Skip to main content
NCBI home page
Cellular and Molecular Neurobiology logoLink to Cellular and Molecular Neurobiology
. 2015 Apr 4;35(6):881–890. doi: 10.1007/s10571-015-0182-x

Local Injection of Lenti–BDNF at the Lesion Site Promotes M2 Macrophage Polarization and Inhibits Inflammatory Response After Spinal Cord Injury in Mice

Xin-Chao Ji 1,2, Yuan-Yuan Dang 2, Hong-Yan Gao 4, Zhao-Tao Wang 2, Mou Gao 2, Yi Yang 2, Hong-Tian Zhang 2,3,✉, Ru-Xiang Xu 1,2,3,✉
PMCID: PMC11486196  PMID: 25840805

Abstract

There is much evidence to suggest that brain-derived neurotrophic factor (BDNF) is a prominent candidate in promoting neuroprotection, axonal regeneration, and synaptic plasticity following spinal cord injury (SCI). Although some evidence indicates that BDNF has potent anti-oxidative effects and may be involved in the regulation of the immune response, the effects of BDNF in the inflammatory response during the course of secondary damage after SCI is still unclear. The present study was designed to investigate the effects of BDNF with a special focus on their effect on macrophage polarization after SCI. Adult C57 mice underwent T10 spinal cord clip compression injury and received lenti-BDNF vector injections at the epicenter of the lesion site. Four days later, total BDNF levels were greatly increased in animals that received lenti-BDNF injections. Confocal imaging showed that more than 80 % of the lenti-virus infected cells were CD11b-positive macrophages. In addition, the expression of arginase-1 and CD206 (associated with M2 macrophage phenotype) significantly increased in the animals that received lenti-BDNF injections compared with those that received lenti-EGFP injections. On the contrary, the expression of CD16/32 and inducible nitric oxide synthase (M1 phenotype marker) was down-regulated as demonstrated using flow cytometry and immunohistochemistry. Furthermore, the production of interleukin 1β and tumor necrosis factor alpha was significantly reduced whereas the levels of interleukin 10 and interleukin 13 were elevated in subjects that received lenti-BDNF vector injections. The time course of functional recovery revealed that gradual recovery was observed in the subacute phase in lenti-BDNF group, little improvement was observed in lenti-EGFP group. At the axonal level, significant retraction of the CST axons were observed in lenti-EGFP injected animals relative to lenti-BDNF group by biotinylated dextran amine tracing. In addition, compared to lenti-BDNF group markedly demyelination was observed in the lenti-EGFP group using luxol fast blue staining. In conclusion, we found that BDNF could promote the shift of M1 to M2 phenotype and ameliorate the inflammatory microenvironment. Furthermore, the roles of BDNF in immunity modulation may enhance neuroprotective effects and partially contribute to the locomotor functional recovery after SCI.

Keywords: Macrophage, Microglia, Polarization, BDNF, Spinal cord injury

Introduction

The poor recovery of spinal cord injury (SCI) is generally attributed to the hostile local milieu created at the trauma site. Local inflammatory response and glial scarring are considered to be two major barriers to repair (Fitch and Silver 2008). The role of inflammation in the injured spinal cord has been regarded as a “double-edged sword”. It can promote tissue repair by eliminating pathogens and clearing debris. However, these beneficial effects may be overshadowed by the excessive accumulation of pro-inflammatory cytokines and neurotoxins (Benowitz and Popovich 2011). Resident microglia and infiltrating hematogenous macrophages are central players in the local inflammatory response after SCI (Jones et al. 2005).

Macrophages/microglia can acquire distinct functional phenotypes in response to different environmental stimuli (Gordon and Martinez 2010). M1 macrophages activated by microbial ligands and cytokines, such as lipopolysaccharide (LPS), prototypical T helper 1 cytokine (TH1), or interferon-gamma (IFN-γ), represent a pro-inflammatory phenotype. These macrophages exhibit increased phagocytic and antigen-processing activity as well as increased production of pro-inflammatory mediators such as tumor necrosis factor alpha (TNF-α), interleukin 12 and 23 (IL-12, IL-23), nitric oxide (NO), and reactive oxygen species (ROS) to improve host defense, but that also cause collateral damage to healthy tissue. Conversely, M2 phenotype is induced by a variety of stimuli with Th2 cytokines such as interleukin 4 and 13 (IL-4, IL-13), as well as M-CSF, in response to promote wound healing and tissue remodeling. They are referred to as anti-inflammatory macrophages and produce IL-4, IL-5, IL-10, and IL-13 (Lawrence and Natoli 2011; Qin et al. 2012; Sica and Mantovani 2012; Spence et al. 2013). Plasticity is one of the key features of macrophages. M1/M2 phenotype macrophages can be converted into each other in response to different microenvironments (Murray et al. 2014).

Abundant evidence confirmed that BDNF is a prominent candidate in spinal cord repair. BDNF delivery has many effects in the nervous system including the enhancement of neurite outgrowth, axonal regeneration, and the promotion of myelination as well as the modulation of synaptic transmission and neuronal excitability (Blesch et al. 2011; Fouad et al. 2013; Kobayashi et al. 1997; Müller et al. 2008). Furthermore, the expression of BDNF and their high affinity TrkB receptors has been detected in different lymphoid tissues (Artico et al. 2008). Although some evidence shows that BDNF has potent anti-oxidant effects and may be involved in the regulation of immune response (Neumann et al. 1998), the effects of BDNF in the inflammatory response during the course of secondary damage following SCI are not well known. For this purpose, in the present study, we examined the effects of BDNF-overexpression on the inflammatory responses, especially focusing on the macrophage polarization.

