C-X-C chemokine receptor type 7 antibody enhances neural plasticity after ischemic stroke

Abstract

Stromal cell-derived factor-1 and its receptor C-X-C chemokine receptor 4 (CXCR4) have been shown to regulate neural regeneration after stroke. However, whether stromal cell-derived factor-1 receptor CXCR7, which is widely distributed in the developing and adult central nervous system, participates in neural regeneration remains poorly understood. In this study, we established rat models of focal cerebral ischemia by injecting endothelin-1 into the cerebral cortex and striatum. Starting on day 7 after injury, CXCR7-neutralizing antibody was injected into the lateral ventricle using a micro drug delivery system for 6 consecutive days. Our results showed that CXCR7-neutralizing antibody increased the total length and number of sprouting corticospinal tract fibers in rats with cerebral ischemia, increased the expression of vesicular glutamate transporter 1 and growth-related protein 43, markers of the denervated spinal cord synapses, and promoted the differentiation and maturation of oligodendrocyte progenitor cells in the striatum. In addition, CXCR7 antibody increased the expression of CXCR4 in the striatum, increased the protein expression of RAS and ERK1/2 associated with the RAS/ERK signaling pathway, and improved rat motor function. These findings suggest that CXCR7 improved neural functional recovery after ischemic stroke by promoting axonal regeneration, synaptogenesis, and myelin regeneration, which may be achieved by activation of CXCR4 and the RAS/ERK1/2 signaling pathway.

Introduction

After ischemic stroke, endogenous neural plasticity, which includes axonal regeneration, synaptogenesis, and remyelination, occurs spontaneously in the brain and restores lost function in stroke patients (Cheng et al., 2017; Sandvig et al., 2018). Many restorative strategies, such as rehabilitation, antioxidants, and stem cell therapy, have been shown to further improve endogenous neural plasticity after ischemic injury in the brain (Bacigaluppi et al., 2009; Rodrigo et al., 2013; Shiromoto et al., 2017).

Stromal cell-derived factor-1 (SDF-1, also known as CXCR12), which is a member of the CXC chemokine subfamily, together with its receptor, chemokine C-X-C motif receptor 4 (CXCR4), is involved in endogenous neural regeneration within the developing and injured central nervous system (CNS) (Zhu and Murakami, 2012; Li et al., 2015, 2021; Tian et al., 2018; Gavriel et al., 2022; Terheyden-Keighley et al., 2022). An alternative receptor of SDF-1, chemokine C-X-C motif receptor 7 (CXCR7), is widely distributed in the CNS during development and in adulthood (Schönemeier et al., 2008b; Thelen and Thelen, 2008). Regeneration of impaired axons after stroke is very difficult, resulting in various deficits of neurological function. Strong evidence indicates that SDF-1 and CXCR4 participate in axonal elongation and outgrowth (Zanetti et al., 2019; Hilla et al., 2021), and promote axonal regeneration in animal models of stroke and spinal cord injury (Shyu et al., 2008; Opatz et al., 2009). Both CXCR4 and CXCR7 signaling pathways may be involved in axonal growth because both receptors have been detected in corticospinal tract (CST) axons (Jaerve and Müller, 2012; Wu et al., 2017). To date, the exact and direct effect of CXCR7 on axonal regeneration after cerebral ischemia remains unexplored.

Remyelination is also pivotal for brain repair after white matter injury induced by ischemic stroke, which often fails due to the insufficient recruitment and differentiation of oligodendrocyte progenitor cells (OPCs) (Li et al., 2015). Previous studies have indicated that CXC chemokines and their receptors are involved in central nervous system (CNS) remyelination (Beigi Boroujeni et al., 2020; Skinner and Lane, 2020). SDF-1 has been shown to regulate remyelination by promoting differentiation and migration of OPCs through the CXCR4 signaling pathway not only in vitro but also in animal models of cerebral ischemia and multiple sclerosis (Li et al., 2015; Marastoni et al., 2021). Previous studies have demonstrated that CXCR7 is expressed on oligodendroglia cells and OPCs in the demyelinated CNS, and promotes the maturation and differentiation of OPCs via SDF-1 stimulation (Göttle et al., 2010; Kremer et al., 2016). Moreover, CXCR7 expression is enhanced in cortical pyramidal cells after cerebral ischemia (Schönemeier et al., 2008a). However, few studies have explored the effect of CXCR7 on remyelination in models of stroke.

The aim of this study was to explore the effects of CXCR7 on axonal regeneration, synaptogenesis, and remyelination as well as motor functional recovery by using an anti-CXCR7 antibody to block the interaction between SDF-1 and CXCR7. We also explored the underlying signaling pathway after cerebral ischemia in rats.

Methods

Animals

Fifty-two male Wistar rats (200–250 g, 1.5–2 months old, specific-pathogen-free level) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd., Beijing, China (license No. SCXK (Liao) 2008-0005). The rats were randomly divided into the following groups: 1) sham-operated rats (Sham; n = 12), 2) sham-operated rats treated with CXCR7-neutralizing antibody (Sham + anti-CXCR7; n = 12), 3) ischemic stroke rats (ISC; n = 14), and 4) ischemic stroke rats treated with CXCR7 antibody (ISC + anti-CXCR7; n = 14). The experimental procedure followed the Guide for the Care and Use of Laboratory Animals (8^th ed) (National Institutes of Health, 2011). All animal experimental protocols and procedures were approved by the Institutional Animal Care and Use Committee of the China Medical University (approval No. CMU2020407) on September 26, 2020. The experimental procedure is illustrated in Figure 1. Anesthesia was conducted initially with 4.5% isoflurane (RWD Life Science Co., Ltd., Shenzhen, China), then maintained using 2% isoflurane.

