Inhibition of bradykinin receptor B1 protects mice from focal brain injury by reducing blood-brain barrier leakage and inflammation.

Inhibition of bradykinin receptor B1 protects mice
from focal brain injury by reducing blood–brain
barrier leakage and inflammation
Furat Raslan1,5, Tobias Schwarz2,5, Sven G Meuth2, Madeleine Austinat2, Michael Bader3,
Thomas Renne´4, Klaus Roosen1, Guido Stoll2, Anna-Leena Sire´n1 and
Christoph Kleinschnitz2
1Department of Neurosurgery, University of Wu¨rzburg, Wu¨rzburg, Germany; 2Department of Neurology,
University of Wu¨rzburg, Wu¨rzburg, Germany; 3Max-Delbru¨ck Center for Molecular Medicine, Berlin-Buch,
Germany; 4Department of Molecular Medicine and Surgery, Karolinska University Hospital, Stockholm,
Sweden
Kinins are proinflammatory and vasoactive peptides that are released during tissue damage and
may contribute to neuronal degeneration, inflammation, and edema formation after brain injury by
acting on discrete bradykinin receptors, B1R and B2R. We studied the expression of B1R and B2R
and the effect of their inhibition on lesion size, blood–brain barrier (BBB) disruption, and
inflammatory processes after a focal cryolesion of the right parietal cortex in mice. B1R and B2R
gene transcripts were significantly induced in the lesioned hemispheres of wild-type mice (P<0.05).
The volume of the cortical lesions and neuronal damage at 24h after injury in B1R�/� mice were
significantly smaller than in wild-type controls (2.5±2.6 versus 11.5±3.9mm3, P<0.001). Treatment
with the B1R antagonist R-715 1h after lesion induction likewise reduced lesion volume in wild-type
mice (2.6±1.4 versus 12.2±6.1mm3, P<0.001). This was accompanied by a remarkable reduction of
BBB disruption and tissue inflammation. In contrast, genetic deletion or pharmacological inhibition
of B2R had no significant impact on lesion formation or the development of brain edema. We
conclude that B1R inhibition may offer a novel therapeutic strategy after acute brain injuries.
Journal of Cerebral Blood Flow & Metabolism advance online publication, 3 March 2010; doi:10.1038/jcbfm.2010.28
Keywords: endothelin-1; Evan’s Blue; IL-1; macrophages; TBI; TNF
Introduction
Traumatic brain injury (TBI) remains a major cause of
disability and death in developed countries (Tagliaferri
et al, 2006). In spite of extensive efforts and advances in
understanding of the acute pathophysiology of TBI, no
specific therapies are available thus far (Beauchamp
et al, 2008; Margulies and Hicks, 2009). The initial
traumatic insult initiates self-propagating deleterious
biochemical and immunologic processes leading to
secondary damage characterized by widespread degen-
eration of neurons and glial cells (Greve and Zink,
2009; Margulies and Hicks, 2009). Key contributing
factors to the secondary brain damage are inflamma-
tion, metabolic disturbances, and cerebrovascular dys-
function that further propagates injury-induced tissue
ischemia and brain edema due to breakdown of the
blood–brain barrier (BBB) (Margulies and Hicks, 2009).
The kallikrein–kinin system represents a potential
endogenous target to combat injury-induced edema
and inflammation (Unterberg et al, 1986; Schilling
and Wahl, 1999). All components of the kallikrein–
kinin system have been identified in the brain
(Camargo et al, 1973; Kariya et al, 1985; Kizuki
et al, 1994) and kinins, for example, bradykinin
and kallidin constitute the end products of this
enzymatic cascade. Kinins are proinflammatory
Received 9 November 2009; revised 18 December 2009; accepted
5 February 2010
Correspondence: Dr C Kleinschnitz, Department of Neurology,
University of Wuerzburg, Josef-Schneider Str. 11, Wuerzburg
97080, Germany.
