Dose-dependent effects of levetiracetam after hypoxia and hypothermia in the neonatal mouse brain

Brain Research 1646 (2016) 116–124

Contents lists available at ScienceDirect

Brain Research journal homepage: www.elsevier.com/locate/brainres

Research report

Dose-dependent effects of levetiracetam after hypoxia and hypothermia in the neonatal mouse brain Katja Strasser a,b, Laura Lueckemann a,c, Verena Kluever a, Sinthuya Thavaneetharajah a, Daniela Hoeber a, Ivo Bendix a, Joachim Fandrey d, Astrid Bertsche a,e,n, Ursula Felderhoff-Mueser a a

Department of Pediatrics 1, Neonatology, Un iversity Hospital Essen, University Duisburg, Essen, Germany Department of Pediatrics, Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, Germany c Institute of Medical Psychology and Behavioral Immunobiology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany d Institute of Physiology, University of Duisburg-Essen, Essen, Germany e Department of Women and Child Health, University Hospital for Children and Adolescents, Centre for Pediatric Research Leipzig, Germany b

art ic l e i nf o

a b s t r a c t

Article history: Received 16 March 2016 Received in revised form 18 May 2016 Accepted 19 May 2016 Available online 20 May 2016

Perinatal asphyxia to the developing brain remains a major cause of morbidity. Hypothermia is currently the only established neuroprotective treatment available for term born infants with hypoxic-ischemic encephalopathy, saving one in seven to eight infants from developing severe neurological deficits. Therefore, additional treatments with clinically applicable drugs are indispensable. This study investigates a potential additive neuroprotective effect of levetiracetam combined with hypothermia after hypoxia-induced brain injury in neonatal mice. 9-day-old C57BL/6-mice (P9) were subjected either to acute hypoxia or room-air. After 90 min of systemic hypoxia (6% O2), pups were randomized into six groups: 1) vehicle, 2) low-dose levetiracetam (LEV), 3) high-dose LEV, 4) hypothermia (HT), 5) HT combined with low-dose LEV and 6) HT combined with high-dose LEV. Pro-apoptotic factors, neuronal structures, and myelination were analysed by histology and on protein level at appropriate time points. On P28 to P37 long-term outcome was assessed by neurobehavioral testing. Hypothermia confers acute and long-term neuroprotection by reducing apoptosis and preservation of myelinating oligodendrocytes and neurons in a model of acute hypoxia in the neonatal mouse brain. Low-dose LEV caused no adverse effects after neonatal hypoxic brain damage treated with hypothermia whereas administration of high-dose LEV alone or in combination with hypothermia increased neuronal apoptosis after hypoxic brain injury. LEV in low- dosage had no additive neuroprotective effect following acute hypoxic brain injury. & 2016 Elsevier B.V. All rights reserved.

Keywords: Perinatal asphyxia Hypothermia Levetiracetam Neonatal mice Neuroprotection Apoptosis

1. Introduction Perinatal asphyxia remains a common cause of neonatal death and long-term disability with an incidence of 20 per 1000 live births (Lee et al., 2013; Rennie et al., 2007). Out of those about 2–3 infants suffer from hypoxic-ischemic encephalopathy (HIE) resulting from significant injury to the developing brain (Murray et al., 2010) with a mortality rate of 20%. Survivors are frequently affected by secondary neurological morbidity, including cerebral palsy (15%), severe cognitive delay (11%), seizure disorders (8%), hearing loss (3%), and visual impairment (3%) (Lee et al., 2013; n Correspondence to: Universitätsklinik und Poliklinik für Kinder und Jugendliche, Liebigstraße 20a, 04103 Leipzig, Germany. E-mail address: [email protected] (A. Bertsche).

http://dx.doi.org/10.1016/j.brainres.2016.05.040 0006-8993/& 2016 Elsevier B.V. All rights reserved.

Rennie et al., 2007). Perinatal hypoxia results in numerous cell-damaging processes such as neuronal cell injury (Blomgren et al., 2006; Yager et al., 1992) and disturbance of myelination (Back et al., 2001). In response to acute hypoxia, selectively vulnerable regions in the developing mouse brain, mainly striatum, ventrobasal thalamus and periventricular zone can undergo continued apoptosis for a prolonged period up to 6–7 days post insult (Trollmann et al., 2014). Therapeutic hypothermia within the first 6 h of postnatal life has been shown in animal studies and randomized clinical trials to improve acute brain lesions, survival and neurological long term outcome and is now an established therapy effective in mild and moderate hypoxic-ischemic brain injury with a number needed to treat (NNT) from 7 to 8 (Azzopardi et al., 2014; Shankaran et al., 2012; Edwards et al., 2010; Azzopardi et al., 2009; Gressens et al., 2008; Thoresen et al., 1996). 40–50% of cooled newborns, however,

K. Strasser et al. / Brain Research 1646 (2016) 116–124

the above mentioned studies in the neonatal brain focused on sole treatment with LEV, combined effects with hypothermia have not been investigated. However, to translate findings into clinical practice a combined treatment strategy with therapeutic hypothermia is obligatory because whole body cooling is the current standard therapy in mild to moderate neonatal hypoxic brain injury. The aim of this study was to investigate acute and long-term effects of treatment with LEV in combination with hypothermia in a neonatal mouse model of mild to moderate hypoxic brain injury.

still suffer from major neurological problems (Jacobs et al. 2013; Tagin et al. 2012; Edwards et al., 2010). Thus, additional treatment strategies are urgently required. Symptoms associated with perinatal asphyxia (e.g. muscular hypotonia, cardiorespiratory failure, hyperexcitability, seizures) may occur within the first hours post injury which makes pharmacological treatment essential (Sarnat and Sarnat, 1976). As such, about 60% of patients suffering from hypoxic-ischemic encephalopathy (HIE) and treated with hypothermia need anticonvulsant medication because of clinical or electrographic seizures within the first hours post injury (Shah et al., 2014). Conventional anticonvulsive drugs (e.g. Phenobarbital and Phenytoin), however, are reported to be ineffective in 50% of treated newborns (Booth and Evans, 2004; Boylan et al., 2002). Moreover, they experimentally trigger neuronal apoptosis in the immature rodent brain (Stefovska et al., 2008; Bittigau et al., 2002), and may induce cognitive impairment in infants (Holmes et al., 2001; Dessens et al., 2000). Levetiracetam (LEV), S-α-ethyl-2-oxo-1-pyrrolidine-acetamide, is an anticonvulsive drug of the second generation which has already been approved for clinical treatment of epilepsy in infants older than 4 weeks of age (Beaulieu et al., 2013) and is also proposed for the intervention in neonatal seizures (Neininger et al., 2015; Ramantani et al., 2011; Silverstein and Ferriero, 2008) with little side effects (Radtke, 2001; Piña-Garza et al., 2009). Furthermore, neuroprotective properties of LEV in high-doses up to 1000 mg/kg body weight are described in the adult rodent brain after stroke (Hanon and Klitgaard, 2001), subarachnoid hemorrhage (Wang et al., 2006) and status epilepticus (Mazarati et al., 2004) as well as in the neonatal brain after hypoxic-ischemic injury (Komur et al., 2014; Kilicdag et al., 2012). Therefore, LEV is a promising therapy to be administered in conjunction with hypothermia after perinatal asphyxia. Whereas Tissue loss: volumetry of side ventricle P10