Materials and Methods

Animals

Adult female C57BL/6 mice were used. All experiments were performed in accordance with the guidelines established by the Animal Care and Use Committee of Third Military Medical University (Chongqing, China). Animals were divided into groups based on the type of vectors received. Lentiviral vectors expressed either EGFP only (lenti-EGFP groups), or BDNF plus EGFP (lenti-BDNF groups). Animals in the control group were injected accordingly with saline (Saline group). Additionally, animals receiving T10 laminectomy without SCI were set up as a sham group.

Surgical Procedure

Mice were kept anesthetized using 3.6 % chloral hydrate. Body temperature was maintained by keeping the mice on a heating pad (37 °C) during the entire procedure. The skin was incised, and the muscle tissue dissected to expose laminae T9–T11. A T10 laminectomy was performed. A modified aneurysm clip with a closing force of 20 g was placed around the cord between T9 and T10 and snapped closed for 60 s (Fehlings and Tator 1995; Joshi and Fehlings 2002). The muscle layers were sutured and the skin was secured with wound clips. For the laminectomy in sham group, the compression was not performed. The mice were placed on soft bedding on a warming blanket held at 37 °C until fully awake.

Vector Injections

We amplified the coding sequence of BDNF-N4-5261(NM_001048139, N-terminal amino acid residues 4–5261 of BDNF) by RT-PCR and ligated them into the pGV208 plasmid (Shanghai Gene Chem) to produce pGV208-BDNF-N4-5261. As a control, we also generated a lentiviral vector that expresses EGFP alone (LV-EGFP). Three days post-surgery, the animals were anesthetized as above and placed in a stereotaxic device. The stitches were taken out and the injury area was exposed. Lentiviral vectors and saline were injected into the epicenter of injured spinal cord at a depth of 0.5 and 1 mm (1 µl each depth), at a rate of 200 nl/min with a 5-µl Hamilton syringe. Pipettes were left in place for 1 min after the injection and then slowly withdrawn. Overlying muscle layers were sutured and the skin was stapled. Animals were sacrificed at 4- and 11-day post-injection (7 and 14 days after SCI).

Biotinylated Dextran Amine (BDA) Tracing

To trace the corticospinal tract, tracer injections were performed at 14 days before being sacrificed (n = 6 for each group). The animals were anesthetized as above and placed in a stereotaxic device. Fur was shaved and the scalp was incised; then sphenotresia was performed with a dental drill. BDA mini-ruby (10 % in 0.01 MPBS; 10,000 MW; Invitrogen) was injected into four sites (AP coordinates from bregma in mm: AP 0.5/1.0, −0.2/1.0, −0.7/1.0, −1.0/1.0 at a depth of 0.5 mm into the right sensorimotor cortex and 0.4 µl per site over a 3–5 min time period) with a 5-µl Hamilton syringe (Steward et al. 2008). After the injections were completed, the skin overlying the skull was sutured with 4-0 silk.

To quantify the dieback of the corticospinal tract (CST), five consecutive sagittal sections from the ventral part near the central canal of the spinal cord were chosen and digital images were taken at the CST end using fluorescence microscopy. The distance is assessed by measuring to a “fiber front” which is defined as the point at which there is a fascicle of contiguous fibers approximately 30 μm wide or more (Shen et al. 2009). The lesion border was localized by the edge of the glial scars which were infected with EGFP. Finally, the distance was determined by using the image scale (Floriddia et al. 2012).

Behavioral Assessments

All animals were acclimated to the open field for the locomotor test once daily before surgery. Motor function was assessed using the Basso Mouse Scale (BMS) open field test (Basso et al. 2006). All animals were allowed to freely move in the open field for 5 min and two independent investigators performed blinded assessment of their performance. The animals in each group (n = 10) were tested at 1, 3, 7, 14, 21, and 28 days following injury.

Preparation for Histology and Immunohistochemistry

Mice were anesthetized with 3.6 % chloral hydrate and transcardially perfused with 4 % paraformaldehyde. Spinal cords were isolated and post-fixed in fixative for 16 h at 4 °C. Tissues were dehydrated using 30 % sucrose at 4 °C for 2 days. After embedding into Tissue-Tek OCT compound, sagittal spinal cord sections (12 μm) were collected using a Leica freezing microtome and stored at −80 °C until samples could be processed. Sections of spinal cords were blocked with 10 % bovine serum albumin (BSA; Sigma) in PBS for 1 h at room temperature and then incubated with Rat anti-CD16/32 (1:100; BD Biosciences Pharmingen) Rabbit anti-GFAP (1:2000; Abcam) Rat anti-CD11b (1:100; Abcam) in 2 % BSA in PBS or Goat anti-arginase-1 (1:100; Abcam) and Mouse anti-NF-200 (1:200; Abcam) antibodies in 5 % BSA with 0.3 % Triton at 4 °C overnight. They were washed three times with PBS over 15 min with gentle agitation. This was followed by incubation with Alexa Flour 555-conjugated donkey anti-goat, rat, rabbit, or mouse (1:500) at room temperature for 1 h. Counterstaining was performed with 4′,6-diamidino-2-phenylindole (DAPI) and images were obtained by confocal microscopy (Leica, SP5II).