E-mail: christoph.kleinschnitz@mail.uni-wuerzburg.de
or Dr A-L Sire´n, Department of Neurosurgery, University of
Wuerzburg, Josef-Schneider Str. 11, Wuerzburg 97080, Germany.
E-mail: siren.a@nch.uni-wuerzburg.de
This work was supported by the Interdisziplina¨res Zentrum fu¨r
Klinische Forschung (IZKF), University of Wuerzburg, Germany
(TP E-35), the Wilhelm Sander-Stiftung, Germany (Grant
209.017.1) and the Deutsche Forschungsgemeinschaft (SFB 688,
TP A13 to CK).
5These authors contributed equally to this work.
Journal of Cerebral Blood Flow & Metabolism (2010), 1–10
& 2010 ISCBFM All rights reserved 0271-678X/10 $32.00
www.jcbfm.com

peptides that mediate the classic symptoms of
inflammation, vascular, and pain responses to tissue
injury by stimulating their two G-protein-coupled
receptors, bradykinin receptor B1 (B1R) and B2R
(Marceau and Regoli, 2004; Leeb-Lundberg et al,
2005). The tissue expression of B2R is ubiquitous
and constitutive (Marceau and Regoli, 2004). This
receptor subtype mediates the physiologic effects of
bradykinin, whereas the B1R is transiently induced
by tissue injury and inflammation. After their
activation, both B1R and B2R mediate the classic
inflammatory processes after tissue injury such
as proinflammatory cytokine release, immune
cell influx, and increased vascular permeability
(Marceau and Regoli, 2004; Leeb-Lundberg et al,
2005). Previous studies have established a function
for the kallikrein–kinin system in the pathophysiol-
ogy of ischemic injury and TBI (Unterberg et al,
1986; Schilling and Wahl, 1999; Plesnila et al,
2001; Austinat et al, 2009). In these earlier studies,
B2R has been characterized as the receptor mediating
detrimental effects of kinins after acute brain injury
(Gorlach et al, 2001; Plesnila et al, 2001; Thal et al,
2009; Trabold et al, 2010; Zweckberger and Plesnila,
2009). In a recent study (Austinat et al, 2009) we
identified the B1R as novel target in experimental
stroke. Thus, a dramatic reduction in infarct
volume, brain edema, and inflammation was seen
in B1R�/� mice subjected to transient middle cerebral
artery occlusion (Austinat et al, 2009). These
protective actions were also present after pharmaco-
logical antagonism of B1R (Austinat et al, 2009).
Remarkably, in our previous study no beneficial
effect on infarct volumes, brain edema, or inflamma-
tion could be observed in B2R-deficient mice or in
wild-type mice treated with a selective B2R inhibitor
(Austinat et al, 2009). To investigate whether this
pattern of bradykinin receptor action could also be
observed in another model of acute brain injury,
we tested the effects of B1R and B2R deficiency or
pharmacological blockade on lesion volume, BBB
permeability, and inflammatory processes after
a cryogenic cortical trauma, an established brain
injury model for profound brain edema and BBB
leakage in mice (Sire´n et al, 2006; Beauchamp et al,
2008).
Materials and methods
Animals
A total of 142 mice were used in this study. All
experiments were approved by and conducted in accor-
dance with the laws and regulations of the regulatory
authorities for animal care and use in Lower Franconia.
Male 8-week-old B1R�/� and B2R�/� mice backcrossed for
more than 10 generations to C57Bl/6 background (Pesquero
et al, 2000; Austinat et al, 2009) or C57Bl/6 wild-type mice
(Charles River, Sulzfeld, Germany) were used. The weight
of mice ranged between 18 and 24 g.
Surgery
We deliberately chose the cryogenic lesion model in mice
to address the function of kinin receptors in focal brain
trauma. This model is advantageous in comparison to more
established models of diffuse TBI (e.g., fluid percussion,
cortical impact) in that the lesions are clearly circum-
scribed and highly reproducible in size and location.
Moreover, cryogenic cortical injury has been shown to
induce early and profound BBB leakage and inflammation
(Sire´n et al, 2006), two key readout parameters of the
present investigation.
