B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

Kainic acid and 3-Nitropropionic acid induced expression oflaminin in vascular elements of the rat brain

Sumit Sarkar, Larry Schmued⁎

Division of Neurotoxicology, National Center for Toxicological Research (NCTR), Jefferson, AR-72079, USA

A R T I C L E I N F O

0006-8993/$ – see front matter. Published bydoi:10.1016/j.brainres.2010.07.011

A B S T R A C T

Article history:Accepted 5 July 2010Available online 5 August 2010

Laminin is a glycoprotein component of the basement membrane and has been reported tobe found in different areas of the nervous system including brain endothelial cells, Schwanncells and peripheral nerves. Although the in-vitro studies suggest that laminin plays animportant role in growth and neurite extension of cultured neurons, localization of lamininin the brain has been controversial and inconsistent results have been reported. Recently,laminin immunoreactivity has been used as a marker for vascular elements in the brain. Inthis study, we have investigated the effect of two mechanistically different neurotoxins,kainic acid (KA), an NMDA agonist and 3-Nitropropionic acid (3-NPA), an inhibitor ofmitochondrial respiration, on brain vascular elements revealed by laminin immunolabeling.We also explored whether administration of these two neurotoxic drugs correlate with theneuronal degeneration observed after neurotoxic insult by staining with Fluoro-Jade C dye.We have employed single immunolabeling to localize laminin in the brains. In KA treatedrats, most of the laminin immunoreactivity is present in the piriform cortex, corpuscallosum (myelinated tracts) amygdala, hippocampus, ventral thalamus and tenia tacta. In3-NPA treated animals, laminin immunoreactivity was confined mostly to the striatum. Incontrast, saline treated rats showed very little laminin immunolabeling around capillaries,arteries and in themeningeal membranes. To determine the effects of these neurotoxins onthe integrity of the blood brain barrier (BBB), endothelial brain barrier antigen (EBA)immunolabeling was also performed. In addition, we performed CD11b immunolabeling toevaluate the effect of 3-NPA and KA on the activation of microglia in the brain. CD11b wasdramatically increased in KA and 3-NPA treated animals. We have also combined lamininimmunolabeling with Fluoro-Jade C labeling to evaluate the spatio-temporal association ofdegenerating neurons and the expression of laminin containing microvessels. Areas whichshowed intense laminin immunolabeling following KA or 3-NPA exposure correlated withthose exhibiting the greatest number of degenerating neurons observed after Fluoro-Jade Cstaining. EBA-laminin double immunolabeling demonstrated that the expressions oflaminin were predominantly localized in the areas (cortex, thalamus and hippocampus)where EBA has been either reduced or is absent. Our results from these experimentsdemonstrate that vascular laminin expression increases after treatment with KA or 3-NPA,suggesting the occurrence of neovascularization. Microglia may also contribute to theneurotoxic induced neovascularization and neurodegeneration.

Published by Elsevier B.V.

Keywords:LamininKainic acid3-NPAFluoro-Jade CNeurodegeneration

⁎ Corresponding author.E-mail address: lschmued@nctr.fda.gov (L. Schmued).

Elsevier B.V.