Figure 1.

Procedure of the study.

Schematic diagram showing the timing of pretraining, ET-1 injection, CXCR7 antibody infusion, BDA injection, beam walking test, BrdU labeling, immunohistology, WB and PCR. BDA: Biotin dextran amine; BrdU: 5-bromo-2’-deoxyuridine; CXCR7: C-X-C chemokine receptor type 7; ET-1: endothelin-1; PCR: polymerase chain reaction; WB: western blotting.

Endothelin-1-induced ischemic stroke model

Endothelin-1 (Sigma-Aldrich, Burlington, MA, USA), which is a vasoconstrictor peptide, has been used to induce focal cerebral ischemia in our previous studies (Zhao et al., 2013a, b; Xu et al., 2016, 2017; Sun et al., 2017; Dong et al., 2020), and offers many advantages for research of functional recovery (Windle et al., 2006). In this model, endothelin-1 was dissolved at a concentration of 0.5 μg/μL in sterile water and injected stereotaxically into two cortical and one striatal locations (Paxinos and Watson, 1998): (1) anteroposterior (AP) +0.7 mm, mediolateral (ML) +2.2 mm, dorsoventral –2.0 mm; (2) AP +2.3 mm, ML +2.5 mm, dorsoventral −2.3 mm; and (3) AP +0.7 mm, ML +3.8 mm, dorsoventral –5.8 mm at a rate of 0.5 μL/min (2 μL/rat). The total volume per rat was 6 μL. Sham-operated rats were given an equal volume of 0.9% normal saline instead. Rats were warmed on an electric homeothermic blanket (37°C) during the operation. After the operation, all rats were maintained in a controlled environment (temperature 20 ± 1°C, 12-hour light/dark cycle) with free access to food and fresh water, and housed with four animals per cage.

CXCR7-neutralizing antibody infusion for blocking the SDF-1/CXCR7 axis

To block the SDF-1/CXCR7 pathway, CXCR7-neutralizing antibody (0.5 μg/μL, anti-GPCR-RDC-1, Abcam, Boston, MA, USA) was diluted in saline and infused into the lateral ventricle (AP –0.8 mm, ML +1.5 mm) via a microinjection system (RWD Life Science Co., Ltd.). Antibody administration began on day 7 after ischemia and lasted for 6 continuous days (3.5 μL per day; Figure 1).

Biotin dextran amine labeling for tracing CST fibers

Rats were injected with 10% biotinylated dextran amine (BDA, Invitrogen, Waltham, MA, USA) in 0.01 M phosphate-buffered saline (PBS) to label CST fibers in the spinal cord 2 weeks after the ischemia operation (Figure 1). BDA was infused into the nonlesional brain hemisphere according to a previous protocol (Sun et al., 2017).

Intraperitoneal injection of 5-bromo-2’-deoxyuridine

To assess the regeneration of oligodendrocyte lineage transcription factor 2 (Olig2)-positive cells, 5-bromo-2′-deoxyuridine (BrdU; Sigma-Aldrich) was dissolved in 0.01 M PBS (10 mg/mL) and injected intraperitoneally (100 mg/kg) twice a day for 2 days before sacrifice (Figure 1) for double staining with Olig2 in the striatum.

Tissue preparation for immunofluorescence staining, reverse transcription-polymerase chain reaction and western blotting

Twenty-four rats (n = 6 rats per group) were sacrificed by perfusion through the heart with 4% paraformaldehyde after anesthesia on postoperative day 33 (Figure 1), which is considered to be the chronic phase after stroke and is consistent with our previous studies (Zhao et al., 2013a, 2015). The whole brain and cervical spinal cord were dissected, fixed in 4% paraformaldehyde overnight, and then stored in 30% sucrose for 7 days. Brain sections (35 μm) were serially sliced and stained for identification of oligodendrogenesis markers. Spinal cord (C6–C8) was cut coronally in sections of 50 and 25 μm. The 50-μm-thick sections were prepared for BDA immunostaining and the 25-μm-thick sections were prepared for staining of synaptic markers. Tissue sections were stored at −20°C for subsequent immunofluorescence staining.

Another 24 rats were anesthetized and the peri-infarcted striatum (about 3 mm × 2 mm × 2 mm) was dissected and rapidly frozen in liquid nitrogen, and then stored at −80°C for western blotting and reverse transcription-quantitative polymerase chain reaction (qRT-PCR) analysis. Three rats died after the ischemic operation and one rat had an incomplete lesion; these rats were excluded from the following experiments.