The mice were anesthetized with intraperitoneal injec-
tions of ketamine (0.1mg/g) and xylazine (0.005mg/g).
Surgery was performed on the right parietal cortex after
exposing the skull through a scalp incision as described
(Sire´n et al, 2006). Briefly, a copper cylinder with a tip
diameter of 2.5mm was filled with liquid nitrogen
(�183 1C) and placed stereotactically on the right parietal
cortex (coordinates from bregma: 1.5mm posterior, 1.5mm
lateral) for 90 secs. Sham-operated animals went through
the same procedure without cooling the copper cylinder.
All operations were performed by the same operator (FR)
masked to the genotype or treatment group.
The selective B1R inhibitor R-715 (Gobeil et al, 1996)
(Ac-Lys-[D-b Nal7, Ile8]-des-Arg9-BK; Biomatik) was
administered intravenously 1h before or after cryolesion
at a dosage of 0.5 or 1mg/kg body weight, respectively. For
the selective blockade of B2R, Hoe140 (Wirth et al, 1991)
(D-Arg0-Hyp3-Thi5-D-Tic7-Oci8-BK; 0.2 or 0.4mg/kg body
weight; Sigma-Aldrich, Munich, Germany) was injected
intravenously 1h after the induction of the cryogenic
lesion. To further address a potential B2R effect in the
development of cryogenic lesion injury, we treated B1R�/�
mice with the selective B2R inhibitor Hoe140 (0.2mg/kg
body weight intravenously) 1 h after setting of the lesion.
Animals were always killed 24h after surgery for analysis.
Determination of Lesion Size
At 24h after lesion induction, the brains were quickly
removed and cut into six 1-mm-thick coronal sections
using a mouse brain slice matrix (Harvard Apparatus,
Holliston, MA, USA). The slices were stained with 2%
2,3,5-triphenyltetrazolium chloride (TTC; Sigma-Aldrich)
in phosphate-buffered saline (PBS) to visualize the lesions.
Planimetric measurements (ImageJ software; National
Institutes of Health, Bethesda, MD, USA) of the slices
were performed masked to the groups and were used to
calculate lesion volumes (Austinat et al, 2009).
Determination of Blood–Brain Barrier Permeability
To determine the permeability of the cerebral vasculature,
we intravenously injected 2% Evan’s Blue tracer (Sigma-
Aldrich) diluted in 0.9% NaCl 2h after the induction of the
cryogenic lesion in R-715- and Hoe140-treated mice or
vehicle-treated controls. After 24h mice were transcar-
dially perfused with 4% paraformaldehyde and brains
were quickly removed and cut in six 1-mm-thick coronal
Bradykinin B1 receptor blockade reduces brain injury
F Raslan et al
2
Journal of Cerebral Blood Flow & Metabolism (2010), 1–10

sections using a mouse brain slice matrix (Harvard
Apparatus). Planimetric measurements (ImageJ software)
of the brain parenchyma stained with Evan’s Blue were
performed to calculate edema volumes.
