/+ RFP/+ CX CR1 and CCR2 cells in all three experimental groups at day 14 after stroke. Illustrative ×20 close-up magnification displays the changes in MHC-II expression by microglia and macrophages. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages; white, MHC-II. b Detailed eGFP/+ RFP/+ phenotypic quantitative analysis of CX CR1 and CCR2 cells expressing MHC-II showed a significant difference in the number of macrophages − + + + − + expressing MHCII, GFP RFP MHC-II , in comparison to the number of microglia expressing MHC-II, GFP RFP MHC-II , in all three experimental groups. In RFP/+ IL13-MSC-treated group, a trend towards less MHC-II expression by CCR2 macrophages was detected, in comparison to both control groups. c The + + corresponding exploded pie charts show the distribution of MHC-II expression by GFP and RFP cells in all three experimental groups. In the pie chart, + + + GFP microglia and RFP macrophages are encircled in green and red, respectively. In all three groups, the MHC-II fraction of macrophages is significantly higher than that of microglia. n =5–6miceineachgroup.Datarepresent mean±SD.Thedatawere compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’sposthoc test astrocytes surrounding the MSC grafts . Expression of the they are randomly spread throughout the striatum. In two M2 markers Arginase1 and Ym1 by microglia and addition to this region-dependent distribution, we here monocytes/macrophages, as a direct result of stimulation demonstrate a strong effect on modulation of the by IL13, was only detected in IL13-MSC grafts, but not in polarization of resident microglia and infiltrated macro- MSC grafts (Additional file 2:Figure S2). phages through the IL13 secretion by transplanted MSCs after ischemic stroke. The locally secreted IL13 modu- Discussion lates the stroke-induced immune reaction by promoting The present study demonstrates that brain-resident an anti-inflammatory, M2-like phenotype of microglia microglia and infiltrated macrophages accumulate in the and macrophages, as extensively visualized by arginase 1 post-ischemic brain with distinct spatial patterns, indi- expression, with the nature of the M2-like phenotype be- cating that they act as functionally different populations ing further confirmed by the demonstration of Ym1 ex- in CNS injury. While these two cell populations dis- pression. This protective phenotype is much stronger in tinctly accumulate in the ischemic lesion in the cortex, macrophages but also exists in microglia. The potent Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 13 of 17 immunomodulatory effect of IL13 encourages to further the peri-infarct area encircling the ischemic lesion core evaluate the application of this cytokine in clinically rele- filled with CCR2 macrophages [39]. Indeed, in a rather vant neurodegenerative disease models. complex study design, these latter authors demon- + − strated that CCR2 CX3CR1 monocytes may, in a Spatial distribution of microglia and infiltrated time-dependent fashion after infiltration, turn into − + macrophages following ischemic stroke CCR2 CX3CR1 macrophages. Thus, we must caution The phenotypic distinction between CNS-infiltrating that our assignment of green fluorescent, CX3CR1-positive macrophages and brain-resident microglia is a major im- microglia may to some extent also contain macrophages. munohistochemical concern. It is not only important in This, however, does not affect our assignment of CCR2 recognizing the origin of these cell populations, but it is red cells as a pure macrophage population. Another study also of fundamental importance in assessing the patho- showed that infiltrated blood-borne monocytes were exclu- genic and therapeutic significance of immune cells sively located at the ischemically injured striatum at days 3, within the damaged brain. The lack of a single specific 7, and 14 after MCAO [40]. Their results represent a ran- membranous and/or biochemical marker allowing de- dom distribution of brain-resident microglia and infiltrated finitive and discriminating identification of these cells is monocyte within the striatum after ischemic stroke, with still a puzzling issue in neurobiology. Fortunately, alter- microglia being the vast majorityof cells invading theische- native methods have been developed to overcome this mic lesion. Moreover, in a mouse model of traumatic spinal +/GFP +/RFP hurdle, including the generation of bone marrow cord injury, CX CR1 and CCR2 cells were ran- chimeric mice, which has proven to provide a powerful domly distributed around and inside the lesion, with +/RFP tool to distinguish microglia and infiltrated macrophages CCR2 cells constituting the greater number of accu- followingischemicstroke[36, 37]. In this study, how- mulated cells in the lesion area [41]. +/GFP +/RFP ever, we made use of the CX CR1 CCR2 These findings suggest an unknown mechanism by transgenic mouse model, which is based on the fact which temporally and spatially distinct ischemic regions that chemokine receptor CX CR1 is predominantly can differentially signal to microglia and macrophages. expressed by CNS resident microglia and that CCR2 is Region-specific differences between cortex and striatum upregulated in activated infiltrated macrophages and in regard to vascular volume [30], neurogenic potential therefore allows distinction of eGFP-expressing micro- [42] and neuroinflammatory responses [43] may cause glia from RFP-expressing infiltrated macrophages [5]. the region-dependent distribution of microglia and mac- Through the use of this transgenic mouse model and rophages after ischemic stroke. +/GFP immunofluorescence microscopy, we were able to demon- In our experimental setup, considering CX CR1 strate a region-dependent distribution of microglia and cells as brain-resident microglia, it might be argued that macrophages at day 14 after stroke. In the cortical ische- perivascular macrophages, supraependymal macrophages, mic lesion, CX CR1 microglia distinctly accumulated at epiplexus cells of the choroid plexus, and meningeal mac- the border of the lesion, while CCR2 macrophages were rophages express CX CR1 too [44]. In addition, we cannot localized at the core of the lesion. At day 14, after ische- neglect the presence of patrolling macrophages Ly6c +/GFP high low mic stroke, CX CR1 microglia were the major con- CX CR1 CCR2 in the ischemic brain with disrupted 3 3 tributors of recruited cells to the ischemic lesion. These blood-brain barrier [41, 45]. Interestingly, in an attempt to observations are in line with our earlier findings following observe chemokine receptor expression changes in micro- MSC transplantation in mouse brain, which demonstrated glia and monocytes/macrophages in development and a similar distinct distribution of microglia and macro- during inflammatory condition, Mizutani and colleagues phages, where infiltrated macrophages invaded the hyp- have indicated that CCR2 is absent in adult naïve and in- oxic/ischemic MSC transplant core, while active microglia flamed CNS resident microglia [5]. Importantly, they have remained in the surrounding border [23, 25, 38]. In con- demonstrated that CX CR1 expression by microglia is sig- low high trast to their orderly distribution in the cortex, both nificantly higher compared to CX CR1 CCR2 or + + high low CX CR1 microglia and CCR2 macrophages were found CX CR1 CCR2 monocytes/macrophages. Therefore, 3 3 +/GFP to be randomly distributed in striatal lesions. In agreement we hypothesize that CX3CR1 cells are mostly of with our own observations, others have also reported brain-resident microglial origin and double-positive cells +/GFP +/RFP that resident microglia and blood-derived macrophages for CX CR1 and CCR2 are blood-derived localize in the ischemic brain with different temporal macrophage populations [33, 41, 46]. It should be empha- and spatial patterns. Garcia-Bonilla et al. describe that sized that the presence of double-positive cells at day 14 at the first week after stroke, mostly diffused CCR2 after ischemic stroke accounts for a very small proportion macrophages were observed throughout the ischemic of infiltrated macrophages in our experiment. Further lesion, while during the second and third week after assessment of earlier or later time point of macro- stroke, they saw that CX CR1 cells were localized in phage recruitment following CNS damage should also 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 14 of 17 RFP/+ be evaluated. Nevertheless, the discovery of more CCR2 double transgenic mice revealed that about microglia-specific markers would contribute to more two thirds of Arg-1 cells are infiltrated macrophages. Evi- efficient discrimination between CNS resident micro- dently, continued secretion of the cytokine IL13 by MSCs glia and infiltrated macrophages in future studies. was able to boost the immunomodulatory properties of MSCs by increasing the number of Arg-1 cells, thus Dynamic modulation of microglia and infiltrated acting as a potent neuroprotective mediator [47]. Several macrophages following ischemic stroke groups have reported that Arg-1 expression by different In this study, we are the first to determine the immuno- cell types, including astrocytes, neurons, and immune modulatory potential of MSC and IL13-expressing MSC cells, notably decreases during the first few days after cere- engraftment in adjusting the polarization of brain-resident bral ischemia, after an early peak at day 3 [6, 48]. By con- microglia and infiltrated macrophages in an ischemic tinuous secretion of IL-13 via MSCs, we could enhance stroke model. We focused on addressing important appli- and prolong the expression of neuroprotective and cations of IL13-expressing MSCs as a new potential thera- anti-inflammatory marker Arg-1 until week 2 after stroke peutic approach for ischemic stroke. The main goal of our induction. In an earlier study by our laboratory, in which study was to enhance the therapeutic effects of MSC we attempted to modulate the polarization of microglia transplantation following ischemic stroke by boosting the and macrophages by microRNA-124 after ischemic stroke, immunomodulatory properties of MSCs. For this reason, we achieved an upregulation of Arg-1 and CD-206 expres- we used MSCs as a biodegradable delivery system for the sion by microglia and macrophages until day 6 after stroke. immune modulating cytokine IL13. With this method, In this study, the increased polarization of microglia and we achieved long-term effect on immune function until macrophages towards the more anti-inflammatory pheno- the second week after stroke induction, during the type was positively correlated to increased neuronal sur- pro-inflammatory peak of stroke [6]. Earlier investiga- vival and functional recovery at day 6 after stroke [32, 49]. tionsbyus, on theuseof MSCs as adeliverysystemfor Another important observation to be taken into ac- the anti-inflammatory cytokine IL13, demonstrated that count here is the expression and accumulation of major continued secretion of IL13 by MSCs is able to limit histocompatibility complex class II (MHC-II) following microgliosis, oligodendrocyte loss, and demyelination MSC and IL13-expressing MSC engraftment after stroke. in cuprizone-treated mice, a model for neuroinflamma- MHC-II expression was predominantly observed on infil- tion and demyelination [23], and promotes histopatho- trated macrophages, localized at the core of the ischemic logical and functional recovery following spinal cord lesion. MHC-II is considered to be a pro-inflammatory injury in mice [33]. In these studies, we also showed mediator, which is overexpressed under a wide range of that transplantation of IL13-expressing MSCs induces pathological conditions [50]. However, some studies have alternative activation in both MSC graft-invading mac- highlighted that MHC-II expression can also provide pro- rophages and in MSC graft-surrounding microglia, tective and neurotrophic functions and is involved in characterized by the expression of Arg-1 and Ym1 as regeneration and axon integrity in neurodegenerative dis- anti-inflammatory-associated markers. orders [51, 52]. As reported previously, accumulation of + + Here, we present for the first time that transplantation CD11b MHC-II cells in the ipsilateral ischemic hemi- of IL13-expressing MSCs at 48 h after ischemia shifts sphere after transient MCAO provides trophic support major players of the immune reaction, namely microglia and is effective in the remyelination process after stroke and macrophages, towards an anti-inflammatory, neuro- [53]. With regard to the expression of Arg-1 and MHC-II, protective phenotype at 14 days after ischemia. Although mostly by macrophages at the core of ischemic lesion in not investigated in detail in this study, our previous data our study, we could argue a substantial role of infil- with regard to grafting IL-13 expressing MSCs in muscle trated macrophages for the regeneration and repair of brain tissue have shown that the anti-inflammatory process of injured brain. We propose that infiltrated + + M2 phenotype can be induced already at 5 and 7 days pro-inflammatory CCR2 MHC-II macrophages may post grafting, respectively [35]. The transplantation of exert some pro-neurogenic functions such as clear- IL13-expressing MSCs at day 2 after ischemia caused a ance of cell debris after brain injury, which could be noticeable increase in the expression of the general activa- efficient at neurogenesis and regeneration process tion marker F4/80, especially by infiltrated macrophages. after ischemic stroke. Based on our data, transplant- Most importantly, MSC graft-associated microglia and in- ation of IL13-expressing MSCs does not broadly sup- filtrated macrophages were strongly driven towards an press M1 phenotype of microglia and macrophages, Arg-1 alternative activation status at day 14 after stroke. but it rather adjusts the balance between pro- and Further distinction between recruited brain-resident anti-inflammatory immune cell processes. As M1-activated microglia and infiltrated systemic macrophages following cells typically resolve injured and/or infected tissue, which eGFP/+ stroke and MSC transplantation in CX CR1 can lead to detrimental exacerbated injury of healthy tissue, 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 15 of 17 they can also assist in the repair/conservation of syn- Additional files aptic connectivity and axonal regeneration [54, 55]. Additional file 1: Figure S1. Motor and sensory performance after When considering M2 activated immune cells, mainly ischemic stroke. Behavioral performance was assessed by the modified anti-inflammatory and regeneration-inducing functions neurological deficit scores (mNDS), before and every 2 days after MCAO. are described, but it is worth noting that long-term main- In all three groups, the mNDS increased significantly at day 2 after MCAO compared to baseline. Graphs of all three experimental groups show no tenance of anti-inflammatory phenotype may adversely significant difference among the