2426 Original article
The b-lactam antibiotic, ceftriaxone, dramatically improves
survival, increases glutamate uptake and induces
neurotrophins in stroke
Christa Thöne-Reinekea, Christian Neumanna, Pawel Namsollecka,
Kristin Schmerbacha, Maxim Krikova, Jan H. Schefea, Kristin Luchta,
Heide Hörtnagla, Michael Godesb, Susanne Müllerc, Kay Rumschüssela,
Heiko Funke-Kaisera, Arno Villringerc, U. Muscha Steckelingsa,1
and Thomas Ungera,1
Objective Ceftriaxone has been reported to reduce
neuronal damage in amyotrophic lateral sclerosis
and in an in-vitro model of neuronal ischaemia through
increased expression and activity of the glutamate
transporter, GLT1. We tested the effects of ceftriaxone on
mortality, neurological outcome, and infarct size in
experimental stroke in rats and looked for underlying
mechanisms.
Methods Male normotensive Wistar rats received
ceftriaxone (200 mg/kg intraperitoneal) as a
single injection 90 min after middle cerebral
artery occlusion (90 min with reperfusion).
Forty-eight hours after middle cerebral artery
occlusion, infarct size (MRI) and neurological
deficits were estimated. GLT1 expression was
determined by real time RT-PCR, immunoblotting and
promoter reporter assay, astrocyte GLT1 activity by
measuring glutamate uptake. Bacterial load in various
organs was measured by real time RT-PCR, neurotrophins
and IL-6 by immunoblotting.
Results Ceftriaxone dramatically reduced early (24-h)
mortality from 34.5% (vehicle treatment, n U 29) to 0%
(P < 0.01, n U 19). In a subgroup, followed up for 4 weeks,
mortality persisted at 0%. Ceftriaxone strongly tended to
reduce infarct size, it significantly improved neuronal
survival within the penumbra, reduced neurological deficits
(P < 0.001) and led to an upregulation of neurotrophins
(P < 0.01) in the peri-infarct zone. Ceftriaxone did not
Introduction
Stroke is one of the major causes of death and disability in
industrialized nations [1]. Preventive measures such as
antihypertensive or anticoagulant treatment or antifibril-
lation have become very effective during recent years [2].
However, acute therapeutic intervention is still basically
limited to the attempt of rapid reperfusion [3]. In recent
years, there has been an extensive search for neuropro-
This work has been presented only at scientific conferences.
1
Both authors contributed equally to this work.
0263-6352 ß 2008 Wolters Kluwer Health | Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
increase GLT1 expression, but increased GLT1 activity
(P < 0.05).
Conclusion Ceftriaxone causes a significant reduction in
acute stroke mortality in a poststroke treatment regimen in
animal studies. Improved neurological performance and
survival may be due to neuroprotection by activation of
GLT1 and a stimulation of neurotrophins resulting in an
increased number of surviving neurons in the penumbra.
J Hypertens 26:2426–2435 Q 2008 Wolters Kluwer Health |
Lippincott Williams & Wilkins.
Journal of Hypertension 2008, 26:2426–2435
Keywords: b-lactam antibiotic, mortality, stroke
Abbreviations: ALS, amyotrophic lateral sclerosis; BDNF, brain derived
neurotrophic factor; CBF, cerebral blood flow; dbcAMP, Dibutyryl cyclic AMP;
ECA, external carotic artery; GLT-1, glutamate transporter 1; IL-6, interleukin
6; MCAO, middle cerebral artery occlusion; MRI, magnet resonance imaging;
NT3, neurotrophin 3; RT-PCR, reverse transcription polymerase chain
reaction; TrkB, tyrosine receptor kinase B
a
Center for Cardiovascular Research (CCR)/Institute of Pharmacology, bInstitute
of Physiology and cClinic and Polyclinic for Neurology, Charité –
Universitätsmedizin Berlin, Berlin, Germany
Correspondence to Professor Thomas Unger, MD, Center for Cardiovascular
Research (CCR)/Institut für Pharmakologie, Charité – Universitätsmedizin Berlin,
Hessische Straße 3-4, 10115 Berlin, Germany
Tel: +49 30 450 525 001; fax: +49 30 450 525 901;
e-mail: thomas.unger@charite.de
Received 17 July 2008 Accepted 6 August 2008
See editorial commentary on page 2274
tective agents with the ability to increase the number of
surviving neurons after stroke, in particular within the so
called penumbra, the tissue at highest risk of delayed cell
death surrounding the necrotic infarct core [4]. Although
a number of such agents proved to be effective in animal
models, most clinical studies testing these agents in man
have been rather disappointing [4].
A number of recent publications suggested antibiotics to
act neuroprotective by mechanisms exceeding their anti-
microbial properties [5–9]. Two of them, minocycline
DOI:10.1097/HJH.0b013e328313e403
 Reduced mortality in stroke by ceftriaxone Thöne-Reineke et al. 2427
and dapsone, have very recently been tested in clinical
pilot-studies in stroke patients, and both were effective in
improving neurological outcome [10,11].