240 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

1. Introduction

Laminin isan important cell adhesionmolecule that constitutesafamily of glycoproteins found in the basement membrane.Structurally, laminins are cross and T shaped hetero-trimersconsisting of α, β and γ chains (Colognato and Yurchenco, 2000).Laminins have tissue specific distribution and this differentialexpression is determined by variations in the expression of the αchain of the hetero-trimer. Laminins play various roles inphysiological function (Miner, 2008). Laminin is also essentialfor embryonic development and organogenesis and has crucialfunctions in several tissues includingmuscle, nerve, skin, kidney,lung and the vasculature (Miner, 2008). Laminin is considered tobe an important component of the basement membrane withcrucial roles in development and disease (Durbeej, 2009).Although recent in-vitro studies suggest that laminin plays animportant role in growth and neurite extension in culturedneurons, localization of laminin in the brain has been controver-sial and inconsistent results have been reported. Thismaybedueto the use of non-specific antibodies, non-specific staining andvariations in fixation conditions. Recently, laminin immunoreac-tivityhasbeenusedasamarker forvascularelements in thebrainand has been localized in the basement membranes associatedwith themeningial vasculature that surroundthecentralnervoussystem and also maintains the BBB (Patton, 2000). Laminin alsoplays an important role in angiogenesis (Form et al., 1986).Increased laminin immunoreactivity has been demonstrated innewly formed vessels in developing or embryonic tissues(Ekblom, 1981; Foidart et al., 1980), in neoplasms (Bohling et al.,1983; Giordana et al., 1985), and recently observed in neuraltransplantations (Shigematsu et al., 1989a). In the present study,we aimed to investigate the effect of two mechanisticallydifferent neurotoxins, kainic acid (KA), an excitotoxin, and 3-Nitropropionic acid (3-NPA), an inhibitor of mitochondrialrespiration, on blood vessels as revealed by laminin immunola-beling. We also explored whether administration of these twoneurotoxic drugs correlate with the neuronal degenerationobserved after neurotoxic insult revealed by staining withFluoro-Jade C. Here we also addressed whether changes in theBBB permeability induced by neurotoxins are caused solely bytight junctionmodifications or concomitant changes in endothe-lial cells and basal lamina. Therefore, we have assessed theintegrity of the BBB by EBA immunolabeling combined withlaminin immunolabeling. EBA is a specific rat antigen localized inendothelial cells of the blood vessels in which the BBB is intact(Sternberger and Sternberger, 1987; Sternberger et al., 1989). EBAwasused to characterize the endothelial cells after KA and 3-NPAtreatment. We also performed CD11b and laminin doubleimmunolabeling to explore the activation of microglia in thesame regions of the brain which expresses laminin followinginsult with either of the two neurotoxins studied.

2. Results

2.1. Behavioral changes observed after KA injection

KA induced neurotoxicity is also associated with well docu-mented behavioral changes (Racine, 1972). Within 1 h after an

injection of KA, animals exhibited behavioral changes such asgrooming, rearing, hind limb scratching, wet dog shakes, jawmovements, salivation, urination, defecation, and head nod-ding. Hyperthermia was also observed in these animals(>40 °C).

2.2. Behavioral changes observed after 3-NPA injection

After 3-NPA administration, rats were separated and theirgeneral behavior was observed. Little effect on behavior wasobserved after the initial administration of 3-NPA. However,after the second dose about 80% of the animals exhibitedabnormal gait within 1 h of the injection. Some animals alsoexhibited other motor symptoms including paddling, rolling,tremor, recumbence and somnolence. 3-NPA administered onday 4 resulted in more severe behavior such as tremor,reduced locomotion and abnormal gait. Hypothermia wasprominent in these 3-NPA treated animals (35 °C–36 °C).

2.3. Fluoro-Jade C labeling

The saline treated animals did not show any Fluoro-Jade Clabeling in the brain (Figs. 1A, D, G and J). The excitotoxin kainicacid resulted in extensive Fluoro-Jade C labeling within thebrain, consistent with the pattern previously described bySchmued and Hopkins (2000) using Fluoro-Jade B. The mostdense labeling was observed throughout the pyramidal cells ofthe hippocampus and piriform cortex (Figs. 1H and I). Otherstructureswhich also showeddense labeling include themedialthalamus (Figs. 1K and L), septumand the central nucleus of theamygdala. Fluoro-Jade C labeling was also observed in neo-cortex regions especially in pyramidal cells of layers II, III, andV.In contrast, injection of 3-NPA resulted in extensive degenera-tion throughout the basal ganglia (Figs. 1B and C) andoccasionally involved the hippocampus as well (Figs. 1E andF). This toxin induces a necrotic type lesion characterized by thestaining of virtually all neurons and neuropil within a specificstructure or substructures (Schmued andHopkins, 2000), exceptfor the penetrating myelinated fasicles.

2.4. Combined Fluoro-Jade C labeling and lamininimmunohistochemistry

In the saline treated animals, faint laminin immunolabelingwas confined to themeningealmembranes and the occasionallarge capillary (Figs. 1A, D,G and J). In KA treated rats, lamininlabeling was extensively observed in the tenia tecta, hippo-campus and cortex, (Figs. 1H and I) correlating with wheremost of FJC positive degenerating neurons were observed.Similarly, in 3-NPA injected animals, degenerating neuronswere observed mostly in the striatum and in some cases, inthe hippocampus using FJC staining (Figs. 1B and C). The sameareas showed extensive laminin immunolabeling. Brainregions with FJC stained degenerating neurons and vascularlaminin expression were highly correlated.