groups. The procedure of cell grafting affect the immune response, which leads to serious nega- into the ischemic brain and the additional surgical procedure did not tive effects, such as tumorigenesis [56]. Induction of an worsen the motor and sensory performance after stroke. n =5–6 mice in each group. Data are mean ± SD. The data were compared between M1/M2 phenotype switch to improve natural repair the three experimental groups using the nonparametric Kruskal-Wallis processes in neurodegenerative disorders should be con- H test. (TIF 837 kb) sidered with caution. In stroke, our earlier studies on Additional file 2: Figure S2. Representative example of MSC and IL13- modulation of the pro-/anti-inflammatory balance by MSC graft site remodeling within the MCAO brain lesion site. Control MSC (upper panel) and IL13-MSC (lower panel) grafts are able to survive in the microRNA-124 demonstrated enhanced neuronal survival pro-inflammatory stroke environment and display a similar remodeling and behavioral improvement [32, 49]. RFP/+ pattern. MSC graft-infiltrating CCR2 monocytes/macrophages (in red) at eGFP/+ In our present investigation, we have not analyzed the in- the core of the MSC grafts and brain-resident CX3CR1 microglia (in green) and astrocytes (in blue, first row) surrounding the MSC grafts. fluence of continuous secretion of the anti-inflammatory Arginase1 expression (in blue, second and third row) and Ym1 expression cytokine IL-13 and/or M2 polarized microglia and mac- (in magenta, third row) by microglia and monocytes/macrophages, as a rophages on the survival of autologous MSC grafts after direct result of stimulation by IL13, was only detected in IL13-MSC grafts, but not in control MSC grafts. Scale bar 100 μm. (TIF 13199 kb) stroke. But in an earlier study, we have demonstrated that transplantation of IL13-expressing MSCs leads to a notable decrease of direct and indirect immune recog- Abbreviations AKT: Alpha serine/threonine-protein kinase; Arg-1: Arginase-1; CCA: Common nition and rejection of cell grafts [35]. Further support carotid artery; CCR2: CC-chemokine receptor 2; CD11b: Clusters of differentiation of functional interaction between alternative activation 11b; CNS: Central nervous system; CX CR1: CX C chemokine receptor 1; 3 3 status of surrounding inflammatory milieu at the graft EAE: Experimental autoimmune encephalomyelitis; ERK: Extracellular signal- regulated kinases; FOXO3: Forkhead box O3; GAPDH: Glyceraldehyde-3-phosphate site and survival of MSCs comes from a study by Yu et dehydrogenase; Iba-1: Ionized calcium-binding adapter molecule 1; IFN- al. They have recently demonstrated that M2 ϒ: Interferon gamma; IHC: Immunohistochemistry; IL: Interleukin; IL13-MSC: IL13- macrophage-secreted OA/GPNMB, osteoactivin/glyco- producing MSCs; LPS: Lipopolysaccharide; MCA: Middle cerebral artery; MCAO: Middle cerebral artery occlusion; MHC-II: Major histocompatibility complex protein in non-metastatic melanoma protein B, class II; MIF: Macrophage migration inhibitory factor; miR-124: MicroRNA-124; positively contributes to viability, proliferation, and mi- mNDS: Modified neurological deficit score; MRI: Magnetic resonance gration of MSCs, through ERK and AKT signaling path- imaging; MSCs: Mesenchymal stem cells; PBS: Phosphate-buffered saline; PFA: Paraformaldehyde; ppiA: Peptidylprolyl isomerase A; RT: Room temperature ways [57]. Similar MSC survival improvement was achieved via an ischemic microenvironment in vitro Acknowledgements study, in which macrophage migration inhibitory factor We thank Ulla Uhlenküken for her professional arrangement of the graphics art work. (MIF), a cytokine expressed by activated monocytes/ macrophages, protects MSCs from apoptosis via a Funding CD74-dependent Akt-FOXO3a-related pathway [58]. This work was financially supported by grants from the EU-FP7 programs Enhancing MSC survival in the face of oxygen and nu- TargetBraIn (HEALTH-F2-2012-279017) and BrainPath (PIAPP-GA-2013-612360), as well as by funds from the University of Antwerp (Belgium), the Fund for trient deprivation that naturally occur in the ischemic Scientific Research (FWO)—Flanders (G091518N), and the Belgian Charcot stroke and cell graft procedure clearly needs further Foundation. Chloé Hoornaert and Alessandra Quarta hold an FWO-funded investigation. Ph.D. studentship and Debbie Le Blon holds an FWO-funded post-doctoral fellowship. All funding bodies had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript. Conclusions Availability of data and materials Our findings will serve as a base for future studies to The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. apply MSCs in a more effective way to improve CNS re- pair. We have identified that MSCs expressing the Authors’ contributions anti-inflammatory cytokine IL13 serve as an excellent SHT designed the project, performed the surgery and the cell implantation, candidate for more effective modulation of inflammatory analyzed the immunohistochemical data, wrote the manuscript. DLB performed the qRT-PCR and immunohistochemical staining and data analysis, wrote the responses in neurodegenerative disorders. This new manuscript. CH performed the immunohistochemical staining and data analysis. insight could have noticeable therapeutic implications, JD performed the immunohistochemical staining and data analysis. AQ performed which demands further studies using human biological the immunohistochemical staining and data analysis. JP performed the MRI experiments. AvdL analyzed the MRI and helped in writing the manuscript. systems, such as human-derived mesenchymal stem PP designed the project, analyzed the data, and wrote the manuscript. MH cells, for translation of immune response modulation re- designed the project, analyzed the data, and wrote the manuscript. All authors search into clinical practice. read and approved the final manuscript. Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 16 of 17 Ethics approval and consent to participate 14. Ankrum JA, Ong JF, Karp JM. Mesenchymal stem cells: immune evasive, not N/A immune priviledged. Nat Biotechnol. 2014;32:252–60. 15. De Vocht N, Lin D, Praet J, Hoornaert C, Reekmans K, Le Blon D, et al. Competing interests Quantitative and phenotypic analysis of mesenchymal stromal cell graft survival The authors declare that they have no competing interests. and recognition by microglia and astrocytes in mouse brain. Immunobiology. 2013;218(5):696–705. https://doi.org/10.1016/j.imbio.2012.08.266. PubMed PMID: Publisher’sNote 16. Praet J, Reekmans K, Lin D, De Vocht N, Bergwerf I, Tambuyzer B, et al. Cell Springer Nature remains neutral with regard to jurisdictional claims in type-associated differences in migration, survival, and immunogenicity published maps and institutional affiliations. following grafting in CNS tissue. Cell Transplant. 2012;21(9):1867–81. https://doi.org/10.3727/096368912X636920. PubMed PMID: 22472278 Author details 17. Le Blanc K. Immunomodulatory effects of fetal and adult mesenchymal stem In-vivo-NMR Laboratory, Max Planck Institute for Metabolism Research, cells. Cytotherapy. 2003;5(6):485–9. https://doi.org/10.1080/14653240310003611. Gleuelerstrasse 50, D-50931 Köln, Germany. Laboratory of Experimental PubMed PMID: 14660044 Hematology, University of Antwerp, Antwerp, Belgium. Vaccine and 18. Krampera M, Glennie S, Dyson J, Scott D, Laylor R, Simpson E, et al. Bone Infectious Disease Institute (Vaxinfectio), University of Antwerp, Antwerp, marrow mesenchymal stem cells inhibit the response of naive and Belgium. Bio-Imaging Laboratory, University of Antwerp, Antwerp, Belgium. memory antigen-specific T cells to their cognate peptide. Blood. 2003; Department of Radiology, Leiden University Medical Center, Leiden, 101(9):3722–9. https://doi.org/10.1182/blood-2002-07-2104. PubMed PMID: Netherlands. 19. Quertainmont R, Cantinieaux D, Botman O, Sid S, Schoenen J, Franzen R. Received: 10 November 2017 Accepted: 21 May 2018 Mesenchymal stem cell graft improves recovery after spinal cord injury in adult rats through neurotrophic and pro-angiogenic actions. PLoS One. 2012;7(6):e39500. https://doi.org/10.1371/journal.pone.0039500. PubMed References PMID: 22745769; PubMed Central PMCID: PMC3380009 1. Donnan GA, Fisher M, Macleod M, Davis SM. Stroke Lancet. 2008;371(9624): 20. Nauta AJ, Fibbe WE. Immunomodulatory properties of mesenchymal 1612–23. https://doi.org/10.1016/S0140-6736(08)60694-7. PubMed PMID: stromal cells. Blood. 2007;110(10):3499–506. https://doi.org/10.1182/blood- 2007-02-069716. PubMed PMID: 17664353 2. Jin R, Yang G, Li G. Inflammatory mechanisms in ischemic stroke: role of 21. Ishikane S, Ohnishi S, Yamahara K, Sada M, Harada K, Mishima K, et al. inflammatory cells. J Leukoc Biol. 2010;87(5):779–89. https://doi.org/10.1189/ Allogeneic injection of fetal membrane-derived mesenchymal stem cells jlb.1109766. PubMed PMID: 20130219; PubMed Central PMCID: PMC2858674 induces therapeutic angiogenesis in a rat model of hind limb ischemia. 3. Patel AR, Ritzel R, McCullough LD, Liu F. Microglia and ischemic stroke: a Stem Cells. 2008;26(10):2625–33. https://doi.org/10.1634/stemcells.2008-0236. double-edged sword. Int J Physiol Pathophysiol Pharmacol. 2013;5(2):73–90. PubMed PMID: 18669910 PubMed PMID: 23750306; PubMed Central PMCID: PMC3669736 22. Tambuyzer BR, Bergwerf I, De Vocht N, Reekmans K, Daans J, Jorens PG, 4. Shechter R, Schwartz M. Harnessing monocyte-derived macrophages to et al. Allogeneic stromal cell implantation in brain tissue leads to robust control central nervous system pathologies: no longer ‘if’ but ‘how’. J Pathol. microglial activation. Immunol Cell Biol. 2009;87(4):267–73. https://doi.org/ 2013;229(2):332–46. https://doi.org/10.1002/path.4106. PubMed PMID: 23007711 10.1038/icb.2009.12. PubMed PMID: 19290016 5. Mizutani M, Pino PA, Saederup N, Charo IF, Ransohoff RM, Cardona AE. The 23. Le Blon D, Guglielmetti C, Hoornaert C, Quarta A, Daans J, Dooley D, et al. fractalkine receptor but not CCR2 is present on microglia from embryonic Intracerebral transplantation of interleukin 13-producing mesenchymal development throughout adulthood. J Immunol. 2012;188(1):29–36. stem cells limits microgliosis, oligodendrocyte loss and demyelination in the https://doi.org/10.4049/jimmunol.1100421. PubMed PMID: WOS: cuprizone mouse model. J Neuroinflammation. 2016;13(1):288. https://doi. org/10.1186/s12974-016-0756-7. PubMed PMID: 27829467; PubMed Central 6. Hu X, Li P, Guo Y, Wang H, Leak RK, Chen S, et al. Microglia/macrophage PMCID: PMC5103449 polarization dynamics reveal novel mechanism of injury expansion after 24. Bergwerf I, De Vocht N, Tambuyzer B, Verschueren J, Reekmans K, Daans J, focal cerebral ischemia. Stroke. 2012;43(11):3063–70. https://doi.org/10.1161/ et al. Reporter gene-expressing bone marrow-derived stromal cells are STROKEAHA.112.659656. PubMed PMID: 22933588 immune-tolerated following implantation in the central nervous system of 7. Hu X, Leak RK, Shi Y, Suenaga J, Gao Y, Zheng P, et al. Microglial and syngeneic immunocompetent mice. BMC Biotechnol. 2009;9:1. https://doi. macrophage polarization-new prospects for brain repair. Nat Rev Neurol. org/10.1186/1472-6750-9-1. PubMed PMID: 19128466; PubMed Central 2015;11(1):56–64. https://doi.org/10.1038/nrneurol.2014.207. PubMed PMID: PMCID: PMC2630974 25385337; PubMed Central PMCID: PMCPMC4395497 25. Le Blon D, Hoornaert C, Daans J, Santermans E, Hens N, Goossens H, et al. 8. Doherty TM, Kastelein R, Menon S, Andrade S, Coffman RL. Modulation of Distinct spatial distribution of microglia and macrophages following murine macrophage function by IL-13. J Immunol. 1993;151(12):7151–60. mesenchymal stem cell implantation in mouse brain. Immunol Cell Biol. 2014; PubMed PMID: 7903102 92(8):650–8. https://doi.org/10.1038/icb.2014.49. PubMed PMID: 24983456 9. Doyle AG, Herbein G, Montaner LJ, Minty AJ, Caput D, Ferrara P, et al. 26. Reekmans K, De Vocht N, Praet J, Le Blon D, Hoornaert C, Daans J, et al. Interleukin-13 alters the activation state of murine macrophages in vitro: Quantitative evaluation of stem cell grafting in the central nervous system comparison with interleukin-4 and interferon-gamma. Eur J Immunol. 1994; of mice by in vivo bioluminescence imaging and postmortem multicolor 24(6):1441–5. https://doi.org/10.1002/eji.1830240630. PubMed PMID: 7911424 histological analysis. Methods Mol Biol. 2013;1052:125–41. https://doi.org/10. 10. Offner H, Subramanian S, Wang C, Afentoulis M, Vandenbark AA, Huan J, 1007/7651_2013_17. PubMed PMID: 23733539 et al. Treatment of passive experimental autoimmune encephalomyelitis in 27. De Vocht N, Reekmans K, Bergwerf I, Praet J, Hoornaert C, Le Blon D, et al. SJL mice with a recombinant TCR ligand induces IL-13 and prevents axonal Multimodal imaging of stem cell implantation in the central nervous system injury. J Immunol. 2005;175(6):4103–11. PubMed PMID: 16148160 of mice. J Vis Exp. 2012;64:e3906. https://doi.org/10.3791/3906. PubMed 11. Ochoa-Reparaz J, Rynda A, Ascon MA, Yang X, Kochetkova I, Riccardi C, et al. PMID: 22733218; PubMed Central PMCID: PMC3471290 IL-13 production by regulatory T cells protects against experimental 28. Praet J, Santermans E, Reekmans K, de Vocht N, Le Blon D, Hoornaert C, autoimmune encephalomyelitis independently of autoantigen. J Immunol. et al. Histological characterization and quantification of cellular events 2008;181(2):954–68. PubMed PMID: 18606647; PubMed Central PMCID: following neural and fibroblast(-like) stem cell grafting in healthy and PMC2599928 demyelinated CNS tissue. Methods Mol Biol. 2014;1213:265–83. https://doi. 12. Lee K, Majumdar MK, Buyaner D, Hendricks JK, Pittenger MF, Mosca JD. org/10.1007/978-1-4939-1453-1_22. PubMed PMID: 25173390 Human mesenchymal stem cells maintain transgene expression during 29. Bahmani P, Schellenberger E, Klohs J, Steinbrink J, Cordell R, Zille M, et al. expansion and differentiation. Mol Ther. 2001;3(6):857–66. https://doi.org/10. Visualization of cell death in mice with focal cerebral ischemia using 1006/mthe.2001.0327. PubMed PMID: 11407899 fluorescent annexin A5, propidium iodide, and TUNEL staining. J Cereb Blood 13. Ozawa K, Sato K, Oh I, Ozaki K, Uchibori R, Obara Y, et al. Cell and gene Flow Metab. 2011;31(5):1311–20. https://doi.org/10.1038/jcbfm.2010.233. therapy using mesenchymal stem cells (MSCs). J Autoimmun. 2008;30(3): PubMed PMID: 21245871; PubMed Central PMCID: PMC3099638 121–7. https://doi.org/10.1016/j.jaut.2007.12.008. PubMed PMID: 18249090 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 17 of 17 30. Adamczak JM, Schneider G, Nelles M, Que I, Suidgeest E, van der Weerd L, 46. Yamasaki R, Lu H, Butovsky O, Ohno N, Rietsch AM, Cialic R, et al. Differential et al. In vivo bioluminescence imaging of vascular remodeling after stroke. roles of microglia and monocytes in the inflamed central nervous system. Front Cell Neurosci. 