Effects of lipopolysaccharide on blood–brain barrier permeability during pentylenetetrazole-induced epileptic seizures in rats

(2006) 1–7www.elsevier.com/locate/lifescie

Life Sciences 79

Effects of lipopolysaccharide on blood–brain barrier permeability duringpentylenetetrazole-induced epileptic seizures in rats

Nadir Arican a, Mehmet Kaya b,⁎, Rivaze Kalayci c, Hafize Uzun d, Bulent Ahishali e,Bilge Bilgic f, Imdat Elmas a,g, Mutlu Kucuk c, Candan Gurses h, Mustafa Uzun i

a Department of Forensic Medicine, Istanbul Faculty of Medicine, Istanbul University, Capa 34390, Istanbul, Turkeyb Department of Physiology, Istanbul Faculty of Medicine, Istanbul University, Capa 34390, Istanbul, Turkey

c Research Institute for Experimental Medicine, Istanbul University, Capa 34390, Istanbul, Turkeyd Department of Biochemistry Cerrahpasa Medical Faculty, Istanbul University, Cerrahpasa 34303, Istanbul, Turkey

e Department of Histology, Istanbul Faculty of Medicine, Istanbul University, Capa 34390, Istanbul, Turkeyf Department of Pathology, Istanbul Faculty of Medicine, Istanbul University, Capa 34390, Istanbul, Turkey

g Institute of Legal Medicine, Istanbul University, Cerrahpasa 34303, Istanbul, Turkeyh Department of Neurology, Istanbul Faculty of Medicine, Istanbul University, Capa 34390, Istanbul, Turkey

i Council of Forensic Medicine, Cerrahpasa 34303, Istanbul, Turkey

Received 24 August 2005; accepted 12 December 2005

Abstract

We investigated the effects of lipopolysachharide (LPS) on functional and structural properties of the blood–brain barrier (BBB) duringpentylenetetrazole (PTZ)-induced epileptic seizures in rats. Arterial blood pressure was significantly elevated during epileptic seizures irrespectiveof LPS pretreatment. Plasma levels of interleukin (IL)-1, interleukin (IL)-6, nitric oxide (NO) and malondialdehyde (MDA) increased whilecatalase concentrations decreased in animals treated with LPS, PTZ and LPS plus PTZ. The significantly increased BBB permeability to Evansblue (EB) dye in the cerebral cortex, diencephalon and cerebellum regions of rats by PTZ-induced seizures was markedly reduced upon LPSpretreatment. Immunoreactivity for tight junction proteins, zonula occludens-1 and occludin, did not change in brain vessels of animals treatedwith PTZ and LPS plus PTZ. Glial fibrillary acidic protein immunoreactivity was increased in LPS, but not in PTZ and LPS plus PTZ. Theseresults indicate that LPS pretreatment reduces the passage of EB dye bound to albumin into the brain, at least partly, by increasing plasma NO andIL-6 levels during PTZ-induced epileptic seizures. We suggest that LPS may provide protective effects on the BBB integrity during epilepticseizures through transcellular pathway, since the paracellular route remained unaffected by LPS and LPS plus PTZ.© 2005 Elsevier Inc. All rights reserved.

Keywords: Lipopolysaccharide; Pentylentetrazole; Epileptic seizures; Blood–brain barrier; Evans blue; Zonula occludens-1; Occludin

Introduction

Epileptic seizures are the second most common neurologicdisorder after stroke and central nervous system inflammationis implicated on the activity of epileptic seizures in cases suchas infection. Epileptic seizures induce neuronal death in limbicstructures accompanied by major elevations of both cerebralblood flow and metabolic rate (Hashizume et al., 1998;Holmes, 2002). Epileptic seizures lead to the disruption of theblood–brain barrier (BBB) allowing the permeation of blood

⁎ Corresponding author. Tel.: +90 0212 414 22 52; fax: +90 0212 414 22 53.E-mail address: [email protected] (M. Kaya).

0024-3205/$ - see front matter © 2005 Elsevier Inc. All rights reserved.doi:10.1016/j.lfs.2005.12.035

borne macromolecules such as proteins that do not cross theBBB under normal conditions (Westergaard et al., 1978; Nitschet al., 1986a,b; Kalayci et al., 2002). The endothelial cells ofbrain microvessels are sealed together by tight junctions thatrender the vascular wall impermeable to macromolecules(Brightman, 1989). On the other hand, the presence of fewpinocytotic vesicles in the endothelial cells of cerebral micro-vessels accounts for a minor transcellular penetration of blood-borne substances (Abbott and Revest, 1991). The two proposedmechanisms by which the BBB opens to macromolecules inepileptic seizure are enhancement of transcellular route byvesicular transport (Westergaard et al., 1978; Petrali et al.,1991) and opening of the tight junctions (Hedley-Whyte et

2 N. Arican et al. / Life Sciences 79 (2006) 1–7

al., 1977). Therefore, determination of the route of macromo-lecular passage across the BBB could be important in elucidat-ing the pathophysiology of epileptic seizures.

