ACKR3 is a hallmark of the diseased, but not developing CNS

Accepted Manuscript Astrocytic expression of the CXCL12 receptor, CXCR7/ACKR3 is a hallmark of the diseased, but not developing CNS

Malte Puchert, Fabian Pelkner, Gregor Stein, Doychin N. Angelov, Johannes Boltze, Daniel-Christoph Wagner, Francesca Odoardi, Alexander Flügel, Wolfgang J. Streit, Jürgen Engele PII: DOI: Reference:

S1044-7431(17)30152-5 doi: 10.1016/j.mcn.2017.09.001 YMCNE 3224

To appear in:

Molecular and Cellular Neuroscience

Received date: Revised date: Accepted date:

10 May 2017 9 August 2017 3 September 2017

Please cite this article as: Malte Puchert, Fabian Pelkner, Gregor Stein, Doychin N. Angelov, Johannes Boltze, Daniel-Christoph Wagner, Francesca Odoardi, Alexander Flügel, Wolfgang J. Streit, Jürgen Engele , Astrocytic expression of the CXCL12 receptor, CXCR7/ACKR3 is a hallmark of the diseased, but not developing CNS, Molecular and Cellular Neuroscience (2017), doi: 10.1016/j.mcn.2017.09.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Astrocytic expression of the CXCL12 receptor, CXCR7/ACKR3 is a hallmark of the diseased, but not developing CNS

Malte Pucherta*, Fabian Pelknera*, Gregor Steinb, Doychin N. Angelovc, Johannes Boltzed, Daniel-Christoph Wagnere, Francesca Odoardif, Alexander

PT

Flügelf, Wolfgang J. Streitg, Jürgen Engelea

SC

RI

*These authors contributed equally

a

Institute of Anatomy, University of Leipzig, Liebigstr. 13, 04103 Leipzig,

b

NU

Germany

Cologne University Hospital, Department of Orthopedic and Trauma Surgery,

MA

Uniklinik Köln, 50924 Köln, Germany c

Institute of Anatomy 1, University of Cologne, Joseph-Stelzmann-Str. 9, 50924

Department of Medical Cell Technology, Fraunhofer Research Institution for

PT E

d

D

Köln, Germany

Marine Biotechnology and Institute of Medical and Marine Biotechnology, University of Lübeck, Mönkhofer Weg 239a, 23562 Lübeck, Germany e

CE

Institute of Pathology, Building 706, Langenbeckstr. 1, 55131 Mainz, Germany

f

Institute for Multiple Sclerosis Research, Department of Neuroimmunology,

g

AC

University Medical Centre Göttingen, Waldweg 33, 37073 Göttingen, Germany Department of Neuroscience, University of Florida, 1149 Newell Dr,

Gainesville, FL 32610, USA

Corresponding Author: Jürgen Engele, Ph.D. E-mail: [email protected]¸; Phone: ..49/341 97 22071; Fax: ..49/341 97 22009

1

ACCEPTED MANUSCRIPT Abstract Based on our previous demonstration of CXCR7 as the major mediator of CXCL12 signalling in cultured astrocytes, we have now compared astrocytic expression of the CXCL12 receptors, CXCR7 and CXCR4, during CNS development and disease. In addition, we asked whether disease-associated

PT

conditions/factors affect expression of CXCL12 receptors in astrocytes. In the late embryonic rat brain, CXCR7+/GFAP+ cells were restricted to the

RI

ventricular/subventricular zone while CXCR4 was widely absent from GFAP-

SC

positive cells. In the early postnatal and adult brain, CXCR7 and CXCR4 were almost exclusively expressed by GFAP-immunoreactive astrocytes forming the

NU

superficial glia limitans. Contrasting the situation in the intact CNS, a striking increase in astrocytic CXCR7 expression was detectable in the cortex of rats

MA

with experimental brain infarcts, in the spinal cord of rats with experimental autoimmune encephalomyelitis (EAE) and after mechanical compression, as well as in the in infarcted human cerebral cortex and in the hippocampus of

PT E

D

Alzheimer’s disease patients. None of these pathologies was associated with substantial increases in astrocytic CXCR4 expression. Screening of various disease-associated factors/conditions further revealed that CXCR7 expression of

CE

cultured cortical astrocytes increases with IFN as well as under hypoxic conditions whereas CXCR7 expression is attenuated following treatment with

AC

IFN. Again, none of the treatments affected CXCR4 expression in cultured astrocytes. Together, these findings support the hypothesis of a crucial role of astrocytic CXCR7 in the progression of various CNS pathologies.

Key words: CXCR7, CXCR4, astrocytes, glia limitans, MCAO, EAE, spinal cord compression, cerebral infarction, Alzheimer’s disease

2

ACCEPTED MANUSCRIPT Introduction The SDF-1/CXCL12 chemokine system is indispensable for proper brain development and becomes (re)activated following brain inflammation and injury (Réaux-Le Goazigo et al., 2013; Li and Ransohoff, 2008). It is currently assumed that most of the effects of CXCL12 on brain development are mediated

PT

by CXCR4, which is considered the genuine CXCL12 receptor (Puchert and Engele, 2014). It is further assumed that during development the previously

RI

identified second CXCL12 receptor, CXCR7/ACKR3, acts primarily as a silent

SC

or scavenging chemokine receptor (Puchert and Engele, 2014). In line with this, animals deficient for either CXCL12 or CXCR4 show disturbed morphologies

NU

of cerebellum and hippocampal dentate gyrus, which result from derailed precursor/progenitor cell migration (Ma et al. 1998; Zou et al. 1998; Lu et al.

MA

2002; Klein et al. 2001). Both cerebellar and hippocampal morphology is, however, unaffected in animals deficient for CXCR7 (Sierro et al., 2007). On

D

the other hand, CXCR7-deficiency results in impaired migration and subsequent

PT E

misplacement of interneurons during cortical development (Sánchez-Alcañiz et al., 2011; Wang et al., 2011; Tiveron et al., 2010). This defect is attributed to the lack of modulatory influences of CXCR7 on the chemotactic function of

CE

CXCR4 (Abe et al., 2014). Contrasting with an apparently silent function in the developing brain, evidence has emerged that CXCR7 shows active cell

AC

signalling in the diseased brain. Indeed, we have previously shown that the chemotactic response of cultured microglia to CXCL12 occurs only in the presence of both CXCR7 and CXCR4 (Lipfert et al., 2013). Consistent with this in vitro finding, we have further documented the concomitant upregulation of CXCR7 and CXCR4 in microglia from the cortex of animals with experimental stroke (MCAO) (Lipfert et al., 2013). Moreover, in addition to CXCR4responsive neural progenitor cells which induce regeneration of the adult brain (Puchert and Engele, 2014), several studies have provided evidence for existence of CXCR7-responsive neural progenitors (Chen et al., 2015; Tiveron 3

ACCEPTED MANUSCRIPT et al., 2010). Finally, CXCR7 expressed by brain endothelial cells turns out to be essential for tumor vascularization (Liu et al., 2014). Based on slight modifications of the DRYLAIV motif, which is essential for recruiting and activating G proteins (Thelen and Thelen 2008), it is assumed that CXCR7 does not activate G-proteins and rather signals through ß-arrestin (Rajagopal et al.

PT

2010; Wang et al. 2011; Lee et al. 2013). Among other functions, astrocytes control synaptic plasticity (Haydon and

RI

Nedergaard, 2014), brain blood flow, brain energy metabolism (MacVicar and

SC

Newman, 2015; Bélanger et al., 2011), and neurotransmitter homeostasis (Schousboe et al., 2013) in the intact brain. Likewise, in the diseased central

NU

nervous system (CNS), (reactive) astrocytes play a multi-facetted role which includes the control of immunological processes, neuronal survival and

MA

regeneration, and (transcellular) permeability of the brain blood barrier (Burda and Sofroniew, 2014). Initial hints for a role of the CXCL12 system in the

D

control of (reactive) astrocytes emerged from the demonstration that cultured

PT E

astrocytes respond to CXCL12 with increased cell proliferation and migration (Ödemis et al., 2010; Bonavia et al., 2003; Tanabe et al., 1997), effects which could well contribute to glial scar formation in the injured brain (Filous and

CE

Silver, 2016). Moreover, we previously demonstrated that although cultured astrocytes express CXCR4 and CXCR7 to similar levels, CXCL12-dependent

AC

astrocytic responses are exclusively mediated through CXCR7 (Ödemis et al., 2010). Our analysis also revealed that despite its modified DRYLAIV motif, CXCR7 allows for the activation of G-proteins in cultured astrocytes, acting as a classical chemokine receptor in these cells (Ödemis et al., 2012). In contrast to cultured astrocytes, our understanding of the role of CXCR7 in astrocytes of the healthy and diseased CNS is still incomplete. This issue is further intrigued by the fact that the expression of the CXCR7 ligand, CXCL12, increases in various brain pathologies (Banisadr et al., 2016; Chen et al., 2016; Gu et al., 2016; Krumbholz et al., 2006; Ma et al., 2015; Robin et al., 2006; Sun et al., 2014). 4

ACCEPTED MANUSCRIPT The few studies presently available on CXCR7 in vivo demonstrate that the CXCR7 gene is transcribed in a subpopulation of astrocytes in the postnatal and adult brain and that this transcriptional process is not affected by experimental brain

infarct

or

experimental

autoimmune

encephalomyelitis

(EAE)

(Schönemeier et al. 2008a,b; Banisadr et al., 2016). The most robust upregulation of CXCR7 and/or CXCR4 observed to date occurs in brain tumors,

PT

especially gliomas, and is associated with enhanced tumor growth and

RI

metastasis (see for example Liu et al., 2013a; Ehtesham et al., 2013; Hattermann

SC

et al., 2010).