2014;8:274. https://doi.org/10.3389/fncel.2014.00274. J Exp Med. 2014;211(8):1533–49. https://doi.org/10.1084/jem.20132477. PubMed PMID: 25249937; PubMed Central PMCID: PMC4155794 PubMed PMID: 25002752; PubMed Central PMCID: PMC4113947 31. Chen J, Sanberg PR, Li Y, Wang L, Lu M, Willing AE, et al. Intravenous 47. Estevez AG, Sahawneh MA, Lange PS, Bae N, Egea M, Ratan RR. Arginase 1 administration of human umbilical cord blood reduces behavioral deficits regulation of nitric oxide production is key to survival of trophic factor- after stroke in rats. Stroke. 2001;32(11):2682–8. PubMed PMID: 11692034 deprived motor neurons. J Neurosci. 2006;26(33):8512–6. https://doi.org/10. 1523/JNEUROSCI.0728-06.2006. PubMed PMID: 16914676; PubMed Central 32. Hamzei Taj S, Kho W, Aswendt M, Collmann FM, Green C, Adamczak J, et al. PMCID: PMC2570095 Dynamic modulation of microglia/macrophage polarization by miR-124 48. Wang G, Zhang J, Hu X, Zhang L, Mao L, Jiang X, et al. Microglia/ after focal cerebral ischemia. J NeuroImmune Pharmacol. 2016;11(4):733–48. macrophage polarization dynamics in white matter after traumatic brain https://doi.org/10.1007/s11481-016-9700-y. PubMed PMID: 27539642; injury. J Cereb Blood Flow Metab. 2013;33(12):1864–74. https://doi.org/10. PubMed Central PMCID: PMC5097787 1038/jcbfm.2013.146. PubMed PMID: 23942366; PubMed Central PMCID: 33. Dooley D, Lemmens E, Vangansewinkel T, Le Blon D, Hoornaert C, Ponsaerts P, PMC3851898 et al. Cell-based delivery of Interleukin-13 directs alternative activation of 49. Hamzei Taj S, Kho W, Riou A, Wiedermann D, Hoehn M. MiRNA-124 induces macrophages resulting in improved functional outcome after spinal cord injury. neuroprotection and functional improvement after focal cerebral ischemia. Stem Cell Rep. 2016;7(6):1099–115. https://doi.org/10.1016/j.stemcr.2016.11.005. Biomaterials. 2016;91:151–65. https://doi.org/10.1016/j.biomaterials.2016.03. PubMed PMID: 27974221 025. PubMed PMID: 27031810 34. Ali I, Aertgeerts S, Le Blon D, Bertoglio D, Hoornaert C, Ponsaerts P, et al. 50. Frank MG, Barrientos RM, Biedenkapp JC, Rudy JW, Watkins LR, Maier SF. Intracerebral delivery of the M2 polarizing cytokine interleukin 13 using mRNA up-regulation of MHC II and pivotal pro-inflammatory genes in mesenchymal stem cell implants in a model of temporal lobe epilepsy in normal brain aging. Neurobiol Aging. 2006;27(5):717–22. https://doi.org/10. mice. Epilepsia. 2017;58:1063–72. 1016/j.neurobiolaging.2005.03.013. PubMed PMID: 15890435 35. Hoornaert CJ, Luyckx E, Reekmans K, Dhainaut M, Guglielmetti C, Le Blon D, 51. Arnett HA, Wang Y, Matsushima GK, Suzuki K, Ting JP. Functional genomic et al. In vivo interleukin-13-primed macrophages contribute to reduced analysis of remyelination reveals importance of inflammation in alloantigen-specific T cell activation and prolong immunological survival of oligodendrocyte regeneration. J Neurosci. 2003;23(30):9824–32. PubMed allogeneic mesenchymal stem cell implants. Stem Cells. 2016;34(7):1971–84. PMID: 14586011 https://doi.org/10.1002/stem.2360. PubMed PMID: 26992046 52. Ohtaki H, Yin L, Nakamachi T, Dohi K, Kudo Y, Makino R, et al. Expression of 36. Tanaka R, Komine-Kobayashi M, Mochizuki H, Yamada M, Furuya T, Migita M, tumor necrosis factor alpha in nerve fibers and oligodendrocytes after et al. Migration of enhanced green fluorescent protein expressing bone transient focal ischemia in mice. Neurosci Lett. 2004;368(2):162–6. https:// marrow-derived microglia/macrophage into the mouse brain following doi.org/10.1016/j.neulet.2004.07.016. PubMed PMID: 15351441 permanent focal ischemia. Neuroscience. 2003;117(3):531–9. PubMed PMID: 53. Kuric E, Ruscher K. Dynamics of major histocompatibility complex class II- positive cells in the postischemic brain—influence of levodopa treatment. 37. Schilling M, Besselmann M, Leonhard C, Mueller M, Ringelstein EB, Kiefer R. J Neuroinflammation. 2014;11:145. https://doi.org/10.1186/s12974-014-0145-z. Microglial activation precedes and predominates over macrophage PubMed PMID: 25178113; PubMed Central PMCID: PMC4149192 infiltration in transient focal cerebral ischemia: a study in green fluorescent 54. Stellwagen D, Malenka RC. Synaptic scaling mediated by glial TNF-alpha. protein transgenic bone marrow chimeric mice. Exp Neurol. 2003;183(1):25– Nature. 2006;440(7087):1054–9. https://doi.org/10.1038/nature04671. 33. PubMed PMID: 12957485 PubMed PMID: 16547515 38. Le Blon D, Hoornaert C, Detrez JR, Bevers S, Daans J, Goossens H, et al. 55. Hanania R, Sun HS, Xu K, Pustylnik S, Jeganathan S, Harrison RE. Classically Immune remodelling of stromal cell grafts in the central nervous system: activated macrophages use stable microtubules for matrix therapeutic inflammation or (harmless) side-effect? J Tissue Eng Regen Med. metalloproteinase-9 (MMP-9) secretion. J Biol Chem. 2012;287(11):8468–83. 2016; https://doi.org/10.1002/term.2188. https://doi.org/10.1074/jbc.M111.290676. PubMed PMID: 22270361; PubMed 39. Garcia-Bonilla L, Faraco G, Moore J, Murphy M, Racchumi G, Srinivasan J, Central PMCID: PMC3318683 et al. Spatio-temporal profile, phenotypic diversity, and fate of recruited 56. Komohara Y, Ohnishi K, Kuratsu J, Takeya M. Possible involvement of the M2 monocytes into the post-ischemic brain. J Neuroinflammation. 2016;13(1): anti-inflammatory macrophage phenotype in growth of human gliomas. J 285. https://doi.org/10.1186/s12974-016-0750-0. PubMed PMID: 27814740; Pathol. 2008;216(1):15–24. https://doi.org/10.1002/path.2370. PubMed PMID: PubMed Central PMCID: PMC5097435 40. Wattananit S, Tornero D, Graubardt N, Memanishvili T, Monni E, Tatarishvili J, 57. Yu B, Sondag GR, Malcuit C, Kim MH, Safadi FF. Macrophage-associated et al. Monocyte-derived macrophages contribute to spontaneous long-term osteoactivin/GPNMB mediates mesenchymal stem cell survival, proliferation, functional recovery after stroke in mice. J Neurosci. 2016;36(15):4182–95. and migration via a CD44-dependent mechanism. J Cell Biochem. 2016; https://doi.org/10.1523/JNEUROSCI.4317-15.2016. PubMed PMID: 27076418 117(7):1511–21. https://doi.org/10.1002/jcb.25394. PubMed PMID: 26442636 41. Evans TA, Barkauskas DS, Myers JT, Hare EG, You JQ, Ransohoff RM, et al. 58. Xia W, Xie C, Jiang M, Hou M. Improved survival of mesenchymal stem cells High-resolution intravital imaging reveals that blood-derived macrophages by macrophage migration inhibitory factor. Mol Cell Biochem. 2015;404(1–2): but not resident microglia facilitate secondary axonal dieback in traumatic 11–24. https://doi.org/10.1007/s11010-015-2361-y. PubMed PMID: 25701358; spinal cord injury. Exp Neurol. 2014;254:109–20. https://doi.org/10.1016/j. PubMed Central PMCID: PMC4544672 expneurol.2014.01.013. PubMed PMID: 24468477; PubMed Central PMCID: PMC3954731 42. Darsalia V, Kallur T, Kokaia Z. Survival, migration and neuronal differentiation of human fetal striatal and cortical neural stem cells grafted in stroke-damaged rat striatum. Eur J Neurosci. 2007;26(3):605–14. https://doi.org/10.1111/j.1460-9568.2007.05702.x. PubMed PMID: 17686040 43. Janata A, Magnet IA, Uray T, Stezoski JP, Janesko-Feldman K, Tisherman SA, et al. Regional TNFalpha mapping in the brain reveals the striatum as a neuroinflammatory target after ventricular fibrillation cardiac arrest in rats. Resuscitation. 2014;85(5):694–701. https://doi.org/10.1016/j.resuscitation.2014. 01.033. PubMed PMID: 24530249; PubMed Central PMCID: PMC4034695 44. Shechter R, Miller O, Yovel G, Rosenzweig N, London A, Ruckh J, et al. Recruitment of beneficial M2 macrophages to injured spinal cord is orchestrated by remote brain choroid plexus. Immunity. 2013;38(3):555–69. https://doi.org/10. 1016/j.immuni.2013.02.012. PubMed PMID: WOS:000330941500019 45. Geissmann F, Jung S, Littman DR. Blood monocytes consist of two principal subsets with distinct migratory properties. Immunity. 2003;19(1):71–82. PubMed PMID: 12871640 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Neuroinflammation Springer Journals

Targeted intracerebral delivery of the anti-inflammatory cytokine IL13 promotes alternative activation of both microglia and macrophages after stroke

Free
17 pages
Loading next page...
 
/lp/springer_journal/targeted-intracerebral-delivery-of-the-anti-inflammatory-cytokine-il13-NBSxVRGG8j
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Neurology; Neurobiology; Immunology
eISSN
1742-2094
D.O.I.
10.1186/s12974-018-1212-7
Publisher site
See Article on Publisher Site

Abstract

Background: Subtle adjustment of the activation status of CNS resident microglia and peripheral macrophages, to promote their neuroprotective and neuroregenerative functions, may facilitate research towards curing neurodegenerative disorders. In the present study, we investigated whether targeted intracerebral delivery of the anti-inflammatory cytokine interleukin (IL)13, by means of transplanting IL13-expressing mesenchymal stem cells (IL13-MSCs), can promote a phenotypic switch in both microglia and macrophages during the pro-inflammatory phase in a mouse model of ischemic stroke. eGFP/+ RFP/+ Methods: We used the CX CR1 CCR2 transgenic mouse model to separately recognize brain-resident microglia from infiltrated macrophages. Quantitative immunohistochemical analyses were applied to characterize polarization phenotypes of both cell types. Results: Distinct behaviors of both cell populations were noted dependent on the anatomical site of the lesion. Immunohistochemistry revealed that mice grafted with IL13-MSCs, in contrast to non-grafted and MSC-grafted control mice, were able to drive recruited microglia and macrophages into an alternative activation state, as visualized by a significant increase of Arg-1 and a noticeable decrease of MHC-II expression at day 14 after ischemic stroke. Interestingly, both Arg-1 and MHC-II were expressed more abundantly in macrophages than in microglia, further confirming the distinct behavior of both cell populations. Conclusions: The current data highlight the importance of controlled and localized delivery of the anti- inflammatory cytokine IL13 for modulation of both microglia and macrophage responses after ischemic stroke, thereby providing pre-clinical rationale for the application of L13-MSCs in future investigations of neurodegenerative disorders. Keywords: Stroke, Microglia/macrophage polarization, Interleukin 13, Mesenchymal stem cells, Neuroinflammation * Correspondence: mathias@sf.mpg.de Somayyeh Hamzei Taj and Debbie Le Blon contributed equally to this work. Peter Ponsaerts and Mathias Hoehn shared senior authorship. In-vivo-NMR Laboratory, Max Planck Institute for Metabolism Research, Gleuelerstrasse 50, D-50931 Köln, Germany Department of Radiology, Leiden University Medical Center, Leiden, Netherlands Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 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. Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 2 of 17 Background can be considered as competent carrier cells, since they Ischemic stroke following cerebral artery occlusion is a (i) are conveniently engineered with viral vectors to pro- major cause of chronic disability worldwide and an effect- vide long-term gene expression [12, 13], (ii) have re- ive therapy to improve functional recovery after stroke is duced immunogenicity because of a low MHC class I missing in the clinical settings [1]. A time-dependent re- expression and the absence of MHC class II molecules cruitment and activation of immune cells, including and co-stimulatory factors [14], (iii) display a robust sur- brain-resident microglia, monocytes/macrophages, granu- vival rate upon transplantation in CNS tissue [15, 16], locytes, and T cells are a hallmark of the secondary dam- (iv) have a potential immunomodulatory effect on natural age after ischemia [2]. Under inflammatory condition killer cell function and B and T cell proliferation [17, 18], caused by ischemia, there is a complex interaction be- and (v) are able to produce immunomodulatory, neuro- tween brain-resident microglia and infiltrated macro- trophic, and angiogenic factors [19–21]. Therefore, the phages. These two phagocytic cell populations share many therapeutic relevance of MSCs in stroke models and other antigenic markers, although rapidly growing literature on neurodegenerative disorders can be applied upon intrace- this subject indicates that both cell types have distinct spe- rebral or intraventricular administration. cialized functions [3, 4]. In this study, we performed stroke The first aim of this study was to investigate the eGFP/+ RFP/+ experiments in CX CR1 CCR2 knock-in fluor- spatial distribution and the behavior of recruited resi- escent protein reporter mice, in order to distinguish dent microglia and infiltrating, peripheral macrophages brain-resident microglia from infiltrated macrophages in a transient focal cerebral ischemia model, to better after ischemic stroke. In these transgenic mice, the che- understand the contribution of these cell populations to mokine receptor CX CR1 is mainly expressed by the inflammatory process after stroke. By using the eGFP/+ RFP/+ brain-resident microglia, while CCR2 is overexpressed CX CR1 CCR2 transgenic mouse model, our in infiltrating activated macrophages [5]. A profound results have provided new insights into the trafficking and characterization of these cell populations in damaged distinct functions of resident microglia and infiltrating CNS may provide the opportunity to manipulate spe- macrophages in the post-ischemic brain. The second aim cific subsets for therapeutic benefits. of this study was to discover whether MSC-based delivery In the early stage of ischemic stroke, these immune of IL13 can modulate the spontaneous polarization shift cells can acquire a protective function known as of anti-inflammatory (M2) to pro-inflammatory (M1) anti-inflammatory (M2) phenotype, i.e., presenting char- phenotypes in microglia and macrophages during the acteristic features comparable to in vitro IL4 and/or second week after ischemic stroke. We agreed on ap- IL13 polarized cells. In the late stage of ischemic stroke, plying the simplified classification of pro-inflammatory they can polarize into a classically activated state known and anti-inflammatory states as M1 and M2 pheno- as pro-inflammatory (M1) phenotype, i.e., presenting types, respectively, in order to best describe our intention characteristic features comparable to in vitro LPS and/or of modulating the polarization of microglia and macro- interferon gamma (IFN-ϒ) activated cells [6]. Emerging phages. Our results have provided confirmatory evidence knowledge in targeting cerebral inflammation argues that transplantation of IL13-MSCs, which continuously se- that therapeutic approaches should move from extensive crete IL13, is able to polarize both microglia and macro- and forcible suppression of immune cells towards careful phages to a neuroprotective M2 phenotype during the adjustment of the balance between their different pheno- pro-inflammatory status in ischemic stroke. These data types [7]. Therefore, to drive microglia and macrophages strengthen the idea that IL13-MSCs could be considered as towards a more protective anti-inflammatory phenotype potent modulators of the cellular and molecular responses after stroke, we selected the immunomodulatory cyto- in neuroinflammation, with strong anti-inflammatory po- kine interleukin 13 (IL13), which is a well-known modu- tential, for further explorationinanimalmodelsofstroke lator of immune responses in vitro and in vivo [8, 9]. and other neurodegenerative disorders. The neuroprotective importance of the cytokine IL13 has been demonstrated in several experimental models of Methods neurodegenerative disorders. It can strongly decrease the Animals and experimental groups pro-inflammatory cytokine secretion, reduce inflammatory Wild type C57BL/6J mice were obtained via Charles cell infiltration, and suppress axonal loss [10, 11]. River Laboratories (strain code 027) and were used Effective delivery of therapeutic proteins, such as IL13 for in vivo detection of IL13 mRNA produced by in the present study, is a main challenge in neuroinflam- grafted IL13-MSCs. In total, 24 mice were included mation research. To efficiently deliver IL13 to the in the experiment, divided over the following three ischemic brain, we used autologous mesenchymal stem/ groups: (i) a non-injected control mice (n = 8), (ii) stromal cells (MSCs) genetically engineered to secrete MSC-grafted mice (n = 8), and (iii) IL13-MSC-grafted IL13. As previously reported by us and others, MSCs mice (n =8). Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 3 of 17 eGFP/eGFP Transgenic CX CR1 mice (strain code medium (IMDM; Lonza) supplemented with 8% fetal bo- RFP/RFP 005582) and CCR2 mice (strain code 017586) vine serum, 8% horse serum, 200 U/mL + 100 μg/mL peni- eGFP/+ were obtained via Jackson Laboratories. CX CR1 cillin/streptomycin, and 1 μg/mL amphotericin B (all RFP/+ CCR2 mice were obtained by breeding products from Thermo Fisher Scientific). Culture medium eGFP/eGFP RFP/RFP CX CR1 mice with CCR2 mice. During for IL13-MSCs was further supplemented with 5 μg/mL the entire study, mice were kept in the animalarium of puromycin (InvivoGen). MSC cultures were split 1:5 the University of Antwerp (UA) under normal day-night twice a week using 0.05% trypsin-EDTA (Thermo cycle (12/12) with free access to food and water. All Fisher Scientific) for cell detachment. animal experimental procedures were approved by the Ethics Committee for Animal Experiments of the UA IL13 protein secretion assay (Approval No 2015-84 and 2018-36). Wild type MSCs and IL13-MSCs were plated at a con- eGFP/+ 5 Ischemic stroke was induced in 32 male CX CR1 centration of 2 × 10 /well in a six-well plate and allowed RFP/+ CCR2 mice (11–13 weeks, 22–26 g) by middle cere- to adhere during overnight incubation. After the following bral artery occlusion (MCAO), as described below. 24 h of culture, supernatant was harvested and analyzed Stroke lesions were checked by magnetic resonance im- for the presence of IL13 protein by means of ELISA, ac- aging (MRI) scans 48 h after MCAO, before transplant- cording to the manufacturer’s instructions (Peprotech). ation of mesenchymal stem cells (MSCs). Two mice died after stroke induction (stroke-only group), one mouse Cell transplantation in healthy mouse brain before stroke during anesthesia. Four mice were excluded All surgical experiments were performed under sterile due to too small or absent stroke lesions. Eight mice died conditions, as previously described [26–28]. Briefly, mice after cell implantation (n = 3 for MSC group; n =5 for were anesthetized by an intraperitoneal injection of a IL13-MSC group). In total, 17 mice were included in the ketamine (80 mg/kg, Pfizer) + xylazine (16 mg/kg, Bayer study, resulting in the following three groups: (i) control Health Care) mixture in 0.9% NaCl solution (Baxter) group subjected to stroke only (n = 6), (ii) MSC-treated and placed in a stereotactic frame (Stoelting). A midline group, which received MSC transplantation 2 days after scalp incision was made and a hole was drilled in the MCAO (n = 5), and (iii) IL13-MSC-treated group, which skull using a dental burr drill (Stoelting). Stereotactic co- received a transplantation of IL13 producing MSCs 2 days ordinates were as follows (relative to bregma): AP 0 mm, after MCAO (n = 6). All mice were perfused at day 14 Lat. 2.3 mm and − 2.3 mm, and DV − 2.3 mm. Next, an after stroke induction. An overview of the experimental automatic microinjector pump (kdScientific) with a protocol is presented in Fig. 1. 10 μl Hamilton Syringe was positioned above the ex- posed dura. A 30-gauge needle (Hamilton), attached to Genetic engineering and culture of MSCs the syringe, was lowered through the intact dura and In this study, we used a previously established and charac- positioned at the respective depth. After 2 min of tissue terized C57BL/6 mouse bone marrow-derived MSC line pressure equilibration, a suspension of 5 × 10 MSCs in [22] and a derivative thereof, genetically engineered to ex- a volume of 0.4 μl was injected. Following 4 min to allow press murine interleukin-13 (further named as IL13-MSCs) pressure equilibration and to prevent backflow of the [23]. The latter MSC line was generated by transduction of injected cell suspension, the needle was fully retracted. parental MSCs with the pCHMWS-IL13-IRES-Pac lenti- The exact same procedure was performed at both left viral vector, according to previously optimized procedures and right side of the brain. The skin was sutured (Vicryl, [24, 25]. For expansion, both MSC lines were cultured in Ethicon), and 100 μl of a 0.9% NaCl solution was adminis- standard cell culture plastic ware in “complete expansion tered subcutaneously in order to prevent dehydration medium” [24] consisting of Iscove’s modified Dulbecco’s while mice were placed under a heating lamp to recover. Fig. 1 A time line of the experimental protocol. Stroke was induced in all mice via the middle cerebral artery occlusion (MCAO). All mice were scanned by MRI at day 2 after MCAO. MSCs or IL13-MSCs were transplanted into the right hemisphere, ipsilateral to the lesion, after performing MRI. Motor and sensory performance were tested and ranked according to the modified neurological deficit scores (mNDS), once before MCAO (baseline) and every 2 days after MCAO until the day of perfusion. All animals were sacrificed at day 14 after MCAO induction. Abbreviations: B = behavior, MCAO = middle cerebral artery occlusion, MRI = magnetic resonance imaging, IHC = immunohistochemistry Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 4 of 17 qRT-PCR for transgenic IL13 mRNA and received a subcutaneous injection of 4.0 mg/kg Car- Mice were perfused with ice-cold 0.9% NaCl solution, profen (Pfizer, Karlsruhe, Germany) for analgesia after directly followed by removal of the brain and dissection the surgical interventions. With a neck incision, the of the transplantation areas (left and right). The ex- common carotid artery (CCA) and its proximal branches tracted tissue sections were snap-frozen in liquid nitro- were exposed. The internal carotid artery (ICA) was oc- gen and kept at − 80 °C until further processing. As cluded for a short time with a metal microvessel clip. A negative and positive control for qRT-PCR analysis, cul- silicon rubber-coated filament with a tip diameter of tured MSCs and IL13-MSCs were harvested, washed, 170 μm (7017PK5Re, Doccol Corporation, Sharon, MA, and resuspended in RNAlater (Qiagen) solution at 4 °C USA) was inserted into the ICA. The filament was ad- for further processing the next day. Total RNA was ex- vanced through the ICA until it blocked the blood flow tracted using the Purelink RNA kit (Invitrogen). RNA to the middle cerebral artery. Animals were allowed to quantity and purity were determined using an ND-1000 recover during the 30 min occlusion in a temperature micro-spectrophotometer (NanoDrop Technologies). stable box (MediHeat, Peco Services Ltd., Brough, UK). Two micrograms of total RNA was reverse-transcribed Afterwards, the animals were re-anesthetized, the fila- using Omniscript RT kit (Qiagen). PCR primers were ment was carefully removed to initiate reperfusion and designed using online available primer design software the CCA was permanently ligated. During 1 week after from Thermo Fisher Scientific and were purchased from the surgery, body weight was daily monitored in all Thermo Fisher Scientific. The forward primer (5′ to 3′) animals. Animals were randomly assigned to the three GAAGCCGCTTGGAATAAGGC and the reverse primer different experimental groups. (5′ to 3′) ACCTTGCATTCCTTTGGCGA cover part of the IRES sequence within the pCHMWS-IL13-IRES-Pac IL13-MSC injection following MCAO lentiviral construct, thereby allowing to detect only trans- All surgical experiments were performed as described genic IL13 mRNA and not endogenous IL13 mRNA. earlier under section cell transplantation in healthy Real-time quantitative RT-PCR analysis was carried out mouse brain. The needle was positioned at a depth of using Power SYBR Green PCR Master Mix (Applied 2.5 mm into the striatum close to the ischemic lesion (as Biosystems) detection, using the StepOnePlus Real-Time determined following MRI analysis). First, a suspension PCR System (Thermo Fisher Scientific). Thermal cycling of 2 × 10 MSCs in a volume of 0.4 μl was injected. Fol- conditions were 10 min at 95 °C and 40 cycles of 15 s at lowing 4 min of pressure equilibration, the needle was 95 °C and 1 min at 60 °C. Melt curves were performed retracted to a depth of 1.0 mm and a second cell infusion upon completion of the cycles to ensure specificity of (2 × 10 MSCs in a volume of 0.4 μl) was performed. the product amplification. Housekeeping genes for normalization were peptidylprolyl isomerase A (ppiA) Behavioral tests for detection of transgenic IL13 mRNA in control and To observe the different aspects of neurological func- MSC- and IL13-MSC-grafted mouse brains, using for- tions, a modified neurological deficit score (mNDS) was ward primer (5′ to 3′) CAGACGCCACTGTCGCTTT performed before and every 2 days after MCAO, based and reverse primer (5′ to 3′) TGTCTTTGGAACTT on a modification of a previous report [31]. In brief, the TGTCTGCAA and GAPDH for detection of transgenic modified NDS consists of a set of motor tests (muscle IL13 mRNA in in vitro cultured MSC and IL13-MSC, status and abnormal movement), sensory tests (tactile, using forward primer (5′ to 3′) AGGTCGGTGTGAAC and proprioceptive), and reflex tests on a scale of 0–18. GGATTTG and reverse primer (5′ to 3′)GGGG Increasing score indicates the severity of the stroke TCGTTGATGGCAACA. Data were analyzed with damage [32]. qbase+ analysis software. For comparison purposes, ob- tained values for MSC-grafted and IL13-MSC-grafted MRI acquisition mice are displayed as fold expression versus the mean At 2 days post induction of stroke, mice were subjected value of control non-injected mice. Similarly, expres- to MRI using a 7T Pharmascan MR scanner with a sion of transgenic IL13 mRNA for IL13-MSC is dis- 16-cm-diameter horizontal bore (Bruker, Ettlingen, played as fold expression versus control MSC. Germany). This system is equipped with a standard Bruker cross-coil setup, using a quadrature volume coil Middle cerebral artery occlusion for excitation and an array mouse surface coil for signal Focal cerebral ischemia was induced by transient occlu- detection. The system was interfaced to a Linux PC sion of the right middle cerebral artery (MCA) in all ani- running Topspin 2.0 and Paravision 5.1 software (Bruker). mals, using the intraluminal filament model as described Mice anesthesia was induced using 2% isoflurane (Forane previously [29, 30]. Briefly, mice were anesthetized with Abbott, UK) in a gas mixture of 30% O and 70% N at a 2 2 1–2% isoflurane in a gas mixture of O /N O (30:70%) flow rate of 600 ml/min. During MRI acquisition, 2 2 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 5 of 17 isoflurane concentration was set at 2%, and the respiration Fisher Scientific, A21245). Before mounting with Prolong rate was continuously monitored using a pressure sensi- Gold antifade reagent, nuclear staining was performed tive pad. In addition, body temperature was monitored via using TOPRO-3 (1/200, Thermo Fisher Scientific). Images a rectal probe and was held constant between 37.0 and were acquired using a BX51 fluorescence microscope 37.3 °C using warm air coupled to a feedback unit (SA equipped with an Olympus DP71 digital camera. Olympus instruments, NY, USA). Both respiration and body cellSens dimension software was used for image process- temperature control systems were controlled by PC-Sam ing. In each brain section, five different regions of interest monitoring software (SA instruments, NY, USA). Following (ROIs) were selected with the constant exposure time: the MR image acquisition, mice were left to recover separately border and the core region of the ischemic hemisphere in under a heating lamp before returning to their respective the striatum, the core region of the ischemic hemisphere cages. Following scout scans, we acquired an axial proton in the cortex, and two ROIs in the cortex and striatum of density weighted Rapid Acquisition and Relaxation En- the intact contralateral hemisphere. hancement (RARE) image using the following parameters: repetition time (TR) = 3000 ms, echo time (TE) = 13.3 ms, Quantitative immunohistochemical analysis matrix size (256 × 256), field of view (FOV) = (17.5 × 17.5) For quantitative phenotypic analysis of microglia and 2 2 mm , resolution = (0.068 × 0.068) mm , 10 coronal slices, macrophage activities after stroke, immunofluorescence slice thickness = 0.8 mm, RARE factor = 8, and aver- images were analyzed using NIH ImageJ analysis soft- ages = 2. Next, for determination of the T2 values, we ware (ImageJ) and TissueQuest 4.0 (TissueGnostics, applied a multi-slice, multi-echo sequence based on the Vienna, Austria). An entire picture of all ROIs, taken at Carr-Purcell-Meiboom-Gill sequence. The following × 20 magnification, was used for quantification. Cells parameters were used: TR = 3000 ms, number of were recognized based on the nuclei staining (nuclei echoes = 10, echo spacing = 11 ms, matrix size (128 × 128), size, staining intensity, and discrimination by area was field of view (FOV) = (17.5 × 17.5) mm ,resolution= optimized manually), followed by the analysis of specific 2 + − (0.137 × 0.137) mm , 10 slices, and slice thickness = staining. The cellular densities of eGFP RFP microglia eGFP/+ − + RFP/+ 0.8 mm. (CX CR1 ), eGFP RFP macrophages (CCR2 ), + + and eGFP RFP double-positive microglia/macrophages eGFP/+ RFP/+ Immunofluorescent staining (CX CR1 CCR2 ) were quantified. To achieve All immunofluorescence analyses were performed ac- optimal cell detection, the background threshold was de- cording to previously described procedures [26, 28]. fined. To discriminate false signals due to the coverage of eGFP/+ Briefly, mice were transcardially perfused with 0.9% CX CR1 cell ramification with total nuclei intensity, NaCl solution followed by 4% paraformaldehyde (PFA). the cut-off was defined for each ROI. Scattergrams were Next, brains were