CCL11 promotes migration and proliferation of mouse neural progenitor cells

Neonatal hypoxia-ischemia induces massive brain damage during the perinatal period, resulting in long-term consequences to central nervous system structural and functional maturation. Although neural progenitor cells (NPCs) migrate through the parenchyma and home in to injury sites in the rodent brain, the molecular mechanisms are unknown. We examined the role of chemokines in mediating NPC migration after neonatal hypoxic-ischemic brain injury. Nine-day-old mice were exposed to a 120-minute hypoxia following unilateral carotid occlusion. Chemokine levels were quantified in mouse brain extract. Migration and proliferation assays were performed using embryonic and infant mouse NPCs. The neonatal hypoxic-ischemic brain injury resulted in an ipsilateral lesion, which was extended to the cortical and striatal areas. NPCs migrated toward an injured area, where a marked increase of CC chemokines was detected. In vitro studies showed that incubation of NPCs with recombinant mouse CCL11 promoted migration and proliferation. These effects were partly inhibited by a CCR3 antagonist, SB297006. Our data implicate an important effect of CCL11 for mouse NPCs. The effective activation of NPCs may offer a promising strategy for neuroregeneration in neonatal hypoxic-ischemic brain injury.


Background
Perinatal hypoxia and ischemia can cause serious complications. In neonatal animal models, accumulating evidence suggests that hypoxic-ischemic brain damage influences the activation of endogenous neural progenitor cells (NPCs) in the subventricular zone (SVZ) [1]. Increased neurogenesis and migration of NPCs toward the injury site has been observed in animal models of epilepsy, stroke, trauma, Alzheimer's disease, Parkinson's disease, and Huntington's disease [2]. Recent studies indicate that neurogenesis may play regenerative roles in response to central nervous system (CNS) injuries. Neurogenesis in the immature mammalian brain is highly modulated by excitotoxic insults [3,4]. Moreover, NPCs have been shown to be induced in the post-stroke human cerebral cortex for more than 1 month based on immunostaining of autopsied tissue from stroke patients [5,6]. Although the origin of these NPCs is uncertain, this raises the possibility of developing therapeutic strategies aimed at activating the neurogenic capacity in order to repair the damaged brain.
Nevertheless, the molecular mechanisms for NPC activation and mobilization have not been clearly identified. It has been demonstrated in rodents that repulsive Slit-Robo signal orientated the migration of newly generated immature neurons through the rostral migratory stream (RMS) under normal conditions. RMS astrocytes express Robo receptors and regulate the rapid migration of Slit1-expressing new neurons [7]. Fibroblast growth factor 2 (FGF-2) and anosmin-1 increase SVZ NPC migration via epidermal growth factor receptor 1 (EGFR1) during pre-and postnatal rat development [8]. In contrast, in the injured brain, NPCs from the SVZ migrate to injury sites through blood vessels for up to 1 year after injury [9,10]. This directed NPC migration is considered the result of chemoattractive cues expressed from the injury site. Inflammatory mediators contribute to the pathological changes, which are sensitive to hypoxic-ischemic brain injury [11].
We hypothesized that inflammatory mediators were involved in the migration of NPCs toward the injury site. The major role of chemokines is to act as a chemoattractant to guide the migration of immune cells. The CC chemokines induce recruitment of eosinophils, basophils, neutrophils, and monocytes, and play important roles in the regulation of many inflammatory conditions. In this study, we demonstrated that CCL11, one of CC chemokines highly expressed at brain injury sites, promoted the migration and proliferation of NPCs. These findings suggest that CCL11 may have important functions in the neonatal CNS.

In vivo study Neonatal ischemia-reperfusion brain injury
The NOD/SCID (NOD.CB17-Prkdc scid /J) mice were purchased from Charles River Laboratories Japan Inc. (Kanagawa, Japan). Experimental mouse models of neonatal ischemia-reperfusion brain injury were used based on a modified Rice-Vannucci model as previously described [12]. Briefly, 9-day-old postnatal NOD/SCID mice of both sexes (n = 48) were anesthetized with 2% isoflurane. The right common carotid artery (CCA) was occluded with an aneurysm clip (Mizuho, Tokyo, Japan). The pups were placed in a hypoxia chamber held at 8% O 2 for 120 minutes. Reperfusion was achieved by unclamping the artery and exposing the pups to normoxic conditions. Pups were returned to their dams after the procedure. Sham-operated mice (n = 8) were subjected to CCA isolation alone and no brain injury. NOD/SCID mice aged 1, 2, 4, 6, and 8 weeks were used as normal control (n = 30). All animal experiments were performed under an institutionally approved protocol for the use of animal research at Kochi University (approval number: J-00002).