Immunofluorescence staining of CST fibers, synaptic markers and oligodendrogenesis markers

Tissue sections were warmed at 20–25°C for 30 minutes, washed using PBS, and blocked in 10% goat serum (Solarbio, Beijing, China) for 1.5 hours. This was followed by incubation with one of the following primary antibodies overnight at 4°C: rabbit anti-vesicular glutamate transporter 1 (vGlut1; 1:100, Abcam, Cat# ab72311, RRID: AB_1271456), rabbit anti-synaptophysin (1:400, Abcam, Cat# ab32127, RRID: AB_2286949), rabbit anti-postsynaptic density-95 (PSD-95; 1:800, Abcam, Cat# ab18258, RRID: AB_444362), rabbit anti-growth associated protein-43 (GAP43; 1:500, Abcam, Cat# ab12274, AB_2247459), rabbit anti-neuron-glial antigen 2 (NG2; 1:200, Abcam, Cat# ab83178, RRID: AB_10672215), rabbit anti-Olig2 (1:500, Abcam, Cat# ab109186, RRID: AB_10861310), mouse anti-myelin basic protein (MBP; 1:300, Abcam, Cat# ab62631, RRID: AB_956157), and sheep anti-BrdU (1:500, Abcam, Cat# ab2284, RRID: AB_302944). Then, the sections were incubated with the following secondary antibodies at room temperature for 2 hours: streptavidin Alexa Fluor 594 (1:200, Invitrogen, Cat#S32356) for BDA; goat anti-rabbit Alexa Fluor 488 (1:200, Invitrogen, Cat# A11034, RRID: AB_2576217) for vGlut1, synaptophysin, PSD-95, and GAP-43; goat anti-rabbit Alexa Fluor 594 (1:200, Invitrogen, Cat# A11012, RRID: AB_2534079) for NG2, and Olig2; goat anti-mouse Alexa Fluor 594 (1:200, Invitrogen, Cat# A11005,RRID: AB_2534073) for MBP; and donkey anti-sheep Alexa Fluor 488 IgG (1:200, Invitrogen, Cat# A11015, RRID: AB_141362) for BrdU. Nuclei were counterstained with 4′,6-diamidino-2′-phenylindole (Solarbio) diluted in PBS (1 μg/mL) overnight at 4°C.

Quantification of axonal regeneration, synaptogenesis and remyelination

To observe axonal regeneration, the total length and number of crossing CST fibers in the spinal cord were detected. Five coronal spinal sections (C6–C8, every tenth section) from each rat were assessed by tracing the BDA-labeled CST fibers using confocal microscopic analysis (Olympus FV-1000, Tokyo, Japan). The total length and number of BDA-labeled CST axons were measured with a simple neurite tracer plugin in ImageJ 1.53t software (National Institutes of Health, Bethesda, MD, USA) (Schneider et al., 2012). BDA-labeled fibers crossing the midline were measured by counting intersections with lines M (midline of the spinal cord gray matter), D1 (one-third of the distance between the midline and the lateral border of the gray matter), and D2 (two-thirds of the distance between the midline and the lateral border of the gray matter) in the spinal cord contralateral to the ischemic site (Figure 2). To detect synaptogenesis, the expressions of synaptic markers (vGlut1, synaptophysin, PSD-95, and GAP-43) were measured. Six sections (C6–C8, every 20^th section) from each rat in the denervated gray matter of the spinal cord were captured at 20× magnification. The integrated density of pixels was measured using ImageJ, and an average density was calculated. Finally, to investigate the effects of CXCR7 antibody on remyelination after cerebral ischemia, the integrated density of NG2⁺ cells in the subventricular zone (SVZ), the proportion of Olig2⁺BrdU⁺ cells among Olig2⁺ cells, and the integrated density of MBP in the perilesional striatum were measured using ImageJ.

Figure 2.

Effects of CXCR7 antibody on the number of crossing CST fibers in the spinal cord of stroke model rats.

(A) Reconstruction of BDA-labeled CST fibers that sprouted across the midline into the denervated spinal cord. The number of CST axons in the ISC + anti-CXCR7 group was significantly increased compared with that in the ISC group in regions M, D1 and D2. (B) BDA-labeled fibers were quantified by counting all intersections with three lines (M, D1, and D2), averaged across multiple sections in the denervated spinal cord. (C) Quantification of the number of CST fibers crossing lines M, D1, and D2. Data are expressed as mean ± SEM (n = 5 per group). *P < 0.05, vs. Sham group; #P < 0.05, ##P < 0.01, vs. ISC group (one-way analysis of variance followed by least significant difference test). BDA: Biotin dextran amine; CST: corticospinal tract; CXCR7: C-X-C chemokine receptor type 7; D1: one-third of the distance between the midline and the lateral border of the gray matter; D2: two-third of the distance between the midline and the lateral border of the gray matter; ISC: cerebral ischemia; M: midline of the spinal cord gray matter.

