Conditioned Medium From M1 Macrophages Induces the Migration of Neural Stem Cells Through MCP-1/CCR2 and ERK Pathway


 After spinal cord injury, grafted neural stem cells (NSCs) can migrate toward injured area, where there are a large number of bone marrow derived macrophage. However, little is known about the effect of M1 macrophage from bone marrow on NSCs migration and the mechanism responsible for migratory responses of grafted NSCs in vitro and in vivo. Migration of NSCs were conducted by using the transwell chamber and SCI model. Conditioned medium from M1 macrophages (M1-CM) can attract NSCs migration in vitro. The number of migrated C-C chemokine receptor 2 (CCR2) -/- NSCs induced by M1-CM were significantly decreased compared with wild type (WT) NSCs (P<0.05). Furthermore, compared with M0 macrophages, the production of MCP-1 by M1 macrophages was significantly increased and M1-CM treated with anti-MCP-1 antibody to neutralize MCP-1 can lead to a significant reduction of NSCs migration (P<0.05). In addition, Western blot showed that ERK1/2 was dramatically activated following the stimuli of M1-CM and then ERK inhibitor can also inhibit M1-CM-mediated MSCs migration. Finally, CCR2 was needed for grafted NSCs migration toward injured area. Bone marrow derived M1 macrophage persisted within the epicenter of injured area and the expression of MCP-1 mRNA was significantly increased 7days after SCI (P<0.05). These results demonstrated the effect of M1 macrophages on NSCs migration and the important role of MCP-1/CCR2 and ERK signal pathway on M1-CM-induced NSCs migration.


Introduction
Neural stem cells (NSCs) are de ned by their ability to self-renew and to generate the main phenotypes of central nervous system (CNS) cells, which are neurons, astrocytes, and oligodendrocytes [1,2].
Transplantation of NS/PCs is promising treatment for spinal cord injury (SCI) [3][4][5]. NSCs migration is an essential process for the development of the central nervous system (CNS). Recently, it's has also been showed that transplanted NSCs migrate to sites of pathological area such as various types of CNS diseases (i.e., spinal cord injury and ischemia) [6][7][8]. The migration capability is very important for endogenous or exogenous NSCs to repair the damaged tissue [9,10]. Identi cation of NSCs migration would not only aid a better understanding of the mechanisms responsible for injury-mediated NSCs migration but also have clinical implications for stem cell-based therapeutic strategies for spinal cord injury (SCI).
Spinal cord injury (SCI) is a traumatic event that involves not just an acute physical injury, but also in ammation driven secondary injury [11]. Macrophages play a very important role in secondary injury.
After SCI, blood monocytes are recruited by primary chemokines and cytokines to injured area 2-3days post-injury and bone-marrow derived macrophages accumulate in the injury site about 1week after SCI [12]. Regardless of the cause of SCI, there are a lot of macrophages persisting at injury site inde nitely. Kigerl et al reported that the rapid response and maintenance of M1 and the transient existence of M2 at the injured spinal cord area leaded to the second damage and regulate the differentiation of NSCs [13]. M1 macrophages could regulate the differentiation and proliferation of neural stem/progenitor cells, and may be relative to migration of NSCs.
Although little is known of the mechanisms responsible for the injury-mediated NSPC migration, there are a variety of cytokines and growth factors, such as monocyte chemoattractant protein (MCP-1), CXC chemokine ligand (CXCL)12, stromal cell-derived factor-1 (SDF-1), broblast growth factor (FGF), and epidermal growth factor (EGF), have been shown to in uence NSCs migration [14][15][16]. Chemokine C-C motif receptor 2 (CCR2) is a G protein-coupled receptor. Previous studies showed NSCs express CCR2 and play a crucial role on NSCs migration. MCP-1 activates the migration capacity of NSCs in vitro [17].
To date, little is known about the role of bone marrow derived M1 macrophages on the migration of NSCs after transplanted into injured spinal cord, and the precise molecular mechanisms involved in this process remain unclear. We hypothesized that M1 macrophage promotes migration of NSCs though MCP-1/CCR2 in vitro as well as in vivo. This study analyzed (1)

Materials And Methods
Animals CCR2-/-mice (B6.129S4-Ccr2tm1Ifc) and wild type C57BL/6 mice were purchased from Jackson Laboratory (Bar harbor, ME) and mice were housed in pathogen-free Animal Experiment Center of Xi'an Jiaotong University's college of medicine. All experimental procedures were approved by Xi'an Jiaotong University's Animals Care and Use Committee (ACUC) and performed in accordance with the guide for the care and use of laboratory animals.