isolated and post-fixed with 4% PFA generated to visualize the corresponding positive cells in for 3 h, then dehydrated through a sucrose gradient of the source ROIs through the real-time back gating com- 5, 10, and 20%. Brains were snap-frozen in liquid nitro- ponent. Mean intensity and the relative number of the eGFP/+ RFP/+ gen and kept at − 80 °C until further processing. co-expressed CX CR1 or CCR2 with pro- or Ten-micrometer-thick cryosections were made using a anti-inflammatory markers F4/80, Arg-1, and MHC-II microm HM500 cryostat. Immunofluorescent staining were obtained. Afterwards, the mean values were deter- was performed on brain slides using the following anti- mined from analyses of at least three brain sections per body combinations: a primary rabbit anti-RFP antibody mouse. Based on the above cell density analysis, the pro- (2.5 μg/ml, Abcam, ab62341) with a secondary donkey portion of microglia or macrophages expressing F4/80, anti-rabbit Alexa Fluor 555 antibody (2 μg/ml, Thermo Arg-1, or MHC-II was calculated. To observe anti- and Fisher Scientific, A31572), a primary rat anti-F4/80 pro-inflammatory representative markers in double stain- + + antibody (4 μg/ml, Bio-Rad, MCA497R) with a second- ing with eGFP or RFP , the above-defined five different ary goat anti-rat Cy5 antibody (10 μg/ml, Thermo ROIs provided the following number of cells/mm :(i) Fisher Scientific, A10525), a primary rat anti-MHC-II total cell density according to the nuclei staining of + + antibody (2.5 μg/ml, eBioscience, 14–5321-82) with a TOPRO-3, (ii) eGFP microglia cell density, (iii) RFP + + secondary goat anti-rat Alexa Fluor 350 antibody macrophage cell density, (iv) eGFP RFP double-positive + + (10 μg/ml, Thermo Fisher Scientific, A21093), and a microglia/macrophages cell density, (v) F4/80 /eGFP or + + + + primary goat anti-Arg1 antibody (4 μg/ml, Santa Cruz, RFP cell density, (vi) Arg-1 /eGFP or RFP cell density, + + + sc-18354) with a secondary donkey anti-goat Alexa and (vii) MHC-II /eGFP or RFP cell density. Fluor 350 (10 μg/ml, Thermo Fisher Scientific, A21081) and a primary rabbit anti-Ym1 antibody (1:50 dilution, Statistics StemCell Technologies 01404) with a secondary goat For statistical analyses of qRT-PCR and ELISA data, anti-rabbit Alexa Fluor 647 antibody (10 g/ml, Thermo GraphPad Prism software was used. For all other Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 6 of 17 experiments, statistical analyses were performed with post-implantation, as compared to control non-injected SPSS version 22 (IBMSPSS statistics, Ehningen, Germany). (p = 0.0216) and MSC-injected mice (p = 0.0214) (Fig. 2c). The Normality test and homogeneity of variances were Overall, these data demonstrate that IL13-MSC can effect- assessed for all data. The nonparametric analysis ap- ively produce transgenic IL13 mRNA in situ following proach, Kruskal-Wallis H, was performed for analysis of grafting in the CNS. behavioral scores (mNDS). Immunohistochemistry (IHC) and qRT-PCR data from in vivo experiments were ana- Characterization of the ischemic lesion lyzed for significant changes between the three groups Occlusion of the right middle cerebral artery with a sili- using one-way analysis of variance (ANOVA) with cone rubber-coated filament resulted in cerebral ische- Bonferroni-corrected post hoc comparisons. qRT-PCR mia. Lesion size and location were assessed by MRI data and ELISA data for comparison of in vitro cul- using quantitative T2 maps. Ischemic territory shows up tured MSCs and IL13-MSCs unpaired t tests were per- as area of increased T2 values on T2 maps, representing formed. Differences were considered statistically significant the area with augmented tissue water content. In the at a p value < 0.05. Data shown represent mean value per hemisphere ipsilateral to the lesion, vasogenic edema in- group ± standard deviation. creased T2 values and a clear lesion was detectable in the right ischemic hemisphere on T2 maps. In the intact Results contralateral hemisphere, T2 maps showed no signs of In vivo transgenic IL13 mRNA production by grafted lesion. Two days after MCAO, two characteristic lesion IL13-MSC types were detected: (1) in 5 mice, lesions were restricted In our previous studies related to in vivo grafting of to the striatum and (2) in 12 mice, lesions involved both IL13-MSCs, we assumed transgenic IL13 mRNA (and striatum and cortex. Subsequently, mice were divided protein) to be effectively produced as determined by the over three experimental groups: (1) control mice with appearance of alternatively activated microglia and mac- untreated stroke, (2) mice with stroke receiving a control rophages [23, 33, 34]. In this first part of our study, we MSC graft, and (3) mice with stroke receiving an aimed to determine whether IL13-MSCs effectively pro- IL13-MSC graft. Representative lesions in each experi- duce transgenic IL13 mRNA after in vivo transplant- mental group are displayed in Fig. 3. ation. At first, results obtained from cultured MSCs and IL13-MSCs demonstrate that both mRNA and protein ex- Neurological deficit pression levels of IL13 are only detected in IL13-MSCs We determined whether the procedure of cell grafting and not in MSCs (respectively; p <0.0001 and p = 0.0003) (MSC or IL13-MSC) into the ischemic brain influences (Fig. 2a, b). Next, following transplantation of MSCs and the motor and sensory performance after stroke in mice. IL13-MSCs in healthy mouse brain, we were able to detect In all three experimental groups, the neurological deficit significant levels of transgenic IL13 mRNA expression scores increased at day 2 after MCAO (up to score 7 in in five out of eight IL13-MSC-injected mice at day 7 mNDS) compared to baseline (score 0 in mNDS), Fig. 2 In vitro and in vivo expression of transgenic IL13 mRNA by IL13-MSCs. a Graph showing the analysis of transgenic IL13 mRNA expression by cultured MSCs and IL13-MSCs in vitro. b Graph showing the analysis of IL13 protein expression by cultured MSCs and IL13-MSCs after 24 h in vitro. c Graph showing the analysis of transgenic IL13 mRNA expression in the non-injected control group, the MSC-injected group, and the IL13-MSC-injected group at 1 week post-injection. Statistical significances are indicated by *** for p ≤ 0.001 and * for p ≤ 0.05 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 7 of 17 Fig. 3 Representative T2 maps for lesion location and size of two representative animals from each experimental group 2 days after ischemic stroke. T2 maps are displayed as coronal brain section. Lesions show up as areas of elevated T2 values on T2 maps, representing the area with increased tissue water content following stroke induction. Two characteristic lesion types are visible in the right hemisphere 48 h after MCAO: a lesions involving both striatum and cortex (n = 12) and b lesions restricted to the striatum (n =5) indicating the impaired motor and sensory functions macrophages were diffusely distributed across the whole due to the stroke. No differences were observed between territory. The only population within the intact contra- eGFP/+ the three experimental groups within the two-week obser- lateral hemisphere contained resident CX CR1 vation period. Nevertheless, our findings clearly indicate microglia (Fig. 4c). The microglia density on the ische- that the additional surgical procedure, intracranial MSC mic hemisphere was significantly higher than on the transplantation at 2 days after MCAO, did not worsen contralateral hemisphere, in all groups and analysis the neurological deficits after stroke (Additional file 1: (Fig. 4d–f). No obvious difference in this distribution pat- Figure S1). tern was observed between the three experimental groups. Next, we assessed whether any differences in cell density Distinct spatial distribution of microglia and infiltrated existed between resident and infiltrating cells 2 weeks macrophages following ischemic stroke after stroke. When evaluating the whole lesion site (i.e., In order to histologically investigate brain trafficking of the cortical and/or striatal region), the vast majority of re- eGFP/+ recruited microglia and macrophages after cerebral is- cruited cells within the lesion are CX CR1 microglia eGFP/+ RFP/+ chemia, we took advantage of CX CR1 CCR2 in all three experimental groups. Statistical analysis con- eGFP/+ double transgenic mice. eGFP and RFP expression in the firmed that CX CR1 cell density, being of microglial RFP/+ brain sections of these mice enables us to simultaneously origin, is significantly higher than CCR2 cell density, + eGFP/+ track resident eGFP microglia (CX CR1 ) and infil- being of macrophage origin, in all three experimental + RFP/+ trating RFP monocytes/macrophages (CCR2 ). Using groups at day 14 after stroke (stroke-only group p =0.013, these transgenic mice, we observed differences in displace- MSC-treated group p = 0.038, IL13-MSC-treated group ment of migration between microglia and peripheral p = 0.017) (Fig. 4d). When analyzing the cell density of monocyte-derived macrophages. Histological analysis re- microglia and macrophages, separately for striatal and vealed a noticeable infiltration of both microglia and mac- for cortico-striatal lesions, the statistical significance rophages throughout the ischemic territory at day 14 after between both cell populations was preserved in all stroke induction, (Fig. 4a–c). From the representative im- cases, except for the two MSC groups in the cortical munofluorescence images in Fig. 4b, we noted that eGFP analysis (Fig. 4e), where only a trend was reached (MSC and RFP expression shows a spatially different distribution group p = 0.06; IL13-MSC group p = 0.063). Despite the in the cortex and the striatum. In the cortical part of the difference in spatial distribution of microglia and RFP/+ ischemic lesion, CCR2 macrophages were located in macrophages between the striatal and cortical lesion eGFP/+ the infarct core and CX CR1 microglia were mainly area, further quantitative histological analysis revealed accumulated at the infarct border. In contrast, in the stri- closely comparable numbers of infiltrating macro- eGFP/+ RFP/+ atal lesion area, CX CR1 microglia and CCR2 phages and of resident microglia between cortical 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 8 of 17 Fig. 4 Localization and cell density of microglia and macrophages 2 weeks after ischemic stroke. a Representative coronal T2-weighted MR images are displayed above the corresponding immunohistochemistry images of all experimental groups. The hyperintense regions in the coronal MR images show the ischemic lesion 2 days post stroke. b–c Fluorescent microscopic images of brain coronal sections of all experimental eGFP RFP + groups represent the distribution of CX CR1 microglia and CCR2 macrophages at day 14 after stroke; b localization of eGFP microglia and RFP macrophages in ipsilateral cortical and striatal ischemic lesions, and c in contralateral cortical and striatal intact regions. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages. d–f Representative stacked column graphs show the cell density quantification of all experimental groups at day 14 after stroke. d Evaluation of microglia/macrophage cell density in the whole hemisphere. No significant difference + + eGFP/+ was observed in the total number of eGFP microglia and RFP macrophages between the three experimental groups. CX CR1 cell density RFP/+ + is significantly higher than CCR2 cell density in all three experimental groups at day 14 after stroke. In intact contralateral hemisphere, eGFP microglia was the main population. e Further evaluation of microglia and macrophage cell density in cortical lesion areas, and f in striatal lesion areas. The cell density of recruited microglia and macrophages was higher in large cortical lesion areas compared to striatal lesion areas. Data represent mean ± SD. The data were compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’s eGFP/+ RFP/+ post hoc test. To compare the CX CR1 cell density to CCR2 cell density in each group, an independent one-tailed Student’s t test was used (Fig. 4e)and striatal (Fig. 4f) lesions over the three (Fig. 5, left column): (i) cells with long thin arborized experimental groups. processes and a small cell body, defined as ramified cells (top row), (ii) cells with swollen processes and elongated eGFP/+ Recruited CX CR1 microglia exhibit various amorphous larger cell body, classified as intermediate morphological phenotypes within the ischemic cells (middle row), and (iii) cells with round shape and hemisphere no plasmalemmal processes, specified as amoeboid cells eGFP/+ eGFP/+ CX CR1 microglia are visible in both the intact (bottom row). Ramified CX CR1 cells were mostly 3 3 and the ischemic hemisphere, but are found in different observed in the healthy periphery of the lesion and in numbers, distribution patterns (see above), and morpho- the contralateral hemisphere, but were not seen in the eGFP/+ logical appearance. In this study, we could observe at core of the lesion. Intermediate type CX CR1 cells least three distinct morphological appearances of were closely associated with peri-infarct regions. eGFP/+ eGFP/+ CX CR1 microglia at day 14 after stroke induction Amoeboid CX CR1 cells were mainly localized in 3 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 9 of 17 Fig. 5 Different morphology of microglia and macrophages at day 14 after ischemic stroke. Representative photomicrographs with subsequent magnified images represent different morphology of microglia/macrophages in a intact contralateral hemisphere, b border of the ischemic lesion of ipsilateral hemisphere, and c the core of the ischemic lesion of ipsilateral hemisphere. eGFP microglia display at least three distinct phenotypes at day 14 after stroke including ramified cells (top row), intermediate cells (middle row), and amoeboid cells (bottom row). RFP macrophages show oval or round shape without apparent processes. F4/80 cells, alternatively activated cells, display intermediate and amoeboid shape. Scale bar 50 μm RFP/+ the ischemic core. In contrast, recruited CCR2 mac- macrophage populations (central and bottom row) within rophages displayed a uniform oval, kidney form, or the lesion area, at least in part, express this F4/80 activa- round shape without apparent processes (Fig. 5, central tion marker. Notably, morphological inspection of F4/ column). Some CCR2 cells showed cellular processes, 80-expressing cells shows that these cells display an inter- suggesting that these cells may acquire an intermediate mediate or amoeboid shape. Next, we assessed whether phenotype in the damaged tissue. Examples in Fig. 5 microglia and/or macrophage activation, based on F4/ were taken from the stroke-only group, but no obvious 80 expression, was altered upon transplantation of differences between the appearance of microglia and IL13-producing MSC, as compared to the situation macrophage morphology was observed between the after transplantation of MSC alone or without any three experimental groups. transplantation after stroke induction (Fig. 6). As shown by the representative immunofluorescent images Transplantation of IL13-producing MSCs increases the in Fig. 6a, and further supported by the quantitative proportion of highly activated infiltrating macrophages analysis provided in Fig. 6b (relative proportion of F4/ Not only the recruitment of microglia and macrophages 80 expression, stacked bars) and Fig. 5c (absolute pro- is a key feature of neuroinflammation, but also the sub- portion of F4/80 expression, pie charts), F4/80 expres- sequent phenotypic alterations associated with their acti- sion was observed on both microglia and macrophages vation status. Here, we first investigated the appearance in all three experimental groups. Interestingly, we ob- of F4/80 expression, a general marker of activation, on served a significantly increased expression of F4/80 by RFP/+ microglia and macrophages following stroke (Fig. 5, right CCR2 macrophages upon IL13-MSC transplantation RFP/+ column). F4/80 expression was not detected contralateral as compared to F4/80 expression by CCR2 macro- to the lesion site (top row), while both microglia and phages following stroke or stroke + control MSC grafting Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 10 of 17 Fig. 6 Transplantation of IL13-producing MSCs results in alternative activation of microglia and macrophages at day 14 after ischemic stroke. a eGFP/+ RFP/+ Representative fluorescent microscopic images of ischemic cortical region display the expression of F4/80 on CX CR1 and CCR2 cells, acquired from the three experimental groups at day 14 after stroke. Illustrative ×20 close-up magnification represents the changes in F4/80 expression by microglia and macrophages. F4/80 biomarker is expressed on both cell populations in the hemisphere ipsilateral to the lesion. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages; white, F4/80. b Detailed analysis of the expression of biomarker F4/80 RFP/+ revealed a significant increase of F4/80 expression of CCR2 macrophages upon IL13-MSCs transplantation in comparison to other two control + + groups. c The corresponding exploded pie charts show the distribution of F4/80 expression by GFP and RFP cells in all three experimental groups. + + In the pie chart, GFP microglia and RFP macrophages are encircled in green and red, respectively. n =5–6 mice in each group. Data represent mean ± SD. The data were compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’s post hoc test (p = 0.008 and p = 0.007). The increased number of F4/80 immunofluorescent images in Fig. 7a, and further sup- microglia detected following IL13-MSC grafting was not ported by the quantitative analysis provided in Fig. 7b significantly different from other groups. (relative proportion of Arg1 expression, stacked bars) and Fig. 6c (absolute proportion of Arg1 expression, pie Transplantation of IL13-producing MSC following stroke charts), significant increase in Arg1 expression was promotes the induction of alternatively activated observed within and surrounding the lesion area on microglia and macrophages both microglia and macrophages following grafting of To further analyze the phenotype of recruited microglia IL13-MSC (for microglia, stroke + IL13-MSC vs. stroke + and macrophages within the ischemic stroke lesion, as MSC and stroke only, p = 0.002 and p = 0.005 respectively; well as the effect of IL13-MSC thereon, we first performed for macrophages, stroke + IL13-MSC vs. stroke + MSC and additional immunofluorescence staining for Arg-1. Not- stroke only, both p < 0.001). Furthermore, the induc- ably, we wanted to investigate whether IL13-producing tion of this alternative activation program, as charac- MSC were able to convert stroke-associated neuroinflam- terized by Arg1 expression, was significantly higher in matory immune responses into an alternatively activated macrophages than in microglia in the IL13-MSC inflammatory response. As shown by the representative group (p = 0.002). Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 11 of 17 Fig. 7 Transplantation of IL13-producing MSCs increases the number of recruited neuroprotective microglia and macrophages at day 14 after eGFP/+ ischemic stroke. a Representative fluorescent microscopic images of ischemic cortical region show the expression of Arg-1 on CX CR1 and RFP/+ CCR2 cells in all three experimental groups at day 14 after stroke. Illustrative 20× close-up magnification display the changes in Arg-1 expression by microglia and macrophages. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages; white, Arg-1. b Detailed eGFP/+ RFP/+ phenotypic quantitative analysis of CX CR1 and CCR2 cells expressing Arg-1 showed a significant increase in the number of + − + − + + GFP RFP Arg-1 and GFP RFP Arg-1 cells in IL13-MSC-treated group, in comparison to both control groups. c The corresponding exploded pie + + + charts show the distribution of Arg-1 expression by GFP and RFP cells in all three experimental groups. In the pie chart, GFP microglia and + + RFP macrophages are encircled in green and red, respectively. In the IL13-MSC group, the Arg-1 fraction of macrophages is significantly higher than that of microglia. n =5–6 mice in each group. Data represent Mean ± SD. The data were compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’s post hoc test Transplantation of IL13-producing MSC following stroke microglia (p = 0.008, p = 0.007 and p = 0.002, respectively, may reduce pro-inflammatory MHC-II expression on in each group). Following IL13-MSC grafting, no signifi- infiltrated macrophages cant difference in MHC-II expression was observed on Likewise, we investigated the degree of MHC-II expression microglia. Even so, infiltrating macrophages seem to dis- within and surrounding the lesion area over the different play less MHC-II expression following IL13-MSC grafting, experimental conditions. As shown by the representative albeit this decrease was not statistically significant. immunofluorescent images in Fig. 8a, MHC-II expression is observed mainly inside the ischemic lesion in all three Characterization of MSC and IL13-MSC grafts experimental groups at day 14 after stroke induction. Fur- Similar to our preceding studies [23, 33–35], MSC and ther quantitative analyses provided in Fig. 8b (relative pro- IL13-MSC grafts were able to survive in the pro-inflam- portion of MHC-II expression, stacked bars) and 7C matory stroke environment and displayed a similar re- RFP/+ (absolute proportion of MHC-II expression, pie charts), in- modeling pattern, MSC graft-infiltrating CCR2 dicated that MHC-II expression is significantly higher in monocytes/macrophages at the core of the MSC RFP/+ eGFP/+ eGFP/+ CCR2 macrophages, compared to CX CR1 grafts and brain-resident CX3CR1 microglia and 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 12 of 17 Fig. 8 Transplantation of IL13-producing MSCs switches the polarization of microglia and macrophages towards less pro-inflammatory condition at day 14 after ischemic stroke. a Representative fluorescent microscopic images of ischemic cortical region show the expression of MHC-II on eGFP/+ RFP/+ CX CR1 and CCR2 cells in all three experimental groups at day 14 after stroke. Illustrative ×20 close-up magnification displays the changes in MHC-II expression by microglia and macrophages. Immunofluorescence colors: blue, TOPRO; green, microglia; red, macrophages; white, MHC-II. b Detailed eGFP/+ RFP/+ phenotypic quantitative analysis of CX CR1 and CCR2 cells expressing MHC-II showed a significant difference in the number of macrophages − + + + − + expressing MHCII, GFP RFP MHC-II , in comparison to the number of microglia expressing MHC-II, GFP RFP MHC-II , in all three experimental groups. In RFP/+ IL13-MSC-treated group, a trend towards less MHC-II expression by CCR2 macrophages was detected, in comparison to both control groups. c The + + corresponding exploded pie charts show the distribution of MHC-II expression by GFP and RFP cells in all three experimental groups. In the pie chart, + + + GFP microglia and RFP macrophages are encircled in green and red, respectively. In all three groups, the MHC-II fraction of macrophages is significantly higher than that of microglia. n =5–6miceineachgroup.Datarepresent mean±SD.Thedatawere compared between the three experimental groups using a parametric one-way ANOVA test with Bonferroni’sposthoc test astrocytes surrounding the MSC grafts . Expression of the they are randomly spread throughout the striatum. In two M2 markers Arginase1 and Ym1 by microglia and addition to this region-dependent distribution, we here monocytes/macrophages, as a direct result of stimulation demonstrate a strong effect on modulation of the by IL13, was only detected in IL13-MSC grafts, but not in polarization of resident microglia and infiltrated macro- MSC grafts (Additional file 2:Figure S2). phages through the IL13 secretion by transplanted MSCs after ischemic stroke. The locally secreted IL13 modu- Discussion lates the stroke-induced immune reaction by promoting The present study demonstrates that brain-resident an anti-inflammatory, M2-like phenotype of microglia microglia and infiltrated macrophages accumulate in the and macrophages, as extensively visualized by arginase 1 post-ischemic brain with distinct spatial patterns, indi- expression, with the nature of the M2-like phenotype be- cating that they act as functionally different populations ing further confirmed by the demonstration of Ym1 ex- in CNS injury. While these two cell populations dis- pression. This protective phenotype is much stronger in tinctly accumulate in the ischemic lesion in the cortex, macrophages but also exists in microglia. The potent Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 13 of 17 immunomodulatory effect of IL13 encourages to further the peri-infarct area encircling the ischemic lesion core evaluate the application of this cytokine in clinically rele- filled with CCR2 macrophages [39]. Indeed, in a rather vant neurodegenerative disease models. complex study design, these latter authors demon- + − strated that CCR2 CX3CR1 monocytes may, in a Spatial distribution of microglia and infiltrated time-dependent fashion after infiltration, turn into − + macrophages following ischemic stroke CCR2 CX3CR1 macrophages. Thus, we must caution The phenotypic distinction between CNS-infiltrating that our assignment of green fluorescent, CX3CR1-positive macrophages and brain-resident microglia is a major im- microglia may to some extent also contain macrophages. munohistochemical concern. It is not only important in This, however, does not affect our assignment of CCR2 recognizing the origin of these cell populations, but it is red cells as a pure macrophage population. Another study also of fundamental importance in assessing the patho- showed that infiltrated blood-borne monocytes were exclu- genic and therapeutic significance of immune cells sively located at the ischemically injured striatum at days 3, within the damaged brain. The lack of a single specific 7, and 14 after MCAO [40]. Their results represent a ran- membranous and/or biochemical marker allowing de- dom distribution of brain-resident microglia and infiltrated finitive and discriminating identification of these cells is monocyte within the striatum after ischemic stroke, with still a puzzling issue in neurobiology. Fortunately, alter- microglia being the vast majorityof cells invading theische- native methods have been developed to overcome this mic lesion. Moreover, in a mouse model of traumatic spinal +/GFP +/RFP hurdle, including the generation of bone marrow cord injury, CX CR1 and CCR2 cells were ran- chimeric mice, which has proven to provide a powerful domly distributed around and inside the lesion, with +/RFP tool to distinguish microglia and infiltrated macrophages CCR2 cells constituting the greater number of accu- followingischemicstroke[36, 37]. In this study, how- mulated cells in the lesion area [41]. +/GFP +/RFP ever, we made use of the CX CR1 CCR2 These findings suggest an unknown mechanism by transgenic mouse model, which is based on the fact which temporally and spatially distinct ischemic regions that chemokine receptor CX CR1 is predominantly can differentially signal to microglia and macrophages. expressed by CNS resident microglia and that CCR2 is Region-specific differences between cortex and striatum upregulated in activated infiltrated macrophages and in regard to vascular volume [30], neurogenic potential therefore allows distinction of eGFP-expressing micro- [42] and neuroinflammatory responses [43] may cause glia from RFP-expressing infiltrated macrophages [5]. the region-dependent distribution of microglia and mac- Through the use of this transgenic mouse model and rophages after ischemic stroke. +/GFP immunofluorescence microscopy, we were able to demon- In our experimental setup, considering CX CR1 strate a region-dependent distribution of microglia and cells as brain-resident microglia, it might be argued that macrophages at day 14 after stroke. In the cortical ische- perivascular macrophages, supraependymal macrophages, mic lesion, CX CR1 microglia distinctly accumulated at epiplexus cells of the choroid plexus, and meningeal mac- the border of the lesion, while CCR2 macrophages were rophages express CX CR1 too [44]. In addition, we cannot localized at the core of the lesion. At day 14, after ische- neglect the presence of patrolling macrophages Ly6c +/GFP high low mic stroke, CX CR1 microglia were the major con- CX CR1 CCR2 in the ischemic brain with disrupted 3 3 tributors of recruited cells to the ischemic lesion. These blood-brain barrier [41, 45]. Interestingly, in an attempt to observations are in line with our earlier findings following observe chemokine receptor expression changes in micro- MSC transplantation in mouse brain, which demonstrated glia and monocytes/macrophages in development and a similar distinct distribution of microglia and macro- during inflammatory condition, Mizutani and colleagues phages, where infiltrated macrophages invaded the hyp- have indicated that CCR2 is absent in adult naïve and in- oxic/ischemic MSC transplant core, while active microglia flamed CNS resident microglia [5]. Importantly, they have remained in the surrounding border [23, 25, 38]. In con- demonstrated that CX CR1 expression by microglia is sig- low high trast to their orderly distribution in the cortex, both nificantly higher compared to CX CR1 CCR2 or + + high low CX CR1 microglia and CCR2 macrophages were found CX CR1 CCR2 monocytes/macrophages. Therefore, 3 3 +/GFP to be randomly distributed in striatal lesions. In agreement we hypothesize that CX3CR1 cells are mostly of with our own observations, others have also reported brain-resident microglial origin and double-positive cells +/GFP +/RFP that resident microglia and blood-derived macrophages for CX CR1 and CCR2 are blood-derived localize in the ischemic brain with different temporal macrophage populations [33, 41, 46]. It should be empha- and