2. Results Animals treated with high-dose (70 mg/kg) LEV had to be excluded from the long-term experiment on P12 due to a significant weight loss of over 20% (p o 0.001; data not shown), according to the Federal Guidelines for the Care and Use of Laboratory Animals. After treatment with high-dose LEV, animals showed reduced movements and reduced food intake. We did not observe any significant sex differences between groups at the various time points investigated. 2.1. LEV influences apoptosis after hypoxic injury in the neonatal brain and treatment with hypothermia Induction of acute systemic hypoxia (90 min, 6% O2) on P10 resulted in periventricular tissue loss with enlarged lateral ventricles as revealed by Cresyl violet staining (Fig. 1(A)/(B)). Furthermore, apoptotic cell-death is demonstrated by increased number of DNA-fragmented cells and increased density of cleaved caspase-3 positive cells was detected in hypoxic brains (Fig. 1(C)– (F)). These alterations persisted on P60 with significantly enlarged Apoptosis: TUNEL-staining P10

Apoptosis: cCasp3 protein expression P10

control

1A

117

1E

1C

control

negative

cCasp3 + LEV 7

+ vehicle + LEV 70

+ LEV 70

β-actin LEV 70

vehicle

H

LEV 7

control

LEV 70

H

H

LEV 7

+ LEV 7

vehicle

+ vehicle

+ LEV 7

+ vehicle

+ LEV 70

H + HT

H + HT

500

1D

+ vehicle + LEV 7 + LEV 70

400 **

300 **

200

§

100

control

H

H + HT

300

1F

+ vehicle + LEV 7 + LEV 70 **

200

**

**

** §

§

100

control

H

H + HT

TUNEL+ cells per hemisphere

600

density ratio cCasp3 / β-actin [%, control = 100%]

ratio ventricle / hemisphere [%, control = 100%]

1B

H + HT

40

+ vehicle + LEV 7 + LEV 70

30

# ** **

** # **

**

20 §

10

control

H

H + HT

Fig. 1. Hypothermia but not levetiracetam reduces hypoxia-induced brain injury. 9-day-old C57BL/6-mice were subjected to either acute hypoxia or room-air (control). After 90 min hypoxia (6%, H), pups were randomized into six groups: (1) vehicle (0.9% NaCl i.p.), (2) low-dose LEV (7 mg/kg i.p.), (3) high-dose LEV (70 mg/kg i.p.), (4) hypothermia (HT, 4 h 32 °C), (5) HT combined with low-dose LEV and (6) HT combined with high-dose LEV. 24 h after hypoxia brains were analysed for sizes of lateral ventricles via Cresylviolet-staining. Photographs show representative images of cresyl violet stainings, scale bar is 1 mm (A). The ratio of lateral ventricle sizes to corresponding hemispheres was calculated and from the mean value of control group was set to¼100% (B). Brain extracts of hemispheres were used for western blot analyses of activated Caspase-3; here in a representative Western Blot series of cCasp3 (C). Normalized ratios of cCasp3 signals to signals of β-actin were calculated and the control group was set to 100% (D). DNA fragmented cells were detected via TUNEL-staining in 6 different brain regions; here exemplarily shown for periventricular striatum (E). The sum of positive cells out of all analysed brain regions was calculated (F). Bars represent mean þSD, post-hoc test Bonferroni in (D/F) and Tamhane's T2 in (B). **p o 0.01 vs. control group, §p o 0.05 vs. hypoxia group, #p o 0.05 vs. Hþ HT. n ¼ 9–10 animals per group for (B) and (F), n ¼10–13 animals per group for (D).

118

K. Strasser et al. / Brain Research 1646 (2016) 116–124

lateral ventricles in hypoxia exposed animals (data not shown). Treatment with moderate whole-body cooling (4 h, 32 °C) significantly impaired acute cerebral tissue loss by reduction of apoptosis in our experimental set-up as detected by volumetry of lateral ventricles (Fig. 1(A)/(B)), immunoblotting of cleaved Caspase-3 (Fig. 1(C)/(D)), and immunohistochemistry of DNA-fragmented cells (Fig. 1(E)/(F)). Administration of LEV did not change apoptotic cell death (Fig. 1 (A)–(F)) under normothermic conditions in a significant manner when compared to vehicle treated animals (Fig. 1(A)–(F)). However, after hypoxia exposure and combined treatment with hypothermia, increased protein expression of cleaved caspase-3 was detected in animals who received a high dose (70 mg/kg) LEV (Fig. 1(C)/(D)). The combined intervention with hypothermia and LEV in both doses increased the number of DNA-fragmented cells (Fig. 1(E)/(F)). However, on P60 lateral ventricles were significantly smaller in the hypothermia group, which remained unchanged in the combined hypothermia/LEV (7 mg/kg) group (data not shown). 2.2. LEV did not influence myelinating oligodendrocytes and neurons after acute hypoxia under normothermic and hypothermic conditions To investigate whether tissue loss and apoptosis correlate

Myelination: MBP protein expression P10 2A

with disturbance of oligodendroglial or neuronal structures, analyses of MBP (Fig. 2(A)/(B)) and MAP-2 (Fig. 3(A)/(B)) were performed by immunoblotting. DNA-fragmented oligodendrocytes (Fig. 2(C)/(D)) and neurons (Fig. 3(C)/(D)) were visualized by immunohistochemistry. We demonstrated a significantly reduced expression of myelin basic protein by immunoblotting (Fig. 2(A)/(B)) and an increase in DNA-fragmented oligodendrocytes (Fig. 2(C)/(D)) after hypoxic injury at P10. Intervention with hypothermia following hypoxia exposure resulted in preservation of oligodendroglia on P10 as demonstrated by protein analyses of MBP (Fig. 2(A)/(B)) as well as in immunohistochemistry of DNA-fragmented oligodendrocytes (Fig. 2(C)/(D)). Intervention with LEV at the low dose of 7 mg/kg and also the higher dose of 70 mg/kg – as a single or combined treatment with hypothermia - did not influence oligodendroglial cells as shown by protein analyses of MBP (Fig. 2(A)/(B)) and counting of DNA-fragmented oligodendrocytes (Fig. 2(C)/(D)). Also, amounts of neuronal dendrites as shown by MAP-2 staining were significantly reduced by systemic hypoxia on P10. Respectively, the number of TUNEL þneurons was increased following acute hypoxia. In parallel to the effect observed in oligodendrocytes, hypothermia treatment preserved neuronal structures. Intervention

Myelination: Olig2-TUNEL-costaining P10 2C

control

MBP + LEV 7

+ vehicle

+ LEV 70

β-actin

H

LEV 70

LEV 7

vehicle

LEV 70

LEV 7

vehicle

control

H

+ LEV 7

+ vehicle

+ LEV 70

H + HT

120 100

2D 30

+ vehicle + LEV 7 + LEV 70 § §

§

80 60

** **

**

40 20

control

H

H + HT

Olig2-TUNEL+ cells per hemisphere

2B density ratio MBP / β-actin [%, control = 100%]

H + HT

+ vehicle + LEV 7 + LEV 70 ** ** **

20

§

§

10

control

H

H + HT

Fig. 2. Hypothermia confers neuroprotection by preservation of myelination while levetiracetam therapy is ineffective. 9-day-old C57BL/6-mice were subjected to either acute hypoxia or room-air (control). After 90 min hypoxia (6%, H), pups were randomized into six groups: (1) vehicle (0.9% NaCl i.p.), (2) low-dose LEV (7 mg/kg i.p.), (3) high-dose LEV (70 mg/kg i.p.), (4) hypothermia (HT, 4 h 32 °C), (5) HT combined with low-dose LEV and (6) HT combined with high-dose LEV. MBP expression was quantified 24 h after hypoxia via western blot analysis (A/B). Brain extracts of hemispheres were used for western blot analyses of myelin basic protein; here in a representative Western Blot series of MBP (A). Normalized ratios of MBP signals to signals of β-actin were calculated and the control group was set to 100% (B). DNAfragmented oligodendrocytes were detected via co-staining of TUNEL (green) and Olig-2 (red) (exemplarily shown for deep white matter, (C)). The sum of TUNEL/Olig2 double positive cells out of all analysed brain regions was calculated (D). Bars represent meanþ SD, post-hoc test Bonferroni in (B) and Tamhane's T2 in (D). **p o 0.01 vs. control group, §po 0.05 vs. hypoxia group. n ¼ 9–10 animals per group for (D), n ¼ 10–13 animals per group for (B).

