ORIGINAL PAPER

Effects of Levetiracetam on neural tube developmentand closure of the chick embryos in ovo

Füsun Demirçivi Özer & Adıgüzel Demirel &Özlem Yılmaz Dilsiz & Murat Aydın & Nail Özdemir &

Yiğit Uyanıkgil & Meral Baka

Received: 1 February 2012 /Accepted: 3 April 2012 /Published online: 9 May 2012# Springer-Verlag 2012

AbstractPurpose Effects of Levetiracetam (LEV) within its therapeu-tic range at a 50 mg/kg dose for the chick embryo in ovo hasbeen studied in order to demonstrate whether LEV wouldeffect neural tube closure at the macroscopic morphology orLEV administered embryos still encounter neuroglial detri-mental effects at the histological level.Methods Embryos were randomly seperated into control(n020) and study (n020) groups. The eggshell was win-dowed at specifically 24 h of incubation, and area underlyingthe membrane was excised to allow injection with 4.5 μl LEVin the study group, while physiologic saline (0.045 ml) wereinjected in the control group and each egg were re-incubatedfor 48 h more. Then, histological and immunohistochemicalevaluation of the subjects were done.Results Macroscopic evaluation revealed immaturity of theplacental vessel network in number and width for the study

group in comparison to the controls. Defects of migration,decrease in the crista neuralis content, delay of the basal platesstructures in the formation of the usual configuration, anddelay in the cellular proliferation and the delay of develop-ment for the central nervous system were determined in theLEV-exposed group. Immunostaining of S100 proteins in thisstudy has clearly demonstrated increased expression patternsof both neuroglial and neuronal cell populations. Toluidineblue stainings revealed mostly bipolar, differentiating neuronsand crista neuralis cells which is concordant with activemigration and differentiation.Conclusions LEV found that delay in the closure of theneural tube and microcephalic fetuses disturb further mor-phological, biochemical, and functional development.

Keywords Chick embryo . Epilepsy . Levetiracetam .

Neural tube defects . Pregnancy

Introductıon

Epileptic women have twice more risk of bearing childrenwith congenital malformations than that of the general popu-lation. Most antiepileptic drugs (AEDs) have been associatedwith such risk. Seizure control must not be neglected in apregnant woman with epilepsy since seizures are associatedwith harm to the fetus as well as the mother. Risk may beminimized by using a single AED at the lowest effectivedosage. It has been known since 1960 that administration ofhigher dosages of antiepileptics, higher concentrations ofthese drugs in the blood, and polytheraphy are all associatedwith higher risks for both anatomical and behavioral terato-genesis in the embryo [1]. For some drugs such as valproate,phenobarbital, phenynitoin, and carbamazepine, the risks arewidely studied and outcomes for the fetus are more or less

F. D. Özer (*) :N. Özdemir (*)Department of Neurosurgery,Tepecik Research and Training Hospital,173 S, No:34/6, Taşdelen Apt, Basin Sitesi,İzmir, Turkeye-mail: fdcviozer@hotmail.come-mail: dr.nailozdemir@gmail.com

A. DemirelDepartment of Neurosurgery,Bozyaka Research and Training Hospital,Izmir, Turkey

Ö. Yılmaz Dilsiz :Y. Uyanıkgil :M. BakaDepartment of Histology and Embryology,Ege University Faculty of Medicine,Izmir, Turkey

M. AydınDepartment of Neurosurgery, Karaman State Hospital,Karaman, Turkey

Childs Nerv Syst (2012) 28:969–976DOI 10.1007/s00381-012-1758-0

clear [1], but consistent data is completely lacking especiallyfor newer drugs.

The rate of major malformation with gabapentin wasfound at 4.5 %; 87 % of the pregnancies resulted in livebirths, 11.3 % in miscarriages, and 2 % in therapeuticabortions [2, 3]. Structural birth defects was reported at1.8 % with lamotrigine monotheraphy according to theresults of the International Lamotrigine Pregnancy Registry[4]. Fewer fetal malformation were reported with the use ofoxcarbazepine, topiramate, and zonisamide [5–13].