Beta-lactam antibiotics represent another class of anti-
biotics that have lately been reported to possess neuro-
protective features [5,9,12,13]. Ceftriaxone, which was
the most effective of several b-lactam antibiotics
screened, reduced neuronal damage in in-vitro models
of ischaemia and motor neurone degeneration and
delayed loss of neurons in an animal model of amyo-
trophic lateral sclerosis (ALS) [5]. Ceftriaxone-related
neuroprotection was reported to be due to an increase
in expression and activity of the glutamate transporter
GLT1 [5]. As excessive release of glutamate from nerve
endings into the extracellular interstitium seems also
causative for neuronal damage in brain ischaemia [14],
we tested the effectiveness of ceftriaxone on neurological
outcome, infarct size and acute mortality in transient
focal cerebral ischaemia in rats. Furthermore, defined
areas of the brains were examined for the impact of
ceftriaxone treatment on the expression of GLT1, the
neurotrophins BDNF and NT3, the neurotrophin recep-
tor TrkB and the cytokine interleukin 6 (IL-6). These
in vivo and ex vivo studies were complemented by the
measurement of the effect of ceftriaxone on GLT1
expression and activity in cultured primary rat astrocytes.
Finally, to exclude that the effect of ceftriaxone was due
to any antibacterial actions, bacterial load was measured
in relevant organs 2 and 6 days after middle cerebral
artery occlusion (MCAO).
Methods
Animals and treatment
Male normotensive Wistar rats (180–200 g; HARLAN
Winkelmann, Borchen, Germany) were kept in a SPF
(specific pathogen free) barrier under standardized
conditions with respect to temperature and humidity,
and were housed on a 12 h light/12 h dark cycle in
groups of 4–5 with food and water ad libitum. Animal
housing, care, and applications of experimental pro-
cedures complied with the Guide for the Care and Use of
Laboratory Animals of the State Government of Berlin,
Germany.
Animals were randomly assigned to the following four
treatment groups: vehicle/sham (n ¼ 8), ceftriaxone/sham
(n ¼ 13), vehicle/stroke [n ¼ 29 (19 survivors 48 h after
MCAO)], ceftriaxone/stroke (n ¼ 19). The dose we used
for ceftriaxone was identical to that used by other studies
looking at the neuroprotective potential of ceftriaxone
in vivo [5,9]. Ceftriaxone (200 mg/kg bodyweight i.p.) was
administered as a single injection 90 min after MCAO.
The majority of animals were followed up for 2 days
including evaluation of their neurological status on day 1
and estimation of infarct size on day 2 (for details see
chapters ‘Neurological score’ and ‘T2 weighted Magnetic
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Resonance Imaging’ below). Eight animals of the stroke/
ceftriaxone group were followed up for 4 weeks to assess
long-term mortality. In a second approach, six (sham-
groups) or ten (stroke-groups) animals per group, respect-
ively, were followed up for 7 days to determine long-term
neurological outcome and a putative effect of ceftriaxone
treatment on stroke-induced immunodeficiency-related
bacterial infections. Finally, two additional groups of
animals were included to examine GLT1 expression in
noninfarcted brains of healthy rats after a 5-day appli-
cation of ceftriaxone (200 mg/kg per day; n ¼ 7) or vehicle
(n ¼ 6), respectively.
Middle cerebral artery occlusion with reperfusion
Focal cerebral ischaemia was induced by right MCAO
with subsequent reperfusion after 90 min as described
previously [15–17]. Briefly, under general anaesthesia
with chloral hydrate (400 mg/kg), the right cervical
carotid bifurcation was exposed through a midline neck
incision. A 4–0 silicon-coated nylon monofilament (Ethi-
con, Belgium) was gently inserted through the proximal
external carotid artery (ECA) into the internal carotid
artery and finally into the middle cerebral artery. After
90 min, the filament was withdrawn into the stump of the
ECA to allow reperfusion. Sham-operated rats underwent
the same surgical procedures except insertion of the
occluding monofilament. Body temperature was main-
tained at 37
0.58C by a heating pad.
Cerebral blood flow
Cerebral blood flow (CBF) was monitored from the
beginning of surgery until 30 min after reperfusion with
a probe attached to the skull above the supply territory of
the MCA (2 mm caudal to bregma and 6 mm lateral to
midline) by laser-Doppler flowmetry (Periflux system
5000, PERIMED, Sweden). The procedure was con-
sidered successful if an over 75% drop in CBF was
observed after MCAO.
Blood parameters and body temperature
Blood samples were taken by retro-orbital puncture to
determine oxygen-partial pressure and carbon dioxide-
partial pressure, pH, haematocrit, glucose-concentration,
sodium-concentration and potassium-concentration
before and after MCAO. Data were quantified using a
RADIOMETER ABL 555 SERIES (Radiometer
medical A/S, Copenhagen, Denmark).
Body temperature was monitored in all animals during
MCAO. In those animals followed-up for 7 days in order
to detect putative bacterial infections, body temperature
was measured during MCAO and once on each consecu-
tive day.
Neurological score
Neurological evaluation was performed by a blinded
observer 24 h and 7 days after MCAO using the 18-point
 2428 Journal of Hypertension 2008, Vol 26 No 12
neurological scoring system of Garcia et al. [18]. This
neurological score tests spontaneous activity, motor
impairments and sensorial function. Severe impairments
were graded 0 or 1, and no observed deficits were
graded 3.
T2 weighted magnetic resonance imaging
MRI was performed 48 h after MCAO on a 7 T Bruker
scanner (Pharmascan 70/16 AS; Bruker Biospin, Ettlin-
gen, Germany). Isoflurane anaesthesia was controlled
using a Biotrig Animal Life Monitoring 9M CPL device
and Bio Trig BT1 software. Cerebral ischemic areas were
visualized with a T2-weighted, fat suppressed 2D turbo
spin-echo sequence (TR 5218.7 ms; TEeff 65.2 ms,
RARE factor 8 and 6 averages). Thirty-five axial slices
with a slice thickness of 0.5 mm and no interslice distance
were positioned to cover the whole brain. The field of
view (FOV) was 3.5  3.5 cm and the matrix was
256  256 resulting in an inplane resolution of 137 mm.