K. Strasser et al. / Brain Research 1646 (2016) 116–124

Neurons: MAP2 protein expression P10 3A

119

Neurons: NeuN-TUNEL-costaining P10 3C

control

MAP2 + LEV 7

+ vehicle

+ LEV 70

β-actin

H

LEV 70

LEV 7

vehicle

LEV 70

LEV 7

vehicle

control

H

+ LEV 7

+ vehicle

+ LEV 70

H + HT H + HT

120 100

3D

+ vehicle + LEV 7 + LEV 70

§

§

80 60

**

**

**

40 20

control

H

H + HT

NeuN-TUNEL+ cells per hemisphere

density ratio MAP2 / β-actin [%,control = 100%]

3B

30

+ vehicle + LEV 7 + LEV 70

20 *

*

10

control

H

H + HT

Fig. 3. Acute hypoxia-induced neuronal injury is reduced by therapeutic hypothermia whereas LEV treatment is ineffective. 9-day-old C57BL/6-mice were subjected to either acute hypoxia or room-air (control). After 90 min hypoxia (6%, H), pups were randomized into six groups: (1) vehicle (0.9% NaCl i.p.), (2) low-dose LEV (7 mg/kg i.p.), (3) high-dose LEV (70 mg/kg i.p.), (4) hypothermia (HT, 4 h 32 °C), (5) HT combined with low-dose LEV and (6) HT combined with high-dose LEV. MAP-2 expression was quantified 24 h after hypoxia via western blot analysis (A/B). Brain extracts of hemispheres were used for western blot analyses of microtubule associated protein-2; here in a representative Western Blot series of MAP-2 (A). Normalized ratios of MAP-2 signals to signals of β-actin were calculated and the control group was set to 100% (B). DNAfragmented neurons were detected via co-staining of TUNEL (green) and NeuN (red) (exemplarily shown for periventricular striatum, (C)). The sum of TUNEL/NeuN double positive cells out of all analysed brain regions was calculated (D). Bars represent meanþ SD, post-hoc test Tamhane's T2. *p o 0.05 vs. control group, **p o 0.01 vs. control group, §p o 0.05 vs. hypoxia group. n ¼9–10 animals per group for (D), n ¼10–13 animals per group for (B).

with both doses of LEV did neither have an effect on neuronal cells in the presence or absence of hypothermia as revealed by protein analyses of MAP-2 and counting of DNA-fragmented neurons (Fig. 3(A)–(D)). 2.3. Long-term neurodevelopment after neonatal hypoxic injury is improved by hypothermia and is not affected by low-dose LEV We quantified effects on long-term sensomotoric function in our animal model and assessed anxiety and curiosity (Elevated Plus Maze), exploration and retentiveness (Novel-Object-Recognition), as well as motoric skills (accelerated RotaRod) in mice on postnatal day P28–P37. Analyses in Elevated Plus Maze are expressed as the percentage of covered distance in the open arms of the whole distance. The control group covered 39.8% þ2.4% of whole distance in open arms which was significantly less in hypoxia-injured animals (18.3% þ 2.9%). After treatment with hypothermia, animals performed similar to the control group (41.8% þ1.2%), (Fig. 4(A)). For Novel-Object-Recognition, following a training period with two identical objects and replacement of one object by a new one, the time mice spent at the different objects was recorded. Data are expressed as percentage of time mice spent with the new object

from the overall time mice spent with any object. With 85.8% þ 3.6%, the control group showed a preference for the novel object compared to the familiar object. After hypoxic injury, animals were significantly less active regarding unknown objects under normothermic conditions than the control group (23.2% þ5.8%) but spent more time with unfamiliar objects after hypothermia compared to normothermia (76.5% þ6.1%), (Fig. 4(B)). Running time on the accelerated RotaRod was significantly less after hypoxia exposure (91.2 s þ 4.6 s) compared to control animals (177.3 s þ3.7 s) kept at room air. Significant improvement of motoric skills was noted after hypothermia treatment compared to hypoxic animals under normothermia (161.4 s þ 7.5 s). Intraperitoneal treatment with 7 mg/kg LEV did not reveal any additional effects compared to hypothermia alone in all behavioral tests applied in this study (Fig. 4(C)).

3. Discussion The present study investigated whether combined treatment with whole-body cooling and the anticonvulsive drug LEV can provide neuroprotection for the developing mouse brain under conditions of systemic hypoxia. To our knowledge this is the first

120

K. Strasser et al. / Brain Research 1646 (2016) 116–124

§§ §§

40

30 **

20

10

control

H

Behaviour: accelerated RotaRod

Behaviour: Novel-Object-Recognition 4B

+ vehicle + LEV 7

H + HT

activity new object NOR P31 [% from whole object activity]

distance open arms EPM P28 [% from whole distance]

50

100

4C

+ vehicle + LEV 7 §§

80

60

40 **

20

control

H

H + HT

running time RotaRod P37 [s]

Behaviour: Elevated Plus Maze 4A

350

+ vehicle + LEV 7

250 §

200 150 ** **

100 50

control

H

H + HT

Fig. 4. Hypothermia but not levetiracetam improve cognitive and sensorimotor functions in adolescent mice after neonatal exposure to hypoxia. 9-day-old C57BL/6-mice were subjected to either acute hypoxia or room-air (control). After 90 min hypoxia (6%, H), pups were randomized into four groups: (1) vehicle (0.9% NaCl i.p.), (2) low-dose LEV (7 mg/kg i.p.), (3) hypothermia (HT, 4 h 32 °C), (4) HT combined with low-dose LEV. At the age of 4 weeks long-term functional outcome was assessed via different behaviour tests. Anxiety-related behaviour was evaluated via Elevated Plus Maze (EPM). Graph shows covered distance in open arms as percentage values of the total distance in EPM (A). Exploration and memory function was determined via Novel-Object-Recognition (NOR). Graph shows time at the new object as percentage of the total time spent at any object in NOR (B). Sensomotoric skills were analysed via accelerated RotaRod. Graph shows absolute running time on the running wheel of the accelerated RotaRod (C). Bars represent mean þSD, post-hoc test Tamhane's T2. **p o 0.01 vs. control group, §po 0.05 vs. hypoxia group; §§p o 0.01 vs. hypoxia group. n ¼ 10–14 animals per group.