Nowadays, Levetiracetam (LEV) is a widely used anti-epileptic agent as a safe and well-tolerated novel drug suchthat it has become one of the four most commonly usedantiepileptic drugs since 2007 for women of childbearingage with epilepsy [14]. LEV is classified as a PregnancyCategory C drug, which means that there are no adequateand well-controlled studies accomplished in humans, butanimal reproduction studies have shown adverse effects onthe fetus. So it should be either not used during pregnancy;or if administration of the drug is a medical necessity, itshould be only used in case the potential benefit justified thepotential risk to the fetus [15]. Prenatal exposure to LEV hasbeen shown to cause skeletal abnormalities and growthretardation in animal studies, but its teratogenicity inhumans is still uncertain. Only a few studies with limitednumber of pregnant epileptic women are present [1, 15–20].Those case studies report that no malformation was detectedin pregnancies with LEV exposure, but low birth weight isan issue for the fetus after the parturition [19] .

Women receiving antiepileptic drug treatment during preg-nancy is most likely known to experience disorders of primaryneurulation, one of which is open neural tube defects of theoffspring. Due to the fact that simple observational period forthe young embryos, shorter developmental period, and pres-ence of easy surgical manipulation, and comparatively inex-pensive and effortless availability characteristics of the eggs,chick embryo model has been widely used for such investi-gations about effects of drugs or various substances on theembryo [21].

The aim of this study is to demonstrate effects of LEVwithin its therapeutic range at a 50 mg/kg dose for the chickembryo in ovo via histological techniques and immunohis-tochemical expression of S100. Our hypothesis is that evenif this LEV dose would not effect neural tube closure of thechick embryos at the macroscopic morphology, LEV-administered embryos still encounter neuroglial detrimentaleffects at the histological level.

Material and methods

The study protocol complied with the European Communityguidelines for the use of experimental animals. All

experiments were approved by the Local Animal CareEthics Committee at Ege University (2010–012). Fertile,specific pathogen-free eggs of the domestic fowl (WhiteLeghorn, Gallus gallus) were obtained from Celal BayarUniversity, Research Institute of Poultry Disease and Vac-cination, Manisa, Turkey. The eggs were incubated for 24 hunder constant temparature at 38–38.5°C in a humidifiedatmosphere (relative humidity, 75–80 %); in a roller incu-bator (Pas Reform C-82, Netherlands) at the ZootechnicsDepartment at Faculty of Agriculture, Ege University, Izmir,Turkey. Incubator automatically repositioned the axis ofeggs every 2 h till the injection time. The eggshell waswindowed at specifically 24 h of incubation and area under-lying the eggshell membrane was excised to allow exposureand injection of the embryo. Embryos were randomly seper-ated into control (n020) and study (n020) groups. Theembryonic disk were seen in every egg, and 4.5 μl LEV(Keppra flacon; UCB Pharma Co, Belgium) diluted in phys-iologic saline were applied once in the study group, whilethe same volume (0.045 ml) of physiologic saline wereinjected in the control group with a 24-gauge syringe underthe embryonic disk. The shell window was closed withadhesive tapes and each egg were re-incubated for 48 hmore till sacrificed.

After being removed from the roller incubator, both controland study group embryos were subsequently removed fromtheir eggs with fine pincers, under the dissection microscope.Embryos were taken into saline solution with their membranesonto the Petri dishes and observed macroscopically first.Then, all embryos were dissected and transferred into 3 %gluteraldehyte with 0.1 M cacodylate buffer for electronmicroscopic investigation and 2 % formaldehyte in 0.1 Mphosphate bufferfor light microscobic examinations.

Histology

Tissue processing and histocytochemical technique

In light microscopic examination period, embryos weretaken in buffer and banished for fixatives. They were thandehydrated in an ascending series of ethanols by adding 1 %eosin for 70 % alcohol. Following becoming transparent byXylene, they were embedded into paraffin. Serial 3–5 μmsections from the caudal embryo including the neuroporusposterior were prepared by microtome(Leica RM 2145),then they were stained with hematoxylin and eosine.