Magnetic resonance imaging
Magnetic resonance imaging (MRI) measurements were performed on a Varian 9.4 T vertical MR system with console Vnmr interface (Agilent Technologies, Santa Clara, CA, USA). The mice were anesthetized with 2% isoflurane and placed in a vertical animal positioning unit. The MRI sequence used was a T2-weighted Spin Echo Multi Slice (SEMS) imaging sequence (T2WI) with the following parameters: repetition time (TR), 2000 ms; echo time (TE), 10 ms; 1 mm thick slices; number of scans, 9; field of view (FOV), 2 × 2 cm; and matrix size, 256 × 128. The total imaging time for each sequence was 16 minutes.

Immunohistochemistry
At each follow-up time point, the brains were removed and fixed in 4% paraformaldehyde. Coronal brain sections (12 μm) were prepared using a microtome (Leica Microsystems Inc., Wetzlar, Germany). The primary and secondary antibodies used in the experiments were: rabbit anti-Doublecortin

Chemokine quantification
The brain tissue of mice were collected at each follow-up time point. The lysate was prepared by homogenization in RIPA Lysis and Extraction Buffer (Thermo Fisher Scientific) containing a complete protease inhibitor cocktail (Roche, Basel, Switzerland). The tissue debris was removed by centrifugation at 1000 × g for 10 minutes at 4°C. Protein concentration was determined with BCA Protein Assay Kit (Thermo Fisher Scientific).
Chemokine levels in RIPA-extracted lysates were quantified using CCL9 ELISA Kit (Sigma-Aldrich, St. Louis, MO, USA), CCL12 Quantikine ELISA Kit (R&D Systems, Minneapolis, MN, USA), and LEGENDplex Multi-Analyte Flow Assay Kit for Mouse Proinflammatory Chemokine (BioLegend, San Diego, CA, USA) according to the manufacturer's instructions. Data were normalized to the expression level in the 1 mg/ml of total protein of the lysates. The relative detection threshold of those assays was 0.2 pg/ml except the CCL12 assay, which was 1.5 pg/ml.

Neutralizing assay
A neutralizing rat monoclonal CCL11 antibody (R&D Systems) at a dose of 25 μg/mouse was administrated twice at 6 and 24 hours after brain injury. At 1 week after injury, the brains were removed and fixed in 4% paraformaldehyde. The primary and secondary antibodies used in the experiments were: rabbit anti-Dcx antibody (1:1000, Abcam) and Alexa Fluor™ 488 goat anti-rabbit IgG (1:1000, Thermo Fisher Scientific). The slides were observed using a fluorescence microscope (BZ-9000; Keyence).

In vitro study
Isolation and culture neural stem/progenitor cells Embryonic NPCs were isolated from pregnant NOD/SCID mice (n = 3) on E13.5 day of gestation. Striata were dissected from each embryo in phosphate-buffered saline (PBS) containing 0.6% glucose (Sigma-Aldrich) and penicillin-streptomycin (50 U/ml; Life Technologies). After dissection, the tissue was triturated to a single cell suspension using a transfer pipette. Cells were cultured in NeuroCult™ Proliferation Medium containing 20 ng/ml epidermal growth factor (EGF) (Stemcell Technologies Inc., Vancouver, BC, Canada). NPCs from infant NOD/ SCID mice were isolated from the ipsilateral and contralateral SVZ at 1 week post-injury (n = 5). The SVZ tissue was triturated with neural tissue dissociation solution (Sumitomo Dainippon Pharma Co., Ltd, Tokyo, Japan). Cells were cultured in NeuroCult™ Proliferation Medium containing 20 ng/ml EGF, 10 ng/ml fibroblast growth factor (FGF) and 2 μl/ml heparin (Stemcell Technologies Inc.). Cells were proliferated to generate neurospheres.

RT-PCR
Total RNA was isolated from neurospheres using an RNeasy Mini kit (Qiagen, Hilden, Germany) according to the manufacturer's instruction. RT-PCR was performed using PrimeScript RT reagent kit (Takara Bio Inc., Otsu, Japan) for reverse transcription, and Ex Taq (Takara Bio Inc.) for PCR amplification. Details of primer sequences are shown in Table 1.