In the current study, we characterized and compared CXCR7 and CXCR4

NU

protein expression in astrocytes of the developing rat CNS as well as in the CNS of rats with MCAO, EAE, and spinal cord compression. In addition, we

MA

analyzed CXCR7 and CXCR4 expression in astrocytes of the hippocampus and cerebral cortex from patients suffering from Alzheimer’s disease and ischemic

D

stroke, respectively. The use of different disease models for our analysis was

PT E

spurred by the previous demonstration that different subpopulations of astrocytes exist within different parts of the CNS (John et al., 2017) and that astrocytic responses within a given part of the CNS may differ with respect to

CE

the type of injury (Zamanian et al., 2012). Finally, we identified disease-

AC

associated conditions controlling expression of CXCL12 receptors in astrocytes.

Materials and methods Developmental studies Brains were obtained from Sprague-Dawley rats (Charles River Laboratories ) at embryonic day 18 (E18), postnatal day 2 (P2), and postnatal day 60 (adult). For each time point at least 3 animals were used. E18 and P2 animals were decapitated and removed brains were immediately fixed in 4% formalin. Adult 5

ACCEPTED MANUSCRIPT animals were anesthetized with 4% isoflurane (Baxter, Unterschleißheim, Germany) and 2% Xylacin (Bayer, Monheim, Germany) /10% ketamine hydrochloride (Merial, Hallbergmoos, Germany) and transcardially perfused with 200 ml phosphate buffered solution followed by perfusion of 250 ml icecold 4% formalin solution. Brains were removed and fixed in 4% formalin solution for another 24 h, drained in ascending concentrations of sucrose for

RI

PT

three days, and frozen at -80°C.

Animal disease models

SC

Experimental stroke

NU

Spontaneously hypertensive rats (SHR, 250 g; Charles River Laboratories) were housed at constant temperature (+23°C) under a 12h light/dark cycle and ad

MA

libitum access to food and water. Experimental focal cerebral ischemia was induced by unilateral permanent occlusion of the middle cerebral artery (MCAO) as previously described (Lipfert et al., 2013). Briefly, animals were

D

anaesthetized with ketamine hydrochloride (100 mg/kg; Merial, Hallbergmoos,

PT E

Germany), Xylacin (10 mg/kg; Bayer, Monheim, Germany) and atropin (0.1 mg/kg; Ratiopharm, Ulm, Germany). The MCA was permanently occluded by

CE

electrocoagulation after sub-temporal craniotomy. Brain sections were prepared from animals 72h, 96h, and 168h after stroke onset, a time frame in which

AC

expression of CXCR7 and CXCR4 peaks (Lipfert et al., 2013). To this end, animals were euthanized with carbon dioxide, transcardially perfused, and processed as described above. Three animals were analyzed for each survival time. The contralateral, uninjured hemispheres of these animals served as controls.

6

ACCEPTED MANUSCRIPT Experimental autoimmune encephalomyelitis (EAE) CD4+ T cells reactive against myelin basic protein (TMBP) were generated and used for disease induction in Lewis rats as previously described (Flügel et al., 2001). All cell lines were CD4+, CD8− and αβTCR+, and were extensively tested for their phenotype, cytokine profile, antigen specificity and encephalitogenic potential. EAE was performed by i.v. injection of 5×106 blasts TMBP cells. The

PT

rats were weighed and scored daily according to a 5-grade scoring scale as

RI

described (Flügel et al., 2001). Animals were perfused for morphological analysis 4-5 days p.t., i.e. at the peak of the clinical symptoms. Analysis was

SC

performed with 3 EAE animals. The spinal cord from healthy animals served as

NU

a control (n = 3).

MA

Spinal cord injury

Spinal cord compression injury was induced in adult Wistar rats as previously described (Wirth et al., 2013; Ozsoy et al., 2012; 2013; Semler et al., 2011).

spinal

cord

compressed

(at

PT E

exposed

D

Briefly, a laminectomy was performed at the level of vertebra T8 and the segment

level

T10)

using

electromagnetically controlled watchmaker forceps. The intact spinal cord was

CE

lesioned by compressing its diameter by 50% with velocity of 100 mm/sec for 1s using a timed current. Rats were then housed individually in standard cages

AC

and the bladder voided manually three times a day. Animal were sacrificed 12 weeks post injury, a stage characterized by prominent increases in numbers of GFAP-positive astrocytes (Manthou et al., 2017). The spinal cords from 3 animals with compression injury were analyzed and compared to the spinal cord from healthy rats.

7

ACCEPTED MANUSCRIPT Human brain samples Formalin-fixed human brain tissue was obtained from the Leipzig brain bank. This included tissue from 3 Alzheimer’s patients, 3 non-Alzheimer’s patients, and 1 patient with subacute cerebral infarction (Table 1).

Histology and Immunohistochemistry

PT

Frozen tissue blocks were sectioned at 12-20 µm on a cryostat, formalin-fixed

RI

tissue was sectioned at 50 µm on a vibratome. For immunohistochemistry to glial fibrillary protein (GFAP), CXCR7, CXCR4 and NeuN tissue sections were

SC

permeabilized with 0.05% saponin and non-specific binding sites were blocked

NU

with 5% (v/v) donkey serum (GIBCO, Invitrogen, Carlsbad, CA) in PBS for 1h. Slices were incubated overnight in a humid chamber at 4 °C with monoclonal

MA

anti-CXCR7 (1:50; clone 11G8, R&D), polyclonal anti-CXCR4 (1:200; ab1670, ab2074, Abcam), polyclonal anti-GFAP (1:200; Z0334, DAKO), monoclonal anti-GFAP (1:200; 556330, BD Pharmingen) and polyclonal anti-NeuN (1:500;

D

ABN78; Millipore) antibodies in combinations as specified in the text. Antibody

PT E

labeling was detected by incubating slices for 2 h at room temperature with appropriate Alexa 488-labeled and Alexa 555-labeled donkey secondary

CE

antibodies (1:200; Invitrogen). Cell nuclei were counterstained with DAPI (DAKO). Sections were mounted with DAPCO/glycerol (DAKO) and examined

AC

under a Zeiss confocal laser scan microscope (LSM). Images represent single optical sections. For preparation of overview pictures, tissue slices were hematoxylin-eosin (HE)-stained using standard protocols. For quantitative protein analysis, tissue samples were lysed by ultrasonication in RIPA buffer (50 mM Tris-HCL pH 8.0, 150 mM NaCl, 1% NP-40, 0,5% sodium deoxycholate, 0.1% SDS), and subjected to Western blot analysis as outlined below.

8

ACCEPTED MANUSCRIPT Cortical astrocytes, treatment, and Western blot analysis Astrocytic cultures were established from postnatal day 3 Sprague Dawley rat cortices as described previously (Ödemis et al., 2012). Confluent cultures were switched to serum-free N2-medium 24 hours prior to experiments and were additionally treated for 24h with the following substances as specified in the text: interferon (IFN)-β (10 U/ml), IFN-γ (10 ng/ml), tumor necrosis factor

PT

(TNF)-α (10 ng/ml), epidermal growth factor (EGF; 10 ng/ml) or H2O2 (0.1

RI

µM). In selected experiments, cultures were maintained for 24 h at 1% O 2. Effects of these treatments on CXCR7, CXCR4 or HIF1-α expression were

SC

determined by Western blotting. To this end, cells were lysed by ultrasonication

NU

in 62.5 mM Tris-HCl, containing 2% SDS and 10% sucrose. Protein content of cell lysates was determined using the BCA protein estimation kit (Pierce;

MA

Rockford, Il) and bovine serum albumin (BSA) as a standard. Obtained proteins (15 µg/lane) were separated by SDS-(10%) polyacrylamide gel electrophoresis and transferred to nitrocellulose. Blots were incubated overnight at 4°C with

D

anti-CXCR7 antibody (1:300; AP17961PU-N; Acris), anti-CXCR4-antibody

PT E

(1:500; ab1670, Abcam) or anti-hypoxia inducible factor (HIF)1-α antibody (1:500; NB100-123F1; Novus Biologicals). Antibody labeling was detected with

CE

a horseradish peroxidase-labeled secondary antibody (1:10000; Jackson ImmunoResearch) and visualized with the enhanced chemiluminescence kit (GE

AC

Healthcare, Waukesha, WI). Staining of the blots with GAPDH (10R-G109A, Fitzgerald Industries International; 1:5000) or actin (1:1000; sc-7210; Santa Cruz Biotechnology) antibodies served as a loading control. Integrated optical densities of immunoreactive protein bands were measured using the Gel-Pro Analyzer software 3.1. (Media Cybernetics) and values were corrected for protein loading.

9

ACCEPTED MANUSCRIPT Immunostaining of cultured cells Cortical astrocytes were established and prepared as outlined above. Immunostaining was performed on 4% formalin fixed cell cultures. For permeabilization, cultured cells were incubated with ice cold ethanol/acetone (1:1) for 5 min. Non-specific binding sites were blocked with 5% (v/v) bovine serum albumin in PBS for 1h. Cultures were subsequently incubated overnight

PT

at 4°C with primary antibodies (anti-CXCR4, ab1670, Abcam, 1:100; anti-

RI

CXCR7, clone 11G8, R&D, 1:50), followed by a 1 h-incubation (37°C) with appropriate Alexa 488-labeled and Alexa 555-labeled secondary antibodies

SC

(1:200, Invitrogen, Carlsbad, CA). Cell nuclei were counterstained with DAPI

NU

(AAT Bioquest, Sunnyvale, CA) and cells were mounted with Glycergel (DAKO). Microscopy was performed on a Zeiss confocal laser scan microscope

MA

(LSM).

D

Statistics

PT E

Data, obtained from at least three independent experiments, are given as mean + SD. Data were subjected to one way analysis of variance (Anova) followed by pairwise multiple comparison procedures (Student-Newman-Keuls method). If

CE

appropriate, intergroup differences were assessed by paired t-test. Differences

Results

AC

with p<0.05 were considered significant.