For light microscopic examination of epon sections,after separation of cranial and caudal parts of embryosbecause of facility and assessing target points, tissues werefixed by 2.5% glutaraldehyde buffer in 0.1 M sodium caco-dylate and post-fixed in 1 % osmium tetroxide in 0.1 sodiumcacodylate buffer for 1 h at 4°C. Sections were dehydratedin a graded alcohol series; following transparancy by

970 Childs Nerv Syst (2012) 28:969–976

propylene oxide, they were embedded in Epon 812.Samples were cut using a rotating-blade ultratome (Leica,Heerbrugg, Switzerland), and stained with Toluidine blue.Caudal parts of embryos were analyzed for focusing poste-rior neuropore.

Application of immunohistochemistry

Immunoreactivity of S100 was analyzed under the lightmicroscope (Olympus BX-51) particularly for the descrip-tion of the neuroglial interaction of the embryo sections.For this, dewaxed tissue sections in xylene were thenrehydrated in sequential 100, 95, 80, and 70 % alcoholseries for 2 min each. After leaving in distilled water for5 min, the tissues samples were delineated on the objectslide, washed in phosphate buffered saline (PBS) for10 min, and than left in trypsin for 15 min. The primaryantibody (S100 Ab-1; Thermo Scientific, 7 ml, ready touse) was then applied in an incubator at 57°C and washedwith PBS. Afterwards, the biotinylated secondary anti-body was applied and washed with PBS before incubatingwith the enzyme conjugate and 3,3-diaminobenzidine tet-rahydrochloride. Then, sections were stained withMayer’s hematoxylin (Zymed Laboratories) and thenmounted. All the sections were examined and photo-graphed with Olympus C-5050 digital camera at OlympusBX51 microscope. Both investigators, blinded to thegroup distinctions of the specimens, obtained five imagesfrom 10 different sections per animal under ×100 magni-fication. Negative control samples were processed like-wise; except that primary antibodies were replaced withPBS alone.

The histological procedures employed in this study arefairly conventional and well-established methods [22].All reagents were purchased from Sigma (St Louis,MO, USA) unless noted otherwise. Known positive con-trol sections were used to ensure correct differentiationhas been achieved. All histological procedures proceededat the same time manually. The slides were immersedin the stain/buffer solutions at all times during theincubations.

The intensity of S100 immunohistochemical staining wasgraded semiquantitatively according to the cytoplasmicimmunoreaction in the sections as follows: (−) no immu-nostaining, (+) weak staining, (++) moderate staining, (+++)strong staining.

Results

Embryos from the control groups and the drug-administeredstudy groups were observed and photographed both macro-scopically and microscopically.

Microscopic evaluation

In transvers sections of neural tube slices, observations werefocused on areas where the chorda dorsalis, the dorsal aorta,mesenchymal connective tissues, migrating cells from cristaneuralis, and somits at neural tube regions were evaluated(Figs. 1, 2, and 3). In the control group, both alar and basalplates surraunding the draft canalis neuralis were observedto be well developed, the majority of cells sorted at the outerbasal lamina was found to be for mesenchymal intensity ofthe cells. When cell populations were considered, distinctnuclear differentiation were observed as well as the enlargedmultipolar neuron cell karyons, frequent mitotic activity,and evident glial–neuronal associations. However, dorsally,a well-developed dorsal and ventral plate and, ventrally,both the mitotic activity and the proper picnotic nuclei ofthe cells of the chorda dorsalis were determined. Not onlythe density of the migrating mesenchymal cells from thecrista neuralis in the connective tissue but also the arrange-ment and the alignment of the cells were favorable on bothsides. Mesenchymal connective tissue around the cap platewere well-developed and showed proper closure of theplate. Both endothelial and mesenchimal connective tissuewere in continuity around the aorta. Except for the densecells originating from the crista neuralis, condensed stromalcells were characteristic for the somit development (Fig. 1).