Calculation of lesion volume was carried out with the
program Analyze 5.0 (AnalyzeDirect, Inc., Lenexa,
Kansas, USA).
Preparation of brain areas
Brains of animals having received a single dose of cef-
triaxone after MCAO were dissected into three parts: the
peri-infarct cortex, the infarct core, and the thalamic
region. For estimation of GLT-1 expression after 5 days
of ceftriaxone treatment, frontal cortex, amygdala/piri-
form cortex, striatum, hippocampus, hypothalamus and
brain stem were dissected from frozen brains.
All preparations were preformed on a cold plate (108C)
according to Paxinos and Watson [19].
Quantitative real-time RT-PCR
Quantitative Real-time RT-PCR was conducted using a
Stratagene Mx 3000 P cycler. Data analysis has been
performed using Gene Expression’s CT Difference
method with 18S rRNA as normalization standard [20].
Primer sequences were as follows: 18S rRNA (forward,
50 -CCG CAG CTA GGA ATA ATG GAA TA-30 ) and
(reverse, 50 -CT AGC GGC GCA ATA CGA AT-30 ).
GLT1 (forward, 50 -GC CAT CTT CAT AGC CCA
GA-30 ) and (reverse, 50 -AT ACT GGC TGC ACC
AAT GCT-30 ).
Primary cell culture
Primary astrocyte-enriched cultures were prepared from
cerebral tissue of 2-day old Wistar rats [21]. After 10–14
days of culture, cell composition was 95% astrocytes and
5% microglia.
GLT1 promoter reporter assay
GLT1 promoter activity was studied in rat primary
neonatal astrocytes. 24 h after seeding in 12-well plates,
the cells were transfected using GeneJuice (Novagen,
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Germany) with pGL3-basic including a 1.5 kb GLT1
promoter-fragment or an empty control plasmid, respect-
ively, and pRL according to the manufacturer’s guide-
lines. After transfection, the cells were treated with
100 mmol/l ceftriaxone for 5 days or 200 mg/ml dbcAMP
for 3 days. Luciferase activity was determined in cell
lysates using the Dual-Luciferase Reporter Assay System
(Promega, Germany) and a Monolight 3010 luminometer
(Pharmingen, Germany).
Immunoblotting
Membranes were incubated with goat anti-GLT1
(1 : 200), rabbit anti-TrkB (1 : 200), goat anti-IL6
(1 : 200), mouse anti-glyceraldehyde-3-phosphate dehy-
drogenase (GAPDH) (1 : 100.000) polyclonal antibodies
(all from Santa Cruz, Heidelberg, Germany) or rabbit
anti-BDNF (1 : 300) or rabbit anti-NT3 polyclonal anti-
bodies (both from Chemicon, Schwalbach, Germany),
respectively, followed by an incubation with rabbit
anti-goat (1 : 2.000), donkey anti-rabbit (1 : 2000) or rabbit
anti-mouse (1 : 3.000) horseradish peroxidase (HRP)-
conjugated secondary antibodies (DAKO, Denmark
and Amersham Biosciences), respectively. Target
proteins were detected by enhanced chemiluminescence
detection system (Amersham Pharmacia Biotech).
Nissl staining
Rat brains were coronally sectioned into slices of 7 mm
thickness in a cryostat and, subsequently, fixed in acetone
for 5 min. Sections were incubated with NeuroTrace
(diluted 1 : 50 in PBS) followed by DAPI staining (both
from Molecular Probes, Eugene, Oregon, USA) according
to the manufacturer’s instructions. The area for counting
the positively stained neurons was determined by com-
paring HE-stained section and MRI sections of all
animals of this experiment to choose an area of approxi-
mately 1 mm2 that was part of the penumbra in all
animals. In this experiment, this area was in a periven-
tricular location. Cells were counted by a blinded inves-
tigator.
Glutamate uptake assay
Primary astrocytes were seeded into 12-well plates and
incubated for 5 days with vehicle (NaCl), ceftriaxone
(1 mmol/l, 10 mmol/l, 100 mmol/l) or 200 mg/ml dbcAMP
(positive control). Subsequently, cells were incubated for
another 10 min with 1 mCi/ml L-[G-3H]-glutamate
(Amersham, Little Chalfont, UK), washed, and lysates
forwarded to determination of intracellular content of
L-[G-3H] glutamate by scintillation counting.
Quantification of bacterial load in tissue samples
For estimation of bacterial load, lung and kidney were
collected 7 days after MCAO and bacterial 16 s rRNA
gene quantified by real-time PCR as described by Millar
et al. with P11P(f) and P13P(r) primer and addition of
SybrGreen to the master mix [22].
 Reduced mortality in stroke by ceftriaxone Thöne-Reineke et al. 2429
Quantification of bacterial load in plasma samples
For determination of endotoxin concentration in plasma,
the Limulus Amebocyte Lysate (LAL) Pyrochrome assay
(Associates of Cape Cod, E. Falmouth, Massachusetts,
USA) was performed according to the manufacturer’s
protocol with diazo endpoint option.
Statistics
Data were analysed using SPSS 12.0 software (SPSS, Inc.,
Chicago, Illinois, USA). Results are expressed as mean

standard error of the mean (SEM). Differences
between groups were compared by Mann-Whitney U-
Test, ANOVA, and student’s t-test. Significance of differ-
ences in mortality was calculated by chi-squared test and
Fisher exact test. All tests were two-sided, and P < 0.05
was considered significant.