Developmental expression of CXCR7 and CXCR4 in cortical astrocytes In the developing rodent cortex, the switch from neurogenesis to gliogenesis occurs between E16 to E18 (Molofsky and Deneen, 2015). Consequently, we focused on brains from E18 animals and older. Immunohistochemial analysis was performed with the CXCR7 antibody, mAb 11G8. The specificity of this antibody was previously confirmed (Berahovich et al, 2010a, b; Walters et al., 10

ACCEPTED MANUSCRIPT 2014). In addition, we used the CXCR4 antibody, ab1670; the selectivity of this latter antibody was confirmed by analyzing CXCR4 expression in cultured astrocytes transfected with predesigned CXCR4-siRNA (supplementary figure 1). To characterize astrocytic CXCR7 expression during brain development,

PT

we double-labeled sections from rat brains ranging from E18 to adult with antibodies against CXCR7 and GFAP. On E18 there was widespread and intense

RI

CXCR7-staining present throughout the cortical hemispheres (Fig. 1A). Double-

SC

labeling defined a distinct number of CXCR7+/GFAP+ cells within the ventricular/subventricular zone (Fig. 2A). In other areas of the cortical anlagen

NU

only very few CXCR7+/GFAP+ cells were detectable (Fig. 1A). Together these findings suggest that most astrocytes lack CXCR7 during prenatal development.

MA

In sections of the postnatal day 2 (P2) brain, cortical CXCR7 staining was widely reduced and mostly confined to a compact layer near the pial surface of

D

the cortical hemispheres (Fig. 1B). Deeper parts of the cerebral hemisphere

PT E

rarely showed specific labelling for CXCR7. The inner part of the superficial layer formed by CXCR7-immunopositive structures additionally exhibited strong staining for GFAP (Fig. 1B), thus, showing that in the early postnatal

CE

brain CXCR7 is expressed predominantly in astrocytes of the glia limitans. The

AC

absence of GFAP-staining from the outer part of the superficial layer formed by CXCR7-immunopositive structures suggests that CXCR7 is additionally present within meninges of the early postnatal brain. In adult cortical hemispheres, CXCR7 expression remained restricted (Fig. 1C). Moderate CXCR7 expression still occurred in GFAP-immunoreactive astrocytes forming the superficial glia limitans as well as in adjacent meninges (Fig 1C). Moderate to high CXCR7-immunostaining was additionally present within the ependymal lining of the ventricles (Fig. 2B). Moreover, CXCR7 was expressed by GFAP-immunoreactive cells, located underneath the ependymal 11

ACCEPTED MANUSCRIPT layer in the subventricular zone (Fig. 2B); these cells most likely represent B1 and/or B2-astrocytes (Platel and Bordey, 2016). Finally, in further accordance with previous studies, using either in-situ hybridization or mice carrying the GFP gene under control of the CXCR7 promoter (Banisadre et al., 2016; CruzOrengo et al., 2011; Schönemeier et al., 2008a), staining for native CXCR7 protein was present within the intima and adventitia of large (peripheral) blood

PT

vessels and partially within the endothelium of brain microvasculature

RI

(supplementary Fig. 2). CXCR7 expression was virtually undetectable in the perivascular glia limitans (supplementary Fig. 2) and neurons (supplementary

SC

Fig. 3).

NU

Additional double-labelling of E18 brain sections with antibodies against CXCR4 and GFAP showed prominent CXCR4-immunostaining in GFAP-

MA

negative cells forming the inner part of the ventricular zone (Fig. 2C). Doublelabelling with antibodies against CXCR4 and CXCR7 further revealed that this

D

layer of CXCR4-expressing cells does not overlap with the layer of CXCR7-

PT E

expressing cells (Fig. 2E). Specific immunostaining for CXCR4 was widely absent from other parts of the cortical anlage (Fig. 1D). In the postnatal and adult

cortex,

CXCR4

expression

was

mainly

restricted

to

GFAP-

CE

immunoreactive astrocytes of the glia limitans and adjacent meninges (Fig. 1E, F). Outside these structures, only few additional cells/astrocytes expressed low

AC

levels of CXCR4. Consistent with previous findings from in situ hybridization analysis (Stumm et al., 2002), CXCR4 expression further occurred over the ependymal lining although staining was extremely faint (Fig. 2D). The developmental decline of both CXCR7 and CXCR4 expression as revealed by immunohistochemistry was fully confirmed by Western blot analysis. Into adulthood, levels of CXCR7 and CXCR4 dropped by 80% and 86%, respectively (supplementary Fig. 4A).

12

ACCEPTED MANUSCRIPT Astrocytic expression of CXCR7 and CXCR4 in the diseased CNS For studying disease-related effects on astrocytic CXCR7 expression, we used the cortical hemisphere from rats with permanent unilateral MCAO as well as the spinal cord (white matter) from rats with EAE and spinal cord compression. We further investigated the cerebral cortex from patients with ischemic

PT

infarction and the hippocampus of patients with Alzheimer’s disease. To initially assess whether experimental brain infarct (MCAO) affects

RI

expression of CXCR7 and/or CXCR4, we stained brain sections obtained from

SC

animals 72h to 168h following unilateral MCAO with antibodies against either CXCR7 or CXCR4 (Fig. 3B, C). In the ipsilateral (injured) hemisphere, CXCR7

NU

and CXCR4 expression was sparse up to 72h following MCAO. Expression of both CXCL12 receptors prominently increased up to 96 h following MCAO and

MA

declined up to 168h. Expression of both CXCL12 receptors remained unchanged in the contralateral (intact) hemisphere during the time-course examined.

D

To determine whether MCAO-induced expression of CXCR7 and/or

PT E

CXCR4 occurs in astrocytes, we double-labeled cortical sections from animals 96h after unilateral MCAO with antibodies against GFAP and CXCR7 or CXCR4. Indicative for reactive gliosis, numbers of GFAP-immunoreactive cells

CE

increased in the area surrounding the infarct core (Fig. 4A, 5A). Similar

AC

increases in GFAP-immunoreactive cell numbers were absent in the contralateral (uninjured) hemisphere (Fig. 4B, 5B). In the infarcted hemisphere, intense and widespread CXCR7-labelling was present within a relatively broad band directly lining the infarct core (Fig. 4A). In contrast, CXCR4immunolabelling was mainly restricted to a layer located more distant to the infarct core (Fig. 5A). Both CXCR7- and CXCR4-immunolabelling was either absent or extremely low in uninjured control hemispheres (Fig. 4B, 5B). Double-labelling demonstrated that the majority of GFAP-immunoreactive astrocytes within the peri-infarct zone expressed CXCR7 whereas only few 13

ACCEPTED MANUSCRIPT GFAP-immunoreactive astrocytes showed staining for CXCR4 (Fig. 4A, 5A). The merged staining further revealed presence of a distinct number of CXCR7positive cells not expressing GFAP (Fig. 4A, C). The CXCR7- or CXCR4positive, but GFAP-negative cells most likely represent microglia/macrophages which are known to accumulate within the peri-infarct area (Lipfert et al., 2013). Finally, double-labelling for CXCR7 and CXCR4 showed that most CXCR4

PT

immunoreactive cells also expressed CXCR7 (Fig. 5C), suggesting that the

RI

small layer of CXCR4 expressing cells represents the outer part of the band

SC

formed by CXCR7 expressing cells.

The spinal cord from healthy rats contained moderate numbers of GFAP-

NU

immunoreactive astrocytes and rarely any of them exhibited immunostaining for CXCR7 or CXCR4 (Fig. 6, 7). In autoimmune CNS lesions of EAE (Fig. 6C, G)

MA

or following mechanical compression (Fig. 7B, H), numbers of GFAPimmunoreactive astrocytes were increased within the spinal cord white matter,

D

an effect that was more pronounced after compression. Most remarkably, both

PT E

pathological conditions were associated with striking increases in the number of CXCR7-immunoreactive cells (Fig. 6C, 7B). Double-labelling with GFAP demonstrated that irrespective of the pathology, CXCR7 expression prominently

CE

increased in astrocytes (Fig. 6C, D, 7B, C). It is noteworthy that in the compressed spinal cord astrocytic CXCR7 expression was highest close to the

AC

site of compression and declined with distance (Fig. 7B, D). In contrast to CXCR7, spinal cord compression remained without obvious effects on CXCR4 expression (Fig. 7H-K). In spinal cord from EAE animals CXCR4 expression increased, however, this increase occurred predominantly in GFAP-negative cells with roundish cell bodies, likely to be microglia/macrophages (Fig. 6G, H). As exemplified for EAE, increases of CXCR7 and CXCR4 in the autoimmune spinal cord were likewise detectable by Western blotting

14

ACCEPTED MANUSCRIPT (supplementary Fig. 4B), again underscoring the specificity of our immunohistochemical analysis. To further assess the size of the population of astrocytes expressing CXCR7 under the various pathological conditions, we determined the fraction of GFAP+/CXCR7+ cells present (1) in the band of reactive astrocytes surrounding

PT

the infarct core of MCAO animals, (2) in areas of the EAE spinal cord white matter exhibiting dense cellular infiltration, (3) in the area adjacent to the

RI

crushed spinal cord, and (4) in the cortex and spinal cord of healthy animals

SC

(Table 2). All disease conditions were associated with a prominent 7-13-fold increase in GFAP/CXCR7 double-labelled cells when compared to healthy

NU

controls (Table 2). Moreover, independent of the pathology, CXCR7 expression was present in a major population (61% to 88%) of (reactive) astrocytes (Table

MA

2).

To investigate whether expression of CXCR7 is likewise increased in

D

astrocytes of the diseased human CNS, we studied human brain material with

PT E

ischemic cerebral infarction (Fig. 8A), a condition closely resembling the MCAO animal model. We noted a clear increase in the number of GFAPimmunoreactive astrocytes in the area adjacent to the infarct/lesion core (Fig.

