www.nature.

com/scientificreports

OPEN Neurotoxic potential of reactive

astrocytes in canine distemper
demyelinating leukoencephalitis
Received: 18 October 2018 J. Klemens1, M. Ciurkiewicz   1,3, E. Chludzinski1,3, M. Iseringhausen1, D. Klotz1,
Accepted: 25 July 2019 V. M. Pfankuche1,3, R. Ulrich2,3, V. Herder1,3, C. Puff1, W. Baumgärtner1,3 & A. Beineke1,3
Published: xx xx xxxx
Canine distemper virus (CDV) causes a fatal demyelinating leukoencephalitis in young dogs resembling
human multiple sclerosis. Astrocytes are the main cellular target of CDV and undergo reactive changes
already in pre-demyelinating brain lesions. Based on their broad range of beneficial and detrimental
effects in the injured brain reactive astrogliosis is in need of intensive investigation. The aim of the
study was to characterize astrocyte plasticity during the course of CDV-induced demyelinating
leukoencephalitis by the aid of immunohistochemistry, immunofluorescence and gene expression
analysis. Immunohistochemistry revealed the presence of reactive glial fibrillary acidic protein (GFAP)+
astrocytes with increased survivin and reduced aquaporin 4, and glutamine synthetase protein levels,
indicating disturbed blood brain barrier function, glutamate homeostasis and astrocyte maladaptation,
respectively. Gene expression analysis revealed 81 differentially expressed astrocyte-related genes
with a dominance of genes associated with neurotoxic A1-polarized astrocytes. Accordingly, acyl-coA
synthetase long-chain family member 5+/GFAP+, and serglycin+/GFAP+ cells, characteristic of A1-
astrocytes, were found in demyelinating lesions by immunofluorescence. In addition, gene expression
revealed a dysregulation of astrocytic function including disturbed glutamate homeostasis and altered
immune function. Observed findings indicate an astrocyte polarization towards a neurotoxic phenotype
likely contributing to lesion initiation and progression in canine distemper leukoencephalitis.

Canine distemper is a fatal disease in dogs, caused by a single-stranded, negative-sense RNA virus of the genus
Morbillivirus, which is closely related to the human measles virus. The canine distemper virus (CDV) host range
includes dogs and other canids, as well as ferrets, raccoons, bears, large felids and as recently described even
macaques1,2. Infection leads to fatal systemic disease with a variety of clinical signs such as respiratory and gastro-
intestinal disorders, skin alterations and severe immunosuppression, which favors secondary infections. In the
central nervous system (CNS), infection causes demyelinating leukoencephalitis (CDV-DL), which is considered
to be a spontaneous animal model for human multiple sclerosis (MS)3. CDV-DL appears to be a biphasic process
with a direct virus-mediated phase in the beginning and an immune-mediated disease progression in the chronic
phase4–7 characterized by viral persistence8. Previous studies revealed increasing levels of MHC class II molecules
within CDV-DL plaques, indicating a fundamental role of immune processes during disease progression9.
In demyelinating distemper lesions, the majority (95%) of infected cells have been identified as astrocytes,
representing the main target for CDV10,11. Moreover, progressive myelin loss in affected dogs is associated with
astrocyte hypertrophy, isomorphic gliosis, reactive astrocytes (gemistocytes), and occasionally the formation of
astrocytic syncytia12–14. However, the functional relevance of astrocyte plasticity in canine distemper remains
to be determined. Astrocytic changes are not restricted to morphologic changes, but include alterations in gene
expression profiles influencing functional properties of this cell type15.
Within the CNS astrocytes are the most abundant cell type. They play fundamental roles in the healthy brain
as well as in pathological conditions. For instance, astrocytes give biochemical support to neurons and oligo-
dendrocytes and maintain the blood brain barrier integrity16. Furthermore astrocytes regulate water transport
through aquaporins16, synthesize metabolic substrates such as glycogen, sterols and lipoproteins17,18 and support
neighboring neurons through the export of glucose and lactate19. They are also able to respond to glutamatergic

1
Department of Pathology, University of Veterinary Medicine Hanover, Hannover, Germany. 2Department of
Experimental Animal Facilities and Biorisk Management, Friedrich-Loeffler-Institut, Greifswald, Insel Riems,
Germany. 3Center for Systems Neuroscience, Hannover, Germany. Correspondence and requests for materials
should be addressed to A.B. (email: andreas.beineke@tiho-hannover.de)

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 1

www.nature.com/scientificreports/ www.nature.com/scientificreports

neurotransmission by influencing the tone of arterioles and thus regulate the local blood supply20. Another key
function is the removal of neurotransmitters, such as glutamate, from the synaptic cleft by specific transporters
and subsequent degradation by glutamine synthetase (glutamate glutamine cycle), thus preventing excitotoxic
cell death of neurons and myelin producing oligodendrocytes17,21. Under disease conditions elevated glutamate
levels can occur due to increased release by neurons and/or glial cells or impaired reuptake by astrocytes. Previous
studies showed that CDV-infected rat hippocampal neuronal cells produce increased amounts of glutamate lead-
ing to neurodegeneration22. Inhibition of AMPA/kainate receptors results in reduction of neuronal death22,23
following CDV-infection in vitro and in experimental autoimmune encephalomyelitis (EAE), a rodent model for
demyelinating diseases23. Similarly, in MS axonal and oligodendrocyte damage is referred to glutamate excess24,
demonstrating the importance of glutamate toxicity in neuroinflammatory diseases. Through secretion of growth
factors and neurotrophins astrocytes enable remyelination and promote neuronal survival25. However, glial scar
formation in response to CNS injuries might hinder neuroregeneration. Additionally, astrocytes facilitate CNS
recruitment of immune cells by releasing chemoattractant cytokines and activate T cells, thus representing impor-
tant immune modulators in the CNS26. Recent publications describe the polarization of reactive astrocytes based
on their gene expression profile into beneficial and detrimental phenotypes. Astrocytes activated by inflammatory
stimuli exhibited a gene expression pattern indicating neurotoxic properties (A1-astrocytes), whereas ischemia
induces astrocytes with neuroprotective functions (A2-astrocytes)27,28.
Whether detrimental or beneficial effects of reactive astrogliosis predominate in the brain of CDV-infected
dogs is discussed controversially. Deeper insights into molecular and functional properties are necessary to
understand better the specific role of astrocytes in CDV-DL. Therefore, aims of this study were to (i) determine
phenotypical changes of astrocytes in demyelinating lesions, (ii) to characterize astrocytic expression pattern by
the aid of gene expression analyses in CDV-infected dogs and (iii) to investigate astrocyte polarization regarding
the A1/A2-phenotype in canine distemper.

Materials and Methods

Ethics statement.  The present study was conducted in accordance with the German Animal Welfare Act.
The authors confirm that for the purpose of this retrospective pathological study no animals were infected or sac-
rificed. This study is not an animal experiment since all animals were dead at the time of submission for necropsy
in order to investigate the causes of death and disease. All tissues used in this study were collected by one of the
authors (WB) during his work at the diagnostic pathology services of the Department of Pathology, University of
Veterinary Medicine Hannover, and the Institute of Veterinary Pathology, Justus-Liebig-University Giessen, and
all animals were used in previous publications29–32. All dog owners provided written consent for the dogs’ tissues
to be collected and used for research purposes.

Animals, histology and neuropathological classification.  For histology, histochemistry and immu-
nohistochemistry, cerebellar tissue of five healthy, CDV-negative control dogs (animal no. 1–5) and 29 sponta-
neously CDV-infected dogs (animal no. 6–34) was investigated. Anamnestic details of dogs used in this study
are listed in Supplemental Table S1. Animals died spontaneously or were euthanized due to poor prognosis.
Control dogs were obtained from an animal experiment, which was approved and authorized by the local author-
ities (Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit (LAVES), Oldenburg,
Germany, permission number 08A580). After necropsy, CNS tissue was fixed in 10% neutral-buffered formalin,
embedded in paraffin and serial sections of 2 µm thickness were prepared for histology and immunohistochem-
istry. Neuropathological diagnosis was based on hematoxylin and eosin (HE) staining and luxol fast blue-cresyl
violet (LFB/KEV) staining for detection of myelin loss. Accordingly, white matter areas were classified into four
groups: group 1 included unaffected brains of healthy control dogs; group 2 comprised acute lesions with vacuoli-
zation and gliosis; group 3 contained subacute demyelinating lesions without perivascular inflammation; and in
group 4 subacute to chronic demyelinating lesions with perivascular inflammation were included31.