Recent studies indicated that lipopolysaccharide (LPS),major component of the gram-negative bacterial cell wall, exertstime- and dose-dependent effects on seizure threshold. It hasbeen shown that LPS treatment causes increased susceptibilityto clonic seizures which is likely mediated by nitric oxide (NO),and interestingly, LPS-induced neuroinflammation does notconstitute a predisposing factor for epileptogenesis (Sayyah etal., 2003). This finding may be in contrast to the clinical evi-dence indicating that CNS injuries and consequent neuroinflam-mation facilitate the occurrence of seizures (Temkin et al., 2001;Herman, 2002; Vezzani, 2005). Although previous studies pro-vided strong evidence about the role of LPS in causing BBBdisruption in vitro and in vivo through its concentration- andtime-dependent effects (Mayhan, 1998; Bickel et al., 1998;Gaillard et al., 2003), it has recently been shown that LPS exertsprotective effects on the endothelium and epithelium of a varietyof tissues (Riehl et al., 2004; Kaya et al., 2004). Moreover, it isextremely important to know that using different serotypes ofLPS and their different application routes could lead to promi-nent differences in the amount of tracer extravasation in variousanimal models (Bickel et al., 1998; Xaio et al., 2001; Nedreboand Reed, 2002; Singh and Jiang, 2004).

Although LPS has been successfully used to reduce ische-mic brain damage (Ahmed et al., 2000) and to prevent BBBdisruption during irradiation (Kaya et al., 2004) and hyperten-sion (Ahishali et al., 2005) in a variety of experimental models,the physiological basis for the use of large doses of LPS againstBBB disruption in epileptic seizures remains to be clarified. Itis also reported that the expression of glucose transporter-1, aspecific BBB marker protein, was reduced in the brain micro-vessels in complex partial seizures (Cornford et al., 1998),however, no experimental data are available yet on the expres-sion tight junction proteins, zonula occludens (ZO)-1 andoccludin, which represent the primary determinant of solutemovement via paracellular pathway through the BBB afterLPS pretreatment in pentylenetetrazole (PTZ)-induced epilepticseizures. On the basis of these data, the objective of this studywas to evaluate the influence of LPS on the functional and thestructural properties of BBB during epileptic seizures.

Materials and methods

Animals

The present study protocol was reviewed and approved bythe Committee on Ethics on Animal Experiments, ResearchInstitute for Experimental Medicine (Istanbul University, Istan-bul), and the experiments were conducted according to theGuidelines for Animal Experiments of this institute fromwhere adult male Wistar rats (230–320 g) used in this studywere obtained. Separate groups of 8 rats for each experimentalgroup have been used for evaluation of functional and structuralchanges with Evans blue (EB) dye extravasation and immuno-histochemistry, respectively.

Experimental protocol

LPS from Escherichia coli (serotype 026:B6, Sigma Chem-ical Company, St. Louis, MO) was injected to rats (5 mg/kg, i.p.)at 0, 6 and 12 h. Control rats were administered an equal volumeof sterile saline with the same intervals. Rats were anaesthetizedwith diethyl ether, and a polyethylene catheter (PE-50) wasplaced in the right femoral artery and vein for the measurementof systemic arterial blood pressure, blood sampling and theinjection of EB dye and PTZ. Lidocain was used as a localanesthetic on the operated side during the experiment. PTZ(80 mg/kg, i.v) was administered to induce epileptic seizuresat 5 min after EB dye injection.

Determination of Evans blue dye extravasation

To evaluate the BBB integrity, EB dye was used as a markerof albumin extravasation. Briefly, EB dye (2% in saline,4 ml/kg) was injected via the right femoral vein at 24 h afterthe first i.p. dose of LPS and allowed to remain in circulationfor 30 min. At the end of experiments, the chest was openedand rats were perfused transcardially with 250 ml of saline at apressure of 110 mm Hg for approximately 15 min until thefluid from the right atrium became colourless. After decapita-tion, the brain was removed and dissected into four regions;left and right cerebral cortex, diencephalon and cerebellum.Then, each brain region was weighed for quantitative mea-surement of EB-albumin extravasation (Kaya et al., 2004).Brain samples were homogenized in 2.5 ml phosphate-buff-ered saline and mixed by vortexing for 2 min after the additionof 2.5 ml of 60% trichloroacetic acid (TCA), to precipitate theprotein. Samples were cooled and then centrifuged for 30 minat 1000×g. The supernatant was measured at 610 nm forabsorbance of EB using a spectrophotometer. EB wasexpressed as μg/mg of brain tissue against a standard curve.

Biochemical parameters

To measure plasma biochemical parameters, blood sampleswere collected in heparinized vacutainer tubes and immediate-ly transported in a cooler with ice to the laboratory. Uponarrival at the laboratory, plasma was separated by centrifuga-tion (+4 °C, 1000×g, 10 min), divided into 0.5–1.0 ml ali-quots, placed in cryovials and stored at −70 °C until analyzed.