CE

8B, D). Most of these GFAP-immunoreactive cells showed double-labelling for

AC

CXCR7 (Fig. 8B, C), but not for CXCR4 (Fig. 8F, G). Moreover, GFAP/CXCR7 double-labelled cells were rarely detectable in the contralateral uninjured hemisphere of the same individual (Fig. 8D, E). The demonstration of astrocytic CXCR7 expression in the infarcted human brain, prompted us to additional determine whether astrocytic CXCR7 expression would likewise occur in the Alzheimer’s brain, which again is characterized by prominent reactive gliosis (Rodriguez-Arellano et al., 2016, for review). In accordance with these previous reports, numbers of GFAPimmunoreactive astrocytes were distinctly higher in the hippocampus from 15

ACCEPTED MANUSCRIPT patients with Alzheimer’s disease than in the hippocampus of age-matched nondemented control individuals (Fig. 9). While CXCR7-immunolabelling was low in control hippocampi, hippocampal CXCR7 staining was readily detectable in Alzheimer’s disease (Fig. 9A, B). Merged images revealed that CXCR7 expression prevails in astrocytes (Fig. 9A). A similar increase in astrocytic CXCR7 expression occurred in the hippocampus from patients with early-onset

PT

AD, which showed no or less pronounced signs of reactive gliosis

RI

(supplementary Fig. 5). Interestingly, CXCR4-immunostaining was hardly detectable in the hippocampi from both AD and control patients (Fig. 9C,

SC

supplementary Fig. 5).

NU

Collectively, these findings establish that in both rodents and humans diverse CNS pathologies result in the predominant expression of CXCR7, but

MA

not of CXCR4, in astrocytes.

D

Influences of disease-associated conditions on astrocytic CXCR7 and

PT E

CXCR4 expression

In additional experiments, we set out to identify factors mediating the diseaseassociated increases in CXCR7 expression in reactive astrocytes. Cultures of

CE

astrocytes from postnatal rat cortex served as an assay system in these experiments. Specifically, we focused on the cytokines, TNF, IFN, INF as

AC

well as on H and hypoxia, factors/conditions known to be involved in numerous CNS pathologies (Wang et al., 2015; Doll et al., 2014; Issazadeh et al., 1995; Sin et al., 2012). We further included EGF in this experiment as it was shown previously to activate astrocytes in the injured brain and to induce expression of CXCR4 in cancer cells (Phillips et al., 2005; Liu et al., 2006). Consistent with our prior findings (Ödemis et al., 2010), Western blot analysis demonstrated presence of similar levels of CXCR7 and CXCR4 in cultures of rat cortical astrocytes (Fig. 10). Reminiscent of the situation in the injured CNS, immunostaining further showed widespread expression of CXCR7 in cultured 16

ACCEPTED MANUSCRIPT astrocytes (suppl. Fig. 6); this observation is consistent with the long-standing view that cultured astrocytes attain distinct reactive traits (Pike et al., 1994). Moreover, discrepant from our findings in the injured CNS, most CXCR7positive astrocytes in culture co-expressed CXCR4 (suppl. Fig. 6). Treatment with IFN (10 U/ml; 24h) resulted in a 90% loss in CXCR7 expression (10A,

PT

B). By contrast, IFN (10 ng/ml; 24 h) and hypoxia (1% O2; 24h) allowed for a 3.5-fold and 2.8-fold increase in CXCR7 expression, respectively (Fig. 10A, B).

RI

Corroborating previous work (Esencay et al., 2013; Liu et al., 2010), the

SC

hypoxia-dependent induction of astrocytic CXCR7 expression was accompanied by an approximately 3-fold increase in HIF1- levels (Fig. 10D). CXCR7

NU

expression was not affected by TNF (10 ng/ml, 24h), H2O2 (0.1 µM; 24h) or EGF (10 ng/ml; 24h) (Fig. 10A, B). Most intriguingly, none of treatments

MA

affected expression of astrocytic CXCR4 (Fig. 10A, C). Collectively, these findings establish that at least in vitro distinct disease-associated conditions either selectively promote or inhibit CXCR7, but not CXCR4 expression in

PT E

D

astrocytes.

CE

DISCUSSION

The expression and function of CXCR7 is well documented in cultured

AC

astrocytes. However, our knowledge on the role of this chemokine receptor in astrocytes in vivo is still sparse. In the present studies, we sought to characterize and compare the expression profiles of CXCR7 and CXCR4 in astrocytes of the developing and adult brain. In addition, we analyzed whether astrocytic CXCR7 and CXCR4 expression would change under various brain pathologies, which with exception of Alzheimer’s disease (Parachikova and Cotman, 2007; Laske et al., 2008) were previously found to result in increased expression of CXCL12 (Banisadr et al., 2016; Chen et al., 2016; Gu et al., 2016; Krumbholz et al., 2006; Ma et al., 2015; Robin et al., 2006; Sun et al., 2014). We demonstrate that 17

ACCEPTED MANUSCRIPT in the postnatal and adult rat brain CXCR7 and CXCR4 are predominantly expressed by astrocytes forming the glia limitans. In addition, we show that in both rodents and humans numerous CNS pathologies are associated with robust increases in the expression of CXCR7, but not of CXCR4 in (reactive) astrocytes. Finally, we identify the disease-associated cytokines, IFNand IFN

PT

as well as hypoxia as potent modulators of astrocytic CXCR7 expression. Collectively, our findings point to a previously unrecognized role of astrocytic

RI

CXCR7 in the diseased CNS

SC

Developmental expression of CXCL12 receptors in astrocytes and other

NU

brain structures

Although immunostaining of the late embryonic rat brain revealed that both

MA

CXCR7 and CXCR4 expression are widely absent from GFAP-immunoreactive cells, implying that the CXCL12 system does not affect early (prenatal) steps of astrocyte development. The exception was a small number of CXCR7+/GFAP+

D

cells, residing within the ventricular/subventricular zone; these cells could

PT E

represent neural/glial progenitor cells (Kriegstein and Alvarez-Buylla, 2009). By contrast, intense CXCR7-immunolabelling was present within numerous GFAP-

CE

negative cells. These cells most likely represent cortical neurons, migrating GABAergic or olfactory interneurons or endothelial cells, which reportedly

AC

express CXCR7 during brain development (Sánchez-Alcañiz et al., 2011; Tiveron et al., 2010; Schönemeier et al., 2008a). One of the rare structures exhibiting intense CXCR4 expression in the late embryonic brain was the ventricular zone/neuroepithelium, which is consistent with previous findings from in situ hybridization analysis of the prenatal mouse brain (Tissir et al., 2004). A recognized function of the CXCL12-CXCR4 axis in the neuroepithelium consists in the modulation of neural stem cells expansion (Li et al., 2011). It is noteworthy that the CXCR4-positive layer within the ventricular zone showed no overlap with CXCR7 staining exhibited by the outer part of the 18

ACCEPTED MANUSCRIPT ventricular and/or subventricular zone. Whether this reflects the differential expression of CXCR4 and CXCR7 by neural stem cells and progenitor cells remains to be established. In accordance with previous in-situ hybridization studies (Stumm et al., 2002; Tissir et al., 2004; Schönemeier et al., 2008a), immunostaining of the

PT

healthy postnatal and adult rat brain revealed a generally restricted expression of CXCR7 and CXCR4. Shortly after birth, expression of CXCR7 and CXCR4

RI

became detectable within a subpopulation of astrocytes forming the superficial

SC

glia limitans, which persisted into adulthood. In the adult brain, few additional CXCR7+/GFAP+ cells were present underneath the ependymal lining of

NU

ventricles which might represent B-astrocytes (Platel and Bordey, 2016), cells reported to possess stem cell properties (Lin and Iacovitti, 2015). Outside these

MA

areas, very few astrocytes within the postnatal and adult brain showed low levels of CXCR7 and hardly any CXCR4. Astrocytic borders are thought to restrict

D

leukocyte migration into the brain (Sofroniew, 2015). Moreover, previous work

PT E

established that CXCR7 controls entry of leukocytes into the brain through endothelial barriers (Cruz-Orengo et al., 2011), a fact reflected by the prominent expression of CXCR7 in microvasculature of the adult brain. It is thus tempting

CE

to speculate that CXCR7 and CXCR4 are additionally involved in the control of leukocyte trafficking through astrocytic brain borders. In the adult rat brain, we

AC

further noticed prominent CXCR7 expression in the ependymal lining of ventricles, a structure previously shown to express CXCR7 transcripts (Banisadre et al., 2016; Schönemeier et al., 2008a). Intriguingly, previous work provided evidence that leukocytes transmigrate through the ependymal lining of ventricles (Alvarez and Teale, 2007). We consequently suggest that CXCL12 receptors not only control leukocyte trafficking through astroglial and endothelial borders, but also through ventricular lining.

19

ACCEPTED MANUSCRIPT Astrocytic expression of CXCL12 receptors in the diseased CNS To date, numerous studies reported the constitutive expression of CXCL12 in the intact CNS and the increased expression of the chemokine in animals suffering from stroke, traumatic brain injury, and EAE (Banisadr et al., 2016; Ma et al., 2015; Robin et al., 2006; Sun et al., 2014). Moreover, all pathological conditions are likewise associated with increased CXCL12 levels in humans as

PT

determined either in the CNS or plasma (Chen et al., 2016; Gu et al., 2016;

RI

Krumbholz et al., 2006). While so far expression of the alternate ligand for CXCR7, CXCL11, was found to increase in EAE-animals (McColl et al., 2004),

SC

expression levels of this chemokine have not yet been analyzed in other CNS

NU

pathologies. In our present analysis, we found restricted expression of the CXCL12 receptor, CXCR7, in astrocytes of the healthy adult rat CNS and robust

MA

increases in astrocytic CXCR7 expression, involving a major population of reactive astrocytes, in various rat disease models, such as MCAO, EAE, and spinal cord compression. We further found a likewise restricted expression of

D

CXCR7 in astrocytes of the uninjured human cerebral cortex and hippocampus

PT E

and pronounced increases in astrocytic CXCR7 expression in the infarcted as well as Alzheimer’s brain. A previous study analyzing MCAO animals by in-

CE

situ hybridization failed to detect increases in CXCR7 transcript levels in astrocytes within the peri-infarct area (Schönemeier et al., 2008b), implying that

AC

the observed disease-associated increases in astrocytic CXCR7 expression might occur predominantly at the translational level. Surprisingly, overall expression as well as astrocytic expression of CXCR4 remained unaffected in rat spinal cord compression, human cerebral infarction, and Alzheimer’s disease. Moreover, whereas in accordance with previous studies (Schönemeier et al., 2008b; Lipfert et al., 2013, Banisadr et al., 2016) overall expression of CXCR4 increased in the cortex from MCAO animals and after EAE induction, CXCR4 expression in GFAP-positive astrocytes again remained largely unaffected. The increased expression of CXCR4 seen in these animals seems to be mainly due to 20

ACCEPTED MANUSCRIPT CXCR4-positive microglia and blood-borne leukocytes invading the peri-infarct area (Lipfert et al., 2013; Ruscher et al., 2013). Collectively, we believe our findings indicate that in vivo CXCL12 signals to astrocytes via CXCR7. This notion fits well with our previous demonstration of CXCR7 as the main mediator of CXCL12-signalling in cultured astrocytes (Ödemis et al., 2010).