Western blotting

Western blotting was used to detect the protein expression of the SDF-1/CXCR4/CXCR7 axis and RAS/ERK signaling pathway in the perilesional striatum or corresponding control areas in Sham rats. Samples from the striatum were lysed in lysis buffer containing a cocktail of protease inhibitors (KeyGen Biotech, Nanjing, China). Protein concentrations were assayed using a bicinchoninic acid reagent kit (Beyotime, Shanghai, China). Protein samples (30 μm) were resolved on 7–12% gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels and then transferred electronically to polyvinylidene fluoride membranes. Subsequently, these samples were blocked and then stained with the following primary antibodies overnight at 4°C: rabbit anti-SDF-1 (1:500, ImmunoWay, Plano, TX, USA, Cat# YT4225), rabbit anti-CXCR4 (1:500, ImmunoWay, Cat# YT1800), rabbit anti-CXCR7 (1:500, ImmunoWay, Cat# YT1162), rabbit anti-RAS (1:500, ImmunoWay, Cat# YT2906), mouse anti-extracellular signal-regulated kinases 1 and 2 (ERK1/2, 1:500, ImmunoWay, Cat#YM3677), and rabbit anti-phospho-extracellular signal-regulated kinases 1 and 2 (1:1000, Abcam, Cat# Ab76299, RRID: AB_1523577). On the second day, polyvinylidene fluoride membranes were stained using the following secondary antibodies at room temperature for 2 hours: goat anti-rabbit IgG (1:8000, Abbkine, Wuhan, China, Cat# A23920) and goat anti-mouse IgG (1:8000, Abbkine, Cat# A23921). Polyvinylidene fluoride membranes were visualized using the Odyssey FC Dual-Mode Imaging System (LI-COR Biosciences, Lincoln, NE, USA).

qRT-PCR

qRT-PCR was used to determine gene expression levels of the SDF-1/CXCR4/CXCR7 axis and RAS/ERK signaling pathway. Total RNA was extracted from peri-infarcted striatum with the miRNeasy Mini Kit (Qiagen, Hilden, Germany), and concentrations were quantified with a NanoDrop spectrophotometer (ND-1000, Thermo Scientific, Waltham, MA, USA). Total mRNA was reverse transcribed to complementary DNA with the GoScript Reverse Transcription System (Promega, Madison, WI, USA). Rat-specific primers were listed in Table 1. Expression levels of mRNA were detected with the 7900HT Fast Real-Time PCR system (Applied Biosystems, Waltham, MA, USA). Ct values were calculated using RQ Manager software (Biosystems Industries) and target gene expressions were calculated with the 2^–ΔΔCT value (Rao et al., 2013). GAPDH mRNA was used as an internal control and three replicates of the same samples were performed.

Table 1.

Sequence of the primers for polymerase chain reaction

Primer	Sequence (5’–3’)	Product length (bp)
SDF-1	Forward: TGT TGC AAG GCT GAA AAG CA	219
	Reverse: GAG CGT ACT GTG GTC CAT CTG
CXCR4	Forward: TGT TGC AAG GCT GAA AAG CA	219
	Reverse: GAG CGT ACT GTG GTC CAT CTG
CXCR7	Forward: TGT GAC ACG AAC CGT ATT GT	84
	Reverse: AAC TTA CAA ATC GAG AGA ATG CTG
RAS	Forward: GGG CTG GCT TTG CTG CTT CT	170
	Reverse: TGG CTT GTC TGG TGC GTT GAA T
Erk1	Forward: GCT CTG GTA GAC GGT TCT GG	88
	Reverse: GCT TAG CCT ACT CAG TGC CC
Erk2	Forward: CAG ATC TTA AAT TGG TCA GGA CAA G	131
	Reverse: CGG CTC AAA GGA GTC AAG AGT
GAPDH	Forward: ACA GCA ACA GGG TGG TGG AC	252
	Reverse: TTT GAG GGT GCA GCG AAC TT

CXCR4: C-X-C chemokine receptor 4; CXCR7: C-X-C chemokine receptor 7; Erk: extracellular signal-regulated kinase; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; SDF-1: stromal cell-derived factor-1.

Tapered/ledged beam-walking test for assessing motor functional recovery

Animal behavioral performance was assessed with the tapered/ledged beam-walking test as in our previous studies (Zhao et al., 2013a, b, 2015; Xu et al., 2017). On a standard beam without ledges, animals with a unilateral motor injury either fall or hang clinging to the beam due to misplacing the contralateral hindlimb with increasing frequency as the beam narrows. The animal must learn how to negotiate the beam skillfully without slipping by extensive practice. The learning of movements include novel use of the non-impaired limbs, tail deviations toward the impaired limbs to provide the balance needed for leaning on the non-impaired limbs, and other weight bearing adjustments to keep all four limbs on the beam, which is a major component of recovery. These behaviors successfully compensate for chronic impairments but make it difficult to assess whether the recovery is due to restorative brain repair or motor learning effects. The ledges of the tapered/ledged walking beam reveal the impairment and the recovery mechanisms or treatment effects by partially reducing the need for motor learning (Krieglstein and Klumpp, 2002). The rats received 3 days of pre-training prior to the stroke. The behavioral test was carried out after 4 weeks of ischemia on postoperative days 30–32 (Figure 1), which is in the chronic stage of stroke, as in our previous studies (Qu et al., 2015; Dong et al., 2020). Placement of a paw on the ledge was counted as a full slip and a paw in contact with the side of the beam was recorded as a half-slip. The performance was evaluated by calculating the slip ratios of the impaired forelimbs and hindlimbs.

Statistical analysis

Power analysis (G*Power 3.1.9.2) (Faul et al., 2007) was performed to predict required sample sizes given the mean difference and standard deviation observed in ischemic rats from our pilot experiment (α = 0.05, β = 0.8). The rater was blinded to the assignments. Data are presented as mean ± standard error of mean and were analyzed using SPSS 23.0 software (IBM, Armonk, NY, USA). One-way analysis of variance followed by least significant difference test was used to assess the statistically significant differences between groups. P < 0.05 was required for results to be considered significant.