To semi-quantify the number of arginase-1, CD16/32-positive cells at 7 and 14 days post-injury, five sagittal sections were serially selected (1:6) in the medial part of each spinal cord from the subjects (n = 4 for each group). Two images were obtained per section from the above-mentioned sections for each spinal cord by confocal microscopy. The number of arginase-1, CD16/32 staining was determined for each image using Image Pro-Plus software.

Luxol Fast Blue Staining

For myelin staining, we used luxol fast blue (LFB). Sections of spinal cords were incubated with 1 % LFB solution at 60 °C overnight. Then we rinsed these sections with 95 % alcohol and distilled water, successively. This was followed by dipping the sections in 0.05 % lithium carbonate solution for 5 s and then washing in 70 % alcohol and distilled water. These steps were repeated until there was a sharp contrast between the white matter and the gray matter. Dehydration was performed using 95 and 100 % alcohol. Sections were cleared in xylene and coverslipped (DeBoy et al. 2007).

For comparison of LFB-positive areas, five sagittal sections of injured spinal cords (n = 6 for each group) randomly selected at a distance up to 4 mm rostral and caudal to the epicenter were stained with LFB and examined at 28 days after SCI. The LFB-positive area was analyzed using grain counting with the light intensity automatically set by the Image Pro-Plus software and the ratio to normal section was calculated (Okada et al. 2006).

Spinal Cord Preparation for Flow Cytometry

Mice were anesthetized with 3.6 % chloral hydrate and perfused with 50 ml of cold saline (n = 5 for each group). The compressed segments (1.5 mm rostral and 1.5 mm caudal to the lesion epicenter in SCI mice) were dissected and grinded. Cells were washed in ice-cold PBS and filtered through a 70-μm strainer. Prior to staining, a cell-count was performed for every sample to ensure a cell density of 1.0 × 106 cells/100 μl. Cells were blocked with 3 % BSA for 30 min at 4 °C. For intracellular staining, the cells were resuspended in fixation buffer and treated with permeabilization buffer followed by re-suspension in ice-cold PBS and incubation for 1 h with arginase-1 (1:200; Santa Cruz) conjugated to PerCP (1:200; Santa Cruz), allophycocyanin (APC) rat anti-CD206 (0.25 μg/ml; Biolegend), phycoerythrin (PE) rat anti-F4/80 (1.0 μg/ml; Biolegend), APC rat anti-CD16/CD32 (2.0 μg/ml; BD Pharmingen), APC anti-mouse iNOS antibody (3.0 g/ml; eBioscience). Samples with cells alone were used as negative controls to eliminate background auto-fluorescence and samples in which the cells had been incubated with a single antibody were used as positive controls to set up the cytometer alignment and to remove any spectral overlap. Flow cytometry was performed immediately using an Accuri C6 (Becton–Dickinson Biosciences, San Jose, CA). Forward scatter was set to further eliminate any cellular debris from analysis. In each test, a minimum of 250,000 cells were analyzed and the data were processed using BD Flow software (Becton–Dickinson Biosciences).

Enzyme-Linked Immunosorbent Assay (ELISA)

At 7 and 14 days post-injury, the animals (n = 5 for each group) were sacrificed and the lesion segments were obtained from the injured spinal cord and centrifuged at 12,000×g for 15 min to remove cellular debris. The supernatant was stored at −80 °C until use. The concentrations of BDNF, IL-1β, IL10, and TNF-α were measured using specific ELISA kits according to the manufacturers’ instructions (Invitrogen). Experiments were performed in duplicate in three independent experiments. The results were expressed as ng/g.

Statistical Analysis

To determine significance between two groups, a two-tailed unpaired Student’s t test was used. Statistical significance between multiple groups was determined using a one-way ANOVA with Tukey’s post hoc analysis. Statistical data are presented as mean ± standard deviation (mean ± SD) and performed using Statistical Package for the Social Sciences ver. 13.0 software (SPSS 13.0). A statistically significant difference was assumed at P < 0.05.

Results

Confirmation of Lenti-BDNF Expression

Four days after injection (7 days after SCI), confocal microscopy reveals that EGFP cells were distributed in the epicenter of the lesion site (Fig. 1a, b). To confirm that what the population of EGFP cells were composed of, the sections were labeled with CD11b (macrophages/microglia marker) and GFAP (astrocytes marker) by immunofluorescent staining. The data showed that most of infected EGFP cells were colocalized with CD11b-positive cells in the lenti-EGFP (83.1 ± 4.71 %) and lenti-BDNF (86.3 ± 5.44 %) groups (Fig. 1c, d). In addition, the percentage of infected GFAP-positive cells accounted for the overall EGFP cells was evaluated in lenti-EGFP (9.9 ± 1.27 %) and lenti-BDNF (9.1 ± 1.16 %) groups (Fig. 1e, f). There was no significant difference in the percentage of infected CD11b/GFAP-positive cells accounted for the overall EGFP cells between the two groups (Fig. 1g, h). In addition, total BDNF levels in the spinal cord segments containing the lesion site (1.5 mm rostral and 1.5 mm caudal to the lesion epicenter), were greatly increased in animals that received lenti-BDNF vector injections (62.63 ± 8.74 ng/g) compared with those that received lenti-EGFP injections (0.59 ± 0.15 ng/g) and those animals within the saline group (0.68 ± 0.13 ng/g) (Fig. 1I).