Malondialdehyde (MDA), as an end product of fatty acidperoxidation in rats, was measured in plasma by the thiobarbi-turic acid (TBA) reactivity assay. One volume of plasma wasmixed thoroughly with 2 volumes of a stock solution of 15% w/v TCA, 0.375% w/v TBA and 0.25 N hydrochloric acid. Themixture was heated for 30 min in a boiling water bath. Aftercooling, the flocculent precipitate was removed by centrifuga-tion at 1000×g for 10 min. TBA reactive substance concentra-tion was calculated using 1.56×10−5 M−1 cm−1 as mol/Lextinction coefficient (Beuge and Aust, 1978). Plasma IL-1and IL-6 levels of animals were determined by ELISA kits(Quantikine, R&D Systems Minneapolis, USA) according tothe manufacturer's instructions. Plasma superoxide dysmutase

0,000

0,002

0,004

0,006

0,008

0,010

0,012

0,014

0,016

0,018

Control LPS PTZ LPS+PTZ

Left Cerebral CortexRight Cerebral CortexDiencephalonCerebellum

ug E

B/m

g br

ain

tiss

ue

*

*

† †

Fig. 1. Effects of LPS and/or PTZ on EB dye extravasation in cerebral cortex,diencephalon and cerebellum regions of brain. †Pb0.01 versus control and LPSplus PTZ. *Pb0.01 versus other groups.

Table 1Effects of LPS and/or PTZ on mean arterial blood pressure in rats

Groups Mean arterial blood pressure (mm Hg)

n Before After treatment

Control 8 104±2.1 –LPS 8 – 96.9±2.5PTZ 8 98±2.5 167±4.6⁎

LPS+PTZ 8 113±2.7 169±5.1⁎

Values are mean±SEM. Wilcoxon Signed Ranks test was performed for differ-ences in data.⁎Pb0.01 versus before treatment. n: number of animals.

3N. Arican et al. / Life Sciences 79 (2006) 1–7

(SOD) activities were measured by inhibition of nitroblue tet-razolium (NBT) reduction with xanthine/xanthine oxidase usedas a superoxide generator. SOD was defined as the amount ofprotein that inhibits the rate of NBT reduction by 50% in rats.Catalase activity in plasma was measured by the breakdown ofhydrogen peroxide catalyzed by catalase and the decrease inabsorbance of hydrogen peroxide at 240 nm was determined.The rates were quantified at 30 °C using 10 mM hydrogenperoxide. The changes in nitric oxide (NO) concentrationswere measured as its stable metabolites nitrate and nitrite.Nitrate was first reduced by nitrate reductase to nitrite andthen nitrite was determined spectrophotometrically by theGriess reaction in rats. Red-violet diazo dye yielded byGriess reagent was measured in the visible range at 540 nmby using a commercially available colorimetric assay (Roche,Cat. #1746081).

Immunohistochemistry

To demonstrate the ZO-1, occludin and glial fibrillary acidicprotein (GFAP) expressions, we used the immunostaining meth-od. For this purpose, rats were anaesthetized with diethyl etherand perfused transcardially with isotonic saline at a pressure of110 mm Hg for 15 s, which was followed by 200 ml fixative(4% paraformaldehyde in phosphate buffer; pH: 7.4) for 10 min.After the perfusion, brains were immersed in the same fixative,kept for 24 h at 4 °C, and embedded in paraffin. 5-μm-thicksections were deparaffinized and pressure cooked in citratebuffer (10 mM, pH: 6) for 2 min for GFAP or incubated withprotease (1 mg/ml; Sigma Inc., USA) for 10 min for ZO-1 andoccludin for antigen retrieval. Endogenous peroxidase activitywas quenched with 0.3% hydrogen peroxide in phosphate buff-ered saline for 20 min. A nonspecific blocking reagent (Ultra-V-Block Lab Vision Co., Westinghouse, CA) was used to preventnonspecific bindings. Polyclonal rabbit anti-occludin (Zymed

Table 2Effects of LPS and/or PTZ on plasma NO, MDA, SOD, catalase, IL-1 and IL-6 lev

Groups n NO (μmol/L) MDA (μmol/L) SOD (IU

Control 8 19.2±0.5 1.1±0.1 25.0±0.LPS 8 52.7±2.1⁎ 4.0±0.5⁎ 18.1±0.PTZ 8 39.7±1.2⁎ 3.9±0.1⁎ 19.9±0.LPS+PTZ 8 52.6±2.2⁎,† 4.0±0.1⁎ 19.3±0.

Values are mean±SEM. Tukey HSD was performed for differences in data.†Pb0.01 versus PTZ. ⁎Pb0.001 versus control values. n: number of animals.