Induction of M1 Macrophage and Conditioned medium preparation
To promote polarization into M1 macrophages, BMDMs were treated with 10ng/ml interferon-γ (IFNγ, Millipore, Billerica, MA, USA) for 24h [19,20]. M0 macrophages weren't treated with IFN-γ except for a change in the macrophage medium. After 24h, macrophages were washed three times with phosphatebuffered saline, changed with fresh medium including 2% B-27 supplement (Gibco) and then cultured for 24h. Conditioned medium (CM) from M0 macrophages and M1 macrophages were prepared by centrifuging collected medium at 2500RPM for 19min and ltering through a 0.4-μm lter (Corning). While, culture medium (without BMDMs) was used as control medium in this study.

Transwell Migration Assay of NSCs
Migration of NSCs were conducted by using the transwell chamber with a pore size of 8μm (Neuro probe) as described previously. After 8-μm pore membrane coated with 0.01% poly-l-lysine, 20μl of 5×10 6 /ml NSCs were added to upper wells of the transwell chamber. DMEM with 2% B-27 supplement (control medium), conditioned medium from M0 macrophages(M0-CM), conditioned medium from M1 macrophages (M1-CM) and other reagents dissolved in DMEM with 2% B-27 supplement were located at the lower wells of the chamber. After 12h, the number of cells, that migrated through the lters were xed and subjected to Giemsa staining, were analyzed by counting three independent eld of each condition.

MTT Assay
MTT (3-(4, 5-Dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide) assay was performed to identify whether M1-CM in uences the proliferation of NSCs. After treated with M1-CM, M0-CM or control medium for 12h, cells were incubated with MMT solution (5 mg/ml, Sigma-Aldrich, St.Louis, MO, USA) for 5h, and then added 200μl of DMSO to dissolve crystals. The absorbance was measured at 550nm using a microplate reader (Bio-Rad, Hercules, CA, USA) and relative proliferation rates were observed by comparing the optical density value of experimental samples to that of control sample.

Enzyme-linked immunosorbent assay (ELISA)
Quantitative measurement of MCP-1 was analyzed and determined using MCP1 ELISA kit (Abcam, Cambridge, MA, USA) following the manufacturer's instruction. The absorbance was measured at 450nm and MCP-1 concentration was observed by comparing absorbance to standard curves.

Western Blot
After treated with M1-CM, NSCs were washed with ice-cold PBS and lysed in RIPA buffer containing a phosphatase inhibitor and proteinase inhibitor cocktail (Amresco, Solon, OH, USA). Western blot assay was carried out following standard procedure. Brie y, total NSCs proteins were adjusted to equal protein concentrations, loaded onto SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to PVDF membranes. After blocking in 5% milk in Tris-buffered saline containing 0.1% Tween 20 (TBST) for 1 h at RT, membranes were incubated with primary antibodies solution overnight at 4°C. Membranes were then placed in appropriate secondary antibody for 1 h at room temperature after rinsing in TBST. Blots were visualized with the ECL plus western blot detection system (GE Healthcare, UK).

Phalloidin staining
The distribution and expression of F-actin was performed using CF488-conjugated phalloidin (89427-132, Biotium, Fremont, CA, USA) according to the manufacturer's protocol. Brie y, NSCs were treated with control medium and M1-CM, then washed three times with PBS and xed with 4% paraformaldehyde (PFA) in PBS for 15 min at room temperature. Following by permeabilization with 0.5% Triton X-100 in PBS for 10 min at room temperature, cells was washed and incubated with phalloidin and DAPI at room temperature for 20 min. Finally, the coverslips were washed, mounted and viewed under EVOS uorescent microscope (Life technologies, Carisbad, CA USA).

SCI Model and NSCs Transplantation
Contusion SCI was induced using a NYU impactor [22]. Adult female C57BL/6 mice (20±2.5g in body weight) were anesthetized with 100mg/kg ketamine and 10mg/kg xylazine through intraperitoneal injection. The tenth thoracic (T10) spinal cord segment was exposed and a contusion injury of the T10 vertebral cord was induced by dropping a 5-g rod 6.25mm onto the spinal cord. Postoperative care such as manual bladder emptying was performed as previously described.
7days after SCI, a total volume of 0.5μl of cell suspension containing 5×10 4 NSCs was microinjected at either side of the spinal cord contusion 2mm away using a Hamilton syringe.