Fig. 1.

Fig. 1

Open in a new tab

Lenti-BDNF vector injection results in the overexpression of BDNF at the lesion site following spinal cord injury. Four days after injection (7 days after SCI), confocal microscopy of sagittal sections showed the EGFP+ cells were distributed in the epicenter of lesion site within lenti-EGFP (a) and lenti-BDNF groups (b) (asterisk indicate epicenter of the lesion site, scale bar 500 μm). Representative immunostaining of CD11b (macrophages/microglia marker) and GFAP (reactive astrocytes marker) in lenti-EGFP and lenti-BDNF groups at 7 days after SCI. EGFP, green; CD11b in (c, d) red, and GFAP in (e, f) red, scale bar 100 μm. g, h Quantitative analysis of the percentage of CD11b+/EGFP+ or GFAP+/EGFP+ cells accounts for overall EGFP+ cells (Data are mean ± SD, n = 4 per group). i Animals that received lenti-BDNF vector injections but not lenti-EGFP or saline injections exhibited significantly higher levels of BDNF protein at the epicenter of the lesion site as measured by ELISA (n = 5 for each group). Values are presented as mean ± SD. *** P < 0.001 (Color figure online)

Lenti-BDNF Promotes the Alternative Activation of the M2 Phenotype

To investigate whether the continuous expression of BDNF at the lesion site could influence the macrophage phenotype, the expression profiles of macrophages were evaluated by immunofluorescence staining with CD16/32 and arginase-1. Seven days post-injury, higher numbers of arginase-1-positive cells were found in the lenti-BDNF group (289 ± 29 cells/mm2) compared with the lenti-EGFP (179 ± 34 cells/mm2) and saline (168 ± 22 cells/mm2) groups (Fig. 2a). Furthermore, the number of CD16/32-positive cells was significantly reduced in animals that received lenti-BDNF injections (668 ± 60 cells/mm2) compared to animals in the lenti-EGFP (897 ± 60 cells/mm2) and saline groups (918 ± 68 cells/mm2) (Fig. 2b). This difference was observed at 7 and 14 days post-injury (Fig. 2c, d). To further elucidate the role of BDNF in the regulation of macrophages polarization, the quantification of the percentage of arginase-1- or CD16/32-positive cells out of EGFP cells was performed. Compared to lenti-EGFP group, the percentage of arginase-1-positive cells out of EGFP cells was significantly increased at 7, 14 days post-SCI. Reversely, the percentage of CD16/32-positive cells out of EGFP cells was significantly reduced (Fig. 2e, f).

Fig. 2.

Fig. 2

Open in a new tab

Lenti-BDNF enhances the expression of arginase-1-positive cells and reduces the number of CD16/32-positive cells. Immunofluorescence staining for arginase-1 (associated with the M2 macrophage phenotype) and CD16/32 (M1 marker) at 4 days post lentivirus vector injection (7 days after SCI). a The number of arginase-1-positive cells significantly increased in animals that received lenti-BDNF injections compared with those that received lenti-EGFP and saline injections. b Overexpression of BDNF significantly down-regulates the number of CD16/32-positive cells compared with lenti-EGFP and saline groups. c, d Quantitation of the expression of arginase-1 and CD16/32 in the epicenter of the lesion site 7 and 14 days after SCI, is presented, respectively. Quantitation of the percentage of arginase-1/CD16/32-positive cells out of EGFP cells in lenti-EGFP and lenti-BDNF groups (e, f). Data are presented as mean ± SD; scale bar 200 µm; n = 4 for each group. *P < 0.05

To further characterize changes in macrophage polarization, we used flow cytometry to assess the M1 markers iNOS and CD16/32, as well as the M2 markers CD206 and arginase-1, at 7 and 14 days post-injury. Phenotypic analysis of macrophages at 7 days post-injury showed that M1 markers iNOS and CD16/32 were significantly reduced in the lenti-BDNF group relative to lenti-EGFP and saline groups (Fig. 3a, b). On the contrary, M2 markers CD206 and arginase-1 were significantly increased in animals that received lenti-BDNF injections compared with those in lenti-EGFP and saline groups (Fig. 3c, d). Although the M2 phenotype macrophages were reduced or eliminated after the first week (consistent with a recent report (Kiger et al. 2009)), animals injected with lenti-BDNF exhibited a significantly higher proportion of M2 phenotype macrophages than those injected with lenti-EGFP or saline at 14 days after SCI (Fig. 3e–h).

Fig. 3.

Fig. 3

Open in a new tab

Overexpression of BDNF induces a shift from M1 to M2 phenotype. a–d Representative flow-cytometry data at 4 days post vector injection (7 days after SCI). Overexpression of BDNF significantly reduced the proportion of iNOS- and CD16/32-positive cells (M1 phenotype) and shifted the macrophage phenotype to arginase-1- and CD206-positive cells (M2 phenotype) compared with lenti-EGFP and saline groups. e–h The differences in iNOS-, CD16/32-, CD206-, and arginase-1-positive macrophages were significant at 7 and 14 days post SCI (n = 5 for each group). Data are presented as mean ± SD; *P < 0.05, **P < 0.01, *** P < 0.001