study investigating long-term effects after mild to moderate hypoxic brain injury in neonatal mice under normothermic and hypothermic conditions. LEV at clinically relevant doses administered following hypoxic injury did not cause any long-term injury but had no additional neuroprotective effect, neither under normothermic nor under hypothermic conditions. Adjuvant administration of LEV in high doses increased apoptotic damage in the hypoxia-exposed neonatal mouse brain and seemed to abrogate hypothermia-induced neuroprotection. Short-term exposure to low oxygen levels is known to induce mild to moderate cell-damaging processes in the neonatal rodent brain. During the period of re-oxygenation secondary energy deficiency, oxidative stress, excitotoxicity, and inflammation contribute to cell death and can persist for days and weeks (Trollmann et al., 2014; Schneider et al., 2012; Mikhailenko et al., 2010; Vannucci and Vannucci, 1980; Hedner and Lundborg, 1980; Towbin, 1970; Rastogi et al., 1968). In our model, acute systemic hypoxia in 9-day-old mice led to increased apoptosis, disturbances in myelination and reduction of neuronal cellular density which leads to changes observed in longterm behavioral outcome regarding anxiety, exploration, cognitive and motor function. Meta analyses from clinical multicenter trials concluded that treatment with prolonged moderate hypothermia applied up to 6 h after perinatal asphyxia reduces acute cerebral injury and improves neurological outcome (Edwards et al., 2010). Furthermore, moderate hypothermia after perinatal asphyxia resulted in improved neurocognitive outcomes in middle childhood (Azzopardi et al., 2014). In order to mimic a clinically relevant situation exposure to systemic hypoxia was combined with hypothermia treatment. Therefore we modified the previously by Carlsson et al. and Edwards et al. described hypoxia-ischemia/hypothermia model in 9-day-old mouse pups and applied only hypoxia to mimic mild to moderate injury (Trollmann et al., 2014; Carlsson et al., 2012; Edwards et al., 1995). In our model we were able to show neuroprotective effects of hypothermia by reduction of apoptosis and preservation of neurons and oligodendrocytes. Furthermore, our study demonstrated for the first time that whole-body-cooling after mild to moderate hypoxic brain damage led to an improvement of examining behaviour, cognitive function, short-term memory and sensorimotoric skills. Neonatal brain injury caused by hypoxic events may lead to neurodevelopmental impairment in later life (Lee et al., 2013; Rennie et al.,

2007). 40–50% of cooled newborns still suffer from major neurological problems (Jacobs et al. 2013; Tagin et al., 2012; Edwards et al., 2010). Thus, adjuvant treatment strategies are urgently warranted. Since HIE is frequently associated with neonatal seizures (Jonsson et al., 2014; Shah et al., 2014), neonates treated with hypothermia after perinatal asphyxia receive additional therapy with anticonvulsive drugs (AED). Although AED used for adult and pediatric patients are also applied in clinical practice for neonates, data regarding potentially adverse effects of their intervention in neonatal seizures especially concerning longterm outcome are limited (van Rooij et al., 2013). A good tolerance of LEV for long lasting treatment of neonatal seizures has been described and despite the lack of randomized controlled trials, LEV is frequently used by neonatologists in clinical practice (Neininger et al., 2015). AEDs have been described to exert neuroprotective effects in the adult brain injury but may trigger apoptotic neurodegeneration in the developing brain during the period of rapid brain growth and active synaptogenesis (Forcelli et al., 2012; Bittigau et al., 2002). Compared to other AEDs, LEV, even given in high doses, seems to have no influence on cell death in healthy immature rodents (Griesmaier et al., 2014; Trollmann et al., 2008; Kim et al., 2007). Previous studies on the intervention with high-dose LEV in vivo and in vitro models for perinatal asphyxia have shown controversial results. Komur et al. (2014) and Kilicdag et al. (2012) reported reduced apoptosis and improved behavioral outcome after LEV application following hypoxic-ischemic brain injury using the Rice-Vannucci model of HIE. In contrast, Griesmaier et al. (2014) found an aggravation of hypoxic-ischemic brain damage under normothermic conditions but no effect in primary hippocampal neurons after oxygen glucose deprivation. In contrast, Sendrowski et al. (2011) observed improved cell survival upon LEV treatment on primary hippocampal neurons exposed to hypoxia. Presumed neuroprotective effects of LEV have been reported in dose ranges of 44–200 mg/kg (Komur et al., 2014; Kilicdag et al., 2012; Marini et al., 2004; Hanon and Klitgaard, 2001). In an adult rat model of status epilepticus dose-dependent protective effects of LEV administered at a high dosage of up to 1000 mg/kg on mitochondrial function during the acute stage of a status epilepticus have been observed (Gibbs et al., 2006). In our study, high-dose LEV (70 mg/kg) was applied in a dose slightly above the anticonvulsant ED50 in rodents (Doheny et al., 1999; Gower et al., 1992) in which sedative or other neurologic adverse effects (Marini et al., 2004; Hanon and Klitgaard, 2001) are not expected from recent experimental

K. Strasser et al. / Brain Research 1646 (2016) 116–124

studies in adult rodents. Here we report a pro-apoptotic effect of LEV in the developing brain after systemic hypoxia and treatment with hypothermia. In agreement with others (Trollmann et al., 2008; Kim et al., 2007), low-dose LEV treatment did not influence apoptotic pathways in oligodendrocytes and neurons. Analysis of protein expression and behaviour of animals treated with LEV showed neither damaging nor cell protective effects 24 h after injury and in long-term neurobehavioural outcome. Unfortunately, analyses at later time points for high-dose administration of LEV were impossible due to significant weight loss ( 4 20% in 3 days) of animals treated with 70 mg/kg LEV. Although no sedative effect for this high dose in adult rodents is known, we speculate about potential tranquilizing influences on immature mice. In the clinical setting, patients with perinatal asphyxia and seizures receive pharmacological anticonvulsive therapy adjunctive to whole-body-cooling. Therefore, we investigated additional effects of LEV in combination with therapeutic hypothermia and started with repetitive LEV administration in combination with hypothermia (Shah et al., 2014; Kim et al., 2007). Combined hypothermia/LEV treatment had no influence on myelination and neurons but led to abrogation of hypothermia-induced neuroprotection in apoptotic parameters, predominantly in higher doses. This conflicts with the findings of Griesmaier et al. (2014) where LEV administration (50 mg/kg) did not increase hypoxic-ischemic brain injury in the hypothermia group. However, in this study there was no neuroprotective effect of hypothermia treatment which may be explained by differences in the experimental setting with a shorter time period of hypothermia (3 h instead of 4 h) and the use of the hypoxiaischemia model (Rice-Vanucci). In order to give clear recommendations regarding the use of LEV following hypoxic brain injury, further dosages need to be investigated.

4. Conclusion To our knowledge this is the first study investigating longterm effects of LEV after mild to moderate neonatal hypoxic brain injury and combined hypothermia treatment. We show that hypothermia confers acute and long-term neuroprotection by reducing neuronal apoptosis, preservation of myelinating oligodendrocytes and improvement of long-term neurocognitive function. LEV treatment revealed a dose-dependent effect after hypoxia and hypothermia with increased toxicity of high doses. LEV in low-dose doses did not confer additional neuroprotection but also did not cause any adverse effects whereas administration of high-dose LEV alone as well as in combination with hypothermia increased TUNEL positive cells after hypoxic brain injury. Therefore, our results call for caution with the use of highdoses of LEV in newborn infants with pre-existing brain injury. However, our data did not indicate any neurotoxic effect of lowdose LEV in this mouse model of mild to moderate hypoxic brain injury under normothermic and hypothermic conditions.