PCR Studies
RNA was always isolated from a 2-mm-thick coronal brain
slice containing the cortical cryogenic lesion (coordinates
from bregma: 1.5mm posterior, 1.5mm lateral) at 12, 24, and
48h after injury. Tissue homogenization, RNA isolation, and
real-time RT-PCR were performed as described (Austinat et
al, 2009). Briefly, total RNAwas prepared with a Miccra D-8
power homogenizer (ART, Mu¨hlheim, Germany) using the
TRIzol reagent (Invitrogen, Karlsruhe, Germany) and was
quantified spectrophotometrically. Then, 250mg of total
RNA was reversely transcribed with the TaqMan Reverse
Transcription Reagents (Applied Biosystems, Darmstadt,
Germany) according to the manufacturer’s protocol using
random hexamers. Relative mRNA levels of cytokines
and B1R and B2R were quantified with the fluorescent
TaqMan technology. PCR primers and probes specific for
murine B1R (assay ID: Mm00432059_s1), B2R (assay ID:
Mm00437788_s1), IL-1b (assay ID: Mm004344228_m1),
Transforming growth factor (TGF)b-1 (assay ID:
Mm00441724_m1), endothelin-1 (ET-1) (assay ID: Mm
00438656_m1), interferon g (assay ID: Mm99999071_m1),
and tumor necrosis factor (TNF)a (assay ID:
Mm00443258_m1) were obtained as TaqMan Gene Expres-
sion Arrays (Applied Biosystems). 18s rRNA (TaqMan
Predeveloped Assay Reagents for gene expression, part
number: 4319413E; Applied Biosystems) was used as an
endogenous control to normalize the amount of sample
RNA. The PCR was performed with equal amounts of cDNA
in the GeneAmp 7700 sequence detection system (Applied
Biosystems) using the TaqMan Universal PCR Master Mix
(Applied Biosystems). Reactions were incubated at 501C for
2mins, at 951C for 10mins followed by 40 cycles of 15 secs
at 951C and 1min at 601C. Water controls were included to
ensure specificity. Each sample was measured in duplicate
and data points were examined for integrity by analysis of
the amplification plot. The DDCt method was used for
relative quantification of gene expression as described
(Livak and Schmittgen, 2001; Austinat et al, 2009).
Histology and Immunohistochemistry
Formalin-fixed brains embedded in paraffin from wild-
type mice, B1R�/� mice, and B2R�/� mice on day 1 after
cryogenic cortical injury were cut into 4-mm-thick sections
across the entire lesion. After deparaffinization and
rehydration, tissues were stained with hematoxylin and
eosin (Sigma-Aldrich). For immunohistochemistry antigen
retrieval was achieved by pretreatment with proteinase
(P8038; Sigma-Aldrich). Thereafter, endogenous peroxi-
dase activity was blocked with 3% H2O2 in methanol for
15mins and unspecific binding was prevented by adding
10% bovine serum albumin for 30mins. For staining of
activated macrophages/microglia, we applied a rat anti-
mouse F4/80 antibody (BM4008; Acris, Hiddenhausen,
Germany) at a dilution of 1:100 in PBS containing 1% BSA
overnight at 41C. Subsequently the slices were incubated
with a biotinylated anti-rat IgG (BA-4001; Vector Labora-
tories, Burlingame, CA, USA) diluted 1:100 in PBS
containing 1% BSA for 45mins at room temperature. The
secondary antibody was linked through streptavidin to a
biotinylated peroxidase and staining was performed
according to the manufacturer’s instructions using Strep-
tABComplex/HRP Duet kit (K0492; Dako-Cytomation,
Hamburg, Denmark) and 3,30-diaminobenzidene (Kem-En-
Tec Diagnostics, Taastrup, Denmark) and counterstained
with aqueous hematoxylin. Negative controls included
omission of primary or secondary antibody and gave no
signals (not shown); tissue sections from murine spleens
served as positive controls (not shown).
Statistical Analysis
Results are presented as mean±s.d. All data were tested
for Gaussian distribution with the D’Agostino and Pearson
omnibus normality test and then analyzed by one-way
analysis of variance (ANOVA) followed by Bonferroni post
hoc test for correction of multiple comparisons. For
statistical analysis Prism Graph 4.0 software (GraphPad
Software, San Diego, CA, USA) was used. P values < 0.05
were considered statistically significant.