Both placental membranes images and transvers sectionsof embryos were different from control group in

Fig. 1 H&E stainings of thecontrol (a) and experimentalgroups (b). M dorsalmesenchyme, NT neural tube.Original magnification, ×20

Childs Nerv Syst (2012) 28:969–976 971

crosssections. Considering the epithelial arrangement, boththe thickness and number of cells around the canalis cen-tralis sketch were found to be decreased. Especially in basalplate, perikaryons of multipolar neuron were seen outnum-bered comparing control group and they showed a scatteredsettlement.

Picnotic nucleoli of both the neuronal and glial cellsprecursors were found more frequent than control group.In terms of configurations of alar and basal plates, imageswere different from control group.

Cap and floor plates were found to be thin; mesenthimalconnective tissue excluding cap plate, which is the embry-onic connective tissue were found to be decreased for boththe cell number per unit area and thickness between ecto-derm and neural tube.

Cells with picnotic nuclei and vacuole neighboring thefloor plate around Chorda dorsalis showed compatible find-ings for the diameter of the neural tube itself. Migrating cellsfrom the crista neuralis were determined as thinner bands onboth sides of the neural tube with neither any signs ofneuronal differentiation nor any increase in the amount ofcells. None of the subjects showed midline closure defectsin any subject of the study group as contrary to expectations.

It is more likely and suitable to describe all the findings ofthis study like the defects of migration, decrease in the cristaneuralis content, delay of the basal plates structures in theformation of the usual configuration, and delay in the cellularproliferation and the delay of development for the centralnervous system all occurred due to the migration defects.

Immunohistochemical evaluation

Distribution of S100 immunoreactivity in the groups is shownat Table 1. Immunohistochemical expression of S100 weredetermined to be increased especially in the neuroglial cellpopulation of the neural tube and increased expression pat-terns were also observed for the crista neuralis cells of thedorsal mesenchyme for the study group (Fig. 4, Table 1).

Evaluation of Toluidine blue stainings

Toluidine blue stainings of the epon sections from the controland experimental groups revealed mostly bipolar, differentiat-ing neurons, and crista neuralis cells (Fig. 5).

Macroscopic evaluation

Macroscopic evaluation of the subjects under the disectionmicroscope revealed the smaller size of the embryos and theimmaturity of the placental vessel network in number andwidth for the study group in comparison to the controls atfirst sight (Fig. 6). In study group, embryos were detectedmacroscopically small as if intrauterine growth retarded.

Discussion

Management of pregnant women with epilepsy is problem-atic. On one hand, seizures need to be prevented because

Fig. 2 H&E stainings of thecontrol (a) and experimentalgroups (b). NT neural tube, Mmesenchyme. Originalmagnification, ×100

Fig. 3 H&E stainings of thecontrol (a) and experimentalgroups (b). NT neural tube, CDchorda dorsalis. Originalmagnification, ×100

972 Childs Nerv Syst (2012) 28:969–976

they clearly increase the risk of maternal and fetal injury,miscarriage, epilepsy in the offspring, and developmentaldelay; on the other hand, fetal exposure to antiepilepticdrugs is associated with increased risk of congenital malfor-mation, congenital anomalies, intrauterine growth retarda-tion, and neonatal hemorrhage [23]. Physicians mostlyencounter the problem of choosing monotheraphy with min-imal teratogenic effect, plus adequate seizure control ofpregnant woman with epilepsy. Major antiepileptic drugssuch as phenytoin, carbamazepine, valproate, and phenobar-bital have been associated with teratogenic effects includinggrowth retardation, developmental delay, microcephaly, oro-facial clefts, digital anomalies, and neural tube defects [24].Carbamazepine appears to be the least developmentallyneurotoxic compound among the major antiepileptic drugs.So, it has been used for a long time till new generationantiepileptic drugs were introduced [24]. However, there isno sufficient knowledge concerning teratogenic effect ofnewer AED’s. Preliminary studies with small sample sizesin humans have limited conclusions on the safety use ofthese newer drugs.