spatial patterns. Garcia-Bonilla et al. describe that sized that the presence of double-positive cells at day 14 at the first week after stroke, mostly diffused CCR2 after ischemic stroke accounts for a very small proportion macrophages were observed throughout the ischemic of infiltrated macrophages in our experiment. Further lesion, while during the second and third week after assessment of earlier or later time point of macro- stroke, they saw that CX CR1 cells were localized in phage recruitment following CNS damage should also 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 14 of 17 RFP/+ be evaluated. Nevertheless, the discovery of more CCR2 double transgenic mice revealed that about microglia-specific markers would contribute to more two thirds of Arg-1 cells are infiltrated macrophages. Evi- efficient discrimination between CNS resident micro- dently, continued secretion of the cytokine IL13 by MSCs glia and infiltrated macrophages in future studies. was able to boost the immunomodulatory properties of MSCs by increasing the number of Arg-1 cells, thus Dynamic modulation of microglia and infiltrated acting as a potent neuroprotective mediator [47]. Several macrophages following ischemic stroke groups have reported that Arg-1 expression by different In this study, we are the first to determine the immuno- cell types, including astrocytes, neurons, and immune modulatory potential of MSC and IL13-expressing MSC cells, notably decreases during the first few days after cere- engraftment in adjusting the polarization of brain-resident bral ischemia, after an early peak at day 3 [6, 48]. By con- microglia and infiltrated macrophages in an ischemic tinuous secretion of IL-13 via MSCs, we could enhance stroke model. We focused on addressing important appli- and prolong the expression of neuroprotective and cations of IL13-expressing MSCs as a new potential thera- anti-inflammatory marker Arg-1 until week 2 after stroke peutic approach for ischemic stroke. The main goal of our induction. In an earlier study by our laboratory, in which study was to enhance the therapeutic effects of MSC we attempted to modulate the polarization of microglia transplantation following ischemic stroke by boosting the and macrophages by microRNA-124 after ischemic stroke, immunomodulatory properties of MSCs. For this reason, we achieved an upregulation of Arg-1 and CD-206 expres- we used MSCs as a biodegradable delivery system for the sion by microglia and macrophages until day 6 after stroke. immune modulating cytokine IL13. With this method, In this study, the increased polarization of microglia and we achieved long-term effect on immune function until macrophages towards the more anti-inflammatory pheno- the second week after stroke induction, during the type was positively correlated to increased neuronal sur- pro-inflammatory peak of stroke [6]. Earlier investiga- vival and functional recovery at day 6 after stroke [32, 49]. tionsbyus, on theuseof MSCs as adeliverysystemfor Another important observation to be taken into ac- the anti-inflammatory cytokine IL13, demonstrated that count here is the expression and accumulation of major continued secretion of IL13 by MSCs is able to limit histocompatibility complex class II (MHC-II) following microgliosis, oligodendrocyte loss, and demyelination MSC and IL13-expressing MSC engraftment after stroke. in cuprizone-treated mice, a model for neuroinflamma- MHC-II expression was predominantly observed on infil- tion and demyelination [23], and promotes histopatho- trated macrophages, localized at the core of the ischemic logical and functional recovery following spinal cord lesion. MHC-II is considered to be a pro-inflammatory injury in mice [33]. In these studies, we also showed mediator, which is overexpressed under a wide range of that transplantation of IL13-expressing MSCs induces pathological conditions [50]. However, some studies have alternative activation in both MSC graft-invading mac- highlighted that MHC-II expression can also provide pro- rophages and in MSC graft-surrounding microglia, tective and neurotrophic functions and is involved in characterized by the expression of Arg-1 and Ym1 as regeneration and axon integrity in neurodegenerative dis- anti-inflammatory-associated markers. orders [51, 52]. As reported previously, accumulation of + + Here, we present for the first time that transplantation CD11b MHC-II cells in the ipsilateral ischemic hemi- of IL13-expressing MSCs at 48 h after ischemia shifts sphere after transient MCAO provides trophic support major players of the immune reaction, namely microglia and is effective in the remyelination process after stroke and macrophages, towards an anti-inflammatory, neuro- [53]. With regard to the expression of Arg-1 and MHC-II, protective phenotype at 14 days after ischemia. Although mostly by macrophages at the core of ischemic lesion in not investigated in detail in this study, our previous data our study, we could argue a substantial role of infil- with regard to grafting IL-13 expressing MSCs in muscle trated macrophages for the regeneration and repair of brain tissue have shown that the anti-inflammatory process of injured brain. We propose that infiltrated + + M2 phenotype can be induced already at 5 and 7 days pro-inflammatory CCR2 MHC-II macrophages may post grafting, respectively [35]. The transplantation of exert some pro-neurogenic functions such as clear- IL13-expressing MSCs at day 2 after ischemia caused a ance of cell debris after brain injury, which could be noticeable increase in the expression of the general activa- efficient at neurogenesis and regeneration process tion marker F4/80, especially by infiltrated macrophages. after ischemic stroke. Based on our data, transplant- Most importantly, MSC graft-associated microglia and in- ation of IL13-expressing MSCs does not broadly sup- filtrated macrophages were strongly driven towards an press M1 phenotype of microglia and macrophages, Arg-1 alternative activation status at day 14 after stroke. but it rather adjusts the balance between pro- and Further distinction between recruited brain-resident anti-inflammatory immune cell processes. As M1-activated microglia and infiltrated systemic macrophages following cells typically resolve injured and/or infected tissue, which eGFP/+ stroke and MSC transplantation in CX CR1 can lead to detrimental exacerbated injury of healthy tissue, 3 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 15 of 17 they can also assist in the repair/conservation of syn- Additional files aptic connectivity and axonal regeneration [54, 55]. Additional file 1: Figure S1. Motor and sensory performance after When considering M2 activated immune cells, mainly ischemic stroke. Behavioral performance was assessed by the modified anti-inflammatory and regeneration-inducing functions neurological deficit scores (mNDS), before and every 2 days after MCAO. are described, but it is worth noting that long-term main- In all three groups, the mNDS increased significantly at day 2 after MCAO compared to baseline. Graphs of all three experimental groups show no tenance of anti-inflammatory phenotype may adversely significant difference among the groups. The procedure of cell grafting affect the immune response, which leads to serious nega- into the ischemic brain and the additional surgical procedure did not tive effects, such as tumorigenesis [56]. Induction of an worsen the motor and sensory performance after stroke. n =5–6 mice in each group. Data are mean ± SD. The data were compared between M1/M2 phenotype switch to improve natural repair the three experimental groups using the nonparametric Kruskal-Wallis processes in neurodegenerative disorders should be con- H test. (TIF 837 kb) sidered with caution. In stroke, our earlier studies on Additional file 2: Figure S2. Representative example of MSC and IL13- modulation of the pro-/anti-inflammatory balance by MSC graft site remodeling within the MCAO brain lesion site. Control MSC (upper panel) and IL13-MSC (lower panel) grafts are able to survive in the microRNA-124 demonstrated enhanced neuronal survival pro-inflammatory stroke environment and display a similar remodeling and behavioral improvement [32, 49]. RFP/+ pattern. MSC graft-infiltrating CCR2 monocytes/macrophages (in red) at eGFP/+ In our present investigation, we have not analyzed the in- the core of the MSC grafts and brain-resident CX3CR1 microglia (in green) and astrocytes (in blue, first row) surrounding the MSC grafts. fluence of continuous secretion of the anti-inflammatory Arginase1 expression (in blue, second and third row) and Ym1 expression cytokine IL-13 and/or M2 polarized microglia and mac- (in magenta, third row) by microglia and monocytes/macrophages, as a rophages on the survival of autologous MSC grafts after direct result of stimulation by IL13, was only detected in IL13-MSC grafts, but not in control MSC grafts. Scale bar 100 μm. (TIF 13199 kb) stroke. But in an earlier study, we have demonstrated that transplantation of IL13-expressing MSCs leads to a notable decrease of direct and indirect immune recog- Abbreviations AKT: Alpha serine/threonine-protein kinase; Arg-1: Arginase-1; CCA: Common nition and rejection of cell grafts [35]. Further support carotid artery; CCR2: CC-chemokine receptor 2; CD11b: Clusters of differentiation of functional interaction between alternative activation 11b; CNS: Central nervous system; CX CR1: CX C chemokine receptor 1; 3 3 status of surrounding inflammatory milieu at the graft EAE: Experimental autoimmune encephalomyelitis; ERK: Extracellular signal- regulated kinases; FOXO3: Forkhead box O3; GAPDH: Glyceraldehyde-3-phosphate site and survival of MSCs comes from a study by Yu et dehydrogenase; Iba-1: Ionized calcium-binding adapter molecule 1; IFN- al. They have recently demonstrated that M2 ϒ: Interferon gamma; IHC: Immunohistochemistry; IL: Interleukin; IL13-MSC: IL13- macrophage-secreted OA/GPNMB, osteoactivin/glyco- producing MSCs; LPS: Lipopolysaccharide; MCA: Middle cerebral artery; MCAO: Middle cerebral artery occlusion; MHC-II: Major histocompatibility complex protein in non-metastatic melanoma protein B, class II; MIF: Macrophage migration inhibitory factor; miR-124: MicroRNA-124; positively contributes to viability, proliferation, and mi- mNDS: Modified neurological deficit score; MRI: Magnetic resonance gration of MSCs, through ERK and AKT signaling path- imaging; MSCs: Mesenchymal stem cells; PBS: Phosphate-buffered saline; PFA: Paraformaldehyde; ppiA: Peptidylprolyl isomerase A; RT: Room temperature ways [57]. Similar MSC survival improvement was achieved via an ischemic microenvironment in vitro Acknowledgements study, in which macrophage migration inhibitory factor We thank Ulla Uhlenküken for her professional arrangement of the graphics art work. (MIF), a cytokine expressed by activated monocytes/ macrophages, protects MSCs from apoptosis via a Funding CD74-dependent Akt-FOXO3a-related pathway [58]. This work was financially supported by grants from the EU-FP7 programs Enhancing MSC survival in the face of oxygen and nu- TargetBraIn (HEALTH-F2-2012-279017) and BrainPath (PIAPP-GA-2013-612360), as well as by funds from the University of Antwerp (Belgium), the Fund for trient deprivation that naturally occur in the ischemic Scientific Research (FWO)—Flanders (G091518N), and the Belgian Charcot stroke and cell graft procedure clearly needs further Foundation. Chloé Hoornaert and Alessandra Quarta hold an FWO-funded investigation. Ph.D. studentship and Debbie Le Blon holds an FWO-funded post-doctoral fellowship. All funding bodies had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript. Conclusions Availability of data and materials Our findings will serve as a base for future studies to The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. apply MSCs in a more effective way to improve CNS re- pair. We have identified that MSCs expressing the Authors’ contributions anti-inflammatory cytokine IL13 serve as an excellent SHT designed the project, performed the surgery and the cell implantation, candidate for more effective modulation of inflammatory analyzed the immunohistochemical data, wrote the manuscript. DLB performed the qRT-PCR and immunohistochemical staining and data analysis, wrote the responses in neurodegenerative disorders. This new manuscript. CH performed the immunohistochemical staining and data analysis. insight could have noticeable therapeutic implications, JD performed the immunohistochemical staining and data analysis. AQ performed which demands further studies using human biological the immunohistochemical staining and data analysis. JP performed the MRI experiments. AvdL analyzed the MRI and helped in writing the manuscript. systems, such as human-derived mesenchymal stem PP designed the project, analyzed the data, and wrote the manuscript. MH cells, for translation of immune response modulation re- designed the project, analyzed the data, and wrote the manuscript. All authors search into clinical practice. read and approved the final manuscript. Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 16 of 17 Ethics approval and consent to participate 14. Ankrum JA, Ong JF, Karp JM. Mesenchymal stem cells: immune evasive, not N/A immune priviledged. Nat Biotechnol. 2014;32:252–60. 15. De Vocht N, Lin D, Praet J, Hoornaert C, Reekmans K, Le Blon D, et al. Competing interests Quantitative and phenotypic analysis of mesenchymal stromal cell graft survival The authors declare that they have no competing interests. and recognition by microglia and astrocytes in mouse brain. Immunobiology. 2013;218(5):696–705. https://doi.org/10.1016/j.imbio.2012.08.266. PubMed PMID: Publisher’sNote 16. Praet J, Reekmans K, Lin D, De Vocht N, Bergwerf I, Tambuyzer B, et al. Cell Springer Nature remains neutral with regard to jurisdictional claims in type-associated differences in migration, survival, and immunogenicity published maps and institutional affiliations. following grafting in CNS tissue. Cell Transplant. 2012;21(9):1867–81. https://doi.org/10.3727/096368912X636920. PubMed PMID: 22472278 Author details 17. Le Blanc K. Immunomodulatory effects of fetal and adult mesenchymal stem In-vivo-NMR Laboratory, Max Planck Institute for Metabolism Research, cells. Cytotherapy. 2003;5(6):485–9. https://doi.org/10.1080/14653240310003611. Gleuelerstrasse 50, D-50931 Köln, Germany. Laboratory of Experimental PubMed PMID: 14660044 Hematology, University of Antwerp, Antwerp, Belgium. Vaccine and 18. Krampera M, Glennie S, Dyson J, Scott D, Laylor R, Simpson E, et al. Bone Infectious Disease Institute (Vaxinfectio), University of Antwerp, Antwerp, marrow mesenchymal stem cells inhibit the response of naive and Belgium. Bio-Imaging Laboratory, University of Antwerp, Antwerp, Belgium. memory antigen-specific T cells to their cognate peptide. Blood. 2003; Department of Radiology, Leiden University Medical Center, Leiden, 101(9):3722–9. https://doi.org/10.1182/blood-2002-07-2104. PubMed PMID: Netherlands. 