PT

Disease-associated conditions affecting expression of CXCL12 receptors in

RI

astrocytes

Screening of several disease-associated conditions for effects on CXCR7

SC

expression in cultured cortical astrocytes demonstrated increased CXCR7

NU

expression following treatment with IFNand a loss of CXCR7 expression following treatment with IFNIFN primarily, but not exclusively, acts as a

MA

proinflammatory cytokine (Ottum et al., 2015). In astrocytes, IFNsignaling seems to contribute to the recruitment of inflammatory cells to the CNS by controlling leukocyte entry through the blood-brain-barrier and/or by the

D

enhanced production of chemokines (Ottum et al., 2015). By contrast, IFN

PT E

promotes production of anti-inflammatory chemokines and cytokines and is used as a first-line treatment in relapsing forms of MS (Kieseier, 2011). The high

levels

of

astrocytic

CXCR7

under

proinflammatory

CE

observed

(IFNtreatment), but not anti-inflammatory conditions (IFN treatment) could

AC

point to the existence of a secondary CXCR7-dependent immune response that contributes to the inflammatory loop (Pugazhenthi et al., 2013). In addition to IFN and IFN, astrocytic CXCR7 expression robustly increased following hypoxia. A similar hypoxia-dependent induction of CXCR7 expression has been previously reported for several other cell types, including glioma cell lines and hippocampal progenitor cells (Esencay et al., 2013; Liu et al., 2013b). Interestingly, hypoxia remained without notable effects on astrocytic CXCR4 expression, which is unexpected because: (1) hypoxia is known to induce CXCR4 expression in various types of tumor cells as well as primary cells (see 21

ACCEPTED MANUSCRIPT for example Zagzag et al., 2006; Guo et al., 2014; Romain et al., 2014; Liu et al., 2010). (2) Hypoxia further increased levels of HIF in cultured astrocytes. (3) The promoter regions of the CXCR7 and CXCR4 genes reportedly contain multiple HREs (Tarnowski et al., 2010). We suggest that the failure of hypoxia to affect astrocytic CXCR4 expression might reflect the cell type-specific

PT

regulation of CXCL12 receptors by oxygen. Indeed, whereas hypoxia promotes expression of CXCR4 and CXCR7 in mesenchymal stem cells (Liu et al., 2010)

RI

it only increases CXCR4 expression in colon cancer cells (Romain et al., 2014).

SC

Following CNS injury, activated (reactive) astrocytes form a border around the lesion area which potentially shields the non-injured CNS from toxic

NU

conditions prevailing in the lesion area (Pekny and Pekna, 2014; Burda and Sofroniew, 2014). Reactive astrocytes further limit neurodegeneration (Liu and

MA

Chopp. 2015) and induce repair of the blood-brain-barrier (Verkhratsky and Butt, 2013). Finally, in the diseased CNS, bidirectional interactions take place

D

between astrocytes and microglia/monocytes which profoundly impact

PT E

immunological responses (Sofroniew, 2014). Previous in vitro studies already demonstrated that CXCL12-CXCR7 promote proliferation and migration of astrocytes and might, thus, control glial scar formation (Ödemis et al., 2010;

CE

2012). Future studies have to address which of above mentioned processes are affected by astrocytic CXCR7 in situ and how these CXCR7-controlled

AC

processes impact the clinical outcome of brain pathologies.

Acknowledgement We thank Florian Kirmse for skillful technical assistance. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. 22

ACCEPTED MANUSCRIPT Statements Informed consent was obtained for experimentation with human subjects. Animal experiments were carried out according to the EU Directive 2010/63/EU

AC

CE

PT E

D

MA

NU

SC

RI

PT

for animal experiments and were approved by local authorities

23

ACCEPTED MANUSCRIPT Table 1 Characteristics of human brain tissue used for analysis. Age

AD

69

Male/Female Amyloid plaques f +

AD

52

m

+

AD

62

f

+

Non-AD

52

m

-

Non-AD

50

m

Non-AD

70

m

-

cerebral infarction (subacute)

49

f

RI

SC

NU MA D PT E CE AC

24

Tau Cause of death plaques pancreas ca, multiple + organ dysfunction esophageal varices, + multiple organ dysfunction breast ca, multiple + organ dysfunction pancreatitis, multiple organ dysfunction pneumonia, heart failure -

PT

Disease

-

heart failure deep vein thrombosis, acute cor pulmonale

ACCEPTED MANUSCRIPT Table 2 Quantification of CXCR7-expressing astrocytes in the various disease models Compressed spinal cord

Healthy cortex

MCAO cortex

7+2

78+5

88+5

9+1

61+6

<0.001

<0.001

PT

EAE spinal cord

<0.001

RI

Fraction (%+SD) of GFAP-positive astrocytes expressing CXCR7 p-value

Healthy spinal cord

SC

GFAP/CXCR7-double labeled cells were counted in histological sections of the spinal cord and cortex from healthy and diseased animals (n=3) as outlined in

NU

the text and referred to the total number of GFAP-positive cells present in the same area. The analysis comprised a total of 42-115 GFAP-positive cells per

AC

CE

PT E

D

MA

section.

25

ACCEPTED MANUSCRIPT Legends to figures Figure 1 Astrocytic expression of CXCR7 and CXCR4 during cortical development Coronal sections were obtained from the cerebral cortex of (A, B) E18, (C, D) P2, and (E, F) adult rats and double-labeled with antibodies against GFAP and

PT

CXCR7 (A, C, E) or CXCR4 (B, D, F). DAPI was used for staining cell nuclei.

RI

At E18, CXCR7, but not CXCR4, is expressed by a large number of GFAPnegative cells within the cortical hemispheres. In postnatal and adult cortical

SC

hemispheres, CXCR7 and CXCR4 expression is mainly restricted to GFAP-

NU

immunoreactive astrocytes forming the glia limitans (arrow) and to adjacent

MA

meninges (m). Dashed lines mark brain surface. Scale bars, 25 µm.

Figure 2 Developmental expression of CXCR7 and CXCR4 within ventricular wall

D

Periventricular area in coronal sections of the E18 (A, B, E) and adult (C, D) rat

PT E

brain, double-labelled with antibodies against GFAP and CXCR7 (A, C), GFAP and CXCR4 (B, D), and CXCR7 and CXCR4 (E). Cell nuclei are visualized with DAPI. During late embryonic development, moderate CXCR7 expression

CE

occurs in GFAP-positive cells within the ventricular (VZ)/subventricular zone (SZ), In the adult brain, moderate to intense CXCR7 staining is exhibited by the

AC

ependymal lining (arrow head) and GFAP-expressing cells, reminiscent of Bastrocytes, within the subventricluar zone (arrow). In the late embryonic brain, prominent CXCR4 expression is present within (neuroepithelial) cells forming the inner part of the ventricular zone (VZ). In the adult cortex very faint CXCR4 staining is restricted to the ependymal lining (arrow). CXCR7/CXCR4 doublelabelling of E18 brain sections further demonstrated that CXCR4 staining does not overlap with CXCR7 staining within the ventricular/subventricular zone. V, ventricle. Scale bars, 25 µm. 26

ACCEPTED MANUSCRIPT Figure 3 Time-course of CXCR7 and CXCR4 expression following experimental brain infarct Rats were subjected to unilateral permanent middle cerebral artery occlusion (MCAO), sacrificed after the indicated times, and brains were removed. (A) Overview of HE-stained coronal brain section obtained 96h following MCAO.

PT

The white dashed line outlines the infarct core (IC). The contralateral, uninjured

RI

area served as control. Boxed areas within the injured and non-injured hemisphere mark the sites were high magnification images were taken. (B, C)

SC

Results from staining coronal brain sections with antibodies against CXCR7 and

NU

CXCR4. In the injured hemisphere, expression of both CXCL12 receptors was still low after 72h following MCAO, clearly increased at 96h, and declined up to

MA

168 h. Scale bars, 100 µm.

Figure 4

D

MCAO is associated with increased expression of CXCR7 in astrocytes

PT E

Coronal sections were obtained from the cortex of rats 96h following MCAO and double-labelled with CXCR7 and GFAP antibodies. (A) In the infarcted

CE

hemisphere, GFAP and CXCR7 expression robustly increased in the area adjacent to the infarct core (IC). Merging of the channels demonstrated that

AC

GFAP-immunoreactive astrocytes within the peri-infarct area co-express CXCR7. (B) GFAP and CXCR7 remained unchanged in the contralateral (control) hemisphere. (C) Close-up view of the peri-infarct area revealing the additional presence of a population of GFAP-/CXCR7+ cells. DAPI was used for nuclear staining. Arrows in (C) denote examples of GFAP+/CXCR7+ cells and arrowheads examples of GFAP-/CXCR7+ cells. Scale bars in (A) and (B), 100 µm; scale bar in (C), 25 µm.