Results

CXCR7 antibody promotes axonal regeneration after cerebral ischemia

After cerebral ischemia, CST fibers generate collateral branches from the intact spinal cord that sprout across the midline into the denervated side, which represents axonal regeneration (Benowitz and Carmichael, 2010). Immunofluorescence of BDA-labeled axons showed that the total length of the crossing CST fibers in the ISC group was significantly longer than that in the Sham group (P < 0.05). CXCR7 antibody treatment further increased the total length of CST fibers after cerebral ischemia (P < 0.05), and did not alter the length in Sham rats (P < 0.05; Figure 3). The crossing fiber number showed similar trends to that of the total length. The number of CST axons in the ISC + anti-CXCR7 group was significantly increased compared with that in the ISC group, both in regions M (P < 0.05) and D1 (P < 0.01). In region D2, the number of CST axons in the ISC group was significantly increased compared with that in the Sham group (P < 0.05), and treatment with CXCR7 antibody further strengthened this trend (P < 0.01; Figure 2).

Figure 3.

Effects of CXCR7 antibody on CST fiber sprouting in the spinal cord of stroke model rats.

(A) Immunofluorescence staining images of BDA-labeled CST fibers that crossed the midline and sprouted into the denervated gray matter. The total length of the crossing CST fibers in the spinal cord of the ischemic rats was significantly increased compared with that in the sham rats, and was further increased by CXCR7 antibody treatment. Dotted-line: midline of the spinal cord; arrows: BDA-positive CST fibers that sprouted into the denervated gray matter; arrowheads: BDA-labeled pyramidal tract in the nonlesional side of the spinal cord. Scale bar: 50 μm. (B) Quantification of the total length of CST fibers crossing the midline. Data are expressed as mean ± SEM (n = 5 per group). *P < 0.05, vs. Sham group; #P < 0.05, vs. ISC group (one-way analysis of variance followed by least significant difference test). BDA: Biotin dextran amine; CC: central canal; CST: corticospinal tract; CXCR7: C-X-C chemokine receptor type 7; ISC: cerebral ischemia.

CXCR7 antibody enhances synaptogenesis after cerebral ischemia

The effect of CXCR7 antibody on synaptogenesis was determined by examining the expression of synaptic markers, including vGlut1, synaptophysin, PSD-95, and GAP43, in the denervated gray matter of the spinal cord. Immunofluorescence staining showed that synaptophysin (P < 0.01), PSD-95 (P < 0.01), and GAP43 (P < 0.05) expression levels in the ischemic rats were all significantly lower, and vGlut1 expression was decreased but not significantly (P > 0.05) compared with those in the sham-operated rats. CXCR7 antibody treatment significantly increased vGlut1 and GAP43 expression levels in the ischemic rats (both P < 0.05; Figure 4), but did not influence synaptophysin and PSD-95 expression levels. None of the synaptic markers in Sham rats were influenced by the CXCR7 antibody.

Figure 4.

Effects of CXCR7 antibody on the levels of synaptic markers in the spinal cord of stroke model rats.

(A–D) Immunofluorescence staining images and quantification of the levels of vGlut1, synaptophysin (green, stained with Alexa Fluor 488), PSD-95 (green, stained with Alexa Fluor 488) and GAP-43 (green, stained with Alexa Fluor 488) in the gray matter of the denervated spinal cord. CXCR7 antibody treatment significantly increased the expression levels of vGlut1 and GAP43, but did not influence the expression levels of synaptophysin and PSD-95 in the ischemic rats. Scale bars: 50 μm. Data are expressed as mean ± SEM (n = 6 per group). *P < 0.05, **P < 0.01, vs. Sham group; #P < 0.05, vs. ISC group (one-way analysis of variance followed by least significant difference test). CXCR7: C-X-C chemokine receptor type 7; GAP-43: growth associated protein-43; ISC: cerebral ischemia; PSD-95: postsynaptic density-95; vGlut1: vesicular glutamate transporter 1.

CXCR7 antibody augments the differentiation and maturation of OPCs after cerebral ischemia

During the process of remyelination after brain injury, OPCs residing in the SVZ proliferate, migrate to the demyelinated area, and differentiate into mature oligodendrocytes to repair the impaired myelin sheaths (Franklin and Ffrench-Constant, 2008; Merino et al., 2015). NG2-positive glia are often equated to OPCs (Song et al., 2017), and are absent during the differentiation and maturation of oligodendrocytes (Goldman and Kuypers, 2015). Olig2, a marker for immature oligodendrocytes, is an essential activator in OPC differentiation and is critical for myelin sheath formation (Ligon et al., 2006). MBP is the marker of mature myelinating oligodendrocytes and a multifunctional protein that maintains the stability and integrity of the myelin sheath (Sarg et al., 2017; Brosolo et al., 2022). Immunofluorescence staining showed no significant differences in the integrated density of NG2 in the SVZ between the four groups (Figure 5A). After focal cerebral ischemia, the proportion of Olig2⁺BrdU⁺ cells among BrdU⁺ cells within the perilesional striatum was significantly higher (P < 0.01), whereas MBP expression was significantly lower (P < 0.05) compared with that in Sham rats. Anti-CXCR7 antibody treatment not only significantly upregulated the proportion of Olig2⁺BrdU⁺ cells (both P < 0.01; Figure 5B) but also greatly increased the MBP expression level in both ischemic rats and sham-operated rats (P < 0.05 and P < 0.01, respectively; Figure 5C).