LEV has a unique mechanism of action and a favorablepharmacokinetic profile that distinguishes it from otherAEDs such that LEV is one of the most commonly pre-scribed antiepileptic drugs, but its mechanism of action isstill uncertain. Based on prior information that LEV binds tothe vesicular protein synaptic vesicle protein 2A (SV2A)and reduces presynaptic neurotransmitter release, wewanted to more rigorously characterize its effect on neural

tube development in chick embryo model. Here, we showthat the neural tube development is impaired microscopical-ly. Inhibition of these functions agreed well with that of thein vivo neurogenesis that has been obtained by us in thechick model, insofar as the same functions are recapitulatedby neural tube development. Overall data suggest that LEVhas a teratogenic component when applied during gestation.This homology between human, rat, and even mice showsthat SV2A increases the confidence in the results obtained inanimal models of epilepsy and compounds interacting withSV2A. The cell functions studied are tightly linked to theintegrity of the cytoskeleton, and hence they are typicallyimpaired by LEV as a consequence of probably its mecha-nism of action.

In 2003, Isoherranen et al. [25] reported their study inSWV/Fnn mice that LEV appeared to be significantly saferthan all other first-generation AEDs for use in pregnancy.Two years later, Manthey et al. [26] investigated LEV andsulthiame on neurotoxic properties in developing rat brainand noted that LEV did not show any neurotoxic effectcontrary to sulthiame.

Özyürek et al. [20] used different doses of LEV (25, 50,and 100 mg/kg/day) in pregnant female rats and show thatLEV had only a transient impact on reflex maturation andno impact on physical and cognitive function in ratoffsprings. Recently, Kim et al. [7] tested the potantialneurotoxicity of three AEDs which included carbamazepine,topiramate, and levetiracetam in the developing rat brain.They found that LEV alone did not induce cell death whenused as monotheraphy; furthermore, only LEV and carba-mazepine combination did not induce neurodegenerationwhen used polytheraphy.

Results of the previous studies indicate that precondition-ing of hippocampal cultures with high concentrations oflevetiracetam (100 and 300 μM) protects neurons againsthypoxia-induced death while lack of neuroprotective actionof the drug on hippocampal neural cultures was determinedwhen a low concentration (10 μM) of levetiracetam wasused [27]. Unique result showing LEV and severedevelopmental abnormalities was reported by Türeci et al.[28] on chicken embryos. They used low and high dose

Table 1 Intensity scores of immunostaining for S100

Evaluated regions/groups Controlgroup

Experimentalgroup

Mesenchymal connectivetissues

++ +

Crista neuralis +++ +++

Neural tube regions ++ ++

Chorda dorsalis ++ +

The intensity of immunostaining is grouped into categories: − nostaining, + weak staining, ++ moderate staining, +++ strong staining

Fig. 4 Immunohistochemicalstaining of S100 for the control(a) and experimental groups(b). Darkly stained cells(arrows) are the neuroglialpopulation of cells that areS100 (+). Originalmagnification is ×100

Childs Nerv Syst (2012) 28:969–976 973

levetiracetam alone, low and high dose valproic acid alone,and combination of two AEDs with low and high dose.Their results showed that both low and high dose LEVcaused developmental anomalies in embryos. Neurotoxicitywith LEV was detected lesser degree comparing valproicacid alone or combination group, but they highlighed thatLEV is not safe for developing embryo and should be usedcautiously in pregnant women with epilepsy.