27

ACCEPTED MANUSCRIPT Figure 5 MCAO has restricted effects on CXCR4 expression in astrocytes (A) Staining of coronal sections from the cortex of rats 96h following MCAO with antibodies against CXCR4 and GFAP, demonstrated increased CXCR4 expression in a cell layer at the outer rim of the peri-infarct region. Merging the staining subsequently demonstrated that CXCR4 expression does occur in a very

PT

small number of GFAP-immunoreactive astrocytes. (B) The contralateral

RI

(control) hemispheres showed unchanged low staining for GFAP and CXCR4. (C) Double-labelling for CXCR4 and CXCR7 revealed that virtually all CXCR4

SC

expressing cells within the outer rim of the peri-infarct region also express CXCR7. DAPI was used for staining cell nuclei. Arrows in (C) denote examples

NU

of CXCR4+/CXCR7+ cells and arrowheads examples of CXCR4-/CXCR7+ cells.

MA

Scale bars in (A) and (B), 100 µm. Scale bar in (C), 25 µm.

Figure 6

D

EAE promotes astrocytic expression of CXCR7, but not of CXCR4

PT E

(A, B) Overview of HE-stained cross sections prepared from the spinal cord of EAE rats at the peak of the disease (A) and healthy controls (B). (A’, B’) show

lesioned sites.

CE

boxed areas at higher magnification. Note the dense cell infiltration within

AC

(C-J) Spinal cord sections were obtained from EAE (C, D, G, H) and healthy animals (E, F, I, J) and double-labelled with antibodies against GFAP and CXCR7 (C-F) or GFAP and CXCR4 (G-J). D, F, H, J show cells at high magnification in double-labelled sections. DAPI was used for staining cell nuclei. EAE animals exhibited a small, but distinct increase in numbers of GFAP-expressing astrocytes when compared to healthy controls. EAE was further associated with a clear cut increase in CXCR7 expression which was to a large part associated with GFAP-immunoreactive astrocytes. Similar to CXCR7, the spinal cord from EAE animals showed increased numbers of CXCR4 28

ACCEPTED MANUSCRIPT expressing cells. Most of these CXCR4 positive cells, however, did not coexpress GFAP, and possessed a roundish cell body, reminiscent of microglia/macrophages. Scale bar in C, E, G, I, 50 µm. Scale bar D, F, H, J, 10 µm.

Figure 7

PT

Spinal cord compression results in the selective upregulation of CXCR7,

RI

but not of CXCR4 in (reactive) astrocytes

Longitudinal sections were obtained from the spinal cord of rats with

SC

mechanical compression (A-E, H, I) and healthy individuals (F, G, J, K). (A)

NU

Overview of a histological section obtained from the compressed spinal cord and stained with HE. Boxed areas mark the sites proximal and distal to the

MA

compressed area (CA) were high magnification pictures have been taken. Sections were stained with a combination of GFAP and CXCR7 (B-G) or GFAP and CXCR4 (H-K) antibodies. (C, E, G, I, K) Close-up views of labelled cells.

D

Robust increases in GFAP immunoreactive cell numbers occurred adjacent to

PT E

the CA. A major portion of GFAP positive cells present in this area also expressed increased levels of CXCR7, but not of CXCR4. Scale bars in B, D, F,

AC

Figure 8

CE

H, J, 50 µm and in C, E, G, I, K, 10 µm.

CXCR7 expression increases in astrocytes of the infarcted human brain Brain tissue was excised from the area adjacent to the lesion core (boxed area in A) and sectioned. Tissue sections from the contralateral uninjured hemisphere of the same individual served as control. Sections were double-labeled with GFAP and either CXCR7 (B-E) or CXCR4 (F, G) antibodies. (C, E, G) Close-up views of labelled cells. Note that GFAP-immunoreactive astrocytes are accumulated adjacent to the infarct core and express CXCR7, but not CXCR4 following

29

ACCEPTED MANUSCRIPT cerebral infarction. Further note that most astrocytes lack CXCR7 expression in the control hemisphere. Scale bars in B, D, F, 50 µm and in C, E, G, 10µm.

Figure 9 Astrocytic CXCR7 expression is increased in the hippocampus from Alzheimer’s patients

PT

Hippocampal sections from patients with AD and from age-matched non-AD

RI

individuals were double labeled with (A, B) GFAP and CXCR7 or (C) GFAP and CXCR4 antibodies. The AD hippocampus contained higher numbers of

SC

GFAP-immunoreactive cells when compared to controls. Likewise, hippocampi

NU

from AD patients exhibited profound increases in CXCR7 expression which mainly occurred in GFAP-positive astrocytes. CXCR4 expression was almost

MA

absent in both AD and non-AD hippocampi. Scale bars, 25 µm.

Figure 10

D

Screening of disease-associated factors/conditions for effects on CXCL12

PT E

receptor expression in cultured cortical rat astrocytes Astrocyte cultures were established from the cerebral hemispheres of P2 rats and

CE

grown to confluency in serum-containing (10%) MEM-medium. Twenty four hours prior to experiments, cultures were switched to serum-free N2-medium,

AC

and maintained for additional 24h in N2-medium additionally supplemented with either INF (10 U/ml), INF (10 ng/ml), TNF (10 ng/ml), EGF (10 ng/ml), or H2O2 (0.1 µM). Cells were subsequently lysed and subjected to Western blot analysis for CXCR7 or CXCR4. GAPDH or actin served as loading controls. In selected experiments, confluent cultures were exposed to hypoxia (1% O2, 24h). (A) Representative blots for the various treatments on CXCR7 and CXCR4 expression. (B, C) Densitometric analyses of (B) CXCR7and (C) CXCR4-immunoreactive protein bands. (D) Characterization of the effects of hypoxia (1% O2, 24h) on astrocytic HIF1- levels. Bars show relative 30

ACCEPTED MANUSCRIPT changes (means + SD) of CXCR7 and CXCR4 protein levels as determined in 35 independent experiments. Levels of CXCR7, CXCR4, and HIF1- present in untreated controls were set to 1 (dashed lines in B and C). **p<0.001, *p<0.05;

PT

treatment vs. control.

References

RI

Abe P, Mueller W, Schütz D, MacKay F, Thelen M, Zhang P, Stumm R. CXCR7 prevents excessive CXCL12-mediated downregulation of CXCR4 in

NU

SC

migrating cortical interneurons. Development 2014; 141: 1857-1863

Alvarez JI, Teale JM. Differential changes in junctional complex proteins

MA

suggest the ependymal lining as the main source of leukocyte infiltration into ventricles in murine neurocysticercosis. J Neuroimmunol 2007; 187: 102-113

D

Banisadr G, Podojil JR, Miller SD, Miller RJ. Pattern of CXCR7 Gene

PT E

Expression in Mouse Brain Under Normal and Inflammatory Conditions. J

CE

Neuroimmune Pharmacol 2016; 11: 26-35

Bélanger M, Allaman I, Magistretti PJ. Brain energy metabolism: focus on

AC

astrocyte-neuron metabolic cooperation. Cell Metab 2011; 14: 724-738

Berahovich RD, Penfold ME, Schall TJ. Nonspecific CXCR7 antibodies. Immunol Lett 2010a; 133: 112-114

Berahovich RD, Zabel BA, Penfold ME, Lewén S, Wang Y, Miao Z, Gan L, Pereda J, Dias J, Slukvin II, McGrath KE, Jaen JC, Schall TJ. CXCR7 protein is not expressed on human or mouse leukocytes. J Immunol 2010b 185: 5130-5139

31

ACCEPTED MANUSCRIPT Bonavia R, Bajetto A, Barbero S, Pirani P, Florio T, Schettini G. Chemokines and their receptors in the CNS: expression of CXCL12/SDF-1 and CXCR4 and their role in astrocyte proliferation. Toxicol Lett 2003; 139: 181-189

Burda JE, Sofroniew MV. Reactive gliosis and the multicellular response to

PT

CNS damage and disease. Neuron 2014; 81: 229-248

RI

Chen TJ, Wu WQ, Ying GR, Fu QY, Xiong K. Serum CXCL12 concentration in patients with severe traumatic brain injury are associated with mortality. Clin

NU

SC

Chim Acta 2016;454:6-9

Chen Q, Zhang M, Li Y, Xu D, Wang Y, Song A, Zhu B, Huang Y, Zheng JC.

MA

CXCR7 Mediates Neural Progenitor Cells Migration to CXCL12 Independent of CXCR4. Stem Cells 2015; 33: 2574-2585

D

Cruz-Orengo L, Holman DW, Dorsey D, Zhou L, Zhang P, Wright M,

PT E

McCandless EE, Patel JR, Luker GD, Littman DR, Russell JH, Klein RS. CXCR7 influences leukocyte entry into the CNS parenchyma by controlling

AC

339

CE

abluminal CXCL12 abundance during autoimmunity. J Exp Med 2011; 208:327-

Doll DN, Barr TL, Simpkins JW. Cytokines: their role in stroke and potential use as biomarkers and therapeutic targets. Aging Dis 2014; 5: 294-306

Ehtesham M, Min E, Issar NM, Kasl RA, Khan IS, Thompson RC. The role of the CXCR4 cell surface chemokine receptor in glioma biology. J Neurooncol 2013; 113: 153-162

32

ACCEPTED MANUSCRIPT Esencay M, Sarfraz Y, Zagzag D. CXCR7 is induced by hypoxia and mediates glioma cell migration towards SDF-1α. BMC Cancer 2013; 13: 347

Filous AR, Silver J. Targeting astrocytes in CNS injury and disease: A translational research approach. Prog Neurobiol 2016; doi: 10.1016/j.pneurobio

PT

Flügel A, Berkowicz T, Ritter T, Labeur M, Li Z, Ellwart JW, Jenne D, Willem

RI

M, Lassmann H, Wekerle H. Migratory activity and functional changes of genetically engineered green fluorescent effector T cells before and during