Immunohistochemistry.  CDV-infection was confirmed by immunohistochemistry (IHC) using a mono-

clonal CDV-nucleoprotein (CDV-NP) antibody. For confirmation of demyelination, antibodies directed against
MBP (myelin sheaths) and Nogo-A (mature oligodendrocytes) were used. In addition, an antibody directed against
S100 (astrocytes and oligodendroglial cells) to characterize white matter lesions was included. For characterization
of astrocytic alterations, the astrocyte markers glial fibrillary acidic protein (GFAP), aquaporin 4 (AQP4), alde-
hyde dehydrogenase 1 family member L1 (ALDH1L1), and glutamine synthetase (GS) as well as a marker for the
astrocyte-related anti-apoptotic protein survivin were used. For confirmation of gene expression data, antibodies
directed against indoleamine 2,3-dioxygenase (IDO), serglycin (SRGN) and acyl-coA synthetase long-chain family
member 5 (ACSL5) were selected. Antibody details are summarized in Supplemental Table S2. Briefly, the sec-
tions were deparaffinized by Roticlear (Roth) and hydrated through graded alcohols. Afterwards the endogenous
peroxidase activity was inhibited by 85% ethanol with H2O2 (0.5%). Sections were washed in phosphate-buffered
saline (PBS) and pretreated with citrate buffer (pH 6.0) for 20 minutes in the microwave (800 W). MBP, GFAP and
AQP4 did not receive any pretreatment. Unspecific bindings were blocked with goat normal serum. Subsequently
tissue was incubated over night at 4 °C with the primary antibody. For negative controls monoclonal and polyclonal
antibodies were substituted with ascites fluid from non-immunized BALB/cJ mice and rabbit normal serum, respec-
tively. Incubation of primary antibodies was followed by incubation with biotinylated secondary antibodies (Vector
Laboratories) for 45 minutes at room temperature. Subsequently the avidin-biotin-complex (VECTASTAIN Elite
ABC Kit; Vector Laboratories) was incubated for 30 minutes also at room temperature. The positive antigen-antibody
reactions were visualized by incubation with 3.3′-diaminobenzidine tetrahydrochloride (DAB) with H2O2 (0.03%,
pH 7.2) for 5 minutes followed by slight counterstaining with Mayer’s hematoxylin (Merck). CDV-NP, Nogo-A,
GFAP, GS, ALDH1L1, S100, survivin, SRGN, and ACSL5 expressing cells were counted in cerebellar white matter

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 2

www.nature.com/scientificreports/ www.nature.com/scientificreports

lesions using a morphometric grid (number of positive cells/0.0625 mm2). Astrocytic AQP4 and IDO protein levels
and oligodendrocytic MBP positivity were quantified morphometrically. Digital photographs of the lesions were
®
taken in 100x magnification and the region of interest (ROI) was selected manually using the analySIS 3.2 Software
(Soft Imaging Solutions GmbH). Within these ROIs the area of immunopositive structures was measured in relation
to the total area (% area).

Immunohistochemistry double labeling.  Double labeling was performed to quantify GFAP and S100
protein levels in ALDH1L1+ astrocytes. Mouse monoclonal ALDH1L1-, rabbit polyclonal GFAP- and rabbit pol-
yclonal S100-antibodies were used (Supplemental Table S2). In brief, following chromogenic reaction with DAB
for localizing ALDH1L1, sections were washed for 5 minutes in PBS buffer, followed by incubation overnight at
4 °C with anti-GFAP or -S100 antibodies, respectively. After washing, sections were incubated with a biotiny-
lated goat-anti-rabbit antibody (Vector Laboratories, Burlingame, dilution 1:200) for 30 minutes, followed by
the avidin-biotin-peroxidase complex (Vector Laboratories) for 30 minutes. Positive reactions were visualized
with the Histogreen substrate system (Linaris). Co-localization of ALDH1L1 (brown) and GFAP (green) or S100
(green) was identified either by the presence of both colors in one cell or by green brown mixed color.

Immunofluorescence double labeling.  For verification of astrocytic SRGN and ACSL5 expression,
revealed by gene expression analysis and immunohistochemistry, immunofluorescence double staining in com-
bination with GFAP specific antibodies was performed. 2 µm thick, formalin-fixed, paraffin-embedded tissue
sections from two control dogs, two dogs with representative, acute (group 2) and two with inflammatory, suba-
cute to chronic (group 4) lesions were used for SRGN and ACSL5 staining. Polyclonal rabbit anti-SRGN (dilution
1:20) and -ACSL5 antibodies (dilution 1:20) together with a polyclonal goat anti-GFAP antibody (dilution 1:200)
were used (see Supplemental Table S2). Secondary antibodies were used in a dilution of 1:200. All antibodies were
diluted in PBS with 1% bovine serum albumin (BSA, Roth) and 0.1% Triton X (Sigma-Aldrich).
Paraffin sections were dewaxed and rehydrated. For blockade of unspecific binding, the slides were treated
with 20% horse normal serum diluted in PBS with 1% BSA and 0.1% of Triton X for 30 minutes. SRGN and
ACSL5 antibodies were incubated overnight. After rinsing with PBS the first secondary antibody (Cy2-conjugated
donkey anti-rabbit, Abcam) was incubated for 1.5 hours. For double staining, the procedure was simultaneously
repeated with 20% horse normal serum for blockade of unspecific bindings, GFAP primary antibody and the
appropriate secondary antibody (Cy3-conjugated donkey anti-goat, Jackson ImmunoResearch). Negative con-
trols received serum of non-immunized goats and rabbits, respectively, instead of primary antibodies. For nuclear
counterstaining bisbenzimide (H 33258, Sigma-Aldrich) was used in a dilution of 1:100 in double distilled water
and incubated for 10 minutes at room temperature. Subsequently sections were mounted with Dako fluorescent
mounting medium (Dako Diagnostika). Evaluation was performed qualitatively by detecting co-localization of
SRGN and ACSL5 with GFAP antigen in representative CDV lesions.

Statistical analysis.  For statistical analysis of non-normal distributed data obtained by immunohistochem-
istry the IBM “Statistic Package for Social Sciences” SPSS program for Windows (version 24) was used, employing
a Mann-Whitney U-test for two independent samples. P-values of less than or equal to 0.05 were considered to
show statistically significant differences between CDV groups. Graphs were designed with GraphPad Prism
(GraphPad Software, version 7.04).
®
Microarray analysis.  For molecular characterization of astrocytic changes, a data set of genes differentially
expressed in the cerebellum of CDV-infected dogs obtained in our previous global gene expression analysis was
used32. Briefly, cerebellar tissues of 14 CDV-infected dogs and 12 control dogs were used and their lesions were
characterized and grouped (group 1–4) based on same morphological criteria as used in the present study. Total
RNA was isolated from the frozen cerebellar specimens using the RNeasy Lipid Tissue Mini Kit (Qiagen) ampli-
fied and labeled employing the 3′IVT express kit (Affymetrix) and hybridized to GeneChip canine genome 2.0
arrays (Affymetrix) as described32. Background adjustment, quantile normalization and probe set summarization
were performed using the GC-RMA algorithm (Bioconductor gcrma for R package, Version 2.3)33. MIAME com-
pliant data sets are deposited in the ArrayExpress database (accession number: E-MEXP-3917; http://www.ebi.
ac.uk/arrayexpress).

Characterization of astrocytic gene expression.  The present analyses focused on a list of manu-
ally selected genes expressed by astrocytes according to peer-reviewed publications27,34–43 and genome data-
bases searching for the term ‘astrocyte’ (http://www.networkglia.eu/en/astrocyte; http://amigo.geneontology.
org/amigo/search/bioentity?q=astrocyte). Gene identifier (ID) conversion was performed using the Gene ID
Conversion Tool (https://david.ncifcrf.gov/conversion.jsp) of the database for annotation, visualization and
integrated discovery (DAVID, version 6.8) with Entrez Gene ID as unified gene identifier and selecting Canis
lupus familiaris as target species. In order to investigate which of the manually selected astrocyte-related genes
(Supplemental Table S3) are differentially expressed during CDV-infection, the data were re-analysed based on
the data set obtained in our previous global gene expression analysis32 employing independent pair-wise t-tests
comparing groups 1–4 followed by adjustment of the p-values according to the method described by Benjamini
and Hochberg44. Significantly differentially expressed genes (DEGs) between CDV-infected and healthy dogs
were selected employing a q-value ≤ 0.05 cutoff combined with a fold change filter (fold change ≥ 2.0 or ≤−2.0).
The fold change was calculated as the ratio of the inverse-transformed arithmetic means of the log2-transformed
expression values of CDV-infected versus healthy control dogs. Downregulations are shown as negative reciprocal
values. Hierarchical clustering of the astrocyte-associated DEGs (Supplemental Table S4) was performed using
TM4 Multi Experiment Viewer with the log2-transformed individual fold change of each dog relative to the mean
of the control dogs, employing Euclidean distance and complete linkage to reveal similar expression patterns45.

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 3

www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 1.  Histologic characterization of the cerebellar white matter of control dogs and canine distemper
virus-infected dogs by hematoxylin and eosin (HE; A–A′′′) and luxol fast blue-cresyl violet (LFB/KEV)
staining (B–B′′′) (A,B) Control animal with intact white matter (group 1). (A′,B′) Acute lesion (group
2) with hypercellularity and vacuolization. (A″,B″) Subacute demyelinated lesion (group 3) with marked
hypercellularity, gemistocytic astrocytes (arrow), gitter cells (arrow head) and decreased intralesional LFB/KEV
staining. (A′′′,B′′′) Subacute to chronic lesion (group 4) with marked perivascular cuffs, gemistocytic astrocytes
(arrows) and demyelination of the white matter. Asterisk: blood vessel. (A–A′′′) Scale bar = 50 µm. (B–B′′′)
Scale bar = 200 µm.

DEGs were visualized through a heat map and analyzed for intersections employing a Venn diagram (http://bio-
informatics.psb.ugent.be/webtools/Venn/). Gene ontology information was assigned to the hierarchical clusters
of DEGs employing the DAVID Functional Annotation Tool. Significantly enriched gene ontology terms were
selected from the biological process category of the gene ontology database at a false discovery rate (FDR) of
1.0%46.
From a previous publication genes expressed by A1- and A2-astrocytes were extracted27 and likewise con-
verted into the respective canine gene identifiers using DAVID Gene ID Conversion Tool (Supplemental
Table S5). For characterization of A1-/A2-polarization, only those genes were selected that are uniquely expressed
by one subset of reactive astrocytes (A1 or A2). Analyses based on the annotation version number 36 of the
canine genome 2.0 array. The relative proportion of A1- compared to A2-related DEGs was compared for each
time point employing Fisher’s exact tests (p-value ≤ 0.05).