Real Time PCR
Mice were sacri ced and spinal cords were immediately extracted 7days after SCI (n=5 each group). Total RNA was isolated from the indicated tissues via TRI reagent (Sigma-Aldrich, St.Louis, MO, USA) and then reverse-transcribed into cDNA using Fist strand cDNA Synthesis kits (Roche, Basel, Switzerland). The following primer sets were as following: MCP-1 (5'-ATGCAGGTCCCTGTCATGCTT-3' and 5'-CATTGGGATCATCTTGCTGGT-3') and GAPDH (5'-ATCAACGACCCCTTCATTGACC-3' and 5'-CCAGTAGACTCCACGACATACTCAGC-3'). We assayed the expression of target gene mRNA by using the ABI 7900HT Fast real-time PCR detection system (Applied Biosystems, UK) and expression was normalized to GAPDH for each sample.

Immunohistochemistry
The spinal cord was xed with 4% paraformaldehyde in phosphate-buffered saline (PBS) overnight at 4°C, incubated with 30% sucrose solution at 4°C until tissue sunk, and then frozen in OCT compound. 8um-thick parasagittal sections were prepared by using a cryostat microtome and attached to poly-llysine-coated glass slides. For immuno uorescence staining, samples were blocked with 1% BSA in PBS containing 0.3% Triton X-100 at room temperature for 1h, and incubated with primary antibodies at 4°C for 12 h. After washed with PBS, the sections were incubated with secondary antibodies at room temperature for 1h. All the procedures for negative controls were performed in the same manner, just except incubation of primary antibodies. EVOS uorescent microscope (Life technologies, Carisbad, CA USA) was used to observe and capture the images.

Statistical Analysis
The results were reported as mean± standard deviation (SD). All statistical data was analyzed using SPSS 22.0 software program (StatSoft, Tulsa, OK). The statistical signi cance was evaluated using Student's unpaired t-test and Statistical signi cance was assumed if P<0.05.

Culture and characterization of NSCs
After 7days cultured in serum free media including B27, EGF and bFGF, NSCs proliferated and formed neurosphere ( Figure 1A). For identi cation of NSCs, the neurosphere after two passages were stained with nestin [23], a marker of NSCs, and DAPI was used to mark nuclei. The result showed that the neurosphere was nestin positive ( Figure 1B). To further examined the capacity of generating the main phenotypes of CNS cells including neurons and astrocytes, NSCs were seeded on the PLL-coated coverslips and cultured for 5days in differentiation medium which were DMEM/F12 medium supplementing with 1% FBS and 2% B27, and then stained with GFAP for astrocyte and Tuj1 for neuron. The result showed that the cells expressed TUJ1 ( Figure 1C) and GFAP ( Figure 1D), which indicated that the multipotency of these NSCs.

Conditioned medium from M1 macrophages (M1-CM) induced NSCs migration
To determine whether M1-CM can induce the migration of NSCs, we induced M1 macrophages by IFNγ (10ng/ml, Millipore) and performed a transwell migration assay in vitro. Compared with control medium, a signi cant increase of migration of NSCs was observed in the presence of M1-CM and M0-CM (P<0.05, Figure 2A and 2B). Meanwhile, the number of migrated NSCs toward M1-CM was signi cantly higher than that of migrated NSCs toward M0-CM (P<0.05, Figure 2A and 2B). MTT assay was performed to exclude possible in uence from the proliferation of NSCs. MTT result showed M1-CM and M0-CM had no signi cant effect on proliferation of NSCs within 12h (P>0.05, Figure 2C). Therefore, these results indicated that M1-CM could attract NSCs migration in vitro.

M1-CM promoted NSCs migration via CCR2
CCR2 is very important for the migration of NSCs [24]. In order to investigate whether CCR2 is involved in the regulation of M1-CM-induced NSCs migration, the effect of M1-CM on the migration of NSCs isolated from CCR2-/-mouse used in migration assay was tested. At the beginning, expression of CCR2 in WT NSCs and CCR2-/-NSCs was con rmed by reverse transcription polymerase chain reaction ( Figure 3A). As showed in Figure 3B and 3C, the migration assay results demonstrated that the chemotactic effect of M1-CM on the migration of NSCs was signi cantly inhibited in the CCR2-/-NSCs group compared with WT NSCs (P<0.05).