Effects of BDNF Overexpression on the Cytokine Expression

To examine whether the overexpressed BDNF could regulate the production of inflammatory cytokines in the injured spinal cord, we measured the levels of pro-inflammatory and anti-inflammatory cytokines such as IL-1β, TNF-α, IL-10, and IL13 by ELISA. As expected, the production of IL-1β and TNF-α was significantly reduced in the animals that received lenti-BDNF vector injections (IL-1β: 13.79 ± 1.31, TNF-α: 9.0 ± 2.21 ng/g, respectively) relative to those in the lenti-EGFP (21.65 ± 2.78, 17.41 ± 2.84 ng/g) and saline groups (20.89 ± 2.35, 18.02 ± 2.41 ng/g) whereas levels of IL-10 and IL-13 were significantly elevated in the lenti-BDNF group (IL-10:18.95 ± 1.74, IL-13:7.67 ± 0.86 ng/g, respectively) relative to lenti-EGFP (11.43 ± 2.58, 0.73 ± 0.11 ng/g) and saline groups (11.89 ± 2.74, 0.82 ± 0.17 ng/g). These significant differences were observed from day 7 to day 14 post-injury (Fig. 4a–d).

Fig. 4.

Fig. 4

Open in a new tab

Effects of overexpressed-BDNF on the cytokine expression. Cytokine levels were measured by ELISA in the injured spinal cord. Levels of IL-1β (a) and TNF-α (b) were significantly reduced in animals that received lenti-BDNF injections, whereas IL-10 (c) and IL-13 (d) levels were significantly increased compared to lenti-EGFP and saline groups at 7 and 14 days after SCI (n = 5 for each group). The results were represented as ng/g. Data are presented as mean ± SD; *P < 0.05, **P < 0.01

Lenti-BDNF Prevents CST Retraction and Promotes Recovery of Motor Functions

The time course of functional recovery and the score of each mouse at 4 weeks after SCI were evaluated using the BMS open field test. Whereas gradual recovery was observed in the subacute phase in lenti-BDNF group, little improvement was observed in lenti-EGFP group (Fig. 5a). In addition, markedly demyelination was observed in the lenti-EGFP group (Fig. 5b) compared to lenti-BDNF group (Fig. 5c, d), at 28 days post-injury using LFB staining. Furthermore, at 28 days post-injury, we observed a significant retraction of the CST axons in lenti-EGFP injected animals (Fig. 5e) relative to lenti-BDNF groups (Fig. 5f). The average length of NF-200-positive nerve fibers in the lenti-BDNF group (Fig. 5h) was fivefold greater than that in the lenti-EGFP group (Fig. 5i).

Fig. 5.

Fig. 5

Open in a new tab

Lenti-BDNF reduces CST retraction and improves myelination following SCI. a The time course of functional recovery was evaluated by the Basso Mouse Scale (BMS) score after SCI. In the open field, significant locomotor recovery was observed in the lenti-BDNF group compared with the lenti-EGFP group (n = 10 for each group). Luxol fast blue (LFB) staining showed marked demyelination in the lenti-EGFP group b compared with the lenti-BDNF group c, d (scale bar 500 µm, n = 6 for each group). Representative images of BDA labeled CST displayed a more pronounced retraction from the lesion site in animals that received lenti-EGFP injections e compared with the lenti-BDNF group f, g. The white dashed lines show the lesion border (scale bar 500 µm, n = 6 for each group). The average length of NF-200-positive nerve fibers were significantly longer in the animals that received lenti-BDNF injections h compared with the lenti-EGFP group (i, j) (scale bar 100 µm, n = 6 for each group). All above measurements were performed in tissues collected 28 days following SCI. In b, f, h: D dorsal, V ventral, R rostral, C caudal. All measures are presented as mean ± SD. *P < 0.05, **P < 0.01

Discussion

Over the past few decades, the function of BDNF has been intensively investigated and its broad effects in the nervous system make it an excellent candidate in spinal cord repair. Such effects mostly focus on the promotion of myelination, neurons survival, neurite outgrowth, and axonal regeneration as well as the modulation of synaptic transmission and neuronal excitability (Boyce et al. 2012; Mantilla et al. 2013). However, the role of BDNF in inflammation and secondary damage is not yet fully understood. In the present study, we demonstrate that high expression of BDNF at the lesion site of SCI can increase the proportion of the M2 macrophage phenotype. We have also demonstrated that these changes are accompanied by the decline of pro-inflammatory cytokines and the upregulation of anti-inflammatory mediators. Furthermore, the early functional recovery and promotion of myelination were observed in the lenti-BDNF group following SCI.

Recent studies indicate that M1 macrophages predominate at the lesion site during the inflammatory response, whereas the M2 phenotype is induced during early injury and then markedly reduced by 7–14 days post-SCI and this “M1 bias” could exacerbate secondary neurodegeneration (Kiger et al. 2009). Our results revealed that overexpressed BDNF at the lesion site could increase the number of M2 macrophages and reduce the expression of M1 phenotype. Furthermore, it is well established that the neuroprotective effects of BDNF are mediated by tropomyosin-related kinase B (TrkB) signaling (Gransee et al. 2013). In addition, the expression of BDNF and their high affinity TrkB receptors has been detected in different lymphoid tissues (Linker et al. 2010). To further elucidate the possible regulatory mechanisms of BDNF in the macrophage polarization, the percentage of arginase-1- or CD16/32-positive cells out of EGFP cells was evaluated. However, the direct effects of BDNF on macrophages were not observed.