Figure 5.

Effects of CXCR7 antibody on OPC proliferation in the SVZ and OPC differentiation and maturation in the perilesional striatum of stroke model rats.

(A) Immunofluorescence staining images and quantification of NG2 staining (red, Alexa Fluor 594) in the SVZ. (B) Representative fluorescent images and quantification of Olig2 (red, Alexa Fluor 594) double staining with BrdU (green, Alexa Fluor 488) in the perilesional striatum. (C) Representative fluorescent images and quantification of MBP staining (red, Alexa Fluor 594) in the perilesional striatum. Scale bars: 100 μm (A), 50 μm (B), 200 μm (C). Anti-CXCR7 antibody treatment upregulated the regeneration of Olig2⁺ cells and increased the MBP expression level in both ischemic rats, but did not alter the integrated density of NG2 in the SVZ after focal cerebral ischemia. Data are expressed as mean ± SEM (n = 6 per group). *P < 0.05, **P < 0.01, vs. Sham group; #P < 0.05, ##P < 0.01, vs. ISC group (one-way analysis of variance followed by least significant difference test). BrdU: 5-Bromo-2′-deoxyuridine; CXCR7: C-X-C chemokine receptor type 7; DAPI: 4′,6-diamidino-2′-phenylindole; ISC: cerebral ischemia; MBP: myelin basic protein; NG2: neuron-glial antigen 2; Olig2: oligodendrocyte lineage transcription factor 2; OPC: oligodendrocyte progenitor cell; SVZ: subventricular zone.

CXCR7 antibody upregulates SDF-1 and CXCR4 expression and downregulates CXCR7 expression after cerebral ischemia

To evaluate the influence of CXCR7 antibody on the SDF-1/CXCR4/CXCR7 axis, SDF-1, CXCR4, and CXCR7 mRNA and protein levels in the perilesional striatum were determined by qRT-PCR and western blotting, respectively. CXCR7 antibody significantly increased SDF-1 mRNA levels (P < 0.01), and the SDF-1 protein level was also increased by 16.8% (P > 0.05) after cerebral ischemia (Figure 6A and B). In addition, CXCR7 antibody increased CXCR4 mRNA levels in both the Sham + anti-CXCR7 and ISC + anti-CXCR7 groups (both P < 0.05; Figure 6). CXCR4 protein levels in these two groups were also significantly increased (Figure 6A and C). CXCR7 antibody treatment significantly upregulated CXCR7 mRNA levels in the sham-operated and ischemic rats (both P < 0.05; Figure 4), and significantly decreased CXCR7 protein expression after cerebral ischemia (P < 0.01; Figure 6A and D).

Figure 6.

Effects of CXCR7 antibody on mRNA and protein expression levels of the SDF-1/CXCR4/CXCR7 axis in the perilesional striatum of stroke model rats.

(A) Representative protein expression bands of SDF-1, CXCR4 and CXCR7 measured by western blot after focal cerebral ischemia. (B) Quantification of SDF-1, CXCR4 and CXCR7 protein expression (fold change relative to sham). (C) Quantification of SDF-1, CXCR4, and CXCR7 mRNA levels (fold change relative to sham). Data are expressed as mean ± SEM (n = 5 per group for western blot assay and n = 6 per group for polymerase chain reaction). *P < 0.05, **P < 0.01, vs. Sham group; #P < 0.05, ##P < 0.01, vs. ISC group (one-way analysis of variance followed by least significant difference test). CXCR4: C-X-C chemokine receptor 4; CXCR7: C-X-C chemokine receptor 7; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; ISC: cerebral ischemia; SDF-1: stromal cell-derived factor-1.

CXCR7 antibody activates the RAS/ERK signaling pathway after cerebral ischemia

To investigate the mechanism underlying the effects of the CXCR7 antibody, we assessed RAS, ERK1/2, and p-ERK1/2 expression by qRT-PCR and western blotting in the perilesional striatum. At 33 days poststroke, CXCR7 antibody treatment significantly upregulated RAS protein expression in ischemia rats (P < 0.05; Figure 7A and B). ERK1, but not ERK2, mRNA expression was elevated in both sham-operated rats (P < 0.01) and ischemic rats (P < 0.05) after CXCR7 antibody treatment (Figure 7A and C). ERK1/2 protein level showed no significant differences between groups. The p-ERK1/2 protein level was significantly elevated in the ISC group compared with the Sham group (P < 0.01), and CXCR7-neutralizing antibody treatment further increased the expression of p-ERK1/2 after cerebral ischemia (P < 0.05; Figure 7A and D).

Figure 7.

Effects of CXCR7 antibody on mRNA and protein expression levels of the RAS/ERK signaling pathway in the perilesional striatum of stroke model rats.