M1-CM induced NSCs migration by secretion of MCP1
CCR2 is known as the only high-a nity receptor for MCP-1 [25]. Widera et al also found that MCP-1 can act as a chemotactic factor for adult NSCs [26]. In order to investigate whether M1 macrophages induce NSCs migration by secreting MCP-1, we examined the production of MCP-1 by M1 macrophages using enzyme-linked immunosorbent assay (ELISA). The ELISA results showed that BMDMs stimulated by IFNγ, dramatically increased the production of MCP-1 compared to BMDMs which were cultured without treatment of INF-IFN-γ (P<0.05, Figure 4A). Next, M1-CM was treated with anti-MCP-1 antibody to neutralize the MCP-1. A signi cant reduction of migration observed in M1-CM+ anti-MCP-1 antibody group in comparison to M1-CM group (P<0.05, Figure 4B and 4C). Taken together, these results suggested M1 macrophages probably secreted MCP-1, Which activated the CCR2 receptor, to attract the migration of NSCs in vitro.
Migration of NSCs toward M1-CM was dependent on ERK pathway ERK pathway has been implicated in the migration of numerous cell types including neural stem/progenitor cells and mesenchymal stem cells. To determine the effect of ERK pathway on the migration of NSCs toward M1-CM, PD98059, an inhibitor of ERK pathway [27], were used in this study and found that it could inhibit the migration of NSCs toward M1-CM (P<0.05, Figure 5A and 5B). Furthermore, the treatment of NSCs with M1-CM could stimulate phosphor-ERK1/2 and the M1-CM-induced expression of P-ERK1/2 in NSCs was markedly decreased after treated with PD98059 ( Figure 5C). Collectively, these results convinced us thatM1-CM induced NSCs migration by activation of ERK1/2.

F-actin participated in the migration of NSCs
F-actin is important for shaping cell morphology and regulating cell migration [28]. Next, to determined whether F-action was involved in NSCs migration induced by M1-CM, uorescent phalloidin staining were used to investigate the distribution of F-actin in NSCs. Phalloidin is a bicyclic peptide that binds to speci cally to the F-actin and doesn't bind to G-actin [29]. F-actin was expressed throughout the cytoplasm. While, compared with the control group and M1-CM group, F-actin structures were distributed around the periphery of the cells and the expression of F-actin were signi cantly increased (Figure 6).

Migration of local transplanted NSCs was dependent on CCR2 after SCI
Furthermore, in order to assess the effect of CCR2 and BMDMs on the migration of grafted NSCs after injury, the microinjection of NSCs besides the injury site was conducted 7days after SCI. As reported, bone marrow derived macrophages (BMDMs) could migrate to the injured area about 3days after the injury and accumulate within the epicenter of the injured site about 7days after SCI [12], so migration of NSCs in vivo was performed at this time point. The results showed that transplanted WT NSCs migrated from injection site into the injured area 5days post transplantation, while the movement of CCR2-/-NSCs was lower than WT NSCs 5days after cell transplantation ( Figure 7B). As showed in Figure 7A1 and 7B1, WT NSCs were found in the injury site identi ed by a lot of F4/80 (a microglia/macrophages marker) positive signal (Red), but no GFP signal were observed within injury area in CCR2-/-NSCs group.
Expression of M1 macrophage and MCP-1 mRNA 7days after SCI Conditioned medium from M1 macrophages, which was induced by IFN-γ, attracted the migration of NSC in vitro possible by secreting MCP-1, so we detected the expression of M1 macrophage marker and mRNA level of MCP-1 7days after SCI. To test the phenotype of macrophage 7days after SCI, M1 macrophages were identi ed by double Mac-2 (a marker of macrophage) and inducible nitric oxide synthase (iNOS, a marker of M1 macrophages) staining. The result showed that most of macrophages were Mac2+/iNOS+ cells 7 days after injury ( Figure 8A). Next, we evaluated the expression of MCP-1 mRNA in injured tissue and found that mRNA level of MCP-1 was signi cantly increased in the SCI group than that of sham group at the same time (P<0.05, Figure 8B). Taken together, M1 macrophages may not only induce NSCs migration in vitro, but also play an important role on the grafted NSCs migration.