Lab Inc., CA, 1:50, 2 h), polyclonal rabbit anti-ZO-1 (ZymedLab Inc., CA, 1:50, overnight), and monoclonal mouse anti-GFAP (Neomarker; Fremont, CA; 1:100, 60 min) were used asprimary antibodies. Secondary antibodies were used as biotiny-lated goat anti-polyvalent for ZO-1 and occludin and biotiny-lated goat anti-mouse (Lab Vision Co., Westinghouse, CA) forGFAP. After washing, peroxidase conjugated streptavidin (LabVision Co., Westinghouse, CA) was applied and aminoethylcarbazole chromogen was used. The sections were counter-stained with Mayer Hematoxylin to enhance the nuclear stain-ing. For negative controls, adjacent sections were processedwith the same steps with the exception of the primary antibo-dies. Images were obtained from hippocampal area by means ofa digital camera (Nikon, Coolpix 4500) attached to lightmicroscope.

Statistical analysis

All data were expressed as mean±S.E.M. One-way analysisof variance (ANOVA), Tukey and Wilcoxon's signed rank testscompared statistical analysis of differences. The statistical anal-ysis was made with SPSS (Statistical Package for SocialSciences 10.0 ver.). A level of Pb0.05 was considered statisti-cally significant.

Results

The mean arterial blood pressure (MABP) of LPS-treatedrats was 96.9±2.5 mm Hg (Table 1). The MABP in PTZ-treatedrats was observed to elevate from 98±2.5 to 167±4.6 mm Hg(Pb0.01). Similarly, the MABP in animals treated with the LPS

els

/mL) Catalase (IU/mL) IL-1 (IU/mL) IL-6 (IU/mL)

77 168.9±3.5 390.0±8.3 172.6±7.09⁎ 156.6±2.8⁎ 1160.3±29.8⁎ 1274.4±16.3⁎

8⁎ 150.4±1.9⁎ 1119.1±24.4⁎ 1149.0±35.3⁎

4⁎ 150.2±3.9⁎ 1097.8±26.3⁎ 1314.8±12.0⁎,†

Fig. 2. Photomicrographs of the hippocampus showing ZO-1 immunoreactivity in the microvessels of saline (A), LPS (B), PTZ (C), LPS plus PTZ (D) treated rats.Note that there are no alterations in ZO-1 immunoreactivity in the endothelial cells of brain microvessels. Arrows indicate positive immunoreactivity for ZO-1 overthe microvessels. Scale bars=10 μm.

4 N. Arican et al. / Life Sciences 79 (2006) 1–7

plus PTZ elevated from 113±2.7 to 169±5.1 mm Hg (Pb0.01)(Table 1).

LPS and/or PTZ significantly increased NO, MDA, IL-1 andIL-6 concentrations in plasma (Pb0.01, Table 2). However,plasma concentrations of NO and IL-6 in animals treated withLPS plus PTZ were found to be higher than those of PTZ(Pb0.05, Table 2). In contrast, plasma SOD and catalase levelsreduced significantly in LPS, PTZ and LPS plus PTZ-treatedanimals (Pb0.01, Table 2).

The content of EB dye in left cerebral cortex, right cerebralcortex, diencephalon and cerebellum regions of LPS-treatedanimals (0.0020±0.0002, 0.0019±0.0001, 0.0021±0.0002,0.0025±0.0002 μg/mg tissue, respectively) slightly increasedover control values (0.0016±0.0002, 0.0017±0.0001, 0.0018±0.0002 and 0.0020±0.0003, respectively, Fig. 1). PTZ-inducedepileptic seizures increased the EB dye extravasation intothese brain regions (0.0035±0.0005, 0.0040±0.0004, 0.014±0.0024, 0.011±0.0015 μg/mg tissue, respectively; Pb0.05,Pb0.01, Fig. 1). Under LPS plus PTZ, EB dye contentdecreased in the brain regions (0.0010±0.0002, 0.0010±0.0003, 0.0012±0.0003 and 0.0013±0.0002 μg/mg tissue,respectively; Pb0.05, Pb0.01, Fig. 2).

In the normal brain, a similar immunostaining pattern wasobserved for ZO-1 and occludin along the endothelial cells innearly all the parenchymal microvessels (Figs. 2A and 3A).Following LPS, the immunoreactivity for both tight junctionproteins was increased on the microvessel walls in brain sec-tions (Figs. 2B and 3B). PTZ and LPS plus PTZ did not cause a

Fig. 3. Photomicrographs of the hippocampus showing occludin immunoreactivity inNote that there are no alterations in occludin immunoreactivity in the endothelial cellover the microvessels. Scale bars=10 μm.

marked alteration in ZO-1 and occludin immunoreactivity(Figs. 2C,D and 3C,D). An increase in GFAP immunoreactivitywas seen in the astrocytes of LPS-treated animals (Fig. 4B),while PTZ and LPS plus PTZ did not lead to a change in theimmunoreactivity of GFAP (Fig. 4C and D).