Results
Expression of B1R and B2R in the Lesioned Cortex of
Wild-Type Mice
We first analyzed the mRNA expression of B1R and
B2R in the ipsilateral cortex of C57Bl/6 mice after
cryogenic brain trauma over time (Figure 1). Both
receptors were constitutively expressed at low levels
in sham-treated animals. B1R and B2R gene tran-
scripts significantly increased after 12 h with a
dominant B1R expression (B1R: 11.4±10.0-fold
induction, P<0.05; B2R: 6.4±4.5-fold induction,
P<0.05). Although B1R induction was transient,
B1R
B2R
R
el
. g
en
e
ex
pr
es
si
on
25
15
10
5
0
20
*
*
* *
Sham 24h 48h12h
Figure 1 Bradykinin receptor B1 (B1R) and B2R are induced
after focal brain damage. Relative gene expression of B1R and
B2R in the lesioned cortex of wild-type mice 12, 24, and 48h
after cryogenic cortical injury or sham operation (n=5 per time
point). *P<0.05, one-way ANOVA, Bonferroni post hoc test
compared with sham-operated mice.
Bradykinin B1 receptor blockade reduces brain injury
F Raslan et al
3
Journal of Cerebral Blood Flow & Metabolism (2010), 1–10

B2R mRNA expression was still significantly inc-
reased at 24 and 48h after injury (P<0.05) (Figure 1).
Taken together, these data indicate that both kinin
receptors are expressed in the murine brain and—
just like their ligand bradykinin (Trabold et al,
2010)—undergo induction after brain trauma, sug-
gesting a functional task of the kallikrein–kinin
system after focal brain injury.
Cortical Lesion Volume in B1R�/� Mice, B2R�/� Mice,
and Wild-Type Controls
We subjected B1R�/� and B2R�/� mice to cryogenic
cortical injury and, after 24h, assessed lesion volumes
by staining brain sections with TTC (Figure 2A).
B1R�/� mice developed significantly smaller cortical
brain lesions compared with wild-type controls
(2.5±2.6 versus 11.5±3.9mm3, P<0.001). In line
with these findings, hematoxylin and eosin stainings
revealed markedly less neurodegeneration in the
cortices of B1R�/� mice than in those of wild-type
controls (Figure 2B). In contrast, B2R-deficient mice
only showed a tendency toward reduced lesion
volumes but the difference did not reach statistical
significance (6.5±5.1 versus 11.5±3.9mm3, P>0.05)
(Figure 2A). Previous reports have emphasized the
pathophysiological function of B2R in diffuse trau-
matic brain damage (Gorlach et al, 2001; Plesnila
et al, 2001; Hellal et al, 2003; Trabold et al, 2010;
Zweckberger and Plesnila, 2009). To further address
a potential B2R effect in the development of
cryogenic brain injury, we treated B1R�/� mice with
the selective B2R inhibitor Hoe140. Application of
Hoe140 had no additive benefit on lesion reduction
in B1R�/� mice (2.5±2.6 versus 2.8±1.8mm3,
P>0.05) (Figure 2A) hence, making a prominent
function of B2R in this model unlikely.
Effect of a Pharmacological Blockade of B1R and B2R
in Wild-Type Mice
As congenital B1R deficiency protected from experi-
mental cryogenic brain lesion, we next tested
whether pharmacological targeting of B1R is equally
effective. In accordance with the observations made
in B1R-deficient mice, the B1R selective blocker R-
715 (1mg/kg body weight) significantly reduced
cortical lesion volumes at 24 h when administered
as a pretreatment 1 h before setting of the lesion
(2.6±1.4 versus 11.7±4.8mm3, P<0.001) (Figure 3).
Importantly, 1mg R-715 per kg body weight was also
able to diminish lesion size when applied in a
therapeutic approach 1h after cryogenic brain trau-
ma (3.9±2.4 versus 11.7±4.8mm3, P<0.001). In
contrast, R-715 administered at a decreased dosage of
0.5mg/kg body weight did not confer neuroprotec-
tion after focal brain injury (Figure 3) (P>0.05). In
line with the results obtained in genetically engi-
neered mice (Figure 2), pharmacological B2R inhibi-
tion using Hoe140 at a dosage of 0.2 or 0.4mg/kg had
no significant impact on lesion volumes on day 1
(Figure 3) (P>0.05).