Thinking over the fact that the genetic mechanisms thatunderlie the specification of macroglial precursors in vivoappear strikingly similar to those that regulate the develop-ment of the diverse neuron types. The switch from produc-ing neuronal to glial subtype-specific precursors have beenpreviously modeled as an interplay between region-restricted components and temporal regulators that deter-mine neurogenic or gliogenic phases of development, con-tributing to glial diversity. Oligodendrocytes and astrocyteshereby have diverse roles in the maintenance of the neuro-logical function [29]. The early-formed neural tube consistsof proliferating, morphologically homogeneous cells,termed “neuroepithelial (NEP) stem cells” which generateneurons, astrocytes, and oligodenrocytes through a series of

intermediate precursor cells [30]. During development, theprecursors for all of these cells reside within the epitheliumof the neural plate and its successor, the neural tube. Theseprecursor cells are the undifferentiated, primitive neuroepi-thelium of the classical literature. Yet, there is still uncer-tainty concerning the developmental stages when glial andneuronal cell lines diverge in the proliferative zone. Com-parison of different brain regions in this study revealed thatcaudal parts of the developing chicken central nervous sys-tem are delayed in development due to LEV exposure suchthat important modifications of the affected glial cells weredetermined. The neuroblast cell population investigatedrevealed alterations as well in the development. Some ofthese alterations included the relationships amongst the cellsof the crista neuralis. Electron microscopy would contributemore to this data if available. Moreover, the asymmetricassociation of development due to LEV exposure seemsmore likely linked to the pattern of expression of adhesionand extracellular matrix molecules. While generally con-firming concepts of astrocyte and oligodendrocyte ontogenyfrom the literature during the neurogenesis, the presentstudy adds considerable detail regarding development. In

Fig. 5 Toluidine blue stainingsof the control (a) andexperimental groups (b). NTneural tube. Mostly bipolar,differentiating neurons andcrista neuralis cells are shown.Original magnification is ×100

Fig. 6 Macroscopic view ofthe control group (left) andexperimental LEV administered(right) embryos. Please note thesmaller size of the embryos andthe immaturity of the placentalvessel network in number andwidth for the study group incomparison to the controls. Thedelay in neck flexuredevelopment is interesting forthe LEV group compared to thecontrols

974 Childs Nerv Syst (2012) 28:969–976

this study, we describe abnormalities occurring in the centralnervous system. These modifications occur mostly duringthe neurulation process.

Various stages of chicken embryos have been previouslyclassified by Hamburger and Hamilton [31] which coincideswith 72 h of embryogenesis in this study. According to thisclassification, migrated crista neuralis cells and closure ofthe caudal neuroporus and formation of the dorsal mesen-chyme, as well as the formation of all the somites point tothe 72 h of embryogenesis. This is probably related to thegrowth retardation and all the other macroscopic findings ofthis study caused by LEV administration. Yet, the delay inthe migrated crista neuralis cells and defect of closure for thecaudal neuroporus and formation of the dorsal mesenchyme,as well as the formation of all the somites point almost to the66–69th hours of development.

S100 proteins are known to regulate processes such asintracellular growth and cell motility, cell cycle regulation,transcription, and differentiation. Recently, S100 proteinshave received increasing attention due to their close associ-ation with several human diseases including cardiomyopa-thy, neurodegenerative disorders and cancer [32]. They havealso been proven to be valuable in the diagnostic of thesediseases, as predictive markers of improving clinical man-agement, outcome and survival of patients and are consid-ered having a potential as drug targets to improve therapies.

Calcium concentration in the cell mechanisms of second-ary messenger are known to serve as a signal for a change.Calcium-binding proteins have essential roles in most of thecellular, signal transduction processes. Adverse outcomes ofantiepileptic drug exposure is related to neuronal apoptosisduring late gestation and the perinatal period. Calcium-dependent interaction of S100 proteins with p63 and p73has been physiologically relevant in both developmental anddisease-related processes [33]. S100 proteins probably playsa role in stimulating neuronal differentiation, proliferation ofastrocytes. These data are consistent with the low amount ofmigrating mesenchymal and crista cells.