Results
Cerebral blood flow
During MCAO, all animals showed a significant
(P < 0.01) and greater than 75% reduction in CBF in
comparison to baseline conditions (data not shown). After
the withdrawal of the occluding filament, ipsilateral blood
flow was restored to approximately 90% of baseline
levels. Reduction in CBF during MCAO and during
reperfusion was identical in vehicle-treated groups and
ceftriaxone-treated groups.
Blood parameters
Oxygen, carbon dioxide, pH, glucose, haematocrit,
sodium and potassium showed no significant differences
at any time point between all four treatment groups (data
not shown).
Infarct size
Infarct size was measured 2 days after MCAO by MRI
and computer-powered image analysis (Fig. 1a). Ceftriax-
one treatment tended to reduce infarct size as compared
with vehicle-treated animals, but this effect did not reach
significance (Fig. 1b).
Estimation of surviving neurons by Nissl staining
Counting of Nissl-positive and, thus, viable neurons in
the periventricular area of the penumbra revealed that
48 h after focal cerebral ischaemia, the number of viable
neurons was significantly reduced, and that this loss of
neurons could be partly and significantly restored by
ceftriaxone treatment (Fig. 1c).
Neurological evaluation
Neurological evaluation was performed 24 h and 7 days
after MCAO, respectively, using an 18-point neurological
score [18]. Successful induction of focal cerebral ischae-
mia led to a significantly reduced test score 24 h after
MCAO in vehicle or ceftriaxone treated rats, respectively
(Fig. 1d). Postinsult treatment with ceftriaxone resulted
in a significant improvement of neurological outcome
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
24 and 48 h after MCAO when compared with vehicle-
treated rats (Fig. 1d). Seven days after MCAO, vehicle-
treated rats presented with a pronounced, spontaneous
recovery – as described in the literature [23] – which
obliterated any differences to the drug-treated group
(Fig. 1d).
Mortality
The most striking finding of our study was that ceftri-
axone treatment dramatically and significantly reduced
mortality (Table 1). As within the first 24 h after MCAO,
mortality in control rats amounted to 35%, a single
injection of ceftriaxone 90 min after MCAO totally pre-
vented any fatalities. There were no further deaths
during the 7-day or 4-week follow-ups, respectively.
Expression of neurotrophins and interleukin-6 ex vivo
Expression of the neurotrophins, BDNF and NT3, or the
neurotrophin receptor, TrkB, and of the proinflammatory
cytokine, IL-6, was measured by immunoblotting 48 h
after MCAO in defined regions of the brains (infarct core,
peri-infarct cortex, thalamic region) of ceftriaxone-trea-
ted animals and vehicle-treated animals. BDNF and
TrkB were significantly upregulated by ceftriaxone treat-
ment when compared with vehicle treatment in the
cortical peri-infarct zone of ischaemic brains
(Fig. 2a,c). NT3 also tended to be upregulated by cef-
triaxone (P ¼ 0.05) (Fig. 2b). Stroke as such significantly
increased TrkB and tended to upregulate BDNF or NT3
expression in stroke/vehicle animals (Fig. 2d). However,
ceftriaxone treatment led to a pronounced additional
increase in neurotrophin expression clearly exceeding
the effect of ischaemia alone (Fig. 2d). Neurotrophins
or TrkB were not upregulated by ceftriaxone in sham-
operated animals (Fig. 2d). In contrast to neurotrophin
pattern, IL-6 concentrations were not altered by ceftriax-
one (data not shown).
GLT1-mRNA and -protein expression ex vivo
GLT1 protein expression was measured 48 h after MCAO.
In contrast to neurotrophin/cytokine pattern, there was
no alteration in GLT1 expression, neither by cerebral
ischaemia, nor by ceftriaxone treatment (Fig. 3a).
Additionally, GLT1 mRNA and protein expression was
determined after 5-day treatment in healthy rats. GLT1
mRNA expression was measured by real-time PCR in
specific brain regions of vehicle compared with ceftri-
axone treated animals (Fig. 3b). Ceftriaxone did not alter
GLT1 expression in hippocampus (a brain region known
to sensitively react to any disturbances in brain function
including ischaemia [24]), frontal cortex, amygdala/
piriform cortex, striatum, hypothalamus or brain stem
(Fig. 3b). These results were verified on protein level
by western blotting (Fig. 3c). Again, no alterations in
GLT1 expression were seen in response to ceftriaxone
treatment.
 2430 Journal of Hypertension 2008, Vol 26 No 12

Fig. 1

Infarct volume and neurological outcome after MCAO. (a) Representative brain MRI (T2) images; left: sham-operated rat; right: focal cerebral infarct
48 h after MCAO; : ischemic area. (b) Infarct volume (MRI 48 h after MCAO) in animals treated with vehicle or ceftriaxone (single dose of 200 mg/kg
i.p. 90 min after MCAO). Ceftriaxone tended to reduce infarct volume but this difference was statistically not significant (n.s.), (c) Number of Nissl-
positive, viable neurons per mm2 in the penumbra 48 h after MCAO (P < 0,05), (d) Reduced neurological score in stroke vs. sham-operated animals
at all time points (P  0.001). Improved neurological outcome by ceftriaxone treatment 24 and 48 h after MCAO when compared to stroke/vehicle
(P < 0.05, P < 0.0001). 7 days after MCAO, vehicle-treated animals showed a profound spontaneous recovery, and the distinction between
vehicle and ceftriaxone-treated animals was not significant anymore. (b, c and d) are mean þ SEM. MCAO, middle cerebral artery occlusion; n.s., non
specific.