Blockade of B1R Reduces Blood–Brain Barrier Leakage
and Inflammation
Edema formation and inflammation critically contri-
bute to secondary tissue damage in the cryogenic brain
injury model (Sire´n et al, 2006). We therefore studied
next the effect of B1R and B2R antagonism on BBB
integrity and the expression of vasoactive and proin-
flammatory mediators in the lesioned brain tissue.
Posttreatment with the selective B1R inhibitor
R-715 reduced Evan’s blue extravasation, that is,
edema formation, in the ipsilateral cortex to approxi-
mately 50% as compared with NaCl-treated animals
(5.8±4.5 versus 12.5±1.0mm3, P<0.05), whereas
the B2R antagonist Hoe140 had no effect on the
lesion-induced increase in brain edema formation
(P<0.05) (Figure 4A). Endothelin-1 is critically
involved in the regulation of vascular permeability
in the central nervous system and can enhance BBB
leakage under pathophysiological conditions (Lo
et al, 2005; Kreipke et al, 2010). Accordingly, gene
expression of ET-1 was induced on day 1 in lesioned
brain tissue of vehicle-treated mice compared with
sham-operated controls (3.1±0.4-fold induction,
P<0.001) (Figure 4A). Administration of R-715 1h
after the induction of focal brain injury was able to
reduce the amount of ET-1 transcripts to the level
observed in the cortices of sham-operated mice
(P<0.001). Again, the B2R antagonist Hoe140 had
no effect on ET-1 expression (Figure 4A).
We also investigated the mRNA levels of several
prototypic pro- and anti-inflammatory cytokines in
the injured brains of R-715- or Hoe140-treated mice
after cryoinjury. IL-1b and TNF-a were significantly
induced on day 1 in vehicle-treated controls com-
pared with sham-operated mice (IL-1b: 44.3±24.0-
fold induction, P<0.001; TNF-a: 31.1±10.6-fold
induction, P<0.05) (Figure 4B). Posttreatment with
R-715 prevented this IL-1b and TNF-a induction
(P<0.05) whereas administration of Hoe140 did not
influence proinflammatory cytokine levels. The
number of interferon-g and TGFb-1 gene transcripts
did not differ between R-715- and Hoe140-treated
mice or vehicle-treated controls (Figure 4B).
We finally analyzed the extent of the cellular
inflammatory response on cryogenic brain damage
by immunohistochemistry (Figure 5). Numerous
activated F4/80-positive cells had invaded the
lesions on day 1 in vehicle- and Hoe140-treated
mice. In contrast, the blockade of B1R using R-715
markedly dampened the invasion of activated macro-
phages/microglia.
Discussion
Here we show that genetic lack of BR1 receptors or
their acute pharmacological antagonism attenuates
Bradykinin B1 receptor blockade reduces brain injury
F Raslan et al
4
Journal of Cerebral Blood Flow & Metabolism (2010), 1–10

WT B1R-/- B2R-/- B1R -/-
+ Hoe140
**
16
12
8
4
0
Le
si
on
v
ol
um
e
(m
m3
)
WT B1R-/-
**
Figure 2 Bradykinin receptor B1 (B1R) deficiency protects from cortical brain damage. (A) (Top) Representative 2,3,5-
triphenyltetrazolium chloride (TTC) stains of six corresponding coronal brain sections from wild-type (WT) mice, B1R�/� mice,
B2R�/� mice, and B1R�/� mice treated with the B2R inhibitor Hoe140 (0.2mg/kg) on day 1 after cryoinjury. Lesions appear to be
smallest in B1R�/� mice (white arrows), which was confirmed by lesion volumetry (n=8 per group) (bottom) and (B) hematoxylin
and eosin staining (broken black lines). Note that the pharmacological blockade of B2R in B1R�/� mice had no additive effect on
lesion volume reduction. **P<0.001, one-way analysis of variance (ANOVA), Bonferroni post hoc test compared with WT mice.
Scale bar represents 200 mm.
Bradykinin B1 receptor blockade reduces brain injury
F Raslan et al
5
Journal of Cerebral Blood Flow & Metabolism (2010), 1–10