Conclusion

Although there is not enough data for the failure of closurein the caudal area in this study, there is still evidence ofdelay in the closure of the neural tube and macroscopicexamination of the embryos have proven to develop intomicrocephalic fetuses. Immunostaining of S100 proteins inthis study has clearly demonstrated active migration anddifferentiation of both neuroglial and neuronal cell popu-lations. However, if acephalic and microcephalic embryosare not taken into account, how one can say that theseabnormalities would not disturb further morphological,biochemical, and functional development and the embryos

would be able to regulate and a majority of normal embry-os would develop? In this concept, LEV is not an innocentdrug to be safely administered to pregnant women andfurther studies are needed to determine whether LEV reallydiffers for its actions on synaptic transmission from otherantiepileptic drugs. Development of drug tolerance andresistance in animal models probably depends on the spe-cific models used and can be highly selective for specificinteractions. Yet, administration of LEVon pregnant wom-en and newborn should be with caution and studies shouldenlighten effects of LEV for the offspring more at thegenetic level.Acknowledgments We thank Mrs. Sezer Özkan and ZootechnicsDepartment of Faculty of Agriculture, Ege University for technicalsupport.

References

1. Meador KJ, Penovich P, Baker GA, Pennel PB, Bromfield E, Pack A,Liporace JD, Sam M, Kalayjian LA, Thurman DJ, Moore E, LoringDW (2009) Antiepileptic drug use in women of childbearing age.Epilepsy Behav 15:339–343

2. Montouris G (2003) Gabapentin exposure in human pregnancy: resultsfrom The Gabapentin Pregnancy Registry. Epilepsy Behav 4:310–317

3. Yerby MS (2003) Clinical care of pregnant women with epilepsy:neural tube defects and folic acid supplementation. Epilepsia44:33–40

4. Ten Berg K, Samren BE, Van Oppen AC, Engelsman M, LindhoutD (2005) Levetiracetam use and pregnancy outcome. ReprodToxicol 20:175–178

5. Bulau P, Paar WD, von Unruh GE (1988) Pharmacokinetics ofoxcarbazepine and 10- hydroxycarbazepine in the newborn child ofan oxcarbazepine-treated mother. Eur J Clin Pharmacol 34:311–313

6. Kawada K, Itop S, Kusaka T, Isobe K, Ishii M (2002) Pharmaco-kinetics of zonisamide in perinatal period. Brain Dev 24:95–97

7. Kim J, Kondratyev A, Gale K (2007) Antiepileptic drug-inducedneuronal cell death in the immature brain: effects of carbamaze-pine, topiramate and levetiracetam as monotheraphy versus poly-theraphy. J Pharmacol Exp Ther 323:165–173

8. Kondo T, Kaneko S, Amano Y, Egewa I (1996) Preliminary reporton teratogenic effects of zonisamide in the offspring of treatedwomen with epilepsy. Epilepsia 37:1242–1244

9. Mountouris G (2005) Safety of the newer antiepileptic drug oxcar-bazepine during pregnancy. Curr Med Res Opin 21:693–701

10. Morrow J, Russell A, Guthrie E, Parsons L, Robertson I, Waddell R,Irwin B, Mc Givern RC, Morrison PJ, Craig J (2006) Malformationrisks of antiepileptic drugs in pregnancy :a prospective study from theUK Epilepsy and Pregnancy Register. J Neurol Neurosurg Psychiatry77:193–198

11. Myllynen P, Pienimaki P, Jouppila P, Vahakangas K (2001) Trans-placental passage of oxcarbazepine and its metabolites in vivo.Epilepsia 42:1482–1485

12. Ohman I, Vitols S, Luef G, Soderfeldt B, Tomson T (2002)Topiramate kinetics during delivery, lactation and in the neonate:preliminary observations. Epilepsia 43:1157–1160

13. Rabinowicz A, Meischenguiser R, Ferraso SM, D’ Giano CH,Carrazana EJ (2002) Single center, 7-year experience of oxcarba-zepine of oxcarbabazepine exposure during pregnancy. Epilepsia43:2008–2009

14. Hunt S, Craig J, Russel A, Gutrie E, Parsons L, Robertson I,Waddel R, Irwin B, Morrison PJ, Morrow J (2006) Levetiracetam