GLT1-mRNA and GLT1-protein expression and

Table 1 Acute stroke mortality GLT1-promoter activity in vitro
Treatment Animals per group (n) Mortality (n) Mortality (%)
We also tested the effect of ceftriaxone on GLT1 synthesis
in primary rat astrocytes in vitro. Astrocytes were chosen
Vehicle 29 10 34,5 because they are known to express GLT1 [25]. Incubation
Ceftriaxone 19 0 0MM
of astrocytes with 100 mM ceftriaxone for 5 days did not
M
P < 0.05, MM
P < 0.001 vs. vehicle, chi-squared test. change GLT1 mRNA or protein expression levels,

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Reduced mortality in stroke by ceftriaxone Thöne-Reineke et al. 2431

Fig. 2

Densitometric analysis of immunoblotting. Protein levels of neurotrophins, BDNF (a) and NT3 (b), and of the neurotrophin receptor, TrkB (c), in
different areas of the rat brain 48 h after MCAO. Ceftriaxone treatment significantly induced BDNF and TrkB expression and tended to increase NT3
(P ¼ 0.05). d) Note that ischaemia as such led to a slight increase in BDNF- (n.s.), NT3- (n.s.) and TrkB-expression (P < 0.05), which was even more
pronounced in response to ceftriaxone administration. (P < 0.05, P < 0.001; n ¼ 3, three independent experiments). BDNF, brain-derived
neurotrophic factor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

respectively (Fig. 4a,b). Additionally, incubation of astro-
cytes with ceftriaxone did not induce GLT1-promoter
activity as estimated by luciferase assay after transient
transfection of a GLT1-promoter-construct (Fig. 4c).
Accuracy and reliability of methods were ensured using
dibutyryl-cAMP as a positive control (Fig. 4a–c).
Glutamate uptake in vitro
Glutamate uptake was measured in primary astrocytes
in vitro as an indicator of GLT1-activity. Ceftriaxone at
concentrations of 10 and 100 mmol/l caused a significant
increase in L-[G-3H] glutamate uptake; an effect which
was comparable in strength with the effect of the positive
control, dbcAMP (200 mg/ml) (Fig. 5).
Body temperature and bacterial load
There were no differences in body temperature between
treatment groups over a period of 7 days after MCAO
(data not shown). Bacterial load in plasma (estimated by
LPS), lung and kidney (estimated by 16s-rRNA detec-
tion) also did not differ between treatment groups.
Neither the induction of stroke, nor ceftriaxone treat-
ment caused any changes in bacterial colonization (data
not shown).
Discussion
Although stroke is a major medical burden in Western
civilization as third leading cause of death and leading
cause of severe disability in adults [1], therapeutic inter-
vention is basically limited to rapid revascularization or
measures for prevention [2]. Numerous studies have
established that in the event of cerebral ischaemia, the
extent of neuronal damage is a dynamic process that
offers a window of a few hours for therapeutic interven-
tion and tissue rescue [14]. As neuroprotective agents are
still a substantial unmet need in the acute poststroke
situation [3], there has been an intensive search for
neuroprotective concepts and substances in recent years.
As a result, a number of promising compounds have been
identified which improved neurological outcome in
animal experiments, but none of these agents has proven
to be effective in clinical trials [26–29]. Moreover, none
of these agents reduced acute mortality – neither in
animal, nor in clinical studies [26–29].
However, most recently, two clinical pilot studies have
confirmed evidence from animal studies in favour of
neuroprotective features of antibiotics (minocycline, dap-
sone) in stroke [10,11]. Although the outcome of these
clinical studies has to be regarded with caution because of
the relatively small number of patients enrolled, they
strengthen the concept of antibiotics as neuroprotective
agents for acute treatment in stroke.
Regarding b-lactam antibiotics, a retrospective case-
control analysis (1888 cases, 9440 controls) revealed a

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 2432 Journal of Hypertension 2008, Vol 26 No 12

Fig. 3

(a) GLT1 protein levels in various regions of infarcted brains measured by western blot. There was no increase in GLT1 expression by ceftriaxone
(n ¼ 3 vehicle, n ¼ 3 ceftriaxone). (b) GLT1 mRNA expression in various regions of noninfarcted brains measured by real time PCR. 5 days treatment
with ceftriaxone (n ¼ 7; light bars) did not alter GLT1 mRNA expression compared with vehicle treated animals (n ¼ 6; dark bars). (c) GLT1 protein
levels in various regions of noninfarcted brains measured by western blot (n ¼ 3 vehicle, n ¼ 4 ceftriaxone). Again, no change in GLT1 expression by
ceftriaxone. BDNF, brain-derived neurotrophic factor.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Reduced mortality in stroke by ceftriaxone Thöne-Reineke et al. 2433
Fig. 4
In-vitro experiments in primary rat astrocytes. (a) dbcAMP (200 mg/ml;
positive control), but not ceftriaxone increased GLT1 mRNA
expression, P < 0.001. (b) dbcAMP, but not ceftriaxone increases
GLT1 protein levels. (c) dbcAMP, but not ceftriaxone increases GLT1
promoter activity as estimated by luciferase assay, P < 0.01. a and c
are mean
SEM, three independent experiments, three independent
measurements. GAPDH, glyceraldehyde-3-phosphate dehydrogenase.
protective association between penicillin intake and the
risk of stroke [13]. Furthermore, b-lactam antibiotics
have recently been demonstrated to act neuroprotective
in animal experiments of ALS and in a prestroke treat-
ment protocol [5,9].