Indeed, the macrophage phenotype can be dynamically regulated by the lesion microenvironment (Nakajima et al. 2012). The potent anti-oxidative and anti-inflammatory effects of BDNF have been widely reported. For example, BDNF can inhibit NO synthase thereby indirectly reducing the production of NO (Sharma et al. 1998). Furthermore, the application of BDNF-collagen at the cavity of the injured spinal cord could reduce lipid peroxidation and affect the microglial response (Joosten and Houweling 2004). Consistent with a recent report, intranasal BDNF might protect the brain against ischemic insult by regulating the levels of cellular cytokine in experimental stroke (Jiang et al. 2011). Our data showed that the overexpression of BDNF at the lesion site could reduce the levels of TNF-α and IL-1β while increasing the expression of IL-10 and IL-13. Together with discussed above, our results support the possibility that BDNF exerts its effects on the macrophage polarization through improving the inflammatory environment.

The activated macrophages/microglia can directly induce axonal retraction by physical cell–cell interactions with injured axons (Horn et al. 2008). Our data indicated that the overexpression of BDNF at the lesion site could prevent axonal retraction and promote locomotor-function recovery. The shift from M1 to M2 phenotype may be one of the underlying mechanisms. Moreover, recent studies indicate that M1, but not M2, is toxic to cortical neurons and that they can directly induce neuronal death by overlaying cortical neuron cultures with activated M1 or M2 macrophages grown on trans-well inserts (Shin et al. 2013). On the other hand, M2 macrophages are referred to as an anti-inflammatory phenotype that promotes wound healing and tissue remodeling (Guerrero et al. 2012; Liu et al. 2013; Yao et al. 2014). Taken together, all these studies support the possibility that the shift from M1 to M2 induced by BDNF could prevent axonal retraction via reducing cell–cell interactions between injured axons and neurotoxic M1 macrophages.

However, we have to be aware that BDNF could directly act on neurons and axons. BDNF exerts its neuroprotective effects by binding to the high affinity TrkB receptor and the low-affinity p75 neurotrophin receptor (p75NTR) which can collaborate with or inhibit each other’s actions (Weishaupt et al. 2012). To stimulate neuronal survival, TrkB signaling through Ras and PI-3K/Akt can suppress p75NTR-induced apoptosis by inhibiting the JNK–p53–Bax or Forkhead pathways (Dolcet et al. 1999; Kaplan and Miller 2000). In addition, BDNF-mediated axonal regeneration is instantiated in the increased expression of the injury/regeneration-associated genes, growth-associated protein-43, and Talpha1 tubulin (Geremia et al. 2010). Furthermore, the application of BDNF could up-regulate the expression of MBP and promote myelinogenesis by inducing the differentiation of oligodendrocyte progenitor cells after SCI (Ikeda et al. 2002). As discussed above, it is well established that the effects of BDNF on neuroprotection result from the integration of multiple mechanisms.

In conclusion, regulation of local inflammation is a promising strategy for promoting SCI repair, our study provides new insight into the role of BDNF in immunity modulation including its regulation of inflammatory cytokine levels and macrophage/microglial polarization. Acquiring a better knowledge of the underlying mechanisms of these events needs further studies.

Acknowledgments

This work was supported by the Natural Scientific Research funds of China (No. 81371345) and Beijing Nova program (XX2013059). We thank Ms. Shan-shan Wang and Ms. Yan Zhang for their technical assistance.

Conflict of interest

None.

Ethical standard

We certify that we have participated sufficiently in the work to take public responsibility for the appropriateness of the experimental design and method, and the collection, analysis, and interpretation of the data.

Contributor Information

Hong-Tian Zhang, Phone: 86-10-66721204, Email: zhanghongtian007@126.com.

Ru-Xiang Xu, Phone: 86-10-66721204, Email: xuruxiang_neuron@126.com.