Discussion

The present study provides new evidence that LPS coulddecrease seizure-related entry of macromolecules from systemiccirculation into brain by maintaining BBB integrity presumablythrough reducing transendothelial transport, since tight junctionproteins, ZO-1 and occludin, remained unaffected under epilep-tic seizures with or without LPS pretreatment. To our knowledgethis is the first study to evaluate the immunohistochemicalprofile of BBB for tight junction proteins and GFAP in thisexperimental setting. The study revealed that there was a sub-stantial elevation in blood pressure along with an increase inBBB permeability under PTZ, although such concurrentchanges in BBB integrity and blood pressure has not beenobserved in LPS plus PTZ. We suggest that the changes inexpression status of tight junction proteins could not provide asubcellular mechanism for the epileptic seizure-induced vascu-lar leakage of macromolecules into brain, and implicate thestimulation of transcellular pathway rather than tight junctiondisruption for the pathogenesis of increased BBB permeabilityin epileptic seizures. Our assumption is supported by a previousstudy indicating that the elevated blood pressure was a primary

the microvessels of saline (A), LPS (B), PTZ (C), LPS plus PTZ (D) treated rats.s of brain microvessels. Arrows indicate positive immunoreactivity for occludin

Fig. 4. Photomicrographs showing GFAP immunoreactivity in the hippocampal area of brain sections in saline (A), LPS (B), PTZ (C), LPS plus PTZ (D) treated rats.Note the increased GFAP immunoreactivity in brain sections of animals treated with LPS (B). Scale bars=25 μm.

5N. Arican et al. / Life Sciences 79 (2006) 1–7

stimulus for the enhancement in pinocytotic activity in cerebralendothelial cells during stroke (Cipolla et al., 2004).

Our results are in well accordance with those previouslyestablished by ultrastructural and biochemical studies indicat-ing that tight junctions did not appear to be affected in seizure-related BBB disruption (Carpentier et al., 1990; Petrali et al.,1991; Grange-Messent et al., 1999). On the other hand, in-creased transcellular transport activity has been demonstratedby a number of studies during epileptic seizures (Westergaard etal., 1978; Nitsch et al., 1986b). The data mentioned above seemto favor vesicular transport rather than opening of endothelialtight junctions in the pathophysiology of BBB disruption inepileptic seizures. Our data demonstrated no evidence of directrelation between tight junction proteins and BBB disruptionafter PTZ-induced epileptic seizures. In accordance with ourresults, endotoxemia has been shown to cause brain edemawithout apparently damaging the tight junctions (Clawson etal., 1966; Papadopoulos et al., 1999). It is logical to concludethat if the tight junctions remain functionally intact, then theother route i.e. transcellular pathway must be involved in BBBdisruption. Therefore we suggest that the transcellular pathwaycould be activated by PTZ and LPS could influence the trans-cellular pathway by decreasing the passage of EB dye to thebrain paranchyma.

The cellular and biochemical mechanism(s) of BBB altera-tions in LPS pretreated epileptic seizures are presently unclear.LPS has been found to disrupt the BBB in vitro and in vivo (deVries et al., 1996; Xaio et al., 2001; Singh and Jiang, 2004). Incontrast, our results showed that i.p. injections of LPS did notsignificantly increase the BBB permeability to EB dye. Thesefindings are consistent with recent data indicating that i.p.injection of endotoxin even at a high dose does not acutelydisrupt the BBB (Bickel et al., 1998; Kaya et al., 2004). Ourdata on ZO-1 and occludin immunohistochemistry indicate thattight junction proteins do not seem to be a major mechanisticcomponent of seizure-induced BBB disruption. Contrary to ourresults, it is reported that claudin-8 is specifically and progres-sively down regulated in the hippocampus and proposed thatepileptic seizures induced changes in local tight junction com-position (Lamas et al., 2002). Therefore, it is likely that theBBB protection against PTZ-induced epileptic seizures by LPScould be the net outcome of its effects on the several potentialpathways that influence endothelial cell function and structurein epileptic seizures.

Contrary to our expectations that LPS could enhance thedegree of BBB disruption during epileptic seizures, we foundthat BBB permeability to EB dye decreased during epilepticseizures following LPS pretreatment. It is reported that theconvulsions induced by a combination of convulsant drugsand LPS did not lead to pathological changes within braintissue (Heida et al., 2004). This finding is in agreement withour results regarding the effects of epileptic seizures on thestructural and functional properties of BBB following LPS.On the other hand, NO and IL-6 are among the inflammatorymediators implicated in the pathogenesis of seizures (Del-Bel etal., 1997; Jelenkovic et al., 2002). In the present study, weobserved alterations in the activity of antioxidant and oxidantstatus during LPS, PTZ and LPS plus PTZ. Oxidative stressmay play an important role in the pathophysiology of epilepticseizure-induced neuronal damage (Erakovic et al., 2000; Frant-seva et al., 2000) and BBB disruption is enhanced by indirecteffects through free radicals and other reactive species duringepileptic seizures (Oztas et al., 2001; Kalayci et al., 2002).Recent data suggest that IL-6 is a molecule with both beneficialand destructive potentials in central nervous system and isinvolved in angiogenesis, induces BBB properties and main-tains the integrity of the BBB (Gadient and Otten, 1997; Ped-chenko and LeVine, 1999; Ali et al., 2000; Krizanac-Bengez etal., 2003). NO and IL-6 are also reported to be involved in themaintenance of BBB integrity after normoxic-normoglycemicflow cessation (Krizanac-Bengez et al., 2003). The presentstudy provides evidence for the involvement of NO and IL-6pathways in LPS-induced BBB protection during epileptic sei-zures and the presence of increased NO and IL-6 levels follow-ing LPS injection could be related, at least partly, to thedecrease in BBB permeability to EB dye in PTZ-induced epi-leptic seizures.