In our study, we tested the effectiveness of the approved
cephalosporin ceftriaxone as a therapeutic option for the
acute treatment of stroke. We found that a single injec-
tion of ceftriaxone (200 mg/kg i.p.) 90 min after MCAO –
a protocol, which resembles the clinical situation of a
stroke patient being treated by an emergency doctor in
the field – significantly improved neurological outcome
and – most strikingly – completely abolished mortality.
There were also no fatalities in the ceftriaxone-treated
animals monitored over longer intervals (7 days or
4 weeks).
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Fig. 5
L-[G-3H] glutamate uptake in primary rat astrocytes. Significant
increase in glutamate uptake in cells treated with 10 mmol/l and
100 mmol/l ceftriaxone, respectively, or with the positive control,
dbcAMP (200 mg/ml). mean
SEM, P < 0,05, 3 independent
experiments.
It is noteworthy that ceftriaxone treatment improved
mortality and neurological outcome without significantly
reducing infarct size (although infarcts tended to be
smaller). However, cerebral infarct volume as estimated
by MRI does not necessarily correlate with functional
deficits in rats [30]. Moreover, despite no reduction in
infarct volume as estimated by T2W-MRI, the number of
surviving neurons within the penumbra was significantly
increased by ceftriaxone. There may be three reasons,
why the increased number of surviving neurons did not
translate into a significantly reduced infarct size: infarct
size measured by T2W-MRI as early as 48 h after ischae-
mic insult may mainly illustrate the area of oedema and
not the area of necrotic tissue, the surviving neurons seem
to be evenly distributed over the whole area of the
penumbra and not to be increased in number towards
the outer edges thus not resulting in a reduced area of
infarction on an MRI image, and there even was a reduc-
tion in mean infarct size in ceftriaxone-treated animals,
but this reduction was statistically not significant – not
because the mean reduction was minor, but because of
the high variability in infarct sizes between animals of the
same group.
A time interval of 24 h for estimation of mortality was
chosen in our study, because, in our hands, more than 90%
of the mortality in the stroke model of 90 min MCAO with
reperfusion in rats occurs within the first 24 h. This obser-
vation is based on data acquired in a total of about 400
animals undergoing MCAO in our previous studies, in
which the therapeutic effect of ACE-inhibitors, angioten-
sin-AT1-receptor blockers or glitazones in stroke was
tested [15–17,31]. In this context, it is of interest to note
that in all these other studies, we did not find any effect of
drug treatment on mortality (as observed in this study),
although most of these compounds reduced neurological
 2434 Journal of Hypertension 2008, Vol 26 No 12
deficits [15–17,31]. In our hands, total mortality in vehicle-
treated animals in the model of 90 min MCAO is around
30% and therefore in the range of what is reported by
others [6,15–17,26,32].
For a long time, it has been speculated that ceftriaxone
may have neuroprotective properties beyond its antibac-
terial effects, and this hypothesis has been examined in a
variety of in vitro, ex vivo and in vivo models [9,12,33–35].
Three years ago, Rothstein et al. reported a beneficial
effect of ceftriaxone in an animal model of amyotrophic
lateral sclerosis (ALS) as well as in an in-vitro model of
cerebral ischaemia (neuronal cultures under glucose-oxy-
gen-deprivation) [5]. They suggested that the molecular
mechanism underlying the beneficial effect of ceftriax-
one consists in an enhanced expression of the glutamate
transporter GLT1 coinciding with an increase in GLT1
transporter activity. As an activation of GLT1 transporter
activity by ceftriaxone was confirmed by Lipski et al. and
our study [12], data on the impact of ceftriaxone on
GLT1 expression are controversial. Rothstein et al.
reported an increase in GLT1 expression by ceftriaxone
in organotypic spinal cord slice cultures, the spinal cord of
G93A mice (ALS model), neuronal cultures under glu-
cose-oxygen-deprivation, human fetal astrocytes, and in
the rat brain ex vivo [5]. This observation was corrobo-
rated by Ouyang et al. in hippocampal slice culture [34],
and by Chu et al. in rat brain ex vivo [9], but it was neither
substantiated in our model in a series of ex vivo and in
vitro experiments including GLT1-mRNA and GLT1-
protein expression studies and GLT1 promoter reporter
assay, nor in experiments performed by the ALS Therapy
Development Institute (ALSTDI) or by Lipski et al. in
hippocampal slices [12,35].
It has to be noted that all studies having examined GLT1
expression so far used noninfarcted brains of healthy rats
and a 5-day application protocol (200 mg/kg ceftriaxone
per day) [5,9,12]. GLT1 expression in the infarcted brain
has so far only been examined by our study, demonstrating
no changes in GLT1 expression in response to ceftriaxone.
But even under those conditions used by the other groups
(5 day treatment of healthy rats), our present results show
no effect of ceftriaxone on GLT1 expression.
Still, a reduction of tissue glutamate levels by GLT1 may
play a role in ceftriaxone-related neuroprotection through
an increase in GLT1 activity as observed by others and us
[12].
Apart from the neuroprotective mechanism of glutamate
reduction in stroke, we found that in ceftriaxone treated
animals with cerebral ischaemia, the neurotrophins,
BDNF, NT3 and the neurotrophin receptor, TrkB, were
increased in the cortical peri-infarct area. Neurotrophins
are known to increase cell survival by reducing apoptosis
[36]. Elevated levels of neurotrophins in the peri-infarct
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
area including the penumbra, which represents an area in
which cells may either survive or demise within the first
hours after ischaemia, may be an important neuroprotec-
tive mechanism by increasing the number of surviving
cells through their antiapoptotic actions.