References

  1. Artico M, Bronzetti E, Felici LM, Alicino V, Ionta B, Bronzetti B, Magliulo G, Grande C, Zamai L, Pasquantonio G, De Vincentiis M (2008) Neurotrophins and their receptors in human lingual tonsil: an immunohistochemical analysis. Oncol Rep 20(5):1201–1206 [PubMed] [Google Scholar]
  2. Basso DM, Fisher LC, Anderson AJ, Jakeman LB, McTigue DM, Popovich PG (2006) Basso mouse scale for locomotion detects differences in recovery after spinal cord injury in five common mouse strains. J Neurotrauma 23:635–659 [DOI] [PubMed] [Google Scholar]
  3. Benowitz L, Popovich PG (2011) Inflammation and axon regeneration. Curr Opin Neurol 24(6):577–583 [DOI] [PubMed] [Google Scholar]
  4. Blesch A, Lu P, Tsukada S, Alto LT, Roet K, Coppola G, Geschwind D, Tuszynski MH (2011) Conditioning lesions before or after spinal cord injury recruit broad genetic mechanisms that sustain axonal regeneration: superiority to camp-mediated effects. Exp Neurol 235:162–173 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Boyce VS, Park J, Gage FH, Mendell LM (2012) Differential effects of brain-derived neurotrophic factor and neurotrophin-3 on hindlimb function in paraplegic rats. Eur J Neurosci 35:221–232 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. DeBoy CA, Zhang J, Dike S, Shats I, Jones M, Reich DS, Mori S, Nguyen T, Rothstein B, Miller RH, Griffin JT, Kerr DA, Calabresi PA (2007) High resolution diffusion tensor imaging of axonal damage in focal inflammatory and demyelinating lesions in rat spinal cord. Brain 130:2199–2210 [DOI] [PubMed] [Google Scholar]
  7. Dolcet X, Egea J, Soler RM, Martin-Zanca D, Comella JX (1999) Activation of phosphatidylinositol 3-kinase, but not extracellular-regulated kinases, is necessary to mediate brain-derived neurotrophic factor-induced motoneuron survival. J Neurochem 73(2):521–531 [DOI] [PubMed] [Google Scholar]
  8. Fehlings MG, Tator CH (1995) The relationships among the severity of spinal cord injury, residual neurological function, axon counts, and counts of retrogradely labeled neurons after experimental spinal cord injury. Exp Neurol 132(2):220–228 [DOI] [PubMed] [Google Scholar]
  9. Fitch MT, Silver J (2008) CNS injury, glial scars and inflammation: inhibitory extracellular matrices and regeneration failure. Exp Neurol 209:294–301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Floriddia EM, Rathore KI, Tedeschi A, Quadrato G, Wuttke A, Lueckmann JM, Kigerl KA, Popovich PG, Di Giovanni S (2012) p53 Regulates the neuronal intrinsic and extrinsic responses affecting the recovery of motor function following spinal cord injury. J Neurosci 32(40):13956–13970 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fouad K, Bennett DJ, Vavrek R, Blesch A (2013) Long-term viral brain-derived neurotrophic factor delivery promotes spasticity in rats with a cervical spinal cord hemisection. Front Neurol 19(4):187 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Geremia NM, Pettersson LM, Hasmatali JC, Hryciw T, Danielsen N, Schreyer DJ, Verge VM (2010) Endogenous BDNF regulates induction of intrinsic neuronal growth programs in injured sensory neurons. Exp Neurol 223(1):128–142 [DOI] [PubMed] [Google Scholar]
  13. Gordon S, Martinez FO (2010) Alternative activation of macrophages: mechanism and functions. Immunity 32(5):593–604 [DOI] [PubMed] [Google Scholar]
  14. Gransee HM, Zhan WZ, Sieck GC, Mantilla CB (2013) Targeted delivery of TrkB receptor to phrenic motoneurons enhances functional recovery of rhythmic phrenic activity after cervical spinal hemisection. PLoS One 8:e64755 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Guerrero AR, Uchida K, Nakajima H, Watanabe S, Nakamura M, Johnson WE, Baba H (2012) Blockade of interleukin-6 signaling inhibits the classic pathway and promotes an alternative pathway of macrophage activation after spinal cord injury in mice. J Neuroinflammation 27(9):40 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Horn KP, Busch SA, Hawthorne AL, van Rooijen N, Silver J (2008) Another barrier to regeneration in the CNS: activated macrophages induce extensive retraction of dystrophic axons through direct physical interactions. J Neurosci 28(38):9330–9341 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ikeda O, Murakami M, Ino H, Yamazaki M, Koda M, Nakayama C, Moriya H (2002) Effects of brain-derived neurotrophic factor (BDNF) on compression induced spinal cord injury: BDNF attenuates down-regulation of superoxide dismutase expression and promotes up-regulation of myelin basic protein expression. J Neuropathol Exp Neurol 61:142–153 [DOI] [PubMed] [Google Scholar]
  18. Jiang Y, Wei N, Lu T, Zhu J, Xu G, Liu X (2011) Intranasal brain-derived neurotrophic factor protects brain from ischemic insult via modulating local inflammation in rats. Neuroscience 172:398–405 [DOI] [PubMed] [Google Scholar]
  19. Jones TB, McDaniel EE, Popovich PG (2005) Inflammatory-mediated injury and repair in the traumatically injured spinal cord. Curr Pharm Des 11(10):1223–1236 [DOI] [PubMed] [Google Scholar]
  20. Joosten EA, Houweling DA (2004) Local acute application of BDNF in the lesioned spinal cord anti-inflammatory and anti-oxidant effects. Neuroreport 15(7):1163–1166 [DOI] [PubMed] [Google Scholar]
  21. Joshi M, Fehlings MG (2002) Development and characterization of a novel, graded model of clip compressive spinal cord injury in the mouse: part 1. Clip design, behavioral outcomes, and histopathology. J Neurotrauma 19(2):175–190 [DOI] [PubMed] [Google Scholar]
  22. Kaplan DR, Miller FD (2000) Neurotrophin signal transduction in the nervous system. Curr Opin Neurobiol 10(3):381–391 [DOI] [PubMed] [Google Scholar]