Results
Characterization of cerebellar lesions.  In total, 128 white matter areas were investigated in canine cere-
bella. Characterization of histopathological alterations was performed using HE and LFB/KEV staining (Fig. 1).
Cerebellar tissue of healthy control dogs (group 1) showed no histopathological alterations. Acute lesions (group
2) were characterized by vacuolization (edema of myelin sheaths) and hypercellularity of the white matter due
to astro- and microgliosis. In some lesions, intranuclear and/or intracytoplasmic eosinophilic inclusion bodies
were found. LFB/KEV-staining revealed no myelin loss or myelinophagia. Group 3 comprised subacute lesions
showing demyelination, astrogliosis with gemistocytes, activated macrophages/microglia, gitter cells and sin-
gle lymphocytes in the neuroparenchyma, but no perivascular inflammation. Demyelination was confirmed by
decreased intralesional LFB/KEV-staining and the presence of LFB+ structures in the cytoplasm of gitter cells
(myelinophagia). In addition to findings observed in group 3 lesions, subacute to chronic lesions (group 4)
showed lymphohistiocytic infiltrations in perivascular spaces. Nogo-A, a marker for mature oligodendrocytes,
showed progressive loss of reactivity in oligodendroglial processes in CDV lesions (Fig. 2). Myelin changes were
also confirmed by MBP-specific immunohistochemistry. Acute lesions (group 2) showed a slight but significant
decrease in morphometric MBP density as a consequence of neuropil vacuolization. In subacute and chronic
lesions (group 3 and 4) a significant MBP loss compared to early phases (group 2) was observed. Additionally
MBP+ intracytoplasmic structures were detected in gitter cells of group 3 and 4 lesions, which confirm demyeli-
nation and myelinophagia (Fig. 2). S100 protein, which is expressed in cells of the astrocytic and oligodendro-
glial lineage47,48 was found throughout the white matter of control dogs. The number of S100+ cells significantly
decreased within CDV lesions of all groups (Fig. 2). Control dogs did not show CDV-nucleoprotein in the brain,
whereas CDV antigen was found in group 2–4 lesions in varying degrees. Group 2 lesions revealed moderate
numbers of CDV-NP+ cells intralesionally, which mostly displayed astrocytic morphology. Most CDV-NP+ cells
were detected in group 3, whereas group 4 revealed a decreased amount of CDV antigen (Fig. 2).
As shown before by experimental infections, CDV-induced leukoencephalitis in dogs develop in a sequential
order49,50. Acute lesions, characterized by white matter vacuolization and glial infection, can be observed 16–24
days post infection50–53. Subacute lesions with demyelination but without perivascular inflammation occur 24–32
days after infection11,50–53. Subacute to chronic lesions with demyelination, perivascular lymphohistiocytic cuffs
and reduced numbers of CDV+ cells can be found after a minimum of 29–63 days post infection in the brain of
experimentally infected dogs11,50–54.

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 4

www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 2.  Immunohistochemistry for detecting myelin basic protein (MBP; A–A′′′), neurite outgrowth
inhibitor A (Nogo-A; B,B′′′), S100 (C–C′′′) and canine distemper virus-nucleoprotein (CDV-NP; D–D′′′). (A)
Control tissue (group 1). (A′) Chronic lesion (group 4) with loss of MBP-positivity. (A, A′) Scale bar = 100 µm.
(A′′) Chronic lesion (group 4) with malacia and MBP+ material within gitter cells (myelinophages,
arrows). Scale bar = 20 µm. (A”’) Statistical analysis of MBP-immunohistochemistry shows decreased
immunopositive area in demyelinating distemper plaques. (B) Control tissue with immunopositive myelinating
oligodendrocytes. (B′,B′′) Subacute lesion (B′, group 3) and chronic inflammatory lesion (B′′, group 4) with
decreased number of Nogo-A+ oligodendrocytes. (B– B′′) Scale bar = 100 µm. (B′′′) Statistical analysis of
Nogo-A immunohistochemistry shows decreased numbers of myelinating oligodendrocytes in demyelinating
distemper lesions. (C) Control tissue with immunopositive glial cells. (C’,C”) Subacute lesion (C′, group 3) and
chronic inflammatory lesion (C′′, group 4) with decreased number of S100+ cells (C,C′′) Scale bar = 20 µm.
(B′′′) Statistical analysis shows decreased numbers of S100+ cells in demyelinating distemper lesions. (D) Group
2 lesion: Most CDV-infected cells show astrocyte morphology. (D’) Subacute lesion (group 3) with numerous
infected astrocytes. (D, D’) Scale bar = 50 µm. (D′′) Group 4 lesion with infected astrocytes in the periphery of
the inflammatory lesion. Scale bar = 200 µm. Inset: Magnification of CDV-NP+ astrocytes. Scale bar = 20 µm.
(D′′′) Statistical analysis shows most numerous CDV-infected cells in group 3. (A′′′,B′′′,C′′′,D′′′) Box and
whisker plots display median and quartiles with maximum and minimum values. Significant differences
(p ≤ 0.05, Mann–Whitney U-test) are labeled by asterisks.

Phenotypic changes of astrocytes in demyelinating leukoencephalitis.  Astrocytic changes dur-

ing the disease course were determined by immunohistochemistry using astrocyte markers (GFAP, ALDH1L1,
AQP4, GS, survivin). In randomly selected white matter areas of control animals (group 1) regularly distributed
GFAP+ fibrous astrocytes were detected (Fig. 3). Within acute lesions (group 2), besides fibrous astrocytes also
enlarged, plump GFAP+ somata (gemistocytes) with a homogenous to finely granulated, light brown, cytoplasmic
signal were observed. In group 3 and 4 the number of GFAP+ gemistocytes continuously increased, whereas the
total number of astrocytes and their processes decreased compared to group 2. Subacute to chronic lesions (group
4) showed a moderate to severe loss of GFAP+ astrocytes in the lesion center, while at the periphery an increased
number of astrocytes built a demarcation line around demyelinating lesions. Statistically, a significant increase
of GFAP+ cells was confirmed in acute lesions (group 2), whereas group 3 and 4 lesions did not show significant
differences compared to control animals (Fig. 4).

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 5

www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 3.  Phenotypical characterization of astrocytes in the cerebellar white matter by immunohistochemistry.
Detection of astrocyte markers glial fibrillary acidic protein (GFAP; A,A′′), aldehyde dehydrogenase 1 L1
(ALDH1L1; B,B′′), aquaporin 4 (AQP4; C,C′′), glutamine synthetase (GS; D,D′′), and survivin (E,E′′). (A,E)
Intact white matter of a control animal (group 1). Scale bar = 50 µm. (A’,E’) Acute distemper lesion (group 2).
Arrows = gemistocytic astrocytes. Scale bar = 50 µm. (A′′,D′′) Subacute to chronic, inflammatory lesion (group
4). Arrows = gemistocytic astrocytes. Scale bar = 50 µm. (E′′) Group 4 lesion with perivascular inflammation
and increase of intralesional survivin+ cells. Scale bar = 200 µm. Inset: Higher magnification. Immunopositive
cells show astrocyte morphology. Scale bar = 20 µm.

Control animals showed an ALDH1L1+ signal in somata and processes of white matter astrocytes (Fig. 3).
Within CDV lesions, astrocytes displayed gemistocytic morphology, but no significant changes in the num-
ber of ALDH1L1+ cells was found in CDV lesions (group 2, 3, 4) compared to control dogs (Fig. 4). Increased
cytoplasmic GFAP protein levels were found in ALDH1L1+ cells of CDV-infected dogs by double labeling.
Moreover, GFAP staining was found in astrocytic cells without detectable ALDH1L1 protein positivity (GFAP+/
ALDH1L1− cells) in demyelinating lesions, indicating a dominating GFAP elevation in reactive astrocytes55
(Fig. 5). Conclusively, unchanged numbers of intralesional ALDH1L1+ cells show that the increased density
of GFAP+ cells is due to higher protein amounts and primarily not a consequence of astrocyte proliferation in
CDV-induced white matter lesions.
Immunohistochemistry revealed a significant reduction of S100+ cells in white matter lesions of CDV-infected
dogs (Fig. 2). Since S100 is not restricted to the astrocytic lineage but also expressed in oligodendrocytes and glial

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 6

www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 4.  Statistical analysis of immunohistochemical evaluation. (A) Increased numbers of glial fibrillary
acidic protein (GFAP)+ cells in acute distemper lesions. (B) Detection of aldehyde dehydrogenase 1 L1
(ALDH1L1) showed no significant differences between control and canine distemper virus-infected dogs. (C)
Progressive loss of aquaporin 4 (AQP4) in distemper lesions. (D) Significant decrease of glutamine synthetase
(GS) in all groups of distemper lesions. (E) Significant increase in survivin protein levels in all phases of canine
distemper virus-induced leukoencephalitis. Box and whisker plots display median and quartiles with maximum
and minimum values. Significant differences (p ≤ 0.05, Mann–Whitney U-test) are labeled by asterisks.

precursor cells47,48, double labeling was performed to analyse S100 protein product in ALDH1L1+ astrocytes
during infection. While the number of ALDH1L1−/S100+ cells declined, the amount of S100+/ALDH1L1+ dou-
ble labeled cells remained unchanged. Data indicate that the S100 decline in infected dogs is due to depletion of
resident glial cells other than astrocytes within demyelinating lesions (Fig. 5).
AQP4 was found in the white matter neuropil of all control animals. Astrocytic foot processes presented as
continuous signal around capillaries and depicted clearly the astrocytic component of the blood brain barrier
(Fig. 3). In acute lesions of CDV-infection (group 2), a decrease in AQP4 protein levels was apparent. In subacute
and chronic lesions (groups 3 and 4), a significant decrease of morphometric density was observed, characterized
by a reduction of signal intensity in the neuropil or total loss in perivascular astrocytic foot processes, respectively
(Fig. 4).
Immunohistochemistry for GS showed an intracytoplasmic signal within numerous astrocytic cells of con-
trol animals (Fig. 3). In CDV lesions of all groups (groups 2, 3, 4) the number of GS+ astrocytes significantly
decreased compared to healthy control dogs (Fig. 4).
Immunohistochemistry for detecting the astrocyte-related anti-apoptotic protein survivin showed single
immunopositive cells within control animals (group 1). A significant upregulation of survivin protein product
was observed in all groups of distemper encephalitis in cells with astrocytic morphology (Figs 3 and 4). In addi-
tion to astrocytes, survivin positivity was also found in inflammatory cells, including gitter cells in advanced
demyelinating lesions (groups 3 and 4).