2.5. EBA and laminin double immunolabeling

In this study we employed endothelial brain barrier antigenlabeling which is commonly known as EBA, to determine the

Fig. 1 – Photomicrographs of representative Fluoro-Jade C stained and laminin immunolabeled tissue following 3-NPA or kainicacid or saline control treatment. In saline treated rats, limited laminin immunoreactivity was occasionally observed around thelarge capillaries (A, D, G and J), white arrow indicates the presence of laminin around a large capillary. In 3-NPA treated animalscombined laminin and FJC labeling reveals the relative localization of both labels in the striatum (B and C) and hippocampus(E and F). In kainic acid treated rats, combining laminin and FJC labelings reveal that the microvessels stained with laminin arealso present in the same location as the FJC containing degenerating neurons observed in the cortex (H and I) and thalamus(K and L). CTX: cortex, S: striatum. Scale bar=100 μm for (A, B, D, E, G, H, J and K); scale bar=200 μm for Figs. C, F, I, and L.

241B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

integrity of the blood brain barrier after neurotoxic insult. EBA isalso a specific marker for normal CNS microvessels. In normalsaline treated animals, EBA immunolabeled vascular structureswereubiquitouslypresent in thebrain. In3-NPAtreatedanimals,

EBA is absent only in the lesioned areas such as in the striatum(Fig. 2B) and in some cases in the hippocampus. In KA injectedanimals, EBA immunolabeling was dramatically reduced in thecortex, thalamus and hippocampus (Figs. 2E, H and K). EBA and

Fig. 2 – Photomicrographs of sections stained with laminin and EBA from animals treated with 3-NPA or kainic acid. (A) Showsintense EBA staining in the entire striatum of control animals. After 3-NPA treatment (B) demonstrates very faint EBA labelingas compared to adjacent non-inured regions of the striatum (S). In (C), the upregulation of laminin can be seen in the sameregion after 3-NPA exposure. In kainic acid treated animals, very few EBA labeled microvessels are present in the thalamus (E)and cortex (H) and hippocampus (L). This was in contrast to controls (D, G and J). Conspicuous laminin immunolabeling wasapparent in the thalamus (F), cortex (I) and hippocampus (L) after KA insult. Scale bar=100 μm.

242 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

laminin double immunolabeling revealed that in 3-NPA injectedrats, lamininwas overexpressed in the striatum and hippocam-pus, with concomitant reduction of EBA labeledmicrovessels. InKA injected rats, laminin was dramatically increased in thecortex, tenia tecta, hippocampus, and in the thalamus (Figs. 2F, Iand L), the same regions that also expressed reduced EBAimmunolabeling.

2.6. CD11b and laminin immunolabeling following3-NPA and KA injection

In the saline treated animals, very littlemicroglial activationwasobserved in the brain andmost of themicroglia are in the restinginactivatedstate (Figs. 3A,D,Gand J). In 3-NPAtreated rats, CD11bcontaining microglia are mostly present in the striatum (Figs. 3B

Fig. 3 – Photomicrographs of sections stainedwith laminin and CD11b from animals treatedwith saline, 3-NPA or kainic acid. Insaline treated animals inactive microglia were observed (A, D, G and J) and limited laminin immunoreactivity was observedaround some of the larger capillaries (white arrow). With the 3-NPA treated animals, CD11b and laminin double labeling revealthe association of laminin and activated microglia (CD11b) in the striatum (B and C) and hippocampus (E and F). In kainic acidtreated animals, CD11b and laminin double labeling showed the association of activated CD11b and laminin in the thalamus (Hand I) and cortex (K and L). Scale bar=100 μm for Figs. (A, B, D, E, G, H, J and K). Scale bar=200 μm for Figs., (C, F, I and L).

243B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

and C), where laminin has dramatically increased. A similarcorrelationwasalsoobserved in thehippocampus (Figs. 3EandF),in both KA and some 3-NPA injected rats, microglia wereoccasionally seen enmeshed within laminin containing micro-vessels (Figs. 3F and I). In KA treated rats, CD11b and laminindouble immunolabeling reveal that laminin containing micro-vessels are in the septum, thalamus, cortex, amygdala, hippo-campus, thalamus (Figs. 3H and I) and cortex (Figs. 3K and L).