5. Experimental procedure 5.1. Animal Procedures All animal experiments were approved and performed in accordance with the guidelines of the University and with permission of the local animal welfare committee (Bezirksregierung Düsseldorf, Germany; TSG1186/11). 9-day-old C57BL/6-mice (Harlan Laboratories, USA) of both sexes with a weight of 4.9 70.3 g served as experimental animals representing a model

121

of term-born mild to moderate oxygen deficiency. Therefore, pups and their dams were exposed to 6% oxygen for 90 min at room temperature in a computer-controlled oxygenchamber (OxyCycler, BioSpherix, USA) whereas control animals stayed with their dams at room-air. 2 h following the hypoxic insult, pups - matched for sex and weight - were randomly assigned to intraperitoneal (i.p.) treatment with (1) LEV (UCB S.A., Belgium) at a dose of 7 mg/kg body weight every 24 h from P9 to P15 (LEV7), (2) LEV at a dose of 70 mg/kg body weight every 24 h from P9 to P15 (LEV70), or (3) equal amounts of vehicle (sodium chloride, 0.9%). High-dose LEV (70 mg/kg) was applied in a dose slightly above the anticonvulsant ED50 in rodents (Doheny et al., 1999; Gower et al., 1992) in which sedative or other neurologic adverse effects (Marini et al., 2004; Hanon and Klitgaard, 2001) are not expected from recent experimental studies in adult rodents. Low-dose LEV (7 mg/kg) was adapted to a recommended daily dosage in clinical practice. Immediately after the first drug administration, pups were randomly assigned to either (1) hypothermia (whole-body-cooling, for this purpose designed apparatus with temperature control by water circulation, 32 °C) for 4 h, or to (2) normothermia (36–37 °C). Skin surface temperatures were regularly monitored using a noncontact infrared laser thermometer (ProScan50, DOSTMANN electronic GmbH, Germany). All animals were kept in a 12:12 h dark/light cycle at 22 °C environmental temperature with adequate food and water supply for dams, and weights of pups were recorded daily. Animals were euthanized by an i.p. dose of chloralhydrate (200 mg/kg body weight) followed by transcardial perfusion. For protein samples, pups were perfused with PBS 24 h after injury, the olfactory bulb and cerebellum were removed and brain hemispheres were immediately snap-frozen in liquid nitrogen and stored at  80 °C until further analysis. For histological studies, pups were transcardially perfused with PBS followed by 4% paraformaldehyde (PFA) 24 h after hypoxia (P10) and at the age of 60 days (P60). Brains were post-fixed for 24 h with 4% PFA and subsequently embedded in paraffin. Total number of animals used in experimental groups were as follows for histopathology and immunohistochemistry at P10: 1) normoxia group:n¼10, 2) hypoxia þvehicle:n¼9, 3) hypoxiaþLEV7:n¼ 9, 4) hypoxia þLEV70:n¼ 9, 5) hypoxia þHT:n¼ 10, 6) hypoxia þHT þLEV70:n¼10, 7) hypoxia þ HT þLEV70:n¼ 9. Total number of animals used in experimental groups were as follows for immunoblotting at P10: 1) normoxia group:n ¼13, 2) hypoxiaþvehicle:n¼12, 3) hypoxia þLEV7:n ¼10, 4) hypoxia þLEV70: n¼10, 5) hypoxia þ HT:n¼11, 6) hypoxia þ HTþ LEV7:n ¼13, 7) hypoxia þHT þLEV70:n¼12. Total number of animals used in experimental groups were as follows for behavioral testing: 1) normoxia group:n¼ 14, 2) hypoxia þvehicle:n¼ 10, 3) hypoxia þLEV7: n¼13, 4) hypoxia þHT:n¼ 10, 5) hypoxia þHT þ LEV7:n ¼10. 5.2. Histopathology and ventricular volumetry After embedding in paraffin, brains were cut into 10 mm coronal sections. Every tenth section was stained with cresyl-violet (Sigma-Aldrich, Germany, Lot #C5042-10G) as previously described (Shrivastava et al., 2012) and the severity of tissue loss was detected by volumetry (Wang et al., 2013; Lin et al., 2005) of lateral ventricles (Allen Mouse brain reference atlas: coronal level bregma 0.145–0.445 mm; http://www.brainatlas.org). Sections were evaluated under a bright light microscope (Axioplan, Zeiss, Germany) connected to a CCD camera (Microfire, AVT Horn, Germany) at a 2  magnification. Photographs of 2 sections per animal were analysed. Volumes of the lateral ventricles and hemispheres were quantified by ImageJ (NIH, USA). Data are expressed as mean-values of the ratio of the size of the lateral ventricles to their corresponding hemispheres and normalized to control values.

122

K. Strasser et al. / Brain Research 1646 (2016) 116–124

5.3. Immunohistochemistry For assessment of apoptotic cell death, 10 mm coronal sections were stained at two different levels (Allen Mouse brain reference atlas: coronal level bregma 0.145–0.445 mm and  2.00 to 2.355 mm) using fluorescence protocols. Cell death was analysed via staining of DNA fragmentation using terminal transferase dUTP nick end labeling (TUNEL) according to the manufacturer´s protocol (In Situ Cell Death Detection Kit, Roche, Switzerland, Lot #11684795910). For immunohistochemical analyses of oligodendrocytes and neurons, the following primary antibodies were used: monoclonal mouse anti-Olig2 (oligodendrocyte transcription factor-2, 1:100, Millipore, USA, Lot #MABN50) and polyclonal rabbit anti-NeuN (neuronal nuclei, 1:200, Millipore, USA, Lot #ABN78). For fluorescence co-staining of Olig2 and NeuN with TUNEL, sections were deparaffinised followed by heat-induced antigen retrieval with citrate buffer (10 mM Tri-Sodium Citrate, 0.05% Tween 20%, pH 6.0). After non-specific blocking (50 mM Tris–HCl, cold fish skin gelatin, bovine serum albumin, auf pH 7.6), sections were incubated overnight with primary antibody in blocking solution as previously described. Thereafter, primary antibody binding was visualized with the appropriate secondary Alexa Fluor 594 conjugated antibody (1:500, Invitrogen, Germany, anti-mouse Lot #A11005 and anti-rabbit Lot #A11037) followed by TUNEL according to the manufacturer's protocol and nuclear staining by 4′,6-diamidine-2-phenylindole (DAPI) (10 ng/ml, Invitrogen, Germany, Lot #D1306) incubation. TUNEL staining without enzyme solution served as negative control. Degenerating cells were analysed in 2 sections per animal by counting either TUNEL þ cells or double-labeled Olig2 þ /TUNEL þ and NeuN/TUNEL þ oligodendrocytes and neurons respectively at 20  magnification under an inverted confocal fluorescence microscope system (A1 Eclipse Ti, Nikon, Germany). TUNEL þ cells as well as DNA-fragmented oligodendrocytes and neurons were evaluated in 6 different regions of interest (ROI): deep cortical white matter (DWM), Hippocampus (HC), Cortex (Cx), Thalamus (Th), Striatum periventricular (Str) and Capsula interna (CI). Data are expressed as the sum of counted co-stained cells per hemisphere. 5.4. Immunoblotting Apoptosis, myelination and neuronal structures were evaluated by immunoblotting of cleaved Caspase-3 (cCasp3), MBP and MAP2. Snap-frozen tissue was homogenized in ice-cooled radioimmunoprecipitation assay (RIPA, Sigma-Aldrich, Germany, Lot #R0278) buffer containing protease inhibitor. The homogenate was centrifuged at 3.000g (4 °C) for 10 min, and the microsomal fraction was subsequently centrifuged at 17.000g (4 °C) for 20 min. Protein concentrations were determined by using a BCA kit (Thermo Fisher Scientific, Rockford, IL, Lot #23225). 40 mg of the resulting cytosolic protein extracts were heat-denatured in Laemmli sample loading buffer (0.5 M Tris–HCl, SDS 10%, 87% glycerol, beta-Mercaptoethanol, bromophenol), separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (15% for cCasp3 and MBP, 6% for MAP2), and electrotransferred onto a nitrocellulose membrane (Protran, Schleicher and Schüll, Germany). Equal loading and transfer of proteins was confirmed by staining the membranes with Ponceau S solution (Fluka, Switzerland, Lot #P7170). 5% nonfat dry milk in TBS/0.1% Tween-20 (TBST) was used for blocking non-specific protein binding. Membranes were incubated overnight (4 °C) with the following primary antibodies in 5% nonfat dry milk in TBST: monoclonal rabbit anti-cleaved caspase-3 (1:1.000, Cell Signaling, USA, Lot #9661), monoclonal mouse anti-MBP (1:2.000, Abcam, UK, Lot #ab78157), monoclonal