Astrocyte-related gene expression in demyelinating leukoencephalitis.  In order to get insights

into alterations of astrocytic gene expression during CDV-DL microarray analyses of cerebellar tissue have been
performed. A total of 2184 astrocyte-related genes were extracted from peer-reviewed publications and genome
databases (Supplemental Table S3). Comparison with the entire data set32 revealed 81 astrocyte-related genes
(Supplemental Table S4) that were differentially expressed in CDV-induced leukoencephalitis. 67 genes were
upregulated and 14 were downregulated. More than half (59.7%) of upregulated genes were upregulated in all
three groups (groups 2–4) of CDV-DL. Among downregulated DEGs, 78.5% were either solely downregulated in
subacute lesions of group 3 (35.7%) and subacute to chronic lesions of group 4 (7.1%) or in both (35.7%). Thus,
downregulation was predominantly observed in late phases of CDV-induced leukoencephalitis. Comparison of
DEGs within defined CDV groups is depicted in Fig. 6B. In order to detect similarities in the expression pattern

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 7

www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 5.  Immunohistochemistry double labeling of astrocytic markers in the cerebellar white matter. (A,A′)
Detection of ALDH1L1 (brown) and GFAP (green) in tissue of a control animal (A) and a CDV-infected dog
(A′). Infection results in increased GFAP protein positivity. Scale bars = 20 µm. (B,B’) Detection of ALDH1L1
(brown) and S100 (green) in tissue of a control animal (B) and a CDV-infected dog (B’). Infected animals show
loss of S100 signal. Scale bars = 20 µm. (C–E) Quantification of ALDH1L1+/GFAP− cells (C), ALDH1L1−/
GFAP+ cells (D) and ALDH1L1+/GFAP+ double-positive cells (E). (F–H) Quantification of ALDH1L1+/
S100− cells (F), ALDH1L1−/S100+ cells (G) and ALDH1L1+/S100+ cells double-positive cells (H). (C–H) Box
and whisker plots display median and quartiles with maximum and minimum values. Significant differences
(p ≤ 0.05, Mann–Whitney U-test) are labeled by asterisks.

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 8

www.nature.com/scientificreports/ www.nature.com/scientificreports

of the 81 astrocyte-related DEGs, the log2 transformed fold changes of control dogs and CDV-infected dogs for
each group were analysed through hierarchical cluster analysis employing Euclidean distance and visualized by a
heat map (Fig. 6A). The resulting four hierarchical clusters (cluster A – D) grouped genes with similar expression.
In order to assign a biological meaning to these genes, the functional annotation tool from DAVID was applied to
all hierarchical clusters (https://david.ncifcrf.gov/summary.jsp). Significantly enriched gene ontology terms (GO
terms, FDR < 1.0%) are listed in Table 1. From each cluster GO terms subjectively giving the best description of
the whole cluster were manually chosen. Cluster A contains 14 downregulated genes. A specific GO term could
not be assigned to this cluster, as the highest ranked ontology term did not meet the cut-off criteria for significant
enrichment (FDR = 8.6%). Interestingly, multiple downregulated genes are involved in glutamate detoxification,
such as GLT-1, SLC7A10, DDO and ATP1A2. In cluster B, mildly upregulated genes were grouped (n = 51),
which were associated to gene ontology terms such as immune system process, positive regulation of immune system
process, apoptotic process and positive regulation of signal transduction (FDR < 0.3). Cluster C included moderately
upregulated genes (n = 14) and was enriched in the ontology term response to cytokine. Hierarchical cluster D
grouped 2 genes, which were severely upregulated in CDV-DL. Functional annotation revealed an association of
the genes with the term regulation of T cell chemotaxis. However, the term was slightly above the cut-off criteria
(FDR = 1.1%).
For characterization of astrocytic A1/A2-polarization associated genes were extracted from a peer-reviewed
publication (Supplemental Table S5)27. 36 genes were assigned to the A1-phenotype and 117 genes were expressed
by the A2-phenotype. Fisher’s exact test was employed for comparison of the relative proportion of A1- and
A2-genes within the CDV-DL groups (Fig. 6C). The test revealed a significantly higher percentage of differen-
tially expressed A1-marker genes for group 2 (p = 0.013) and group 4 (p = 0.004). In addition a statistical ten-
dency towards the A1-phenotype was observed for group 3 (p = 0.073). Compared to controls 19.44%, 27.78%
and 19.44% of A1-related genes were differentially expressed in group 2, 3 and 4, respectively. In contrast, the
percentage of A2-related DEGs accounts for 5.13%, 13.68%, and 3.42% for respective group comparison. The
dominance of A1-related DEGs indicates a shift towards a neurotoxic astrocyte phenotype during CDV-induced
leukoencephalitis.
The immune modulating enzyme IDO was upregulated in all CDV groups in microarray analysis.
Immunohistochemistry confirmed upregulation of IDO mainly in cells displaying astrocyte morphology
throughout the cerebellar white matter. Some of these cells exhibited a reactive phenotype (gemistocytes).
Densitometry revealed an increased IDO+ area in all groups of CDV lesions with highest values in chronic lesions
of group 4 (Fig. 7).
SRGN and ACSL5, representing markers for the neurotoxic A1-phenotype of astrocytes28, were upregulated
in CDV-infected animals in microarray analysis and in immunohistochemistry (Fig. 7). SRGN was increased
in infected lesions of all groups. ACSL5+ cells was significantly increased in group 3 and 4 lesions. In group 2
lesions, a statistical trend (p = 0.07) of an ACSL5 increase was found (Fig. 7). Most of SRGN+ and ACSL5+ cells
exhibited astrocyte morphology. In addition also some infiltrating leukocytic cells and microglia displayed SRGN
and ACSL5 positivity, respectively. Immunofluorescence double labeling with GFAP revealed co-localization with
SRGN (GFAP+/SRGN+ cells) and ACSL5 (GFAP+/ACSL5+ cells) in CDV lesions characteristic of A1-astrocytes
(Fig. 8).

Discussion
Astrocytes play a central role in maintaining normal CNS physiology and critically control the response to brain
injury and neurological diseases. Phenotypical and molecular analyses of the present study revealed an altered
metabolism and neurotoxic properties of reactive astrocytes in CDV lesions, which have the potential to disturb
neurotransmitter uptake and blood brain barrier homeostasis.
Reactive astrocyte responses, characterized by increased GFAP protein levels and mRNA expression, were
found already before the onset of overt demyelination in acute brain lesion of CDV-infected dogs. Similarly,
reactive astrocytes are present in the normal-appearing white matter of MS patients and in pre-demyelinating
lesions in the EAE model, suggesting an early contribution of glial cells to lesion development by chemokine
release and initiation of inflammatory responses56–59. As observed in astrocytes within developing MS lesions,
gene expression analyses revealed an upregulation of pro-inflammatory cytokines such as CCL5 and CXCL10
in the early phase of CDV-DL57. Moreover, several other genes associated with activation of immune response
and complement cascade, such as TLR2, TLR3, CCL2, CD44, VCAM1, C1QA, C1S, C3 and STAT3, were upreg-
ulated32. The activation of the STAT3 pathway by interleukin-6 results in astrogliosis in demyelinating lesions of
Theiler’s murine encephalomyelitis (TME)60. The presented findings indicate that reactive astrocytes contribute
to the inflammatory process in CDV-induced leukoencephalitis by pro-inflammatory stimuli.
Immunohistochemistry showed a significant loss of AQP4 together with a GFAP increase in acute lesions and
GFAP retention in chronic lesions, respectively. AQP4, a water channel exclusively expressed on astrocytes in the
brain, is located at astrocytic end feet surrounding blood vessels and thus regulates water homeostasis at the blood
brain barrier. It enables fast water influx or efflux and facilitates reabsorption of excess fluid in vasogenic brain
edema61,62. Indicating an early blood brain barrier dysfunction, AQP4 reduction was observed before the onset
of overt demyelination in CDV-infected dogs. Dysfunction and loss of AQP4 plays a crucial role in development
of brain edema and was shown to cause prolonged seizure duration through impaired K+ buffering in AQP4−/−
mice63. Moreover, loss of AQP4 is a hallmark of active demyelinating lesions in neuromyelitis optica (NMO) in
human patients64.
Immunophenotyping revealed decreased protein levels of the astrocyte-specific enzyme GS in acute lesions
and foci of progressive myelin loss in CDV-DL. GS catalyzes the rapid degradation of glutamate to non-neurotoxic
amino acid glutamine, thereby preventing neurodegenerative processes65,66. In several pathological conditions,
reduction of GS activity has been detected, for example schizophrenia67, Alzheimer’s disease68, epilepsy69,70,

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 9

www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 6.  Astrocytic gene expression analyses. (A) Expression profile of 81 differentially expressed astrocyte-
associated genes in the white matter during the course of CDV-induced demyelinating leukoencephalitis. The
heat map displays log2-transformed individual fold changes relative to the mean expression of control animals
indicated by a color scale ranging from −5 (relative low expression) in blue to 5 (relative high expression)
in red. Each row represents one of 81 differentially expressed genes (DEGs) and each column one of the 26
biological replicates (cerebellar specimens of individual dogs) sorted according to the histologically defined
groups of CDV-induced leukoencephalitis. Fold changes were grouped by hierarchical cluster analysis to
reveal similar expression patterns. DEGs were subdivided into four clusters with distinct expression profiles:
cluster A (green bar) contains downregulated genes, whereas cluster B (yellow bar) shows mildly upregulated
genes, cluster C (orange bar) moderately upregulated and cluster D (red bar) severely upregulated genes.
(B) Venn diagram comparing up- and downregulated astrocyte-related DEGs within the defined groups of
CDV leukoencephalitis. Group 2–4 are compared for shared (intersection) and unique DEGs, depicted by
total numbers and proportion (percentage). Upregulated genes are marked by an upward directed arrow,
downregulated by a downward directed one. (C) Comparison of the relative proportion of A1- and A2-related
differentially expressed genes employing the Fisher’s exact test revealed significant dominance of A1-related
genes in group 2 and group 4. Significant differences (p ≤ 0.05) are labeled by asterisks. Statistical tendency is
labeled by a square (p = 0.073).