Migration assay
Carrier-free forms of mouse recombinant CCL2, CCL3, CCL4, CCL5, CCL9, CCL11, CCL12, CCL20, and CCL22 were purchased from BioLegend (San Diego, CA, USA). Assays utilized a poly-D-lysine coating μ-slide Chemotaxis 3D (ibidi GmbH, Munich, Germany). 3 × 10 6 cells/ml were seeded onto the μ-slides and incubated at 37°C. Migration was stimulated by addition of the chemokine (20 μg/ml) dissolved in PBS to the indicated port. Timelapse imaging was performed using BioStation IM-Q (Nikon, Tokyo, Japan) with a × 20 objective. Images were captured every 15 minutes for 24 hours. Twelve cells were randomly chosen in each chemokine assay for data analysis. The migration trajectory and distance were measured with ImageJ using the Manual Tracking plugin for NIH ImageJ software (Bethesda, MD, USA).

Cell culture assay
Embryonic and infant NPCs were plated onto 96-well plates at a cell density of 1 × 10 4 cells/well. Each chemokine (2 μg/ml) was added to cell cultures. The culture medium used was NeuroCult™ Proliferation Medium without EGF, FGF, and heparin. After 1 and 3 days of culture, 10 μl of CCK-8 (Dojindo, Kumamoto, Japan) solution was added in triplicate to wells, and incubated for 2 hours at 37°C. Absorbance at 450 nm was then measured in a microplate reader (Thermo Fisher Scientific).

Statistical analysis
The data are presented as means ± S.D. Results from different groups were compared using the Student's t test, or one-way ANOVA followed by Dunnett's multiple comparison tests. P < 0.05 was considered to indicate a statistically significant difference.

Results
NPCs migration in the mouse brain The experimental design is showed in Fig. 1a. The mouse model of neonatal ischemia-reperfusion injury was induced at post-natal day 9. MRI was performed 1 week after the brain injury. T2 hyperintensity abnormalities reflecting tissue edema were observed in the cortex and striatum of the mouse brain ( Fig. 1b). We evaluated the migration of NPCs (Dcxpositive cells) in the neonatal mice brain (Fig. 1c). In the injured SVZ side, the number of Dcx-positive cells began to increase rapidly at 1 week, and was stable thereafter for up to 7 weeks following the brain injury (Fig. 1c, f ). At 5 weeks post-injury, Dcx-positive cells had migrated a long distance and had reached the core injury area (Fig. 1d). In contrast, in the intact SVZ side, the number of Dcx-positive cells was diminished with advancing age (Fig. 1e). The number of PSA-NCAM-positive cells was also significantly increased in the injured side (Additional file 2: Figure S2a, b). In the sham-operated mice, there was no difference in the number of Dcx-positive cells between the intact side and the injury side (Additional file 2: Figure S2c, d).
Chemokine expression in the mouse brain Chemokine levels were quantified using ELISA Kit and Multi-Analyte Flow Assay Kit for Mouse Proinflammatory Chemokine. We focused on the expression profile of nine CC chemokines upregulated in the injured mouse brain: CCL2, CCL3, CCL4, CCL5, CCL9, CCL11, CCL12, CCL20, and CCL22. At 24 hours post-injury, most chemokines were highly expressed even in the intact side of the injured brain compared with the undamaged brain (CNT). At 1 week post-injury, CCL2, CCL3, CCL4, CCL5, CCL11, CCL12, CCL20, and CCL22 were significantly upregulated in the injury side, although there was little difference between the injury and intact side in most chemokines at 3 weeks post-injury. Only CCL9 expression was maintained at high levels in both sides of brain up to 3 weeks post-injury (Fig. 2).

In vitro migration assay
Neurospheres were generated from the embryo and infant mice brains, and these showed high expression of sex-determining region Y-box 2 (SOX2), Ki-67, and no expression of the mature neuronal marker, NeuN (Fig. 3a). NPCs cultured on poly-D-lysine expressed the neuroblast maker, PSA-NCAM (PSA-NCAM + cells: 68.7 ± 5.2%, PSA-NCAMcells: 30.2 ± 2.9%) and Dcx (Fig. 3b, c). We then evaluated the effect of chemokines on NPC migration. Cell tracking with CCL11 over 24 hours of monitoring is shown in Fig. 3d. The minus (leftward) direction of cell movement was defined as chemokine-induced migration (Fig. 3e). Each color represents the trajectory of an individual cell. NPCs actively proliferated and migrated toward the CCL11 injected side (Fig. 3e, Additional file 1: Figure S1). When migration distance and end points of embryonic NPCs between each chemokine were compared, CCL11 significantly promoted embryonic NPCs migration compared with PBS or other chemokines (Fig. 3g, PBS: -0.77 ± 4.3 μm; CCL11: -65.9 ± 11.6 μm). Although CCL11 also promoted migration of NPCs derived from the brain-injured infant mice, the end points of migration were decreased in the infant NPCs compared to embryonic NPCs. There was no significant difference between the injury side and intact side of NPCs from infant brain (Fig. 3f, h).