  23. Kiger K, Gensel JC, Ankeny DP, Alexander JK, Donnelly DJ, Popovich PG (2009) Identification of two distinct macrophage subsets with divergent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord. J Neurosci 29:13435–13444 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kobayashi NR, Fan DP, Giehl KM, Bedard AM, Wiegand SJ, Tetzlaff W (1997) BDNF and NT-4/5 prevent atrophy of rat rubrospinal neurons after cervical axotomy, stimulate GAP-43 and Talpha1-tubulin mRNA expression, and promote axonal regeneration. J Neurosci 17:9583–9595 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lawrence T, Natoli G (2011) Transcriptional regulation of macrophage polarization: enabling diversity with identity. Nat Rev Immunol 11(11):750–761 [DOI] [PubMed] [Google Scholar]
  26. Linker RA, Lee DH, Demir S, Wiese S, Kruse N, Siglienti I, Gerhardt E, Neumann H, Sendtner M, Lühder F, Gold R (2010) Functional role of brain-derived neurotrophic factor in neuroprotective autoimmunity: therapeutic implications in a model of multiple sclerosis. Brain 133(Pt 8):2248–2263 [DOI] [PubMed] [Google Scholar]
  27. Liu C, Li Y, Yu J, Feng L, Hou S, Liu Y, Guo M, Xie Y, Meng J, Zhang H, Xiao B, Ma C (2013) Targeting the shift from M1 to M2 macrophages in experimental autoimmune encephalomyelitis mice treated with fasudil. PLoS One 8:e54841 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mantilla CB, Gransee HM, Zhan WZ, Sieck GC (2013) Motoneuron BDNF/TrkB signaling enhances functional recovery after cervical spinal cord injury. Exp Neurol 247:101–109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Müller HD, Hanumanthiah KM, Diederich K, Schwab S, Schäbitz WR, Sommer C (2008) Brain derived neurotrophic factor but not forced arm use improves long-term outcome after photothrombotic stroke and transiently upregulates binding densities of excitatory glutamate receptors in the rat brain. Stroke 39:1012–1021 [DOI] [PubMed] [Google Scholar]
  30. Murray PJ, Allen JE, Biswas SK, Fisher EA, Gilroy DW, Goerdt S, Gordon S, Hamilton JA, Ivashkiv LB, Lawrence T, Locati M, Mantovani A, Martinez FO, Mege JL, Mosser DM, Natoli G, Saeij JP, Schultze JL, Shirey KA, Sica A, Suttles J, Udalova I, van Ginderachter JA, Vogel SN, Wynn TA (2014) Macrophage activation and polarization: nomenclature and experimental guidelines. Immunity 41(1):14–20 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nakajima H, Uchida K, Guerrero AR, Watanabe S, Sugita D, Takeura N, Yoshida A, Long G, Wright KT, Johnson WE, Baba H (2012) Transplantation of mesenchymal stem cells promotes an alternative pathway of macrophage activation and functional recovery after spinal cord injury. J Neurotrauma 29(8):1614–1625 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Neumann H, Misgeld T, Matsumuro K, Wekerle H (1998) Neurotrophins inhibit major histocompatibility class II inducibility of microglia: involvement of the p75neurotrophin receptor. Proc Natl Acad Sci USA 95:5779–5784 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Okada S, Nakamura M, Katoh H, Miyao T, Shimazaki T, Ishii K, Yamane J, Yoshimura A, Iwamoto Y, Toyama Y, Okano H (2006) Conditional ablation of Stat3 or Socs3 discloses a dual role for reactive astrocytes after spinal cord injury. Nat Med 12(7):829–834 [DOI] [PubMed] [Google Scholar]
  34. Qin H, Holdbrooks AT, Liu Y, Reynolds SL, Yanagisawa LL, Benveniste EN (2012) SOCS3 deficiency promotes M1 macrophage polarization and inflammation. J Immunol 189(7):3439–3448 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sharma HS, Nyberg F, Westman J, Alm P, Gordh T, Lindholm D (1998) Brain derived neurotrophic factor and insulin like growth factor-1 attenuate upregulation of nitric oxide synthase and cell injury following trauma to the spinal cord. An immunohistochemical study in the rat. Amino Acids 14(1–3):121–129 [DOI] [PubMed] [Google Scholar]
  36. Shen Y, Tenney AP, Busch SA, Horn KP, Cuascut FX, Liu K, He Z, Silver J, Flanagan JG (2009) PTP sigma is a receptor for chondroitin sulfate proteoglycan, an inhibitor of neural regeneration. Science 326(5952):592–606 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Shin T, Ahn M, Moon C, Kim S, Sim KB (2013) Alternatively activated macrophages in spinal cord injury and remission: another mechanism for repair? Mol Neurobiol 47(3):1011–1019 [DOI] [PubMed] [Google Scholar]
  38. Sica A, Mantovani A (2012) Macrophage plasticity and polarization: in vivo veritas. J Clin Invest 122(3):787–795 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Spence S, Fitzsimons A, Boyd CR, Kessler J, Fitzgerald D, Elliott J, Gab-hann JN, Smith S, Sica A, Hams E, Saunders SP, Jefferies CA, Fallon PG, McAuley DF, Kissenpfennig A, Johnston JA (2013) Suppressors of cytokine signaling 2 and 3 diametrically control macrophage polarization. Immunity 38:66–78 [DOI] [PubMed] [Google Scholar]
  40. Steward O, Zheng B, Tessier-Lavigne M, Hofstadter M, Sharp K, Yee KM (2008) Regenerative growth of corticospinal tract axons via the ventral column after spinal cord injury in mice. J Neurosci 28(27):6836–6847 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Weishaupt N, Blesch A, Fouad K (2012) BDNF: the career of a multifaceted neurotrophin in spinal cord injury. Exp Neurol 238:254–264 [DOI] [PubMed] [Google Scholar]
  42. Yao A, Liu F, Chen K, Tang L, Liu L, Zhang K, Yu C, Bian G, Guo H, Zheng J, Cheng P, Ju G, Wang J (2014) Programmed death 1 deficiency induces the polarization of macrophages/microglia to the M1 phenotype after spinal cord injury in mice. Neurotherapeutics 11:636–650 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Cellular and Molecular Neurobiology are provided here courtesy of Springer

RESOURCES