Discussion
Many in vivo studies have shown that NSCs, which are able to migrate toward injury sites of the CNS, may act as a delivery tool for tissue repair and regeneration [8,30]. In this study, we investigated the effects on conditioned medium from M1 macrophage (M1-CM) on migration of NSCs and the possible signal molecules in this migratory process.
Results of the transwell migration assay showed that NSCs were seeded in the upper chamber, control medium or M-CM were placed in the lower chamber, and then the number of migrated NSCs toward M1-CM was signi cantly increased compared with M0-CM and control medium. The results con rmed that M1-CM had the capacity to induce NSCs migration in vitro. In vivo result showed that when BMDMs accumulated within the injury site 7days after injury and most of them were M1 macrophage which were identi ed by expressing both iNOS (a M1 macrophage marker) and Mac-2 (a macrophage marker), WT NSCs were local injected into spinal cord and found that they can migrate toward injury area. Therefore, we speculated that the injury-mediated NSCs migration toward damaged area may be partly due to the effect of M1 macrophages 7days after SCI. The migration of endogenous and exogenous NSCs may be affected by other factors in vivo. Previous studies have reported that hypoxia-induced astrocytes and activated microglia can induce NSCs migration in vitro [14,35]. Further studies are needed in order to get more evidences showing M1 macrophages induce NSCs migration in vivo.
Chemokines are major regulators of cell tra cking and adhesion. Numerous previous studies had shown that interaction between MCP-1 and its receptor CCR2. The binding of MCP-1 to CCR2 initiated divergent signaling pathways downstream of ligand binding, which had been already known to control varioustypes of normal and pathological cell migrations [36][37][38], also required for the migration of stem cells [39,40]. In this study, to investigate the mechanism responsible for M1-CM-mediated NSCs migration, we compared the migration between WT NSCs and CCR2-/-NSCs in vitro and in vivo. The results showed that CCR2 played a very important role on the M1-CM-induced NSCs migration in vitro, as well as on the injury-mediated NSCs migration at sub-acute SCI. CCR2 is the only known high-a nity receptor for MCP-1. We found that BMDMs stimulated by IFN-γ could increase the secretion of MCP-1 and lead to increasing migration of NSCs. Our results are in line with a previous report showing CCR2 expression in nestin-positive neurosphere cultures and MCP-1 can activate the migration capacity of NSCs [26]. Thus, we con rmed that M1-CM can induce NSCs migration by means of MCP-1/CCR2. It's known that chemokines are produced at sites of injury. Our RT-PCR result showed that the expression of MCP-1 mRNA was signi cantly increased at 1week after SCI. Similarly, MCP-1 expression can be induced by mechanical injury within the brain [41,42]. Therefore, MCP-1 and CCR2 were needed for grafted NSCs migrated from injection site to injury area after SCI.
Mitogen-activated protein kinase (MAPK) signaling pathway including ERK1/2, p38and JNK molecules have been reported to control major cell function [43,44]. ERK1/2 molecule is an important molecule in the bridge between cell membrane and nucleus which could be activated via integrins, growth factors and so on, and plays an essential role in controlling gene transcription [45]. ERK1/2 control cell migration of human neural progenitor cells [46,47]. In our study, we found that M1-CM strongly activated ERK1/2 phosphorylation. At the same time, PD98059 can abolish the strong phosphorylation of ERK1/2. Result from transwell assay con rmed our hypothesis, because PD98059 can signi cantly prevent M1-CMinduced migration of NSCs. Collectively, these results implied that ERK1/2 was an important signal molecule mediating M1-CM-induced NSCs migration in vitro.

Conclusion
In summary, we demonstrated that M1-CM could promote NSCs migration in vitro. MCP-1/CCR2 and ERK1/2 signal pathway were involved in M1-CM-mediated NSCs migration. Furthermore, the migration of grafted NSCs toward injured area may be due to M1 macrophage within injured area depending on the MCP-1 and CCR2. Deep and detailed investigations on effects of BMDMs on the migration of NSCs will not only lead to a better understanding of the molecular mechanism of the interaction between NSCs and macrophages, but also provide an important theoretical guidance to develop NSCs as an e cient treatment for SCI. Figure 1 Identi cation of NSCs from embryonic mouse brain. After 7days cultured in serum free media including B27, EGF and bFGF, NSCs proliferated and formed neurosphere (A). Meanwhile, the neurosphere expressed nestin (Red, B), a marker of neural stem/progenitor cells. After 5days differentiation in the media without the mitogens, the cells expressed the neuronal marker TUJ1 (Green, C), the astrocytic marker glial brillary acidic protein (GFAP) (Red, D). So, NSCs exhibited the capacity to differentiate into neurons, astrocytes and oligodendrocytes. Nuclei in C-D was stained with DAPI (Blue).    The expression and distribution of F-actin. Immuno uorescence analysis was performed using CF488conjugated phalloidin (F-actin, green). Nuclei were stained with DAPI (Blue). Scale bar 20 μm.