NU

SC

experimental autoimmune encephalomyelitis. Immunity 2001; 14: 547-560

Gu XL, Liu L, Lu XD, Liu ZR. Serum CXCL12 Levels as a Novel Predictor of

Neurobiol 2016; 53:2807-2814

MA

Future Stroke Recurrence in Patients with Acute Ischemic Stroke. Mol

D

Guo M, Cai C, Zhao G, Qiu X, Zhao H, Ma Q, Tian L, Li X, Hu Y, Liao B, Ma

PT E

B, Fan Q. Hypoxia promotes migration and induces CXCR4 expression via HIF-

CE

1α activation in human osteosarcoma. PLoS One 2014; 9: e90518

Hattermann K, Held-Feindt J, Lucius R, Müerköster SS, Penfold ME, Schall TJ,

AC

Mentlein R. The chemokine receptor CXCR7 is highly expressed in human glioma cells and mediates antiapoptotic effects. Cancer Res 2010: 70: 32993308

Haydon PG, Nedergaard M. How do astrocytes participate in neural plasticity? Cold Spring Harb Perspect Biol 2014; 7: a020438

Issazadeh S, Ljungdahl A, Höjeberg B, Mustafa M, Olsson T. Cytokine production in the central nervous system of Lewis rats with experimental 33

ACCEPTED MANUSCRIPT autoimmune encephalomyelitis: dynamics of mRNA expression for interleukin10, interleukin-12, cytolysin, tumor necrosis factor alpha and tumor necrosis factor beta. J Neuroimmunol 1995; 61: 205-212

John Lin CC, Yu K, Hatcher A, Huang TW, Lee HK, Carlson J, Weston MC, Chen F, Zhang Y, Zhu W, Mohila CA, Ahmed N, Patel AJ, Arenkiel BR,

PT

Noebels JL, Creighton CJ, Deneen B. Identification of diverse astrocyte

RI

populations and their malignant analogs. Nat Neurosci 2017; 20: 396-405

NU

sclerosis. CNS Drugs 2011; 25: 491-502

SC

Kieseier BC. The mechanism of action of interferon-β in relapsing multiple

Klein RS, Rubin JB, Gibson HD, DeHaan EN, Alvarez-Hernandez X, Segal RA,

MA

Luster AD. SDF-1 alpha induces chemotaxis and enhances sonic hedgehoginduced proliferation of cerebellar granule cells. Development 2001; 128: 1971–

D

1981

PT E

Kriegstein A, Alvarez-Buylla A. The glial nature of embryonic and adult neural

CE

stem cells. Annu Rev Neurosci 2009; 32: 149-184

Krumbholz M, Theil D, Cepok S, Hemmer B, Kivisäkk P, Ransohoff RM,

AC

Hofbauer M, Farina C, Derfuss T, Hartle C, Newcombe J, Hohlfeld R, Meinl E. Chemokines in multiple sclerosis: CXCL12 and CXCL13 up-regulation is differentially linked to CNS immune cell recruitment. Brain 2006; 129:200-211

Laske C, Stellos K, Stransky E, Seizer P, Akcay O, Eschweiler GW, Leyhe T, Gawaz M. Decreased plasma and cerebrospinal fluid levels of stem cell factor in patients with early Alzheimer's disease. J Alzheimers Dis 2008; 15:451-460

34

ACCEPTED MANUSCRIPT Lee E, Han J, Kim K, Choi H, Cho EG, Lee TR. CXCR7 mediates SDF1induced melanocyte migration. Pigment Cell Melanoma Res 2013; 26: 58–66

Li M, Chang CJ, Lathia JD, Wang L, Pacenta HL, Cotleur A, Ransohoff RM. Chemokine receptor CXCR4 signaling modulates the growth factor-induced cell cycle of self-renewing and multipotent neural progenitor cells. Glia 2011; 59:

RI

PT

108-118

Li M, Ransohoff RM. Multiple roles of chemokine CXCL12 in the central

SC

nervous system: a migration from immunology to neurobiology. Prog Neurobiol

NU

2008; 84: 116-131

MA

Lin R, Iacovitti L. Classic and novel stem cell niches in brain homeostasis and repair. Brain Res 2015; 1628: 327-342

D

Lipfert J, Ödemis V, Wagner DC, Boltze J, Engele J. CXCR4 and CXCR7 form

PT E

a functional receptor unit for SDF-1/CXCL12 in primary rodent microglia.

CE

Neuropathol Appl Neurobiol 2013; 39: 667-680

Liu B, Chen H, Johns TG, Neufeld AH. Epidermal growth factor receptor

AC

activation: an upstream signal for transition of quiescent astrocytes into reactive astrocytes after neural injury. J Neurosci 2006; 26: 7532-7540

Liu C, Pham K, Luo D, Reynolds BA, Hothi P, Foltz G, Harrison JK. Expression and functional heterogeneity of chemokine receptors CXCR4 and CXCR7 in primary patient-derived glioblastoma cells. PLoS One 2013a; 8:e59750

35

ACCEPTED MANUSCRIPT Liu H, Xue W, Ge G, Luo X, Li Y, Xiang H, Ding X, Tian P, Tian X. Hypoxic preconditioning advances CXCR4 and CXCR7 expression by activating HIF-1α in MSCs.Biochem Biophys Res Commun 2010; 401: 509-515

Liu S, Jia X, Li C, Han X, Yan W, Xing Y. CXCR7 silencing attenuates cell adaptive response to stromal cell derived factor 1α after hypoxia. PLoS One

RI

PT

2013b; 8: e55290

Liu Y, Carson-Walter E, Walter KA. Chemokine receptor CXCR7 is a

SC

functional receptor for CXCL12 in brain endothelial cells. PLoS One 2014; 9:

NU

e103938

MA

Liu Z, Chopp M. Astrocytes, therapeutic targets for neuroprotection and neurorestoration in ischemic stroke. Prog Neurobiol 2015; pii: S0301-

D

0082(15)30001-0

PT E

Lu M, Grove EA, Miller RJ. Abnormal development of the hippocampal dentate gyrus in mice lacking the CXCR4 chemokine receptor. Proc Natl Acad Sci U S

CE

A 2002; 99: 7090–7095

AC

Ma J, Liu N, Yi B, Zhang X, Gao BB, Zhang Y, Xu R, Li X, Dai Y. Transplanted hUCB-MSCs migrated to the damaged area by SDF-1/CXCR4 signaling to promote functional recovery after traumatic brain injury in rats. Neurol Res 2015; 37:50-56

Ma Q, Jones D, Borghesani PR, Segal RA, Nagasawa T, Kishimoto T, Bronson RT, Springer TA. Impaired B-lymphopoiesis, myelopoiesis, and derailed cerebellar neuron migration in CXCR4- and SDF-1-deficient mice. Proc Natl Acad Sci U S A 1998; 95: 9448–9453 36

ACCEPTED MANUSCRIPT

MacVicar BA, Newman EA. Astrocyte regulation of blood flow in the brain. Cold Spring Harb Perspect Biol 2015; 7:pii: a020388

Manthou M, Abdulla DS, Pavlov SP, Jansen R, Bendella H, Nohroudi K, Stein G, Meyer C, Ozsoy O, Ozsoy U, Behram Kandemir Y, Sarikcioglu L, Semler O,

PT

Schoenau E, Dunlop S, Angelov DN. Whole body vibration (WBV) following

RI

spinal cord injury (SCI) in rats: Timing of intervention. Restor Neurol Neurosci.

SC

2017; 35:185-216

Gladue

RP,

Neote

Willenborg

during

central

DO.

nervous

Expression system

of

rat

I-

inflammation:

MA

TAC/CXCL11/SCYA11

KS,

NU

McColl SR, Mahalingam S, Staykova M, Tylaska LA, Fisher KE, Strick CA,

comparison with other CXCR3 ligands. Lab Invest 2004; 84:1418-1429

PT E

Glia 2015; 63: 1320-1329

D

Molofsky AV, Deneen B. Astrocyte development: A Guide for the Perplexed.

CE

Ödemis V, Boosmann K, Heinen A, Küry P, Engele J. CXCR7 is an active component of SDF-1 signalling in astrocytes and Schwann cells. J Cell Sci

AC

2010; 123: 1081-1088

Ödemis V, Lipfert J, Kraft R, Hajek P, Abraham G, Hattermann K, Mentlein R, Engele J. The presumed atypical chemokine receptor CXCR7 signals through G(i/o) proteins in primary rodent astrocytes and human glioma cells. Glia 2012; 60: 372-381

37

ACCEPTED MANUSCRIPT Ottum PA, Arellano G, Reyes LI, Iruretagoyena M, Naves R. Opposing Roles of Interferon-Gamma on Cells of the Central Nervous System in Autoimmune Neuroinflammation. Front Immunol 2015; 6: 539

Ozsoy O, Ozsoy U, Schubert H, Abdulla D, Dunlop SA, Angelov DN, Irintchev A, Schönau E. Whole-body vibration improves functional recovery in spinal

RI

PT

cord injured rats. J Neurotrauma 2013; 30:453-468

Ozsoy O, Ozsoy U, Stein G, Semler O, Skouras E, Schempf G, Wellmann K,

SC

Wirth F, Angelova S, Ankerne J, Ashrafi M, Schönau E, Papamitsou-

NU

Sidoropolou T, Jaminet P, Sarikcioglu L, Irintchev A, Dunlop SA, Angelov DN. Functional deficits and morphological changes in the neurogenic bladder match

MA

the severity of spinal cord compression. Restor Neurol Neurosci 2012; 30: 363381

D

Parachikova A, Cotman CW. Reduced CXCL12/CXCR4 results in impaired

PT E

learning and is downregulated in a mouse model of Alzheimer disease.