Protection from infections resulting from the recently
described ‘stroke induced immunosuppression syndrome’
(SIDS) may be discussed to represent another potential
mechanism responsible for the dramatically reduced
mortality in ceftriaxone-treated rats. SIDS is characterized
by a stroke induced, rapid and long-lasting suppression of
the cellular immune response [37]. It is thought to be
responsible for the increased incidence in life-threatening
infections in stroke patients, in particular pneumonia,
which represents the most frequent cause of death in
the postacute phase. Consequently, the effectiveness of
preventive antibacterial therapy (600 mg/kg moxifloxacin)
has been tested in a mouse model of experimental
stroke [6]. Moxifloxacin treatment resulted in a significant
reduction of mortality and neurological deficits, which was
apparent 5 or 3 days after MCAO, respectively [6]. This
delayed beneficial effect can most likely be explained by
the prevention of serious, SIDS-related, infections. In
contrast, the reduction of mortality and neurological def-
icits by ceftriaxone therapy, as observed in our study, was
significant as early as 24 h after MCAO pointing to a
different protective molecular mechanism of b-lactam
antibiotics with an earlier onset compared with chinolones.
Moreover, we did not find any differences in body
temperature, plasma LPS or bacterial load in lung and
kidneys between vehicle or ceftriaxone-treated animals
48 h and 7 days after MCAO which again speaks against an
antibacterial effect of ceftriaxone as the main mechanism
responsible for the better survival rate after MCAO.
The lack of change of IL-6 concentration after ceftriax-
one treatment in our model and in the study by Chu et al.
also supports the notion that the immediate neuropro-
tective actions of ceftriaxone do not involve anti-inflam-
matory, antibacterial mechanisms [9].
Since b-lactam antibiotics are a long-established and well
tolerated class of drugs, the clinical relevance of our data
for acute poststroke treatment can and should be tested in
humans in the near future. If ceftriaxone should turn out
to be as effective in humans as it was in our study, it might
have the potential to become a novel, well tolerated, life-
saving, first-line therapeutic option in acute stroke.
Acknowledgement
There are no conflicts of interest.
References
1 Thom T, Haase N, Rosamond W, Howard VJ, Rumsfeld J, Manolio T, et al.
Heart Disease and Stroke Statistics–2006 Update: A Report From the
American Heart Association Statistics Committee and Stroke Statistics
Subcommittee. Circulation 2006; 113:e85–e151.
 Reduced mortality in stroke by ceftriaxone Thöne-Reineke et al. 2435

2 Kjeldsen SE, Fossum E, Reims HM, Hoieggen A. Hypertension treatment 26 Önal MZ, Li F, Tatlisumak T, Locke KW, Sandage BW Jr, Fisher M.
and stroke prevention. Blood Pressure 2003; 12:264–268. Synergistic effects of citicoline and MK-801 in temporary experimental
3 Adams HP Jr., del Zoppo G, Goldstein LB, Alberts MJ, Bhatt DL, Brass L, focal ischemia in rats. Stroke 1997; 28:1060–1065.
Furlan A, et al. Guidelines for the early management of adults with ischemic 27 Yu GL, Wei EQ, Wang ML, Zhang WP, Zhang SH, Weng JQ, et al.
stroke: a guideline from the American HeartAssociation/American Stroke Pranlukast, a cysteinyl leukotriene receptor-1 antagonist, protects against
Association Stroke Council, Clinical Cardiology Council, Cardiovascular chronic ischemic brain injury and inhibits the glial scar formation in mice.
Radiology and Intervention Council, and the Atherosclerotic Peripheral Brain Res 2005; 1053:116–125.
Vascular Disease and Quality of Care Outcomes in Research 28 Chang CF, Lin SZ, Chiang YH, Morales M, Chou J, Lein P, et al. Intravenous
Interdisciplinary Working Groups: The American Academy of Neurology administration of bone morphogenetic protein-7 After ischemia improves
affirms the value of this guideline as an educational tool for neurologists.: motor function in stroke rats. Stroke 2003; 34:558–564.
2005 Guidelines Update A Scientific Statement From the Stroke Council 29 Huang J, Agus DB, Winfree CJ, Kiss S, Mack WJ, McTaggart RA.
of the American Heart Association/American Stroke Association. Stroke Dehydroascorbic acid, a blood–brain barrier transportable form of vitamin
2007; 38:1655–1711. C, mediates potent cerebroprotection in experimental stroke. Proc Natl
4 Cheng YD, Al-Khoury L, Zivin JA. Neuroprotection for ischemic stroke: two Acad Sci U S A 2001; 98:11720–11724.
decades of success and failure. NeuroRx 2004; 1:36–45. 30 Wegener S, Weber R, Ramos-Cabrer P, Uhlenkueken U, Wiedermann D,
5 Rothstein JD, Patel S, Regan MR, Haenggeli C, Huang YH, Bergles DE, Kandak K, et al. Subcortical lesions after transient thread occlusion in the
et al. b-lactam antibiotics offer neuroprotection by increasing glutamate rat: T2-weighted magnetic resonance imaging findings without
transporter expression. Nature 2005; 433:73–77. corresponding sensorimotor deficits. J Magn Reson Imaging 2005;
6 Meisel C, Prass K, Braun J, Victorov I, Wolf T, Megow D, et al. Preventive 21:340–346.
antibacterial treatment improves the general medical and neurological 31 Krikov M, Thöne-Reineke r C, Müller S, Villringer A, Unger T. Candesartan
outcome in a mouse model of stroke. Stroke 2004; 35:2–6. but not ramipril pretreatment improves outcome after stroke and stimulates
7 Elewa HF, Hilali H, Hess DC, Machado LS, Fagan SC. Minocycline for neurotrophin BNDF/TrkB system in rats. J Hypertens 2008; 26:544–552.
short-term neuroprotection. Pharmacotherapy 2006; 26:515–521. 32 Maier CM, Sun GH, Kunis D, Yenari MA, Steinberg GK. Delayed induction
8 Rios C, Nader-Kawachi J, Rodriguez-Payan AJ, Nava-Ruiz C. and long-term effects of mild hypothermia in a focal model of transient
Neuroprotective effect of dapsone in an occlusive model of focal ischemia cerebral ischemia: neurological outcome and infarct size. J Neurosurg
in rats. Brain Res 2004; 999:212–215. 2001; 94:90–96.