mouse anti-MAP2 (1:1.000, Sigma-Aldrich, USA, Lot #M9942) and monoclonal mouse anti-β-actin (1:1.000, Sigma-Aldrich, USA, Lot #A5316). Horseradishperoxidase-conjugated secondary antibodies (anti-rabbit and anti-mouse, Dako, Denmark, Lot #0217 und #0260) were diluted 1:1.000 in 5% nonfat dry milk in TBST for cCasp3 and MBP respectively in TBST for MAP2 and β-actin. Chemiluminescent detection was performed by enhanced chemiluminescence (ECL; Amersham Biosciences, UK, Lot #RPN2132). For visualization and densitometric analysis Chemi-Doc XRS þ imaging system and ImageLab software (Bio-Rad, Germany) were used. Data were expressed as ratio of density of analysed protein and house-keeping protein β-actin. The control group was set to 100%. 5.5. Behavioral testing To analyze long-term outcome of experimental animals, anxiety, cognitive function and sensomotoric performance were evaluated on postnatal days 28–37. Animals did not repetitively perform the same behavioral task to preserve novelty and mitigate stress. Mice were allowed to acclimate to the surroundings for one week. All tests were conducted during dark cycle under red light. Animals were kept in a 12:12 h dark/light cycle at 22 °C environmental temperature with adequate food and water supply, and weights were recorded daily. 5.5.1. Elevated plus maze The 4 arms of Elevated Plus Maze (TSE systems, USA) are arranged in the shape of a cross, with 2 open arms and 2 arms with surrounding walls in opposing pairs (length 300 mm, width 50 mm, height 150 mm, grey polyvinyl chloride; TSE-systems, USA). Elevated Plus Maze was developed to assess anxiety-related and examining behaviour of rodents. Mice which are more anxious respectively less examining will spend less time in open arms of this maze (Walf et al., 2008; Rägo et al., 1988; Lister, 1987). At the age of four weeks (P28) and the beginning of testing, mice were placed at the junction of the 4 arms, facing an open arm. The entries and duration in each arm of all mice were recorded for 5 min by a video-tracking system (VideoMot 2, TSE-systems, USA). Data are expressed as percentage of covered distance in open arms from the whole distance. 5.5.2. Novel-object-recognition Mice were tested for cognitive function and memory at P29– P31 by Novel-Object-Recognition (Kishimoto et al., 2015; Umpierre et al., 2014; Nagai et al., 2003). Animals were habituated to the empty square arena (length/width 400 mm, height 300 mm, grey polyvinyl chloride) 2 consecutive days prior to testing for 5 min per day. On the following day, mice were in phase 1 exposed to 2 identical objects in 2 opposing corners over a span of 5 min. After a break of 30 min, one of the familiar objects was replaced by a new one. Locomotion and exploration time on different objects were tracked by VideoMot2. Data are expressed as percentage of the time mice spent with the new object from the overall time mice spent at any object. 5.5.3. RotaRod The RotaRod consists of a rotating drum with a speed accelerating from 4 to 40 rounds per minute (Ugo Basile, model 47600, Italy) to assess motor coordination skills (Kilic et al., 2008; Kamei et al., 1975). Maximum speed was reached after 120 s. After two consecutive training days for a maximum time of 120 s, mice were timed on the third day for latency to fall off the RotaRod at P37. Data are expressed as absolute running time the animals spent at the wheel.

K. Strasser et al. / Brain Research 1646 (2016) 116–124

6. Data analysis Statistical analysis was performed using SPSS statistics 21.0 (IBM, USA). For histological and protein expression as well as for behavioral tests, effects between single groups were analysed by one-way ANOVA followed by Bonferroni or Tamhane's T2 post hoc tests depending on variance homogeneity. Data are presented as mean and standard deviation (SD).

Sources of funding This work was supported by UCB S.A. (Belgium), the “IFORESProgram, Interne Forschungsförderung” of the University Hospital Essen (Germany) and the C.D. Stiftung.

Disclosures None.

Acknowledgments The authors thank Karina Kempe, Christian Koester and Mandana Rizazad for their excellent technical assistance and the Department of Precision Mechanics of the University Hospital Essen (Germany) for the construction of the hypothermia apparatus and mazes for behavioural testing.

References Azzopardi, D., Strohm, B., Marlow, N., Brocklehurst, P., Deierl, A., Eddama, O., Goodwin, J., Halliday, H.L., Juszczak, E., Kapellou, O., Levene, M., Linsell, L., Omar, O., Thoresen, M., Tusor, N., Whitelaw, A., Edwards, A.D., TOBY Study Group, 2014. Effects of hypothermia for perinatal asphyxia on childhood outcomes. N. Engl. J. Med. 371 (2), 140–149. Azzopardi, D.V., Strohm, B., Edwards, A.D., Dyet, L., Halliday, H.L., Juszczak, E., Kapellou, O., Levene, M., Marlow, N., Porter, E., Thoresen, M., Whitelaw, A., Brocklehurst, O., TOBY Study Group, 2009. Moderate hypothermia to treat perinatal asphyxial encephalopathy. N. Engl. J. Med. 361 (14), 1349–1358. Back, S.A., Luo, N.L., Borenstein, N.S., Levine, J.M., Volpe, J.J., Kinney, H.C., 2001. Late oligodendrocyte progenitors coincide with the developmental window of vulnerability for human perinatal white matter injury. J. Neurosci. 21 (4), 1302–1312. Beaulieu, M.J., 2013. Levetiracetam. Neonatal Netw. 32 (4), 285–288. Bittigau, P., Sifringer, M., Genz, K., Reith, E., Pospischil, D., Govindarajalu, S., Dzietko, M., Pesditschek, S., Mai, I., Dikranian, K., Olney, J.W., Ikonomidou, C., 2002. Antiepileptic drugs and apoptotic neurodegeneration in the developing brain. Proc. Natl. Acad. Sci. USA 99 (23), 15089–15094. Blomgren, K., Leist, M., Groc, L., 2007. Pathological apoptosis in the developing brain. Apotosis 12 (5), 993–1010. Booth, D., Evans, D.J., 2004. Anticonvulsants for neonates with seizures. Cochrane Database Syst. Rev. (4), CD004218 Boylan, G.B., Rennie, J.M., Pressler, R.M., Wilson, G., Morton, M., Binnie, C.D., 2002. Phenobarbitone, neonatal seizures, and video-EEG. Arch. Dis. Child. Fetal Neonatal Ed. 86 (3), F165–F170. Carlsson, Y., Wang, X., Schwendimann, L., Rousset, C.I., Jacotot, E., Gressens, P., Thoresen, M., Mallard, C., Hagberg, H., 2012. Combined effect of hypothermia and caspase-2 gene deficiency on neonatal hypoxic-ischemic brain injury. Pediatr. Res. 71 (5), 566–572. Dessens, A.B., Cohen-Kettenis, P.T., Mellenbergh, G.J., Koppe, J.G., van De Poll, N.E., Boer, K., 2000. Association of prenatal phenobarbital and phenytoin exposure with small head size at birth and with learning problems. Acta Paediatr. 89 (5), 533–541. Doheny, H.C., Ratnaraj, N., Whittington, M.A., Jefferys, J.G., Patsalos, P.N., 1999. Blood and cerebrospinal fluid pharmacokinetics of the novel anticonvulsant levetiracetam (ucb L059) in the rat. Epilepsy Res. 34 (2–3), 161–168. Edwards, A.D., Yue, X., Squier, M.V., Thoresen, M., Cady, E.B., Penrice, J., Cooper, C.E., Wyatt, J.S., Reynolds, E.O., Mehmet, H., 1995. Specific inhibition of apoptosis after cerebral hypoxia-ischaemia by moderate post-insult hypothermia. Biochem. Biophys. Res. Commun. 217 (3), 1193–1199. Edwards, A.D., Brocklehurst, P., Gunn, A.J., Halliday, H., Juszczak, E., Levene, M., Strohm, B., Thoresen, M., Whitelaw, A., Azzopardi, D., 2010. Neurological outcomes at 18