Cell culture assay
Proliferation of embryonic NPCs was enhanced after incubation with recombinant mouse CCL11 (Fig. 4a). Moreover, the proliferative effect of CCL11 was concentration-dependent. Only NPCs incubated with  CCL11 formed neurospheres, and other CC chemokines neither induced neurosphere formation nor enhanced NPC proliferation (Fig. 4b, c). CCL11 also enhanced the proliferation of infant NPCs. There was no significant difference in proliferation activity between NPCs derived from the injury side and those from the intact side of the infant mouse brain (Fig. 4d).

Receptor blocking assay
The major receptor for CCL11 is CCR3, which is found on leukocytes [13]. We then analyzed whether chemokine/receptor signaling influenced the proliferation of NPCs. NPCs expressed a functional chemokine receptor CCR3, and CCR3 receptor antagonist, SB297006 significantly inhibited proliferation and neurosphere formation in CCL11-treated NPCs (Fig. 4e-h).

In vivo neutralizing assay
A neutralizing CCL11 antibody was administrated in the model mice at 6 and 24 hours after injury. There was no significant difference of NPC migration and proliferation between the antibody-treated group and the PBS-treated group (Fig. 5a, b).

NPCs migration and chemokines in an injured brain
A number of studies have shown activation and recruitment of NPCs in animal models of ischemic brain injury [14][15][16]. The number of NPCs correlated with the volume of brain injury in the mouse model of stroke [17]. While low oxygen increased the proliferation of neural stem cells in vitro [18], oxygen glucose deprivation (OGD) decreased the growth of NPCs and also decreased the phosphorylation of extracellular signal regulated kinase (ERK) [19]. The hypoxic-ischemic condition depletes the ATP content of the cells, compromising their metabolic activity. Various inflammatory mediators (cytokines, chemokines, and adhesion molecules) are produced from immune cells, endothelial cells, fibroblasts, and brain cells after ischemic injury. Directed NPC migration is considered to be the result of chemoattractive cues expressed from the injury site. Our chemokine quantification data demonstrated that CC chemokines were markedly elevated at 24 hours after injury, and were gradually reduced over 3 weeks post-injury in the infant mouse brain. Peculiarly, CCL2 and CCL11 were significantly upregulated in the injury side compared with the intact side up to 3 weeks post-injury. Chemokines are 8-14 kDa small molecules mainly regulating immune cells trafficking during inflammatory responses. Chemokine-mediated signaling leads to cytoskeletal rearrangements that allow cell polarization toward the chemokine gradient. In recent years, chemokines have been reported to have nonimmunological effects in the CNS, including regulation of cell proliferation, migration, survival, and synaptic activity [20]. Adult NPCs have been shown to express chemokine receptors, including CCR1-8, 10, and CXCR1-6 [21]. Among chemokine receptors, CXCL12/CXCR4 signaling is important for neural progenitor cell guidance and orientation in the developing mammalian brain [22,23]. Following neural damage, CXCL12/CXCR4 signaling mediates specific migration of NPCs to the ischemic damaged site [24]. Several signaling pathways may mediate CXCL12-induced NPC migration, including inositol 1,4,5-triphosphate, extracellular signal-regulated kinases 1/2, Akt, c-Jun N-terminal kinase, and intracellular calcium [25]. CCL2, CCL3, and CXCL1 also promote NPC migration to the striatum following quinolinic acidinduced lesion [26]. IFN-γ and IFN-β both inhibit cultured adult NPCs proliferation, but only IFN-γ promotes neuronal differentiation [27]. In the present study, CCL2 and CCL3 were upregulated in the injury side in the hypoxicischemic brain of neonatal mice. The migration assay showed that CCL2 and CCL3 slightly promoted NPC migration. Since NPCs are placed into the gap of μ − slide Chemotaxis 3D and exposed to linear concentration gradients, the concentration of chemokines in the NPC area might be quite diluted. In consideration of a local inflammatory response, the induction effect may vary with the concentration of each chemokine in the injured mouse brain. The results of in vivo neutralizing assay suggests that the activation and migration of NPCs seems to be controlled by an orchestration of cytokines include interleukins, interferons, growth factors and chemokines in the brain-injured mice.
In another study, induction of striatal neurogenesis by the intraventricular administration of brain-derived neurotrophic factor (BDNF) and EGF promoted functional recovery in a mouse model of neonatal hypoxic-ischemic brain injury. In contrast, this BDNF/EGF-associated functional recovery was abolished in mice receiving a co-infusion of 2% cytosine-b-D-arabinofuranoside (Ara-C), a mitotic inhibitor. These results indicate that the effect of functional recovery may be the result of newly generated neurons [28]. Identification of more factors responsible for recruitment of NPCs to the injury site may lead to the development of therapies designed to promote functional recovery after neonatal ischemia-reperfusion brain injury.