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 10

www.nature.com/scientificreports/ www.nature.com/scientificreports

Hierarchical cluster* Gene ontology terms Number of genes FDR [%]

A Not assigned
immune system process 22 <0.1
positive regulation of immune system process 15 <0.1
B
apoptotic process 14 0.1
positive regulation of signal transduction 13 0.3
C response to cytokine 5 0.3
D Not assigned

Table 1.  Significantly enriched gene ontology terms related to differentially expressed astrocyte-associated
genes. *Hierarchical clusters refer to the respective cluster of genes with a similar expression pattern obtained
from hierarchical cluster analysis as displayed in Fig. 5A. Functional annotation was performed for each
hierarchical cluster and specific gene ontology terms that subjectively gave the best description of the cluster
and met the cut-off criteria of FDR ≤ 1.0% were selected.

hypoxia71, diabetes72 and MS24 as well as in the EAE model73. Therefore, GS decline indicates disturbed home-
ostasis of the glutamatergic system in CDV-DL and highlights the importance of glutamate toxicity in demyeli-
nation following morbillivirus infection. In accordance with this, transcriptional analysis of the present study of
astrocyte-related DEGs revealed significant downregulation of genes involved in glutamate detoxification, such as
GLT-1 (=SLC1A2), SLC7A10, D-aspartate oxidase (DDO) and ATP1A2. The excitatory amino acid transporter
GLT-1 accounts for the majority (90%) of glutamate uptake in the CNS. Decreased transport activity contrib-
utes to impaired glutamate uptake and raised extracellular glutamate concentration74. Dysfunction of GS and
reduced astrocytic GLT-1 protein leading to glutamate excitotoxicity can be observed in EAE74,75 and in active
MS lesions76. Interestingly, internalization and redistribution of AQP4 in NMO is also accompanied by downreg-
ulation of its physically associated glutamate transporter GLT-177. Subsequent reduced glutamate uptake is sup-
posed to induce glutamate toxicity to myelin-producing oligodendrocytes78. Gene expression analysis revealed
an upregulation of the neuronal glutamate transporter SLC1A1 in the chronic phase of CDV-DL32. In agreement
with this finding, previous studies revealed an upregulation of SLC1A1 in the hippocampus of CDV-infected dogs
showing seizures compared to those without epileptic seizures, which is supposed to be a compensatory neuronal
mechanism following elevated extracellular glutamate levels to prevent excitotoxicity79.
Astrocytic SLC7A10 is an amino acid transporter mediating transport of the NMDA receptor co-agonists
glycine and D-serine in the CNS and thus functions as regulator of NMDA receptor activity at glutamatergic
synapses80. Binding of co-agonists to this glutamate receptor increases the affinity to glutamate. Thus, decreased
clearance of glycine and D-serine by SLC7A10 causes overstimulation of NMDA receptors and thereby contrib-
utes to glutamate excitotoxicity81–83. Mice lacking this transporter develop tremors, ataxia and seizures in conse-
quence to neuronal hyperexcitability84,85.
Similarly to GS, DDO selectively degrades D-aspartate, which is also known to be an agonist of the NMDA
receptor. DDO-knockout mice show substantially increased extracellular glutamate levels in the CNS 86.
Downregulation of the sodium-potassium-ATPase ATP1A2, as observed by gene expression analysis of the pres-
ent study, leads to breakdown of the electrochemical gradient, which is a prerequisite for neuronal excitability
and activity of glutamate transporters, such as GLAST and GLT-1. Thus, decreased expression of ATP1A2 leads
secondarily to increased extracellular amounts of glutamate and likewise contributes to glutamate excitotoxicity87.
Consequently, findings indicate disturbed astrocytic glutamate homeostasis, which potentially causes excitotoxic
effects. Besides neurons, oligodendrocytes are particularly vulnerable to glutamate toxicity23,88–90. Thus, dam-
age of myelinating oligodendrocytes by glutamate excess might contribute to the initiation and progression in
CDV-induced leukoencephalitis, as described in the EAE model and human MS91,92.
Regarding A1/A2-polarization of astrocytes, gene expression analysis was performed and confirmed by
ACSL5- and SRGN-specific immunohistochemistry and immunofluorescence. Findings strongly indicate
astrocytic polarization towards a neurotoxic A1-phenotype. A1-astrocytes are present in different neurode-
generative diseases, such as MS and Alzheimer’s disease, and show decreased phagocytic capacity, which leads
to disturbed clearance of myelin debris. In contrast to neuroprotective A2-astrocytes, A1-astrocytes lose their
neurotrophic function, promote pro-inflammatory responses, trigger neurotoxicity, and thereby contribute to
neuronal and oligodendroglial death28. The observation of an imbalanced astrocyte polarization towards the
neurotoxic A1-phenotype further supports the notion of maladaptive astrogliosis in CDV-DL. GO-annotation
of astrocyte-related DEGs showed upregulation of several genes assigned to the apoptotic process. The present
study showed over-expression of the astrocyte-related, anti-apoptotic protein survivin, that is a feature of active
MS lesions93. In TME, survivin prevents apoptosis of infected astrocytes, favoring viral persistence94,95. Moreover,
astrocyte apoptosis resistance was demonstrated in TME associated with glial scarring and chronic demyeli-
nation94,96. Lack of apoptosis has been shown also in primary astrocyte cultures infected with CDV, which is
supposed to support cell-to-cell transmission of the virus in the brain of infected dogs8,97. Similarly, non-cytolytic
spread is a putative prerequisite for persistent measles virus infection and subacute sclerosing panencephalitis in
human patients97.
Within gene expression analyses, IDO was upregulated in all phases of CDV-DL. Immunohistochemistry
confirmed an intralesional increase of IDO protein within astrocytic cells. The enzyme catalyzes the first step in
the degradation of tryptophan through the kynurenine pathway98. In the brain, IDO is expressed by astrocytes,
neurons and microglia99,100. It exerts immune modulating functions and suppresses replication and spread of

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 11

www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 7.  Confirmation of gene expression analysis by immunohistochemical detection of indoleamine

2,3-dioxygenase (IDO, A,A′′′), serglycin (SRGN, B,B′′′) and acyl-CoA synthetase long-chain family member 5
(ACSL5, C,C′′′) protein in the cerebellar white matter. (A) Control tissue with few small immunopositive cells
in the healthy white matter. (A′) Numerous IDO+ cells showing astrocyte morphology (arrows) within an acute
lesion (group 2). (A′′) Inflammatory lesion (group 4) with even higher density of IDO+ cells with astrocyte
morphology (arrows). (B) Control tissue with SRGN+ astrocytes. (B′) Acute (group 2) and (B′′) subacute to
chronic lesions (group 4) with increased numbers of SRGN+ cells. (C) Control tissue with few ACSL+ cells.
(C′) Increased numbers of ACSL+ cells with gemistocytic morphology in acute (group 2) and (C′′) subacute
to chronic lesions (group 4). Scale bars = 50 µm (A′′′–C′′′) Statistical analysis shows significant increase of
IDO+ area (A′′′) and significantly increased numbers of SRGN+ cells (B′′′) and ACSL5+ cells (C′′′) in infected
dogs. Box and whisker plots display median and quartiles with maximum and minimum values. Significant
differences (p ≤ 0.05, Mann–Whitney U-test) are labeled by asterisks.

Figure 8.  Immunofluorescence double labeling for detecting SRGN (A) and ACSL5 (B) protein in GFAP+
astrocytes in demyelinating lesions. SRGN+ and ACSL5+ cells (green) and GFAP+ cells (red) show co-
localization (yellow). Blue: nuclear counterstaining.

infectious agents by deprivation of tryptophan, as demonstrated in measles virus infection101–107. However, tryp-
tophan degradation through the kynurenine pathway leads to metabolites such as 3-hydroxykynurenine and
quinolinic acid, which show neurotoxic effects in the CNS through production of reactive radical species and

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 12

www.nature.com/scientificreports/ www.nature.com/scientificreports

activation of glutamate receptors108–110. Thus, similar to human measles, ambivalent functions of IDO with benefi-
cial effects by supporting antiviral immunity and reducing immunopathology and detrimental neurotoxic effects
contributing to neuronal and oligodendroglial damage in CDV-DL have to be considered106,111.
The present study provides a comprehensive database of astrocyte-related gene expression during the initi-
ation and progression of CDV-DL. Reactive astrocytes in canine distemper show neurotoxic properties, which
have the potential to cause neurodegeneration, demyelination, and impaired remyelination. There is cumulative
evidence that astrocytopathies with disturbed astrocyte function and maladaptive astrogliosis are crucial factors
in the pathogenesis of different inflammatory neurological diseases112. Thus, understanding the complex nature
of astrocyte plasticity in CDV-DL represents a prerequisite for targeted therapeutic strategies in canine CNS
disorders.