CE

Neurobiol Dis. 2007; 28:143-153

Pekny M, Pekna M. Astrocyte reactivity and reactive astrogliosis: costs and

AC

benefits. Physiol Rev 2014; 94: 1077-1098

Phillips RJ, Mestas J, Gharaee-Kermani M, Burdick MD, Sica A, Belperio JA, Keane MP, Strieter RM. Epidermal growth factor and hypoxia-induced expression of CXC chemokine receptor 4 on non-small cell lung cancer cells is regulated by the phosphatidylinositol 3-kinase/PTEN/AKT/mammalian target of rapamycin signaling pathway and activation of hypoxia inducible factor-1alpha. J Biol Chem 2005; 280: 22473-22481

38

ACCEPTED MANUSCRIPT Pike CJ, Cummings BJ, Monzavi R, Cotman CW. Beta-amyloid-induced changes in cultured astrocytes parallel reactive astrocytosis associated with senile plaques in Alzheimer's disease. Neuroscience. 1994; 63: 517-531

Platel JC, Bordey A. The multifaceted subventricular zone astrocyte: From a metabolic and pro-neurogenic role to acting as a neural stem cell. Neuroscience

RI

PT

2016; 323: 20-28

Puchert M, Engele J. The peculiarities of the SDF-1/CXCL12 system: in some

SC

cells, CXCR4 and CXCR7 sing solos, in others, they sing duets. Cell Tissue Res

NU

2014; 355: 239-253

Pugazhenthi S, Zhang Y, Bouchard R, Mahaffey G. Induction of an

MA

inflammatory loop by interleukin-1β and tumor necrosis factor-α involves NFkB and STAT-1 in differentiated human neuroprogenitor cells. PLoS One 2013;

PT E

D

8: e69585

Rajagopal S, Kim J, Ahn S, Craig S, Lam CM, Gerard NP, Gerard C, Lefkowitz

CE

R. Beta-arrestin- but not G protein-mediated signaling by the "decoy" receptor

AC

CXCR7. Proc Natl Acad Sci U S A 2010; 107: 628–632

Réaux-Le Goazigo A, Van Steenwinckel J, Rostène W, Mélik Parsadaniantz S. Current status of chemokines in the adult CNS. Prog Neurobiol 2013; 104: 6792

Robin AM, Zhang ZG, Wang L, Zhang RL, Katakowski M, Zhang L, Wang Y, Zhang C, Chopp M. Stromal cell-derived factor 1alpha mediates neural progenitor cell motility after focal cerebral ischemia. J Cereb Blood Flow Metab 2006; 26:125-134 39

ACCEPTED MANUSCRIPT

Rodríguez-Arellano JJ, Parpura V, Zorec R, Verkhratsky A. Astrocytes in physiological aging and Alzheimer's disease. Neuroscience 2016; 323:170-82

Romain B, Hachet-Haas M, Rohr S, Brigand C, Galzi JL, Gaub MP, Pencreach E, Guenot D. Hypoxia differentially regulated CXCR4 and CXCR7 signaling in

RI

PT

colon cancer. Mol Cancer 2014; 13: 58

Ruscher K, Kuric E, Liu Y, Walter HL, Issazadeh-Navikas S, Englund E,

SC

Wieloch T. Inhibition of CXCL12 signaling attenuates the postischemic immune

NU

response and improves functional recovery after stroke. J Cereb Blood Flow

MA

Metab 2013; 33: 1225-1234

Sánchez-Alcañiz JA, Haege S, Mueller W, Pla R, Mackay F, Schulz S, LópezBendito G, Stumm R, Marín O. Cxcr7 controls neuronal migration by regulating

PT E

D

chemokine responsiveness. Neuron 2011; 69: 77-90

Schönemeier B, Kolodziej A, Schulz S, Jacobs S, Hoellt V, Stumm R. Regional

CE

and cellular localization of the CXCl12/SDF-1 chemokine receptor CXCR7 in

AC

the developing and adult rat brain. J Comp Neurol 2008a; 510: 207-220

Schönemeier B, Schulz S, Hoellt V, Stumm R. Enhanced expression of the CXCl12/SDF-1 chemokine receptor CXCR7 after cerebral ischemia in the rat brain. J Neuroimmunol 2008b; 198 :39-45

Schousboe A, Bak LK, Waagepetersen HS. Astrocytic Control of Biosynthesis and Turnover of the Neurotransmitters Glutamate and GABA. Front Endocrinol (Lausanne) 2013; 4: 102

40

ACCEPTED MANUSCRIPT Semler J, Wellmann K, Wirth F, Stein G, Angelova S, Ashrafi M, Schempf G, Ankerne J, Ozsoy O, Ozsoy U, Schönau E, Angelov DN, Irintchev A. Objective measures of motor dysfunction after compression spinal cord injury in adult rats: correlations with locomotor rating scores. J Neurotrauma 2011; 28: 1247-1258

Sierro F, Biben C,Martinez-Munoz L,Mellado M, Ransohoff RM, Li M, Woehl

PT

B, Leung H, Groom J, Batten M, Harvey RP, Martínez-A C, Mackay CR,

RI

Mackay F. Disrupted cardiac development but normal hematopoiesis in mice deficient in the second CXCL12/SDF-1 receptor, CXCR7. Proc Natl Acad Sci U

SC

S A 2007; 104: 14759–14764

NU

Sin WX, Li P, Yeong JP, Chin KC. Activation and regulation of interferon-β in

MA

immune responses. Immunol Res 2012; 53: 25-40

PT E

Neurosci 2015; 16: 249-263

D

Sofroniew MV. Astrocyte barriers to neurotoxic inflammation. Nat Rev

Sofroniew MV. Multiple roles for astrocytes as effectors of cytokines and

CE

inflammatory mediators. Neuroscientist 2014; 20: 160-172

AC

Stumm RK, Rummel J, Junker V, Culmsee C, Pfeiffer M, Krieglstein J, Höllt V, Schulz S. A dual role for the SDF-1/CXCR4 chemokine receptor system in adult brain: isoform-selective regulation of SDF-1 expression modulates CXCR4dependent neuronal plasticity and cerebral leukocyte recruitment after focal ischemia. J Neurosci 2002; 22: 5865-5878

Sun W, Liu J, Huan Y, Zhang C. Intracranial injection of recombinant stromalderived factor-1 alpha (SDF-1α) attenuates traumatic brain injury in rats. Inflamm Res 2014; 63:287-297 41

ACCEPTED MANUSCRIPT

Tanabe S, Heesen M, Yoshizawa I, Berman MA, Luo Y, Bleul CC, Springer TA, Okuda K, Gerard N, Dorf ME. Functional expression of the CXCchemokine receptor-4/fusin on mouse microglial cells and astrocytes. J Immunol 1997; 159: 905-911

PT

Tarnowski M, Grymula K, Reca R, Jankowski K, Maksym R, Tarnowska J,

stromal-derived

factor-1

receptors:

CXCR4

RI

Przybylski G, Barr FG, Kucia M, Ratajczak MZ. Regulation of expression of and

CXCR7

in

human

NU

SC

rhabdomyosarcomas. Mol Cancer Res 2010; 8: 1-14

Thelen M, Thelen S. CXCR7, CXCR4 and CXCL12: an eccentric trio? J

MA

Neuroimmunol 2008; 198: 9-13

Tissir F, Wang CE, Goffinet AM. Expression of the chemokine receptor Cxcr4

D

mRNA during mouse brain development. Brain Res Dev Brain Res 2004; 149:

PT E

63-71

Tiveron MC, Boutin C, Daou P, Moepps B, Cremer H. Expression and function

CE

of CXCR7 in the mouse forebrain. J Neuroimmunol 2010; 224: 72-79

AC

Verkhratsky A, Butt A. Astroglia. In: Glial Physiology and Pathophysiology.Eds A Verkhratsky, AM Butt. Hoboken, NJ: Wiley, 2013, 105–231.

Walters MJ, Ebsworth K, Berahovich RD, Penfold ME, Liu SC, Al Omran R, Kioi M, Chernikova SB, Tseng D, Mulkearns-Hubert EE, Sinyuk M, Ransohoff RM, Lathia JD, Karamchandani J, Kohrt HE, Zhang P, Powers JP, Jaen JC, Schall TJ, Merchant M, Recht L, Brown JM. Inhibition of CXCR7 extends survival following irradiation of brain tumours in mice and rats. Br J Cancer 2014; 110: 1179-1188 42

ACCEPTED MANUSCRIPT

Wang WY, Tan MS, Yu JT, Tan L. Role of pro-inflammatory cytokines released from microglia in Alzheimer's disease. Ann Transl Med 2015; 3: 136

Wang Y, Li G, Stanco A, Long JE, Crawford D, Potter GB, Pleasure SJ, Behrens T, Rubenstein JL. CXCR4 and CXCR7 have distinct functions in

RI

PT

regulating interneuron migration. Neuron 2011; 69: 61-67

Wirth F, Schempf G, Stein G, Wellmann K, Manthou M, Scholl C, Sidorenko

SC

M, Semler O, Eisel L, Harrach R, Angelova S, Jaminet P, Ankerne J, Ashrafi M,

NU

Ozsoy O, Ozsoy U, Schubert H, Abdulla D, Dunlop SA, Angelov DN, Irintchev A, Schönau E. Whole-body vibration improves functional recovery in spinal

MA

cord injured rats. J Neurotrauma 2013; 30: 453-468

Zagzag D, Lukyanov Y, Lan L, Ali MA, Esencay M, Mendez O, Yee H, Voura

D

EB, Newcomb EW. Hypoxia-inducible factor 1 and VEGF upregulate CXCR4

PT E

in glioblastoma: implications for angiogenesis and glioma cell invasion. Lab

CE

Invest 2006; 86: 1221-1232

Zamanian JL, Xu L, Foo LC, Nouri N, Zhou L, Giffard RG, Barres BA.

AC

Genomic analysis of reactive astrogliosis. J Neurosci 2012; 32: 6391-6410

Zou YR, Kottmann AH, Kuroda M, Taniuchi I, Littman DR. Function of the chemokine receptor CXCR4 in haematopoiesis and in cerebellar development. Nature 1998; 393: 595–599

43

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

44

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

45

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

46

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

47

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

48

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

49

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

50

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

51

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

52

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

53

ACCEPTED MANUSCRIPT Highlights - CXCR7 and CXCR4 are normally only present in astrocytes forming the glia limitans. - In the diseased CNS, astrocytes prominently express CXCR7, but not CXCR4.

PT

- IFNIFNand hypoxia modulate astrocytic expression of CXCR7, but

AC

CE

PT E

D

MA

NU

SC

RI

not of CXCR4.

54