123

months of age after moderate hypothermia for perinatal hypoxic ischaemic encephalopathy: synthesis and meta-analysis of trial data. BMJ 340, c363. Forcelli, P.A., Kozlowski, R., Snyder, C., Kondratyev, A., Gale, K., 2012. Effects of neonatal antiepileptic drug exposure on cognitive, emotional, and motor function in adult rats. J. Pharmacol. Exp. Ther. 340 (3), 558–566. Gibbs, J.E., Walker, M.C., Cock, H.R., 2006. Levetiracetam: antiepileptic properties and protective effects on mitochondrial dysfunction in experimental status epilepticus. Epilepsia 47 (3), 469–478. Gower, A.J., Noyer, M., Verloes, R., Gobert, J., Wülfert, E., 1992. ucb L059, a novel anti-convulsant drug: pharmacological profile in animals. Eur. J. Pharmacol. 222 (2–3), 193–203. Gressens, P., Dingley, J., Plaisant, F., Porter, H., Schwendimann, L., Verney, C., Tooley, J., Thoresen, M., 2008. Analysis of neuronal, glial, endothelial, axonal and apoptotic markers following moderate therapeutic hypothermia and anesthesia in the developing piglet brain. Brain Pathol. 18 (1), 10–20. Griesmaier, E., Stock, K., Medek, K., Stanika, R.I., Obermair, G.J., Posod, A., Wegleiter, K., Urbanek, M., Kiechl-Kohlendorfer, U., 2014. Levetiracetam increases neonatal hypoxic-ischemic brain injury under normothermic, but not hypothermic conditions. Brain Res. 1556, 10–18. Hanon, E., Klitgaard, H., 2001. Neuroprotective properties of the novel antiepileptic drug levetiracetam in the rat middle cerebral artery occlusion model of focal cerebral ischemia. Seizure 10 (4), 287–293. Hedner, T., Lundborg, P., 1980. Catecholamine metabolism in neonatal rat brain during asphyxia and recovery. Acta Physiol. Scand. 109 (2), 169–175. Holmes, L.B., Harvey, E.A., Coull, B.A., Huntington, K.B., Khoshbin, S., Hayes, A.M., Ryan, L.M., 2001. The teratogenicity of anticonvulsant drugs. N. Engl. J. Med. 344 (15), 1132–1138. Jacobs, S.E., Berg, M., Hunt, R., Tarnow-Mordi, W.O., Inder, T.E., Davis, P.G., 2013. Cooling for newborns with hypoxic ischaemic encephalopathy. Cochrane Database Syst. Rev., 1. Jonsson, M., Agren, J., Nordén-Lindeberg, S., Ohlin, A., Hanson, U., 2014. Suboptimal care and metabolic acidemia is associated with neonatal encephalopathy but not with neonatalseizures alone: a population-based clinical audit. Acta Obstet. Gynecol. Scand. 93 (5), 477–482. Kilic, E., Kilic, U., Bacigaluppi, M., Guo, Z., Abdallah, N.B., Wolfer, D.P., Reiter, R.J., Hermann, D.M., Bassetti, C.L., 2008. Delayed melatonin administration promotes neuronal survival, neurogenesis and motor recovery, and attenuates hyperactivity and anxiety after mild focal cerebral ischemia in mice. J. Pineal Res. 45 (2), 142–148. Kilicdag, H., Daglıoglu, K., Erdogan, S., Guzel, A., Sencar, L., Polat, S., Zorludemir, S., 2012. The effect of levetiracetam on neuronal apoptosis in neonatal rat model of hypoxic ischemic brain injury. Early Hum. Dev. 89 (5), 355–360. Kim, J.S., Kondratyev, A., Tomita, Y., Gale, K., 2007. Neurodevelopmental impact of antiepileptic drugs and seizures in the immature brain. Epilepsia 48 (5), 19–26. Komur, M., Okuyaz, C., Celik, Y., Resitoglu, B., Polat, A., Balci, S., Tamer, L., Erdogan, S., Beydagi, H., 2014. Neuroprotective effect of levetiracetam on hypoxic ischemic brain injury in neonatal rats. Childs Nerv. Syst. 30 (6), 1001–1009. Kishimoto, Y., Cagniard, B., Yamazaki, M., Nakayama, J., Sakimura, K., Kirino, Y., Kano, M., 2015. Task-specific enhancement of hippocampus-dependent learning in mice deficient in monoacylglycerol lipase, the major hydrolyzing enzyme of the endocannabinoid 2-arachidonoylglycerol. Front. Behav. Neurosci. 9, 134. Kamei, C., Masuda, Y., Oka, M., Shimizu, M., 1975. Effects of antidepressant drugs on amygdaloid after-discharge in rats. Jpn. J. Pharmacol. 25 (4), 359–365. Lee, A.C., Kozuki, N., Blencowe, H., Vos, T., Bahalim, A., Darmstadt, G.L., Niermeyer, S., Ellis, M., Robertson, N.J., Cousens, S., Lawn, J.E., 2013. Intrapartum-related neonatal encephalopathy incidence and impairment at regional and global levels for 2010 with trends from 1990. Pediatr. Res. 74 (1), 50–72. Lin, S., Fan, L.W., Pang, Y., Rhodes, P.G., Mitchell, H.J., Cai, Z., 2005. IGF-1 protects oligodendrocyte progenitor cells and improves neurological functions following cerebral hypoxia-ischemia in the neonatal rat. Brain Res. 1063 (1), 15–26. Lister, R.G., 1987. The use of a plus-maze to measure anxiety in the mouse. Psychopharmacol. Berl. 92 (2), 180–185. Marini, H., Costa, C., Passaniti, M., Esposito, M., Campo, G.M., Ientile, R., Adamo, E.B., Marini, R., Calabresi, P., Altavilla, D., Minutoli, L., Pisani, F., Squadrito, F., 2004. Levetiracetam protects against kainic acid-induced toxicity. Life Sci. 74 (10), 1253–1264. Mazarati, A.M., Baldwin, R., Klitgaard, H., Matagne, A., Wasterlain, C.G., 2004. Anticonvulsant effects of levetiracetam and levetiracetam-diazepam combinations in experimental status epilepticus. Epilepsy Res. 58 (2–3), 167–174. Mikhailenko, V.A., Butkevich, I.P., Bagaeva, T.R., Makukhina, G.V., Otellin, V.A., 2010. Early and delayed effects of hypoxia during the infantile period on behavioral and hormonal reactions of rats. Bull. Exp. Biol. Med. 149 (4), 405–408. Murray, D.M., Bala, P., O’Connor, C.M., Ryan, C.A., Connolly, S., Boylan, G.B., 2010. The predictive value of early neurological examination in neonatal hypoxicischaemic encephalopathy and neurodevelopmental outcome at 24 months. Dev. Med. Child. Neurol. 52 (2), 55–59. Nagai, T., Yamada, K., Kim, H.C., Kim, Y.S., Noda, Y., Imura, A., Nabeshima, Y., Nabeshima, T., 2003. Cognition impairment in the genetic model of aging klotho gene mutant mice: a role of oxidative stress. FASEB J. 17 (1), 50–52. Neininger, M.P., Ullmann, M., Dahse, A.J., Syrbe, S., Bernhard, M.K., Frontini, R., Kiess, W., Merkenschlager, A., Thome, U., Bertsche, T., Bertsche, A., 2015. Use of levetiracetam in neonates in clinical practice: a retrospective study at a german university hospital. Neuropediatrics 46 (5), 329–334. Piña-Garza, J.E., Nordli Jr, D.R., Rating, D., Yang, H., Schiemann-Delgado, J., Duncan, B., Levetiracetam N01009 Study Group, 2009. Adjunctive levetiracetam in infants and young children with refractory partial-onset seizures. Epilepsia 50