Data Availability
MIAME compliant data sets are deposited in the ArrayExpress database (accession number: E-MEXP-3917;
http://www.ebi.ac.uk/arrayexpress). All other datasets generated and analysed during the current study are avail-
able from the corresponding author on reasonable request.

References
1. Qiu, W. et al. Canine distemper outbreak in rhesus monkeys, China. Emerging infectious diseases 17, 1541–1543, https://doi.
org/10.3201/eid1708.101153 (2011).
2. Beineke, A., Baumgärtner, W. & Wohlsein, P. Cross-species transmission of canine distemper virus—an update. One Health 1,
49–59, https://doi.org/10.1016/j.onehlt.2015.09.002 (2015).
3. Dal Canto, M. C. & Rabinowitz, S. G. Experimental models of virus-induced demyelination of the central nervous system. Annals
of neurology 11, 109–127, https://doi.org/10.1002/ana.410110202 (1982).
4. Baumgärtner, W. & Alldinger, S. In Experimental Models of Multiple Sclerosis (ed. C. S. Constantinescu E. Lavi) Ch. p. 871–887
(Springer US, 2005).
5. Beineke, A., Puff, C., Seehusen, F. & Baumgärtner, W. Pathogenesis and immunopathology of systemic and nervous canine
distemper. Veterinary immunology and immunopathology 127, 1–18, https://doi.org/10.1016/j.vetimm.2008.09.023 (2009).
6. Alldinger, S., Baumgärtner, W. & Orvell, C. Restricted expression of viral surface proteins in canine distemper encephalitis. Acta
neuropathologica 85, 635–645 (1993).
7. Alldinger, S., Wunschmann, A., Baumgartner, W., Voss, C. & Kremmer, E. Up-regulation of major histocompatibility complex class
II antigen expression in the central nervous system of dogs with spontaneous canine distemper virus encephalitis. Acta Neuropathol
92, 273–280 (1996).
8. Wyss-Fluehmann, G., Zurbriggen, A., Vandevelde, M. & Plattet, P. Canine distemper virus persistence in demyelinating
encephalitis by swift intracellular cell-to-cell spread in astrocytes is controlled by the viral attachment protein. Acta
neuropathologica 119, 617–630, https://doi.org/10.1007/s00401-010-0644-7 (2010).
9. Zeinstra, E., Wilczak, N., Streefland, C. & De Keyser, J. Astrocytes in chronic active multiple sclerosis plaques express MHC class
II molecules. Neuroreport 11, 89–91 (2000).
10. Mutinelli, F., Vandevelde, M., Griot, C. & Richard, A. Astrocytic infection in canine distemper virus-induced demyelination. Acta
neuropathologica 77, 333–335 (1989).
11. Summers, B. A. & Appel, M. J. Demyelination in canine distemper encephalomyelitis: an ultrastructural analysis. J Neurocytol 16,
871–881 (1987).
12. Vandevelde, M., Kristensen, F., Kristensen, B., Steck, A. J. & Kihm, U. Immunological and pathological findings in demyelinating
encephalitis associated with canine distemper virus infection. Acta neuropathologica 56, 1–8 (1982).
13. Summers, B. A., Greisen, H. A. & Appel, M. J. Canine distemper and experimental allergic encephalomyelitis in the dog:
comparative patterns of demyelination. Journal of comparative pathology 94, 575–589 (1984).
14. Summers, B. A. & Appel, M. J. Aspects of canine distemper virus and measles virus encephalomyelitis. Neuropathology and applied
neurobiology 20, 525–534 (1994).
15. Eddleston, M. & Mucke, L. Molecular profile of reactive astrocytes–implications for their role in neurologic disease. Neuroscience
54, 15–36 (1993).
16. Abbott, N. J., Ronnback, L. & Hansson, E. Astrocyte-endothelial interactions at the blood-brain barrier. Nature reviews.
Neuroscience 7, 41–53, https://doi.org/10.1038/nrn1824 (2006).
17. Correale, J. & Farez, M. F. The Role of Astrocytes in Multiple Sclerosis Progression. Frontiers in Neurology 6, https://doi.
org/10.3389/fneur.2015.00180 (2015).
18. Goritz, C., Mauch, D. H., Nagler, K. & Pfrieger, F. W. Role of glia-derived cholesterol in synaptogenesis: new revelations in the
synapse-glia affair. Journal of physiology, Paris 96, 257–263 (2002).
19. Magistretti, P. J. Neuron-glia metabolic coupling and plasticity. The Journal of experimental biology 209, 2304–2311, https://doi.
org/10.1242/jeb.02208 (2006).
20. Zonta, M. et al. Neuron-to-astrocyte signaling is central to the dynamic control of brain microcirculation. Nature neuroscience 6,
43–50, https://doi.org/10.1038/nn980 (2003).
21. Seifert, G., Schilling, K. & Steinhauser, C. Astrocyte dysfunction in neurological disorders: a molecular perspective. Nature reviews.
Neuroscience 7, 194–206, https://doi.org/10.1038/nrn1870 (2006).
22. Brunner, J. M. et al. Canine distemper virus infection of primary hippocampal cells induces increase in extracellular glutamate and
neurodegeneration. Journal of neurochemistry 103, 1184–1195, https://doi.org/10.1111/j.1471-4159.2007.04819.x (2007).
23. Lin, C. L., Kong, Q., Cuny, G. D. & Glicksman, M. A. Glutamate transporter EAAT2: a new target for the treatment of
neurodegenerative diseases. Future medicinal chemistry 4, 1689–1700, https://doi.org/10.4155/fmc.12.122 (2012).
24. Werner, P., Pitt, D. & Raine, C. S. Multiple sclerosis: altered glutamate homeostasis in lesions correlates with oligodendrocyte and
axonal damage. Annals of neurology 50, 169–180 (2001).
25. Montgomery, D. L. Astrocytes: form, functions, and roles in disease. Veterinary pathology 31, 145–167 (1994).
26. Dong, Y. & Benveniste, E. N. Immune function of astrocytes. Glia 36, 180–190 (2001).
27. Zamanian, J. L. et al. Genomic analysis of reactive astrogliosis. The Journal of neuroscience: the official journal of the Society for
Neuroscience 32, 6391–6410, https://doi.org/10.1523/JNEUROSCI.6221-11.2012 (2012).
28. Liddelow, S. A. et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature 541, 481–487, https://doi.
org/10.1038/nature21029 (2017).
29. Seehusen, F., Orlando, E. A., Wewetzer, K. & Baumgartner, W. Vimentin-positive astrocytes in canine distemper: a target for canine
distemper virus especially in chronic demyelinating lesions? Acta neuropathologica 114, 597–608, https://doi.org/10.1007/s00401-
007-0307-5 (2007).