Astrocytes are particularly important glial components of theBBB and are essential for the BBB integrity (Janzer and Raff,1987; Liedtke et al., 1996; Abbott, 2002; Kaya et al., 2004;Kalayci et al., 2005). It is reported that in stab spinal cord injuries,BBB is damaged and has failed to repair in the absence of reactiveastrocytes (Faulkner et al., 2004). Moreover, astrocytes increaseblood-retinal barrier properties at least in part by increasing ZO-1tight junction protein expression (Gardner et al., 1997). Tio et al.published data showing that astrocytes can also induce tight-junction formation in primary HUVEC cultures (Tio et al.,1990). From our data, it can be concluded that the beneficial

6 N. Arican et al. / Life Sciences 79 (2006) 1–7

response on BBB integrity induced by LPS pretreatment in PTZ-induced epileptic seizure may involve upregulation of BBB prop-erties and GFAP activity via IL-6 production.

In conclusion, our data on immunohistochemical profiles ofZO-1 and occludin indicate that tight junction proteins do notseem to be a major mechanistic component of seizure-inducedBBB disruption. It is likely that the BBB protection againstPTZ-induced seizures by LPS could be the net outcome of itseffects on the several potential pathways that influence endo-thelial cell function and structure in seizures. Therefore, it isreasonable to hypothesize that LPS could have beneficial effectson the endothelial cells of brain microvessels and act throughtranscellular pathway to maintain BBB integrity by increasingNO and IL-6 levels in PTZ-induced epileptic seizures. However,the exact mechanism by which LPS may act on transcellularpathway during epileptic seizures remains to be elucidated.

Acknowledgement

This work was supported by the Research Fund of theUniversity of Istanbul (Project No. UDP-489/24052005).

References

Abbott, N.J., 2002. Astrocyte-endothelial interactions and blood–brain barrierpermeability. Journal of Anatomy 200 (6), 629–638.

Abbott, N.J., Revest, P.A., 1991. Control of brain endothelial permeability.Cerebrovascular Brain Metabolism Review 3 (1), 39–72.

Ahishali, B., Kaya, M., Kalayci, R., Uzun, H., Bilgic, B., Arican, N., Elmas,I., Aydýn, S., Kucuk, M., 2005. Effects of lipopolysaccharide on theblood–brain barrier permeability in prolonged nitric oxide blockade-induced hypertensive rats. International Journal of Neuroscience 115 (2),151–168.

Ahmed, S.H., He, Y.Y., Nassief, A., Xu, J., Xu, X.M., Hsu, C.Y., Faraci, F.M.,2000. Effects of lipopolysaccharide priming on acute ischemic brain injury.Stroke 31 (1), 193–199.

Ali, C., Nicole, O., Docagne, F., Lesne, S., MacKenzie, E.T., Nouvelot, A.,Buisson, A., Vivien, D., 2000. Ischemia-induced interleukin-6 as a potentialendogenous neuroprotective cytokine against NMDA receptor-mediatedexcitotoxicity in the brain. Journal of Cerebral Blood Flow Metabolism 20,956–966.

Beuge, J.A., Aust, S.D., 1978. Microsomal lipid peroxidation. MethodEnzymology 53, 302–310.

Bickel, U., Grave, B., Kang, Y.S., del Rey, A., Voigt, K., 1998. No increase inblood–brain barrier permeability after intraperitoneal injection of endotoxinin the rat. Journal of Neuroimmunology 85 (2), 131–136.

Brightman, M.W., 1989. The anatomical basis of blood–brain barrier. In:Neuwelt, E.A. (Ed.), Implications of the Blood–Brain Barrier and itsManipulation. Plenum Medical Book Company, New York, p. 53.

Carpentier, P., Delamanche, I.S., Le Bert, M., Blanchet, G., Bouchaud, C., 1990.Seizure-related opening of the blood–brain barrier induced by soman:possible correlation with the acute neuropathology observed in poisonedrats. Neurotoxicology 11 (3), 493–508.

Cipolla, M.J., Crete, R., Vitullo, L., Rix, R.D., 2004. Transcellular transport as amechanism of blood–brain barrier disruption during stroke. Frontiers inBioscience 9, 777–785.

Clawson, C.C., Hartmann, J.F., Vernier, R.L., 1966. Electron microscopy of theeffect of gram negative endotoxin on the blood–brain barrier. JournalCompared Neurology 127, 183–198.