9 Chu K, Lee ST, Sinn DI, Ko SY, Kim EH, Kim JM, et al. Pharmacological 33 Rothstein JD, Kuncl RW. Neuroprotective strategies in a model of chronic
induction of ischemic tolerance by glutamate transporter-1 (EAAT2) glutamate mediated motor neuron toxicity. J Neurochem 1995; 65:643–
upregulation. Stroke 2007; 38:177–182. 651.
10 Nader-Kawachi J, Gongora-Rivera F, Santos-Zambrano J, Calzada P, Rios 34 Ouyang YB, Voloboueva LA, Xu LJ, Giffard RG. Selective dysfunction of
C. Neuroprotective effect of dapsone in patients with acute ischemic hippocampal CA1 astrocytes contributes to delayed neuronal damage after
stroke: a pilot study. Neurol Res 2007; 29:331–334. transient forebrain ischemia. J Neurosci 2007; 27:4253–4260.
11 Lampl Y, Boaz M, Gilad R, Lorberboym M, Dabby R, Rapoport A, et al. 35 ALS Therapy Development Institue (ALSTDI): www.als.net.
Minocycline treatment in acute stroke. Neurology 2007; 69:1404–1410. 36 Saito A, Tominaga T, Chan PH. Neuroprotective role of neurotrophins:
12 Lipski J, Wan CK, Bai JZ, Pi R, Li D, Donnelly D. Neuroprotective potential of relationship between nerve growth factor and apoptotic cell survival
ceftriaxone in in vitro models of stroke. Neuroscience 2007; 146:617– pathway after cerebral ischemia. Curr Atheroscler Rep 2005; 7:268–273.
629. 37 Meisel C, Schwab JM, Prass K, Meisel A, Dirnagl U. Central nervous system
13 Brassard P, Bourgault C, Brophy J, Kezouh A, Suissa S. Antibiotics in injury-induced immune deficiency syndrome. Nat Rev 2005; 6:775–786.
primary prevention of stroke in the elderly. Stroke 2003; 34:e163–e167.
14 Kato H, Kogure K. Biochemical and molecular characteristics of the brain
with developing cerebral infarction. Cell Mol Neurobiol 1999; 19:93–108.
15 Dai WJ, Funk A, Herdegen T, Unger T, Culman J. Blockade of central
angiotensin AT(1) receptors improves neurological outcome and reduces
expression of AP-1 transcription factors after focal brain ischemia in rats.
Stroke 1999; 30:2391–2398.
16 Groth W, Blume A, Gohlke P, Unger T, Culman J. Chronic pretreatment
with candesartan improves recovery from focal cerebral ischaemia in rats.
J Hypertens 2003; 21:2175–2182.
17 Li J, Culman J, Hörtnagl H, Zhao Y, Gerova N, Timm M, et al. Angiotensin
AT2 receptor protects against cerebral ischemia-induced neuronal injury.
FASEB J 2005; 19:617–619.
18 Garcia JH, Wagner S, Liu KF, Hu XJ. Neurological deficit and extent of
neuronal necrosis attributable to middle cerebral artery occlusion in rats.
Statistical validation. Stroke 1995; 26:627–634.
19 Paxinos G, Watson C. The rat brain in stereotaxic coordinates, third edition.
London, Boston, New York, Sydney, Tokyo, Toronto: Academic Press,
Harcourt Brace & Company, Publishers.
20 Schefe JH, Lehmann KE, Buschmann IR, Unger T, Funke-Kaiser H.
Quantitative real-time RT-PCR data analysis: current concepts and the
novel ‘gene expression’s CT difference’ formula. J Mol Med 2006; 84:901–
910.
21 De Groot CJ, Langeveld CH, Jongenelen CA, Montagne L, Van der Valk P,
Dijkstra CD. Establishment of human adult astrocyte cultures derived from
postmortem multiple sclerosis and control brain and spinal cord regions:
immunophenotypical and functional characterization. J Neurosci Res
1997; 49:342–354.
22 Millar BC, Jiru X, Moore JE, Earle JAP. A simple and sensitive method to
extract bacterial, yeast and fungal DNA from blood culture material.
J Microbiol Meth 2000; 42:139–147.
23 Durukan A, Tatlisumak T. Acute ischemic stroke: overview of major
experimental rodent models, pathophysiology, and therapy of focal cerebral
ischemia. Pharmacol Biochem Behaviour 2007; 87:179–197.
24 McCarthy MM. Stretching the truth. Why hippocampal neurons
are so vulnerable following traumatic brain injury. Exp Neurol 2003;
184:40–43.
25 Berger UV, deSilva TM, Chen W, Rosenberg PA. Cellular and subcellular
mRNA localization of glutamate transporter isoforms GLT1a and GLT1b in
rat brain by in situ hybridization. J Comp Neurol 2005; 492:78–89.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.