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 13

www.nature.com/scientificreports/ www.nature.com/scientificreports

30. Seehusen, F. & Baumgartner, W. Axonal pathology and loss precede demyelination and accompany chronic lesions in a
spontaneously occurring animal model of multiple sclerosis. Brain patholog y 20, 551–559, https://doi.
org/10.1111/j.1750-3639.2009.00332.x (2010).
31. Qeska, V. et al. Dynamic changes of Foxp3(+) regulatory T cells in spleen and brain of canine distemper virus-infected dogs.
Veterinary immunology and immunopathology 156, 215–222, https://doi.org/10.1016/j.vetimm.2013.10.006 (2013).
32. Ulrich, R. et al. Transcriptional Changes in Canine Distemper Virus-Induced Demyelinating Leukoencephalitis Favor a Biphasic
Mode of Demyelination. PLoS ONE 9, e95917, https://doi.org/10.1371/journal.pone.0095917 (2014).
33. Irizarry, R. A., Wu, Z. & Jaffee, H. A. Comparison of Affymetrix GeneChip expression measures. Bioinformatics 22, 789–794,
https://doi.org/10.1093/bioinformatics/btk046 (2006).
34. Cahoy, J. D. et al. A transcriptome database for astrocytes, neurons, and oligodendrocytes: a new resource for understanding brain
development and function. The Journal of neuroscience: the official journal of the Society for Neuroscience 28, 264–278, https://doi.
org/10.1523/jneurosci.4178-07.2008 (2008).
35. Gimsa, U., Mitchison, N. A. & Brunner-Weinzierl, M. C. Immune privilege as an intrinsic CNS property: astrocytes protect the
CNS against T-cell-mediated neuroinflammation. Mediators of inflammation 2013, 320519, https://doi.org/10.1155/2013/320519
(2013).
36. Lim, R., Liu, Y. X. & Zaheer, A. Cell-surface expression of glia maturation factor beta in astrocytes. FASEB journal: official
publication of the Federation of American Societies for Experimental Biology 4, 3360–3363 (1990).
37. Lovatt, D. et al. The transcriptome and metabolic gene signature of protoplasmic astrocytes in the adult murine cortex. The Journal
of neuroscience: the official journal of the Society for Neuroscience 27, 12255–12266, https://doi.org/10.1523/
JNEUROSCI.3404-07.2007 (2007).
38. Moore, C. S., Abdullah, S. L., Brown, A., Arulpragasam, A. & Crocker, S. J. How factors secreted from astrocytes impact myelin
repair. Journal of neuroscience research 89, 13–21, https://doi.org/10.1002/jnr.22482 (2011).
39. Stahlberg, A. et al. Defining cell populations with single-cell gene expression profiling: correlations and identification of astrocyte
subpopulations. Nucleic acids research 39, e24, https://doi.org/10.1093/nar/gkq1182 (2011).
40. Wu, H., Friedman, W. J. & Dreyfus, C. F. Differential regulation of neurotrophin expression in basal forebrain astrocytes by
neuronal signals. Journal of neuroscience research 76, 76–85, https://doi.org/10.1002/jnr.20060 (2004).
41. Bsibsi, M., Ravid, R., Gveric, D. & van Noort, J. M. Broad expression of Toll-like receptors in the human central nervous system.
Journal of neuropathology and experimental neurology 61, 1013–1021 (2002).
42. Daginakatte, G. C. et al. Expression profiling identifies a molecular signature of reactive astrocytes stimulated by cyclic AMP or
proinflammatory cytokines. Experimental neurology 210, 261–267, https://doi.org/10.1016/j.expneurol.2007.10.016 (2008).
43. Yeh, T. H., Lee, D. Y., Gianino, S. M. & Gutmann, D. H. Microarray analyses reveal regional astrocyte heterogeneity with
implications for neurofibromatosis type 1 (NF1)-regulated glial proliferation. Glia 57, 1239–1249, https://doi.org/10.1002/
glia.20845 (2009).
44. Medina, I. et al. Babelomics: an integrative platform for the analysis of transcriptomics, proteomics and genomic data with
advanced functional profiling. Nucleic acids research 38, W210–213, https://doi.org/10.1093/nar/gkq388 (2010).
45. Eisen, M. B., Spellman, P. T., Brown, P. O. & Botstein, D. Cluster analysis and display of genome-wide expression patterns.
Proceedings of the National Academy of Sciences of the United States of America 95, 14863–14868 (1998).
46. Dennis, G. Jr. et al. DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome biology 4, P3 (2003).
47. Gos, T. et al. S100B-immunopositive astrocytes and oligodendrocytes in the hippocampus are differentially afflicted in unipolar
and bipolar depression: a postmortem study. J Psychiatr Res 47, 1694–1699, https://doi.org/10.1016/j.jpsychires.2013.07.005
(2013).
48. Hachem, S. et al. Spatial and temporal expression of S100B in cells of oligodendrocyte lineage. Glia 51, 81–97, https://doi.
org/10.1002/glia.20184 (2005).
49. Vandevelde, M., Fankhauser, R., Kristensen, F. & Kristensen, B. Immunoglobulins in demyelinating lesions in canine distemper
encephalitis. An immunohistological study. Acta Neuropathol 54, 31–41 (1981).
50. Vandevelde, M. et al. Demyelination in experimental canine distemper virus infection: immunological, pathologic, and
immunohistological studies. Acta Neuropathol 56, 285–293 (1982).
51. Vandevelde, M., Zurbriggen, A., Higgins, R. J. & Palmer, D. Spread and distribution of viral antigen in nervous canine distemper.
Acta Neuropathol 67, 211–218 (1985).
52. Summers, B. A., Greisen, H. A. & Appel, M. J. Early events in canine distemper demyelinating encephalomyelitis. Acta Neuropathol
46, 1–10 (1979).
53. Higgins, R. J., Krakowka, S. G., Metzler, A. E. & Koestner, A. Primary demyelination in experimental canine distemper virus
induced encephalomyelitis in gnotobiotic dogs. Sequential immunologic and morphologic findings. Acta Neuropathol 58, 1–8
(1982).
54. Krakowka, S. & Koestner, A. Age-related susceptibility to infection with canine distemper virus in gnotobiotic dogs. J Infect Dis
134, 629–632 (1976).
55. Yoon, H., Walters, G., Paulsen, A. R. & Scarisbrick, I. A. Astrocyte heterogeneity across the brain and spinal cord occurs
developmentally, in adulthood and in response to demyelination. PLoS One 12, e0180697, https://doi.org/10.1371/journal.
pone.0180697 (2017).
56. Brosnan, C. F. & Raine, C. S. The astrocyte in multiple sclerosis revisited. Glia 61, 453–465, https://doi.org/10.1002/glia.22443
(2013).
57. Ponath, G. et al. Myelin phagocytosis by astrocytes after myelin damage promotes lesion pathology. Brain: a journal of neurology
140, 399–413, https://doi.org/10.1093/brain/aww298 (2017).
58. Pham, H. et al. The astrocytic response in early experimental autoimmune encephalomyelitis occurs across both the grey and white
matter compartments. Journal of neuroimmunology 208, 30–39, https://doi.org/10.1016/j.jneuroim.2008.12.010 (2009).
59. Wang, D. et al. Astrocyte-associated axonal damage in pre-onset stages of experimental autoimmune encephalomyelitis. Glia 51,
235–240, https://doi.org/10.1002/glia.20199 (2005).
60. Rubio, N., Cerciat, M., Unkila, M., Garcia-Segura, L. M. & Arevalo, M. A. An in vitro experimental model of neuroinflammation:
the induction of interleukin-6 in murine astrocytes infected with Theiler’s murine encephalomyelitis virus, and its inhibition by
oestrogenic receptor modulators. Immunology 133, 360–369, https://doi.org/10.1111/j.1365-2567.2011.03448.x (2011).
61. Papadopoulos, M. C., Manley, G. T., Krishna, S. & Verkman, A. S. Aquaporin-4 facilitates reabsorption of excess fluid in vasogenic
brain edema. FASEB journal: official publication of the Federation of American Societies for Experimental Biology 18, 1291–1293,
https://doi.org/10.1096/fj.04-1723fje (2004).
62. Papadopoulos, M. C. & Verkman, A. S. Aquaporin-4 and brain edema. Pediatric nephrology 22, 778–784, https://doi.org/10.1007/
s00467-006-0411-0 (2007).
63. Binder, D. K. et al. Increased seizure duration and slowed potassium kinetics in mice lacking aquaporin-4 water channels. Glia 53,
631–636, https://doi.org/10.1002/glia.20318 (2006).
64. Popescu, B. F. et al. Diagnostic utility of aquaporin-4 in the analysis of active demyelinating lesions. Neurology 84, 148–158, https://
doi.org/10.1212/WNL.0000000000001126 (2015).