Cornford, E.M., Hyman, S., Cornford, M.E., Landaw, E.M., Delgado-Escueta,A.V., 1998. Interictal seizure resections show two configurations ofendothelial Glut1 glucose transporter in the human blood–brain barrier.Journal of Cerebral Blood Flow and Metabolism 18 (1), 26–42.

de Vries, H.E., Blom-Roosemalen, M.C., de Boer, A.G., van Berkel, T.J.,Breimer, D.D., Kuiper, J., 1996. Effect of endotoxin on permeability ofbovine cerebral endothelial cell layers in vitro. Journal PharmacologyExperimental Therapeutics 277, 1418–1423.

Del-Bel, E.A., Oliveira, P.R., Oliveira, J.A., Mishra, P.K., Jobe, P.C., Garcia-Cairasco, N., 1997. Anticonvulsant and proconvulsant roles of nitric oxidein experimental epilepsy models. Brazilian Journal of Medical andBiological Research 30, 971–979.

Erakovic, V., Zupan, G., Varljen, J., Radosevic, S., Simonic, A., 2000.Electroconvulsive shock in rats: changes in superoxide dismutase andglutathione peroxidase activity. Brain Research Molecular Brain Research76 (2), 266–274.

Faulkner, J.R., Herrmann, J.E., Woo, M.J., Tansey, K.E., Doan, N.B., Sofroniew,M.V., 2004. Reactive astrocytes protect tissue and preserve function afterspinal cord injury. The Journal of Neuroscience 24, 2143–2155.

Frantseva, M.V., Velazquez, J.L., Hwang, P.A., Carlen, P.L., 2000. Free radicalproduction correlates with cell death in an in vitro model of epilepsy. TheEuropean Journal of Neuroscience 12 (4), 1431–1439.

Gadient, R.A., Otten, U.H., 1997. Interleukin-6—a molecule with bothbeneficial and destructive potentials. Progress in Neurobiology 52,379–390.

Gaillard, P.J., de Boer, A.B., Breimer, D.D., 2003. Pharmacological investiga-tions on lipopolysaccharide-induced permeability changes in the blood–brain barrier in vitro. Microvascular Research 65 (1), 24–31.

Gardner, T.W., Lieth, E., Khin, S.A., Barber, A.J., Bonsall, D.J., Lesher, T.,Rice, K., Brennan Jr., W.A., 1997. Astrocytes increase barrier properties andZO-1 expression in retinal vascular endothelial cells. InvestigativeOphthalmology and Visual Science 38 (11), 2423–2427.

Grange-Messent, V., Bouchaud, C., Jamme, M., Lallement, G., Foquin, A.,Carpentier, P., 1999. Seizure-related opening of the blood–brain barrierproduced by the anticholinesterase compound, soman: new ultrastructuralobservations. Cellular and Molecular Biology (Noisy-le-Grand, France)45 (1), 1–14.

Hashizume, K., Tanaka, T., Yonemasu, Y., 1998. Change in cerebral glucosemetabolism during limbic seizures elicited from lateral septal nucleus.Epilepsy Research 30 (3), 167–176.

Hedley-Whyte, E.T., Lorenzo, A.V., Hsu, D.W., 1977. Protein transport acrosscerebral vessels during metrazole-induced convulsions. American Journal ofPhysiology 233 (3), C74–C85.

Heida, J.G., Boisse, L., Pittman, Q.J., 2004. Lipopolysaccharide-induced febrileconvulsions in the rat: short-term sequelae. Epilepsia 45, 1317–1329.

Herman, S.T., 2002. Epilepsy after brain insult: targeting epileptogenesis.Neurology 59 (9 Suppl.5), S21–S26.

Holmes, G.L., 2002. Seizure-induced neuronal injury: animal data. Neurology59 (9 Suppl. 5), S3–S6.

Janzer, R.C., Raff, M.C., 1987. Astrocytes induce blood–brain barrier propertiesin endothelial cells. Nature 325, 253–257.

Jelenkovic, A., Jovanovic, M., Ninkovic, M., Maksimovic, M., Bokonjic, D.,Boskovic, B., 2002. Nitric oxide (NO) and convulsions induced bypentylenetetrazol. Annals of the New York Academy of Sciences 962,296–305.

Kalayci, R., Kaya, M., Cimen, V., Kucuk, M., Gurses, C., Arican, N., Elmas, I.,2002. Catalase and alpha-tocopherol attenuate blood–brain barrier break-down in pentylenetetrazole-induced epileptic seizures in acute hypergly-caemic rats. Pharmacological Research 45 (2), 129–133.

Kalayci, R., Kaya, M., Elmas, I., Arican, N., Ahishali, B., Uzun, H., Bilgic, B.,Kucuk, M., Kudat, H., 2005. Effects of atorvastatin on blood–brain barrierpermeability during L-NAME hypertension followed by angiotensin-II inrats. Brain Research 1042 (2), 184–193.