19. Quertainmont R, Cantinieaux D, Botman O, Sid S, Schoenen J, Franzen R. Received: 10 November 2017 Accepted: 21 May 2018 Mesenchymal stem cell graft improves recovery after spinal cord injury in adult rats through neurotrophic and pro-angiogenic actions. PLoS One. 2012;7(6):e39500. https://doi.org/10.1371/journal.pone.0039500. PubMed References PMID: 22745769; PubMed Central PMCID: PMC3380009 1. Donnan GA, Fisher M, Macleod M, Davis SM. Stroke Lancet. 2008;371(9624): 20. Nauta AJ, Fibbe WE. Immunomodulatory properties of mesenchymal 1612–23. https://doi.org/10.1016/S0140-6736(08)60694-7. PubMed PMID: stromal cells. Blood. 2007;110(10):3499–506. https://doi.org/10.1182/blood- 2007-02-069716. PubMed PMID: 17664353 2. Jin R, Yang G, Li G. Inflammatory mechanisms in ischemic stroke: role of 21. Ishikane S, Ohnishi S, Yamahara K, Sada M, Harada K, Mishima K, et al. inflammatory cells. J Leukoc Biol. 2010;87(5):779–89. https://doi.org/10.1189/ Allogeneic injection of fetal membrane-derived mesenchymal stem cells jlb.1109766. PubMed PMID: 20130219; PubMed Central PMCID: PMC2858674 induces therapeutic angiogenesis in a rat model of hind limb ischemia. 3. Patel AR, Ritzel R, McCullough LD, Liu F. Microglia and ischemic stroke: a Stem Cells. 2008;26(10):2625–33. https://doi.org/10.1634/stemcells.2008-0236. double-edged sword. Int J Physiol Pathophysiol Pharmacol. 2013;5(2):73–90. PubMed PMID: 18669910 PubMed PMID: 23750306; PubMed Central PMCID: PMC3669736 22. Tambuyzer BR, Bergwerf I, De Vocht N, Reekmans K, Daans J, Jorens PG, 4. Shechter R, Schwartz M. Harnessing monocyte-derived macrophages to et al. Allogeneic stromal cell implantation in brain tissue leads to robust control central nervous system pathologies: no longer ‘if’ but ‘how’. J Pathol. microglial activation. Immunol Cell Biol. 2009;87(4):267–73. https://doi.org/ 2013;229(2):332–46. https://doi.org/10.1002/path.4106. PubMed PMID: 23007711 10.1038/icb.2009.12. PubMed PMID: 19290016 5. Mizutani M, Pino PA, Saederup N, Charo IF, Ransohoff RM, Cardona AE. The 23. Le Blon D, Guglielmetti C, Hoornaert C, Quarta A, Daans J, Dooley D, et al. fractalkine receptor but not CCR2 is present on microglia from embryonic Intracerebral transplantation of interleukin 13-producing mesenchymal development throughout adulthood. J Immunol. 2012;188(1):29–36. stem cells limits microgliosis, oligodendrocyte loss and demyelination in the https://doi.org/10.4049/jimmunol.1100421. PubMed PMID: WOS: cuprizone mouse model. J Neuroinflammation. 2016;13(1):288. https://doi. org/10.1186/s12974-016-0756-7. PubMed PMID: 27829467; PubMed Central 6. Hu X, Li P, Guo Y, Wang H, Leak RK, Chen S, et al. Microglia/macrophage PMCID: PMC5103449 polarization dynamics reveal novel mechanism of injury expansion after 24. Bergwerf I, De Vocht N, Tambuyzer B, Verschueren J, Reekmans K, Daans J, focal cerebral ischemia. Stroke. 2012;43(11):3063–70. https://doi.org/10.1161/ et al. Reporter gene-expressing bone marrow-derived stromal cells are STROKEAHA.112.659656. PubMed PMID: 22933588 immune-tolerated following implantation in the central nervous system of 7. Hu X, Leak RK, Shi Y, Suenaga J, Gao Y, Zheng P, et al. Microglial and syngeneic immunocompetent mice. BMC Biotechnol. 2009;9:1. https://doi. macrophage polarization-new prospects for brain repair. Nat Rev Neurol. org/10.1186/1472-6750-9-1. PubMed PMID: 19128466; PubMed Central 2015;11(1):56–64. https://doi.org/10.1038/nrneurol.2014.207. PubMed PMID: PMCID: PMC2630974 25385337; PubMed Central PMCID: PMCPMC4395497 25. Le Blon D, Hoornaert C, Daans J, Santermans E, Hens N, Goossens H, et al. 8. Doherty TM, Kastelein R, Menon S, Andrade S, Coffman RL. Modulation of Distinct spatial distribution of microglia and macrophages following murine macrophage function by IL-13. J Immunol. 1993;151(12):7151–60. mesenchymal stem cell implantation in mouse brain. Immunol Cell Biol. 2014; PubMed PMID: 7903102 92(8):650–8. https://doi.org/10.1038/icb.2014.49. PubMed PMID: 24983456 9. Doyle AG, Herbein G, Montaner LJ, Minty AJ, Caput D, Ferrara P, et al. 26. Reekmans K, De Vocht N, Praet J, Le Blon D, Hoornaert C, Daans J, et al. Interleukin-13 alters the activation state of murine macrophages in vitro: Quantitative evaluation of stem cell grafting in the central nervous system comparison with interleukin-4 and interferon-gamma. Eur J Immunol. 1994; of mice by in vivo bioluminescence imaging and postmortem multicolor 24(6):1441–5. https://doi.org/10.1002/eji.1830240630. PubMed PMID: 7911424 histological analysis. Methods Mol Biol. 2013;1052:125–41. https://doi.org/10. 10. Offner H, Subramanian S, Wang C, Afentoulis M, Vandenbark AA, Huan J, 1007/7651_2013_17. PubMed PMID: 23733539 et al. Treatment of passive experimental autoimmune encephalomyelitis in 27. De Vocht N, Reekmans K, Bergwerf I, Praet J, Hoornaert C, Le Blon D, et al. SJL mice with a recombinant TCR ligand induces IL-13 and prevents axonal Multimodal imaging of stem cell implantation in the central nervous system injury. J Immunol. 2005;175(6):4103–11. PubMed PMID: 16148160 of mice. J Vis Exp. 2012;64:e3906. https://doi.org/10.3791/3906. PubMed 11. Ochoa-Reparaz J, Rynda A, Ascon MA, Yang X, Kochetkova I, Riccardi C, et al. PMID: 22733218; PubMed Central PMCID: PMC3471290 IL-13 production by regulatory T cells protects against experimental 28. Praet J, Santermans E, Reekmans K, de Vocht N, Le Blon D, Hoornaert C, autoimmune encephalomyelitis independently of autoantigen. J Immunol. et al. Histological characterization and quantification of cellular events 2008;181(2):954–68. PubMed PMID: 18606647; PubMed Central PMCID: following neural and fibroblast(-like) stem cell grafting in healthy and PMC2599928 demyelinated CNS tissue. Methods Mol Biol. 2014;1213:265–83. https://doi. 12. Lee K, Majumdar MK, Buyaner D, Hendricks JK, Pittenger MF, Mosca JD. org/10.1007/978-1-4939-1453-1_22. PubMed PMID: 25173390 Human mesenchymal stem cells maintain transgene expression during 29. Bahmani P, Schellenberger E, Klohs J, Steinbrink J, Cordell R, Zille M, et al. expansion and differentiation. Mol Ther. 2001;3(6):857–66. https://doi.org/10. Visualization of cell death in mice with focal cerebral ischemia using 1006/mthe.2001.0327. PubMed PMID: 11407899 fluorescent annexin A5, propidium iodide, and TUNEL staining. J Cereb Blood 13. Ozawa K, Sato K, Oh I, Ozaki K, Uchibori R, Obara Y, et al. Cell and gene Flow Metab. 2011;31(5):1311–20. https://doi.org/10.1038/jcbfm.2010.233. therapy using mesenchymal stem cells (MSCs). J Autoimmun. 2008;30(3): PubMed PMID: 21245871; PubMed Central PMCID: PMC3099638 121–7. https://doi.org/10.1016/j.jaut.2007.12.008. PubMed PMID: 18249090 Hamzei Taj et al. Journal of Neuroinflammation (2018) 15:174 Page 17 of 17 30. Adamczak JM, Schneider G, Nelles M, Que I, Suidgeest E, van der Weerd L, 46. Yamasaki R, Lu H, Butovsky O, Ohno N, Rietsch AM, Cialic R, et al. Differential et al. In vivo bioluminescence imaging of vascular remodeling after stroke. roles of microglia and monocytes in the inflamed central nervous system. Front Cell Neurosci. 2014;8:274. https://doi.org/10.3389/fncel.2014.00274. J Exp Med. 2014;211(8):1533–49. https://doi.org/10.1084/jem.20132477. PubMed PMID: 25249937; PubMed Central PMCID: PMC4155794 PubMed PMID: 25002752; PubMed Central PMCID: PMC4113947 31. Chen J, Sanberg PR, Li Y, Wang L, Lu M, Willing AE, et al. Intravenous 47. Estevez AG, Sahawneh MA, Lange PS, Bae N, Egea M, Ratan RR. Arginase 1 administration of human umbilical cord blood reduces behavioral deficits regulation of nitric oxide production is key to survival of trophic factor- after stroke in rats. Stroke. 2001;32(11):2682–8. PubMed PMID: 11692034 deprived motor neurons. J Neurosci. 2006;26(33):8512–6. https://doi.org/10. 1523/JNEUROSCI.0728-06.2006. PubMed PMID: 16914676; PubMed Central 32. Hamzei Taj S, Kho W, Aswendt M, Collmann FM, Green C, Adamczak J, et al. PMCID: PMC2570095 Dynamic modulation of microglia/macrophage polarization by miR-124 48. Wang G, Zhang J, Hu X, Zhang L, Mao L, Jiang X, et al. Microglia/ after focal cerebral ischemia. J NeuroImmune Pharmacol. 2016;11(4):733–48. macrophage polarization dynamics in white matter after traumatic brain https://doi.org/10.1007/s11481-016-9700-y. PubMed PMID: 27539642; injury. J Cereb Blood Flow Metab. 2013;33(12):1864–74. https://doi.org/10. PubMed Central PMCID: PMC5097787 1038/jcbfm.2013.146. PubMed PMID: 23942366; PubMed Central PMCID: 33. Dooley D, Lemmens E, Vangansewinkel T, Le Blon D, Hoornaert C, Ponsaerts P, PMC3851898 et al. Cell-based delivery of Interleukin-13 directs alternative activation of 49. Hamzei Taj S, Kho W, Riou A, Wiedermann D, Hoehn M. MiRNA-124 induces macrophages resulting in improved functional outcome after spinal cord injury. neuroprotection and functional improvement after focal cerebral ischemia. Stem Cell Rep. 2016;7(6):1099–115. https://doi.org/10.1016/j.stemcr.2016.11.005. Biomaterials. 2016;91:151–65. https://doi.org/10.1016/j.biomaterials.2016.03. PubMed PMID: 27974221 025. PubMed PMID: 27031810 34. Ali I, Aertgeerts S, Le Blon D, Bertoglio D, Hoornaert C, Ponsaerts P, et al. 50. Frank MG, Barrientos RM, Biedenkapp JC, Rudy JW, Watkins LR, Maier SF. Intracerebral delivery of the M2 polarizing cytokine interleukin 13 using mRNA up-regulation of MHC II and pivotal pro-inflammatory genes in mesenchymal stem cell implants in a model of temporal lobe epilepsy in normal brain aging. Neurobiol Aging. 2006;27(5):717–22. https://doi.org/10. mice. Epilepsia. 2017;58:1063–72. 1016/j.neurobiolaging.2005.03.013. PubMed PMID: 15890435 35. Hoornaert CJ, Luyckx E, Reekmans K, Dhainaut M, Guglielmetti C, Le Blon D, 51. Arnett HA, Wang Y, Matsushima GK, Suzuki K, Ting JP. Functional genomic et al. In vivo interleukin-13-primed macrophages contribute to reduced analysis of remyelination reveals importance of inflammation in alloantigen-specific T cell activation and prolong immunological survival of oligodendrocyte regeneration. J Neurosci. 2003;23(30):9824–32. PubMed allogeneic mesenchymal stem cell implants. Stem Cells. 2016;34(7):1971–84. PMID: 14586011 https://doi.org/10.1002/stem.2360. PubMed PMID: 26992046 52. Ohtaki H, Yin L, Nakamachi T, Dohi K, Kudo Y, Makino R, et al. Expression of 36. Tanaka R, Komine-Kobayashi M, Mochizuki H, Yamada M, Furuya T, Migita M, tumor necrosis factor alpha in nerve fibers and oligodendrocytes after et al. Migration of enhanced green fluorescent protein expressing bone transient focal ischemia in mice. Neurosci Lett. 2004;368(2):162–6. https:// marrow-derived microglia/macrophage into the mouse brain following doi.org/10.1016/j.neulet.2004.07.016. PubMed PMID: 15351441 permanent focal ischemia. Neuroscience. 2003;117(3):531–9. PubMed PMID: 53. Kuric E, Ruscher K. Dynamics of major histocompatibility complex class II- positive cells in the postischemic brain—influence of levodopa treatment. 37. Schilling M, Besselmann M, Leonhard C, Mueller M, Ringelstein EB, Kiefer R. J Neuroinflammation. 2014;11:145. https://doi.org/10.1186/s12974-014-0145-z. Microglial activation precedes and predominates over macrophage PubMed PMID: 25178113; PubMed Central PMCID: PMC4149192 infiltration in transient focal cerebral ischemia: a study in green fluorescent 54. Stellwagen D, Malenka RC. Synaptic scaling mediated by glial TNF-alpha. protein transgenic bone marrow chimeric mice. Exp Neurol. 2003;183(1):25– Nature. 2006;440(7087):1054–9. https://doi.org/10.1038/nature04671. 33. PubMed PMID: 12957485 PubMed PMID: 16547515 38. Le Blon D, Hoornaert C, Detrez JR, Bevers S, Daans J, Goossens H, et al. 55. Hanania R, Sun HS, Xu K, Pustylnik S, Jeganathan S, Harrison RE. Classically Immune remodelling of stromal cell grafts in the central nervous system: activated macrophages use stable microtubules for matrix therapeutic inflammation or (harmless) side-effect? J Tissue Eng Regen Med. metalloproteinase-9 (MMP-9) secretion. J Biol Chem. 2012;287(11):8468–83. 2016; https://doi.org/10.1002/term.2188. https://doi.org/10.1074/jbc.M111.290676. PubMed PMID: 22270361; PubMed 39. Garcia-Bonilla L, Faraco G, Moore J, Murphy M, Racchumi G, Srinivasan J, Central PMCID: PMC3318683 et al. Spatio-temporal profile, phenotypic diversity, and fate of recruited 56. Komohara Y, Ohnishi K, Kuratsu J, Takeya M. Possible involvement of the M2 monocytes into the post-ischemic brain. J Neuroinflammation. 2016;13(1): anti-inflammatory macrophage phenotype in growth of human gliomas. J 285. https://doi.org/10.1186/s12974-016-0750-0. PubMed PMID: 27814740; Pathol. 2008;216(1):15–24. https://doi.org/10.1002/path.2370. PubMed PMID: PubMed Central PMCID: PMC5097435 40. Wattananit S, Tornero D, Graubardt N, Memanishvili T, Monni E, Tatarishvili J, 57. Yu B, Sondag GR, Malcuit C, Kim MH, Safadi FF. Macrophage-associated et al. Monocyte-derived macrophages contribute to spontaneous long-term osteoactivin/GPNMB mediates mesenchymal stem cell survival, proliferation, functional recovery after stroke in mice. J Neurosci. 2016;36(15):4182–95. and migration via a CD44-dependent mechanism. J Cell Biochem. 2016; https://doi.org/10.1523/JNEUROSCI.4317-15.2016. PubMed PMID: 27076418 117(7):1511–21. https://doi.org/10.1002/jcb.25394. PubMed PMID: 26442636 41. Evans TA, Barkauskas DS, Myers JT, Hare EG, You JQ, Ransohoff RM, et al. 58. Xia W, Xie C, Jiang M, Hou M. Improved survival of mesenchymal stem cells High-resolution intravital imaging reveals that blood-derived macrophages by macrophage migration inhibitory factor. Mol Cell Biochem. 2015;404(1–2): but not resident microglia facilitate secondary axonal dieback in traumatic 11–24. https://doi.org/10.1007/s11010-015-2361-y. PubMed PMID: 25701358; spinal cord injury. Exp Neurol. 2014;254:109–20. https://doi.org/10.1016/j. PubMed Central PMCID: PMC4544672 expneurol.2014.01.013. PubMed PMID: 24468477; PubMed Central PMCID: PMC3954731 42. Darsalia V, Kallur T, Kokaia Z. Survival, migration and neuronal differentiation of human fetal striatal and cortical neural stem cells grafted in stroke-damaged rat striatum. Eur J Neurosci. 2007;26(3):605–14. https://doi.org/10.1111/j.1460-9568.2007.05702.x. PubMed PMID: 17686040 43. Janata A, Magnet IA, Uray T, Stezoski JP, Janesko-Feldman K, Tisherman SA, et al. Regional TNFalpha mapping in the brain reveals the striatum as a neuroinflammatory target after ventricular fibrillation cardiac arrest in rats. Resuscitation. 2014;85(5):694–701. https://doi.org/10.1016/j.resuscitation.2014. 01.033. PubMed PMID: 24530249; PubMed Central PMCID: PMC4034695 44. Shechter R, Miller O, Yovel G, Rosenzweig N, London A, Ruckh J, et al. Recruitment of beneficial M2 macrophages to injured spinal cord is orchestrated by remote brain choroid plexus. Immunity. 2013;38(3):555–69. https://doi.org/10. 1016/j.immuni.2013.02.012. PubMed PMID: WOS:000330941500019 45. Geissmann F, Jung S, Littman DR. Blood monocytes consist of two principal subsets with distinct migratory properties. Immunity. 2003;19(1):71–82. PubMed PMID: 12871640

Journal

Journal of NeuroinflammationSpringer Journals

Published: Jun 4, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off