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 14

www.nature.com/scientificreports/ www.nature.com/scientificreports

65. Zhang, W. et al. Neuroprotective effects of ischemic postconditioning on global brain ischemia in rats through upregulation of
hippocampal glutamine synthetase. Journal of clinical neuroscience: official journal of the Neurosurgical Society of Australasia 18,
685–689, https://doi.org/10.1016/j.jocn.2010.08.027 (2011).
66. Zou, J. et al. Glutamine synthetase down-regulation reduces astrocyte protection against glutamate excitotoxicity to neurons.
Neurochemistry international 56, 577–584, https://doi.org/10.1016/j.neuint.2009.12.021 (2010).
67. Steffek, A. E., McCullumsmith, R. E., Haroutunian, V. & Meador-Woodruff, J. H. Cortical expression of glial fibrillary acidic
protein and glutamine synthetase is decreased in schizophrenia. Schizophrenia research 103, 71–82, https://doi.org/10.1016/j.
schres.2008.04.032 (2008).
68. Le Prince, G. et al. Glutamine synthetase (GS) expression is reduced in senile dementia of the Alzheimer type. Neurochemical
research 20, 859–862 (1995).
69. Eid, T. et al. Loss of glutamine synthetase in the human epileptogenic hippocampus: possible mechanism for raised extracellular
glutamate in mesial temporal lobe epilepsy. Lancet 363, 28–37 (2004).
70. van der Hel, W. S. et al. Reduced glutamine synthetase in hippocampal areas with neuron loss in temporal lobe epilepsy. Neurology
64, 326–333, https://doi.org/10.1212/01.WNL.0000149636.44660.99 (2005).
71. Lee, A. et al. Rapid loss of glutamine synthetase from astrocytes in response to hypoxia: implications for excitotoxicity. Journal of
chemical neuroanatomy 39, 211–220, https://doi.org/10.1016/j.jchemneu.2009.12.002 (2010).
72. Yu, X. H. et al. Time-dependent reduction of glutamine synthetase in retina of diabetic rats. Experimental eye research 89, 967–971,
https://doi.org/10.1016/j.exer.2009.08.006 (2009).
73. Hardin-Pouzet, H. et al. Glutamate metabolism is down-regulated in astrocytes during experimental allergic encephalomyelitis.
Glia 20, 79–85 (1997).
74. Ohgoh, M. et al. Altered expression of glutamate transporters in experimental autoimmune encephalomyelitis. Journal of
neuroimmunology 125, 170–178 (2002).
75. Castegna, A. et al. Oxidative stress and reduced glutamine synthetase activity in the absence of inflammation in the cortex of mice
with experimental allergic encephalomyelitis. Neuroscience 185, 97–105, https://doi.org/10.1016/j.neuroscience.2011.04.041
(2011).
76. Werner, P., Pitt, D. & Raine, C. S. Glutamate excitotoxicity–a mechanism for axonal damage and oligodendrocyte death in Multiple
Sclerosis? Journal of neural transmission. Supplementum, 375–385 (2000).
77. Hinson, S. R. et al. Molecular outcomes of neuromyelitis optica (NMO)-IgG binding to aquaporin-4 in astrocytes. Proceedings of
the National Academy of Sciences of the United States of America 109, 1245–1250, https://doi.org/10.1073/pnas.1109980108 (2012).
78. Hinson, S. R. et al. Aquaporin-4-binding autoantibodies in patients with neuromyelitis optica impair glutamate transport by down-
regulating EAAT2. The Journal of experimental medicine 205, 2473–2481, https://doi.org/10.1084/jem.20081241 (2008).
79. D’Intino, G. et al. A molecular study of hippocampus in dogs with convulsion during canine distemper virus encephalitis. Brain
research 1098, 186–195, https://doi.org/10.1016/j.brainres.2006.04.051 (2006).
80. Rutter, A. R. et al. Evidence from gene knockout studies implicates Asc-1 as the primary transporter mediating d-serine reuptake
in the mouse CNS. The European journal of neuroscience 25, 1757–1766, https://doi.org/10.1111/j.1460-9568.2007.05446.x (2007).
81. Mothet, J. P. et al. D-serine is an endogenous ligand for the glycine site of the N-methyl-D-aspartate receptor. Proceedings of the
National Academy of Sciences of the United States of America 97, 4926–4931 (2000).
82. Fadda, E., Danysz, W., Wroblewski, J. T. & Costa, E. Glycine and D-serine increase the affinity of N-methyl-D-aspartate sensitive
glutamate binding sites in rat brain synaptic membranes. Neuropharmacology 27, 1183–1185 (1988).
83. Choi, D. W. & Rothman, S. M. The role of glutamate neurotoxicity in hypoxic-ischemic neuronal death. Annual review of
neuroscience 13, 171–182, https://doi.org/10.1146/annurev.ne.13.030190.001131 (1990).
84. Xie, X. et al. Lack of the alanine-serine-cysteine transporter 1 causes tremors, seizures, and early postnatal death in mice. Brain
research 1052, 212–221, https://doi.org/10.1016/j.brainres.2005.06.039 (2005).
85. Ehmsen, J. T. et al. The astrocytic transporter SLC7A10 (Asc-1) mediates glycinergic inhibition of spinal cord motor neurons.
Scientific reports 6, 35592, https://doi.org/10.1038/srep35592 (2016).
86. Cristino, L. et al. d-Aspartate oxidase influences glutamatergic system homeostasis in mammalian brain. Neurobiology of aging 36,
1890–1902, https://doi.org/10.1016/j.neurobiolaging.2015.02.003 (2015).
87. Rose, E. M. et al. Glutamate transporter coupling to Na,K-ATPase. The Journal of neuroscience: the official journal of the Society for
Neuroscience 29, 8143–8155, https://doi.org/10.1523/JNEUROSCI.1081-09.2009 (2009).
88. Matute, C., Domercq, M. & Sanchez-Gomez, M. V. Glutamate-mediated glial injury: mechanisms and clinical importance. Glia 53,
212–224, https://doi.org/10.1002/glia.20275 (2006).
89. Matute, C., Sanchez-Gomez, M. V., Martinez-Millan, L. & Miledi, R. Glutamate receptor-mediated toxicity in optic nerve
oligodendrocytes. Proceedings of the National Academy of Sciences of the United States of America 94, 8830–8835 (1997).
90. McDonald, J. W., Althomsons, S. P., Hyrc, K. L., Choi, D. W. & Goldberg, M. P. Oligodendrocytes from forebrain are highly
vulnerable to AMPA/kainate receptor-mediated excitotoxicity. Nature medicine 4, 291–297 (1998).
91. Macrez, R., Stys, P. K., Vivien, D., Lipton, S. A. & Docagne, F. Mechanisms of glutamate toxicity in multiple sclerosis: biomarker
and therapeutic opportunities. The Lancet. Neurology 15, 1089–1102, https://doi.org/10.1016/S1474-4422(16)30165-X (2016).
92. Bolton, C. & Paul, C. Glutamate receptors in neuroinflammatory demyelinating disease. Mediators of inflammation 2006, 93684,
https://doi.org/10.1155/MI/2006/93684 (2006).
93. Sharief, M. K., Noori, M. A., Douglas, M. R. & Semra, Y. K. Upregulated survivin expression in activated T lymphocytes correlates
with disease activity in multiple sclerosis. European journal of neurology 9, 503–510 (2002).
94. Rubio, N., Garcia-Segura, L. M. & Arevalo, M. A. Survivin prevents apoptosis by binding to caspase-3 in astrocytes infected with
the BeAn strain of Theiler’s murine encephalomyelitis virus. Journal of neurovirology 18, 354–363, https://doi.org/10.1007/s13365-
012-0112-3 (2012).
95. Rubio, N. & Sanz-Rodriguez, F. Overexpression of caspase 1 in apoptosis-resistant astrocytes infected with the BeAn Theiler’s virus.
Journal of neurovirology 22, 316–326, https://doi.org/10.1007/s13365-015-0400-9 (2016).
96. Gerhauser, I. et al. Dynamic changes and molecular analysis of cell death in the spinal cord of SJL mice infected with the BeAn
strain of Theiler’s murine encephalomyelitis virus. Apoptosis: an international journal on programmed cell death 23, 170–186,
https://doi.org/10.1007/s10495-018-1448-9 (2018).
97. Alves, L. et al. SLAM- and nectin-4-independent noncytolytic spread of canine distemper virus in astrocytes. Journal of virology
89, 5724–5733, https://doi.org/10.1128/JVI.00004-15 (2015).
98. Stone, T. W. & Darlington, L. G. Endogenous kynurenines as targets for drug discovery and development. Nature reviews. Drug
discovery 1, 609–620, https://doi.org/10.1038/nrd870 (2002).
99. Dai, X. & Zhu, B. T. Indoleamine 2,3-dioxygenase tissue distribution and cellular localization in mice: implications for its biological
functions. The journal of histochemistry and cytochemistry: official journal of the Histochemistry Society 58, 17–28, https://doi.
org/10.1369/jhc.2009.953604 (2010).
100. Guillemin, G. J., Smythe, G., Takikawa, O. & Brew, B. J. Expression of indoleamine 2,3-dioxygenase and production of quinolinic
acid by human microglia, astrocytes, and neurons. Glia 49, 15–23, https://doi.org/10.1002/glia.20090 (2005).
101. Fallarino, F. et al. T cell apoptosis by tryptophan catabolism. Cell death and differentiation 9, 1069–1077, https://doi.org/10.1038/
sj.cdd.4401073 (2002).

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 15

www.nature.com/scientificreports/ www.nature.com/scientificreports

102. Fallarino, F. et al. Modulation of tryptophan catabolism by regulatory T cells. Nature immunology 4, 1206–1212, https://doi.
org/10.1038/ni1003 (2003).
103. Uyttenhove, C. et al. Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine
2,3-dioxygenase. Nature medicine 9, 1269–1274, https://doi.org/10.1038/nm934 (2003).
104. Carlin, J. M., Ozaki, Y., Byrne, G. I., Brown, R. R. & Borden, E. C. Interferons and indoleamine 2,3-dioxygenase: role in
antimicrobial and antitumor effects. Experientia 45, 535–541 (1989).
105. Murray, H. W. et al. Role of tryptophan degradation in respiratory burst-independent antimicrobial activity of gamma interferon-
stimulated human macrophages. Infection and immunity 57, 845–849 (1989).
106. Obojes, K., Andres, O., Kim, K. S., Daubener, W. & Schneider-Schaulies, J. Indoleamine 2,3-dioxygenase mediates cell type-specific
anti-measles virus activity of gamma interferon. Journal of virology 79, 7768–7776, https://doi.org/10.1128/JVI.79.12.7768-
7776.2005 (2005).
107. Kwidzinski, E. et al. Indolamine 2,3-dioxygenase is expressed in the CNS and down-regulates autoimmune inflammation. FASEB
journal: official publication of the Federation of American Societies for Experimental Biology 19, 1347–1349, https://doi.org/10.1096/
fj.04-3228fje (2005).
108. Okuda, S., Nishiyama, N., Saito, H. & Katsuki, H. Hydrogen peroxide-mediated neuronal cell death induced by an endogenous
neurotoxin, 3-hydroxykynurenine. Proceedings of the National Academy of Sciences of the United States of America 93, 12553–12558
(1996).
109. Stone, T. W. & Perkins, M. N. Quinolinic acid: a potent endogenous excitant at amino acid receptors in CNS. European journal of
pharmacology 72, 411–412 (1981).
110. Braidy, N., Grant, R., Adams, S., Brew, B. J. & Guillemin, G. J. Mechanism for quinolinic acid cytotoxicity in human astrocytes and
neurons. Neurotoxicity research 16, 77–86, https://doi.org/10.1007/s12640-009-9051-z (2009).
111. Kwidzinski, E. & Bechmann, I. IDO expression in the brain: a double-edged sword. Journal of molecular medicine (Berlin, Germany)
85, 1351–1359, https://doi.org/10.1007/s00109-007-0229-7 (2007).
112. Pekny, M. & Pekna, M. Reactive gliosis in the pathogenesis of CNS diseases. Biochimica et biophysica acta 1862, 483–491, https://
doi.org/10.1016/j.bbadis.2015.11.014 (2016).

Acknowledgements
The authors wish to thank Petra Grünig and Bettina Buck for their excellent technical assistance. The used anti-
CDV nucleoprotein antibody was kindly provided by Dr. C. Örvell (Division of Clinical Microbiology, Karolinska
University Hospital, Karolinska Institute, Stockholm, Sweden). This study was funded by the German Research
Foundation (DFG, BE 4200/4-1; 398066876/GRK 2485/1). This publication was supported by DFG and University
of Veterinary Medicine Hannover, Foundation within the funding programme Open Access Publishing.

Author Contributions
J.A.K., M.C. and E.C. performed and evaluated the immunohistochemistry, immunofluorescence double labeling
and gene expression analysis. D.K. and V.P. performed immunofluorescence double labeling. M.I. performed
histopathological characterization and immunohistochemical staining and evaluation. Statistical analysis of
immunohistochemical investigations and preparation of figures was performed by J.A.K. and M.C. The presented
manuscript was written by J.A.K. R.U. conducted microarray analysis and Fisher’s exact tests and prepared figures.
All work was planned, guided and supervised by A.B., W.B., V.H. and C.P. All authors reviewed the manuscript.

Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-48146-9.
Competing Interests: The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© The Author(s) 2019

Scientific Reports | (2019) 9:11689 | https://doi.org/10.1038/s41598-019-48146-9 16