Kaya, M., Palanduz, A., Kalayci, R., Kemikler, G., Simsek, G., Bilgic, B.,Ahishali, B., Arican, N., Kocyildiz, Z.C., Elmas, I., Kucuk, M., Karadeniz,A., 2004. Effects of lipopolysaccharide on the radiation-induced changes inthe blood–brain barrier and the astrocytes. Brain Research 1019 (1–2),105–112.

Krizanac-Bengez, L., Kapural, M., Parkinson, F., Cucullo, L., Hossain, M.,Mayberg, M.R., Janigro, D., 2003. Effects of transient loss of shear stress onblood–brain barrier endothelium: role of nitric oxide and IL-6. BrainResearch 977, 239–246.

7N. Arican et al. / Life Sciences 79 (2006) 1–7

Lamas, M., Gonzalez-Mariscal, L., Gutierrez, R., 2002. Presence of claudinsmRNA in the brain. Selective modulation of expression by kindlingepilepsy. Brain Research. Molecular Brain Research 104 (2), 250–254.

Liedtke, W., Edelmann, W., Bieri, P.L., Chiu, F.C., Cowan, N.J., Kucherlapati,R., Raine, C.S., 1996. GFAP is necessary for the integrity of CNS whitematter architecture and long-term maintenance of myelination. Neuron 17(4), 607–615.

Mayhan, W.G., 1998. Effect of lipopolysaccharide on the permeability andreactivity of the cerebral microcirculation: role of inducible nitric oxidesynthase. Brain Research 792, 353–357.

Nedrebo, T., Reed, R.K., 2002. Different serotypes of endotoxin (lipopoly-saccharide) cause different increases in albumin extravasation in rats.Shock 18 (2), 138–141.

Nitsch, C., Goping, G., Laursen, H., Klatzo, I., 1986a. The blood–brain barrierto horseradish peroxidase at the onset of bicuculline-induced seizures inhypothalamus, pallidum, hippocampus, and other selected regions of therabbit. Acta Neuropathology (Berl.) 69 (1–2), 1–16.

Nitsch, C., Goping, G., Klatzo, I., 1986b. Pathophysiological aspects of blood–brain barrier permeability in epileptic seizures. Advances in ExperimentalMedicine and Biology 203, 175–189.

Oztas, B., Kilic, S., Dural, E., Ispir, T., 2001. Influence of antioxidants on theblood–brain barrier permeability during epileptic seizures. Journal ofNeuroscience Research 66 (4), 674–678.

Papadopoulos, M.C., Lamb, F.J., Moss, R.F., Davies, D.C., Tighe, D., Bennett,E.D., S., 1999. Faecal peritonitis causes oedema and neuronal injury in pigcerebral cortex. Clinical Science 46, 461466.

Pedchenko, T.V., LeVine, S.M., 1999. IL-6 deficiency causes enhancedpathology in Twitcher (globoid cell leukodystrophy) mice. ExperimentalNeurology 158, 459–468.

Petrali, J.P., Maxwell, D.M., Lenz, D.E., Mills, K.R., 1991. Effect of ananticholinesterase compound on the ultrastructure and function of the ratblood–brain barrier: a review and experiment. Journal SubmicroscopicCytology and Pathology 23 (2), 331–338.

Riehl, T.E., Newberry, R.D., Lorenz, R.G., Stenson, W.F., 2004. TNFR1mediates the radioprotective effects of lipopolysaccharide in the mouseintestine. American Journal of Physiology. Gastrointestinal and LiverPhysiology 286 (1), G166–G173.

Sayyah, M., Najafabadi, I.T., Beheshti, S., Majzoob, S., 2003. Lipopolysac-charide retards development of amygdala kindling but does not affect fully-kindled seizures in rats. Epilepsy Research 57 (2–3), 175–180.

Singh, A.K., Jiang, Y., 2004. How does peripheral lipopolysaccharide inducegene expression in the brain of rats? Toxicology 201 (1–3), 197–207.

Temkin, N.R., Jarell, A.D., Anderson, G.D., 2001. Antiepileptogenic agents:how close are we? Drugs 61 (8), 1045–1055.

Tio, S., Deenen, M., Marani, E., 1990. Astrocyte-mediated induction of alkalinephosphatase activity in human umbilical cord vein endothelium: an in vitromodel. European Journal of Morphology 28, 289–300.

Vezzani, A., 2005. Inflammation and epilepsy. Epilepsy Current 5 (1), 1–6.Westergaard, E., Hertz, M.M., Bolwig, T.G., 1978. Increased permeability to

horseradish peroxidase across cerebral vessels, evoked by electricallyinduced seizures in the rat. Acta Neuropathology (Berl.) 41 (1), 73–80.

Xaio, H., Banks, W.A., Niehoff, M.L., Morley, J.E., 2001. Effect of LPS on thepermeability of the blood–brain barrier to insulin. Brain Research 896 (1–2),36–42.