124

K. Strasser et al. / Brain Research 1646 (2016) 116–124

(5), 1141–1149. Rägo, L., Kiivet, R.A., Harro, J., Pŏld, M., 1988. Behavioral differences in an elevated plus-maze: correlation between anxiety and decreased number of GABA and benzodiazepine receptors in mouse cerebral cortex. Naunyn Schmiede. Arch. Pharmacol. 337 (6), 675–678. Radtke, R.A., 2001. Pharmacokinetics of levetiracetam. Epilepsia 42 (4), 24–27. Ramantani, G., Ikonomidou, C., Walter, B., Rating, D., Dinger, J., 2011. Levetiracetam: safety and efficacy in neonatal seizures. Eur. J. Paediatr. Neurol. 15 (1), 1–7. Rastogi, R.N., Prichard, J.S., Lowden, J.A., 1968. Elevation of phosphorus levels in serum and decreased rain content of gangliosides in rats following neonatal asphyxia. Pediatr. Res. 2 (2), 125–130. Rennie, J.M., Hagmann, C.F., Robertson, N.J., 2007. Outcome after intrapartum hypoxic ischaemia at term. Semin. Fetal Neonatal Med. 12 (5), 398–407. Sarnat, H.B., Sarnat, M.S., 1976. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch. Neurol. 33 (10), 696–705. Schneider, C., Krischke, G., Rascher, W., Gassmann, M., Trollmann, R., 2012. Systemic hypoxia differentially affects neurogenesis during early mouse brain maturation. Brain Dev. 34 (4), 261–273. Sendrowski, K., et al., 2011. Levetiracetam protects hippocampal neurons in culture against hypoxia-induced injury. Folia Histochem. Cytobiol. 49, 148–152. Shah, D.K., Wusthoff, C.J., Clarke, P., Wyatt, J.S., Ramaiah, S.M., Dias, R.J., Becher, J.C., Kapellou, O., Boardman, J.P., 2014. Electrographic seizures are associated with brain injury in newborns undergoing therapeutic hypothermia. Arch. Dis. Child. Fetal Neonatal Ed. 99 (3), F219–F224. Shankaran, S., Pappas, A., McDonald, S.A., Vohr, B.R., Hintz, S.R., Yolton, K., Gustafson, K.E., Leach, T.M., Green, C., Bara, R., Petrie Huitema, C.M., Ehrenkranz, R. A., Tyson, J.E., Das, A., Hammond, J., Peralta-Carcelen, M., Evans, P.W., Heyne, R. J., Wilson-Costello, D.E., Vaucher, Y.E., Bauer, C.R., Dusick, A.M., Adams-Chapman, I., Goldstein, R.F., Guillet, R., Papile, L.A., Higgins, R.D., Eunice Kennedy Shriver NICHD Neonatal Research Network, 2012. Childhood outcomes after hypothermia for neonatal encephalopathy. N. Engl. J. Med. 366 (22), 2085–2092. Shrivastava, K., Chertoff, M., Llovera, G., Recasens, M., Acarin, L., 2012. Short and long-term analysis and comparison of neurodegeneration and inflammatory cell response in the ipsilateral and contralateral hemisphere of the neonatal mouse brain after hypoxia/ischemia. Neurol. Res. Int. 2012, 781512. Silverstein, F.S., Ferriero, D.M., 2008. Off-label use of antiepileptic drugs for the treatment of neonatal seizures. Pediatr. Neurol. 39 (2), 77–79. Stefovska, V.G., Uckermann, O., Czuczwar, M., Smitka, M., Czuczwar, P., Kis, J., Kaindl,

A.M., Turski, L., Turski, W.A., Ikonomidou, C., 2008. Sedative and anticonvulsant drugs suppress postnatal neurogenesis. Ann. Neurol. 64 (4), 434–445. Tagin, M.A., Woolcott, C.G., Vincer, M.J., Whyte, R.K., Stinson, D.A., 2012. Hypothermia for neonatal hypoxic ischemic encephalopathy: an updated systematic review and meta-analysis. Arch. Pediatr. Adolesc. Med. 166 (6), 558–566. Thoresen, M., Bågenholm, R., Løberg, E.M., Apricena, F., Kjellmer, I., 1996. Posthypoxic cooling of neonatal rats provides protection against brain injury. Arch. Dis. Child. Fetal Neonatal Ed. 74 (1), F3–F9. Towbin, A., 1970. Central nervous system damage in the human fetus and newborn infant. Mechanical and hypoxic injury incurred in the fetal-neonatal period. Am. J. Dis. Child. 119 (6), 529–542. Trollmann, R., Richter, M., Jung, S., Walkinshaw, G., Brackmann, F., 2014. Pharmacologic stabilization of hypoxia-inducible transcription factors protects developing mouse brain from hypoxia-induced apoptotic cell death. Neuroscience 278, 327–342. Trollmann, R., Schneider, J., Keller, S., Strasser, K., Wenzel, D., Rascher, W., Ogunshola, O.O., Gassmann, M., 2008. HIF-1-regulated vasoactive systems are differentially involved in acute hypoxic stress responses of the developing brain of newborn mice and are not affected by levetiracetam. Brain Res. 1199, 27–36. Umpierre, A.D., Remigio, G.J., Dahle, E.J., Bradford, K., Alex, A.B., Smith, M.D., West, P. J., White, H.S., Wilcox, K.S., 2014. Impaired cognitive ability and anxiety-like behavior following acute seizures in the Theiler’s virus model of temporal lobe epilepsy. Neurobiol. Dis. 64, 98–106 (J Neurochem). van Rooij, L.G., Hellström-Westas, L., de Vries, L.S., 2013. Treatment of neonatal seizures. Semin. Fetal Neonatal Med. 18 (4), 209–215. Vannucci, S.J., Vannucci, R.C., 1980. Glycogen metabolism in neonatal rat brain during anoxia and recovery. J. Neurochem. 34 (5), 1100–1105. Walf, A.A., Koonce, C.J., Frye, C.A., 2008. Estradiol or diarylpropionitrile decrease anxiety-like behavior of wildtype, but not estrogen receptor beta knockout mice. Behav. Neurosci. 122 (5), 974–981. Wang, H., Gao, J., Lassiter, T.F., McDonagh, D.L., Sheng, H., Warner, D.S., Lynch, J.R., Laskowitz, D.T., 2006. Levetiracetam is neuroprotective in murine models of closed head injury and subarachnoid hemorrhage. Neurocrit. Care 5 (1), 71–78. Wang, Y., Li, B., Li, Z., Huang, S., Wang, J., Sun, R., 2013. Improvement of hypoxiaischemia-induced white matter injury in immature rat brain by ethyl pyruvate. Neurochem. Res. 38 (4), 742–752. Yager, J.Y., Brucklacher, R.M., Vannucci, R.C., 1992. Cerebral energy metabolism during hypoxia-ischemia and early recovery in immature rats. Am. J. Physiol. 262 (3 Pt 2), H672–H677.