Childs Nerv Syst (2012) 28:969–976 975

in pregnancy: preliminary experience from the UK Epilepsy andPregnancy Register. Neurology 67:1876–1879

15. French J (2001) Use of levetiracetam in special populations.Epilepsia 42:40–43

16. Long L (2003) Levetiracetam monotheraphy during pregnancy: acase series. Epilepsy Behav 4:447–448

17. Longo B, Forinash AB, Murphy JA (2009) Levetiracetam use inpregnancy. Ann Pharmacother 43:1692–1695

18. Shallcross R, Bromley RL, Irvin B, Bonnett LJ, Morrow J, BakerGA (2011) Child development following in utero exposure: leve-tiracetam vs sodium valproate. Neurology 76:383–389

19. Tennis P, Eldridge RR (2002) Preliminary results on pregnancyoutcomes in women using lamotrigine. Scientitif Advisory Com-mittee. International Lamotrigine Pregnancy Registry. Epilepsia43:1161–1167

20. Tomson T, Palm R, Kallen K, Ben-Menacheim E, Suderfeldt B,Danielsson B, Johansson R, Luef G, Ohman I (2007) Pharmaco-kinetics of levetiracetam during pregnancy, delivery, in the neonatalperiod, and lactation. Epilepsia 48:1111–1116

21. You H, Sim KB, Wang KC, Kim DG, Kim H (1994) Morphologicalstudy of surgically induced open neural tube defects in chick embry-os. J Korean Med Sci 2:116–122

22. Sheehan DC, Hrapchak BB (1987) Theory and practice ofhistotechnology. Battelle Memorial Institude, Ohio

23. Kaindal AM, Asimiadou S, Manthey D, Hagen M, Turskı L,Ikonomidou C (2006) Antiepileptic drugs and the developingbrain. Cell Mol Life Sci 63:399–413

24. Isoherranen N, Spiegelstein O, Roeder M, Triplett AA, Schurig V,Finnel RH (2003) Developmental outcome of levetiracetam, its

major metabolite in humans, 2-pyrrlidinone N-butyric acid, and itsenantiomer (R)-alpha-ethyl-oxo-pyrolidine acetamide in a mousemodel of teratogenicity. Epilepsia 44:1280–1288

25. Manthey D, Asimiadou S, Stefovska V, Kaindal AM, Fassbender J,Ikonomidou C, Bittigau P (2005) Sulthiame but not levetiracetamexerts neurotoxic effect in the developing rat brain. Exp Neurol193:497–503

26. Özyürek H, Bozkurt A, Bilge S, Çiftçioğlu E, İlkaya F, Bas DB(2010) Effect of prenatal exposure on motor and cognitive functionof rat offspring. Brain Dev 32:396–403

27. Sendrowski K, Bockowski L, Sobaniec W, Ilendo E, Jaworoska B,Smigielka-Kuzia J (2011) Levetiracetam protects hippocampal neu-rons in culture against hypoxia-induced injury. Folia HistochemCytobiol 49:148–152

28. Türeci E, Asan B, Eser M, Tanrıverdi T, Alkan F, Erdinçler P(2011) The effects of valproic acid and levetiracetam on chickenembryos. J Clin Neurosci 18:816–820

29. Rowitch DH, Kriegstein AR (2010) Developmental genetics ofvertebrate glial–cell specifications. Nature 468:214–222

30. Rao M (2004) Stem and precursor cells in the nervous system. JNeurotrauma 21:415–427

31. Hamburger V, Hamilton HL (1951) A series of stages in thedevelopment of the chick embryo. J Morphol 88:49–92

32. Heizmann CW, Fritz G, Schafer BW (2002) S100 proteins: structure,functions and pathology. Front Biosci 1:1356–1368

33. Van Dieck J, Brandt T, Teufel DP, Veprintsew DB, JoergerAC, Fersht AR (2010) Molecular basis of S100 proteinsinteracting with the p53 homologs p63 and p73. Oncogene29:2024–2035

976 Childs Nerv Syst (2012) 28:969–976