(A) Representative protein expression bands of RAS, ERK1/2 and p-ERK1/2 measured by western blot after focal cerebral ischemia. (B) Quantification of RAS, ERK1/2 and p-ERK1/2 protein expression (fold change relative to sham). (C) Quantification of RAS, Erk1, and Erk2 mRNA levels (fold change relative to sham). The relative expression was normalized by GAPDH. Data are expressed as mean ± SEM (n = 5 per group for western blot and n = 6 per group for polymerase chain reaction). *P < 0.05, **P < 0.01, vs. Sham group; #P < 0.05, vs. ISC group (one-way analysis of variance followed by least significant difference test). CXCR7: C-X-C chemokine receptor 7; ERK1/2: extracellular signal-regulated kinases1 and 2; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; ISC: cerebral ischemia; p-ERK1/2: phospho-extracellular signal-regulated kinases 1 and 2.

CXCR7 antibody promotes motor functional recovery after cerebral ischemia

A beam-walking test was performed on postoperative days 30–32 to evaluate motor functional recovery. Significant increases in slip ratios were observed in both impaired forelimbs and hindlimbs in the ISC group compared with the Sham group (both P < 0.01). CXCR7-neutralizing antibody reduced the slip ratios in the ISC + anti-CXCR7 group compared with the ISC group in both forelimbs and hindlimbs (both P < 0.01, Figure 8).

Figure 8.

Effects of CXCR7 antibody on the behavioral performance of stroke model rats using a beam-walking test.

Data are expressed as mean ± SEM (n = 8 per group). **P < 0.01, vs. Sham group; ##P < 0.01, vs. ISC group (one-way analysis of variance followed by least significant difference test). CXCR7: C-X-C chemokine receptor 7; ISC: cerebral ischemia.

Discussion

In this study, we showed for the first time that CXCR7 antibody increased the total length and number of regenerated CST fibers, enhanced the expression levels of synaptic markers in the spinal cord, and augmented the differentiation and maturation of OPCs in the perilesional striatum after focal cerebral ischemia in rats. Furthermore, the protein expressions of SDF-1, CXCR4, RAS and p-ERK1/2 in the perilesional striatum were increased, and the impaired beam walking test performance in ischemic rats was reversed. Our results indicate that CXCR7 antibody has the potential to enhance axonal regeneration, synaptogenesis and remyelination, and improve motor functional recovery after cerebral ischemia, which might occur through the RAS/ERK signaling pathway via CXCR4 activation.

In our study, CXCR7 antibody was used to block the binding of CXCR7 to its chemokine ligand SDF-1, as we have carried out previously (Dong et al., 2020). The results showed that infusion of CXCR7 antibody enhanced axonal regeneration in rats with cerebral ischemia, and had no influence on axonal regeneration in the sham-operated rats. Some evidence shows that SDF-1 can induce axon elongation and promote oligodendroglia differentiation through CXCR7-mediated ERK phosphorylation, which was reversed by CXCR7 mRNA silencing (Göttle et al., 2010; Liu et al., 2013). Our findings suggest that CXCR7 antibody promoted axonal regeneration through the ERK signaling pathway via CXCR4 activation when the function of CXCR7 was blocked by CXCR7-neutralizing antibody. In demyelinated animal models, CXCR7 antagonist has been shown to preserve axonal integrity by preventing leukocyte entry into the CNS parenchyma and limiting demyelination (Cruz-Orengo et al., 2011a). These conflicting effects of CXCR7 inhibition on axonal outgrowth and remyelination in different studies may be due to the different microenvironments in vitro and in vivo, and the different animal models, in which the formation of CXCR7/CXCR4 heterodimer and its downstream pathways are also different. Whether the effects of CXCR7 antibody on neural plasticity are mediated by activating the CXCR4 signaling pathway needs to be further investigated via experiments inhibiting CXCR4.

In addition to its role in axonal growth, CXCR4 function in synaptic formation has been investigated (Guyon et al., 2006; Bhattacharyya et al., 2008). However, the role of the CXCR7 signaling pathway in this process is still unclear. CXCR7 has been shown to possibly regulate synaptic currents directly by interacting with CXCR4 through heterodimerization, or indirectly through activating CXCR4 signaling pathways (Guyon, 2014). Investigation of essential synaptic markers, including presynaptic protein vGlut1, PSD-95, GAP-43, and synaptophysin can provide insight on synaptogenesis. GAP43 is considered to be the most important and representative structural protein, which is essential for presynaptic terminal formation, synaptic plasticity, axonal growth and regeneration (Merino et al., 2020). In the present study, anti-CXCR7 antibody administration reversed the decreased levels of GAP43 and vGlut1 within the denervated spinal cord in ischemic rats, suggesting that CXCR7 is at least indirectly involved in synaptogenesis after stroke.

Currently, most of the evidence on the effects of the SDF-1/CXCR7 axis on remyelination focuses on demyelinating disease. Previous studies showed that CXCR7 antagonism preserved axonal integrity and increased the proliferation of mature oligodendrocytes in demyelinated areas by increasing levels of CXCL12 and enhancing CXCR4 activation in animals (Cruz-Orengo et al., 2011a, b; Williams et al., 2014). OPC proliferation has been shown to occur during demyelination to repair the demyelinated lesions, which is then followed by a decrease of OPC number during remyelination (Levine and Reynolds, 1999). Our results demonstrated that the proliferation of NG2-positive cells in the SVZ was reduced by CXCR7 antibody in ischemic rats, but the effect was not significant, suggesting that demyelination was complete when the remyelination was occurring. Moreover, the regenerated Olig2⁺ cells and MBP expression were both increased in the peri-infarcted striatum of the ischemic rats after anti-CXCR7 antibody treatment, suggesting that CXCR7 promoted OPC differentiation and preserved myelin sheath integrity in the peri-infarcted area, thus providing a favorable local microenvironment for remyelination.