Role of CCL11 in CNS
Whereas chemokine signaling in immune cells has been studied in detail, the molecular mechanisms of chemokine-induced NPC activation are not completely understood. This study has shown that a chemokine, CCL11, could activate NPCs in vitro. The CCL11 activity for NPCs was exerted under EGF-and FGF-free conditions. CCL11 is traditionally associated with eosinophil recruitment and pro-inflammatory responses [29,30]. The mechanism of how CCL11 affects the brain has not been determined. Recently, increased blood levels of CCL11 in aging have been shown to negatively regulate adult hippocampal neurogenesis. Systemic CCL11 administration decreased the numbers of Dcx-positive cells in the dentate gyrus of young mice. Young mice that received CCL11 exhibited impaired learning and memory deficits [31]. In aged mice, plasma CCL11 protein levels were significantly increased in 18-month-old mice than in 21-month-old mice [32]. In the mice model of experimental cerebral malaria, CCL11 was elevated in the hippocampus and frontal cortex. Dcx-positive cells in the dentate gyrus were decreased in parallel with increased CCL11 [33]. Furthermore, CCL11 promoted the migration of microglia, and induced microglial production of reactive oxygen species. CCL11 triggers oxidative stress via microglial activation and potentiates glutamate-mediated neurotoxicity, which may be involved in the pathogenesis of various neurological disorders [34]. In one physiological study, the striatum had an early rapid-uptake phase for CCL11, which was the fastest among different brain regions [35]. CCL11 in the circulation can cross many regions of the blood-brain barrier (BBB). This suggests that blood-borne CCL11 may have important physiological functions in the CNS and implicates the BBB as an important regulator of the physiological versus pathological effects of this chemokine. Our results may reflect opposite effects of CCL11 on NPCs, as recently reported in previous studies. In our study, the proliferation of NPCs was promoted at a much higher concentration (1-5 μg/ml) of CCL11 compared to Villeda's study (10 ng/ml) [31]. Moreover, CCL11 might have differential effects in SVZ NPCs compared to hippocampal NPCs.
Meanwhile, positive effects of CCL11 have also been reported in several studies. Adzemovic et al. demonstrated that the CCL11/CCR3 pathway was associated with immune response modulation in a rat model of multiple sclerosis (MS). The milder EAE phenotype in the Eae18b congenic strain of rat was accompanied by significantly lower myelin loss and reduced cumulative microglia activation. The congenic rats significantly upregulated CCL11 mRNA in lymph nodes and spinal cord tissue [36]. Moreover, CCL11/CCR3 promotes the proliferation of oligodendrocyte precursor cells (OPCs) and migration of smooth muscle cells (SMCs) [37,38]. Both of these cells are involved in the regeneration of damaged tissue. In demyelinating diseases such as MS, remyelination is preceded by the division of endogenous NG2-positive OPCs. The OPCs migrate to sites of injury and differentiate into mature oligodendrocytes [39,40]. In atherosclerosis, SMCs proliferate and migrate to form part of the intimal plaque [41]. Furukawa et al. showed that IL-10, an antiinflammatory cytokine elevation was correlated with milder symptoms and that CCL11 elevation was correlated with slower disease progression in patients with amyotrophic lateral sclerosis (ALS), suggesting that those cytokines may confer neuroprotection against ALS [42]. Therefore, CCL11 may have both beneficial and harmful effects in the CNS that depend on factors such as concentration, anatomical location, and pathological condition.

Conclusions
We have investigated the mechanisms for NPC activation in a mouse model of neonatal ischemia-reperfusion brain injury. Our main findings are that CCL11 affects the migration and proliferation of NPCs. Further studies are needed to delineate the regulatory pathways and downstream signaling. The activation of NPCs may offer a promising strategy for neuroregeneration in neonatal hypoxic-ischemic brain injury and general neurological disorders as well.

Additional files
Additional file 1: Figure S1. The time-lapse movie of NPC migration. Images were captured every 15 minutes for 24 hours. (A) NPCs did not migrate toward the PBS-injected side. (B) After CCL11 injection, NPCs actively proliferated and migrated toward the CCL11-injected side. (ZIP 144651 kb)