To further elucidate the underlying mechanism of the beneficial effects on neural plasticity induced by CXCR7 antibody, we investigated the expression of SDF-1, CXCR4, CXCR7, RAS, ERK1/2, and p-ERK1/2. CXCR7, as a scavenger receptor of SDF-1, reduces extracellular concentrations of SDF-1 and decreases CXCR4 levels and cellular sensitivity against SDF-1 by forming heterodimers with CXCR4 (Uto-Konomi et al., 2013; Abe et al., 2014). Our results demonstrated that SDF-1 and CXCR4 expression levels were increased, whereas CXCR7 protein expression was decreased after CXCR7 antibody treatment compared with ischemia alone, indicating that the CXCR7 antibody inhibited the function of the SDF-1/CXCR7 axis, and activated SDF-1/CXCR4 and its downstream pathway. ERK1/2, a downstream mediator of mitogen-activated protein kinases, is an important regulator for neurite outgrowth, axonal extension and integrity, and the formation and long-term maintenance of myelin (Ishii et al., 2014; Huang et al., 2017; Liu et al., 2017a; Jeffries et al., 2020; Shum et al., 2020). ERK1/2 has also been shown to participate in OPC regeneration, including proliferation, migration, differentiation, and survival (Liu et al., 2017b; Chen et al., 2018; Ilyasov et al., 2018; Pan et al., 2019). The MEK/ERK pathway is an essential downstream pathway of CXCR7 and CXCR4 heterodimers (Décaillot et al., 2011; Kumar et al., 2012), and often requires CXCR4 activation (Tian et al., 2018; Spinosa et al., 2021; Zhao et al., 2022). The RAS/MEK/ ERK pathway is thought to play an essential role in remyelination, neurogenesis, angiogenesis, and reducing apoptosis, thus improving functional recovery in the chronic phase after cerebral ischemia (Sawe et al., 2008; Lin et al., 2013; Yao et al., 2013). Our results showed that the protein levels of p-ERK1/2 and RAS were significantly upregulated after treatment with CXCR7 antibody compared with ischemia alone. Importantly, beam-walking test performance, which represents motor functional recovery, was also improved by CXCR7 antibody treatment after cerebral ischemia. Together, the above findings suggest that the CXCR7 antibody in this study exerted its effects through the activation of CXCR4 and its downstream RAS/ERK pathway in response to CXCR7 blockade, thus providing a more favorable local microenvironment for neural plasticity and functional recovery in the chronic stage after endothelin-1-induced focal cerebral ischemia.

Some inconsistencies were found between the protein and mRNA levels in our results. The discrepancy between CXCR7 protein level and mRNA level in the Sham + anti-CXCR7 and ISC + anti-CXCR7 groups might be related to the significant reduction of CXCR7 protein level induced by the CXCR7-neutralizing antibody, which possibly led to an increase of CXCR7 mRNA through feedback regulation. The discrepancy between CXCR7 protein and mRNA levels in the ISC group may be due to the CXCR7 mRNA having already degraded when the CXCR7 protein synthesis was at its peak after ischemia.

Finally, our study had several limitations. First, studies on other possible signaling pathways, as well as gain and loss experiments, are needed to elucidate the exact molecular mechanisms underlying the effects of CXCR7 antibody on neural plasticity. Second, CXCR7-neutralizing antibody used in this study may not completely inhibit CXCR7 function. A more comprehensive clarification of the specific effects of CXCR7 blockade in neural plasticity and whether these effects are mediated by activating the CXCR4 signaling pathway will be confirmed by experiments using RNA silencing or conditional knockout of these receptors. Third, many scientists consider female rodents to be messier and more variable compared with male rodents due to the fluctuation of hormones associated with the reproductive cycle, which could interfere with the costs and results of the experiments (Hughes, 2007; Beery and Zucker, 2011). However, females have recently been demonstrated to not be more variable than males in diverse biological traits or in gene expression (Prendergast et al., 2014; Itoh and Arnold, 2015; Becker et al., 2016). Exclusion of female animals from studies may limit generalization of the findings from males to females. Thus, both male and female rats should be included in future experiments to obtain more meaningful results. Our current results suggest a potential therapeutic approach in the treatment of brain ischemic injury based upon an animal model, but the limitations described above need to be addressed before translating the results into preclinical practice.

In conclusion, in this study, we demonstrated that CXCR7 antibody exerts multiple effects after cerebral ischemia, including promoting axonal regeneration, enhancing synaptogenesis, and participating in the remyelination process, which might occur through the CXCR4-activated RAS/ERK signaling pathway, and promotes motor functional recovery within the ischemic brain. The findings suggest that inhibition of CXCR7 may provide a promising new therapeutic target for neural plasticity and neurological functional recovery in the chronic stage of stroke.

Additional file: Open peer review reports 1 ^{(96.1KB, pdf)} –3 ^{(106.6KB, pdf)} .