3. Discussion

Although there are few reports available on the effect ofintracerebral administration of KA on laminin expression(Shigematsu et al., 1989b), the vascular changes in the brainfollowing acute peripheral injection of KA have not beenelucidated. In the present study, the observation that laminin

244 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

immunoreactivity dramatically increased in the areas ofneurodegeneration is quite intriguing. As noted in theResults section, laminin immunoreactive vascular elementshave been observed in the pyriform cortex, septum, hippo-campus and medial thalamus, which are the same regionsthat also contain numerous FJC positive neurons. Althoughthe exact cause of the laminin expression in the brain is notwell understood, several possible explanations should beconsidered. These results suggest some association betweenangiogenesis and neurodegeneration. It is conceivable thatneuronal damage could be involved in the disruption of theblood brain barrier, stimulation of angiogenic activity, andinfiltration of inflammatory mediators (de Vries et al., 1997).

The administration of KA can result in various responsesincluding immune reactions which potentially increase theexpression of laminin immunoreactivity in the lesionedregions of the brain (Akiyama et al., 1988; McGeer et al.,1988). In this regard, some direct effect of KA can not be ruledout. A plausible explanation for the increase of lamininimmunoreactive elements in the brain is neovascularization.Neovascularization may be induced by a variety of stimuliincluding ischemia (Nukada and Dyck, 1986) and trauma (Becket al., 1983). Laminin, being a non-collagenous extracellularmatrix glycoprotein, plays an important role in endothelialorganization during angiogenesis (Form et al., 1986). Lamininhas the ability to promote growth of various cell typesincluding peripheral and central neurons (Carbonetto et al.,1987) and endothelial cells (Shigematsu et al., 1989b). The P1ifragment of laminin has a growth promoting activity onmicrovascular endothelial cells. Type IV collagen is anotherimportant component of the basement membrane. Duringangiogeneisis in vivo, type IV collagen appears before lamininin the development of the basement membrane (Form et al.,1986). During neovascularization, laminin is actively involvedin themigration and proliferation of endothelial cells, whereasthe appearance of type IV collagen correlatesmorewith lumenformation and maintenance of the differentiated endothelialcell phenotype (Form et al., 1986). These authors (Shigematsuet al., 1989b) have reported that vascular changes detected bylaminin immunolabeling roughly correlate with gliosis asrevealed by GFAP immunolabeling, but the appearance ofvessels with laminin immunolabeling precedes the appear-ance of gliosis, and the laminin immunoreactivity in the glialelements is negligible. Shigematsu et al. (1989a) also demon-strated in a separate study that neovascularization could beobserved in transplants by using laminin immunolabeling. Inthe present study we used a fluorescent marker of neurode-generation, FJC, to reveal the neurodegeneration in the brainfollowing KA insult. Increased laminin immunolabeling in thebrain regions where FJC positive degenerating neurons werealso present ushers in the possibility that laminin immuno-labeling can be used as a marker for newly formed bloodvessels in the brain following neurotoxic insults including KAinduced neurodegeneration. In the present study we haveshown that the effect of 3-NPA on the expression of laminincorrelates with those regions where FJC positive neurons arepreponderant. Most of the laminin expression was in thestriatum while roughly 50% of the cases exhibited lamininimmunolabeling in the hippocampus as well. Althoughrecently Duran-Vilaregut et al. (2009) reported the disintegra-

tion of laminin stained vessels following neurotoxic insult, weobserved the converse, intense laminin labeling of vessels inthe lesioned areas with no laminin immunoreactivity in thenon lesioned areas.

The discrepancy in the results might be due to the differentexperimentalparadigmadoptedbyDuran-Vilaregutetal. (2009).This group injected 3-NPA (20 mg/kg) into rats for 3 consecutivedays (i.p) and thenwaited an additional 3 days before perfusion,resulting in a relatively longer survival interval. Whereas, weinjected (s.c) 3 doses of 3-NPA (20 mg/kg)with a one day intervalbetween doses, Duran-Vilaregut et al. (2009) waited 3more daysbefore perfusion. In contrast, we perfused 1 day after the lastdose. As we have consistently observed neurodegeneration inthe striatum and hippocampus in our paradigm and concom-itant upregulation of laminin immunoreactive microvesselsboth in striatum and hippocampus, it is possible that in thestudy conducted by Duran-Vilaregut et al., employed a survivalinterval at which time the blood vessels might already havedisintegrated, hence less laminin immunoreactivity.

In the present KA induced neurotoxicity study, we alsoemployed EBA immunolabeling and CD11b immunolabelingto reveal the respective integrity of the blood brain barrier(BBB) and the activation of microglia. EBA plays an importantrole in maintaining BBB integrity (Ghabriel et al., 2000). EBA isalso a specific marker for normal CNS microvessels and itsstaining intensity is reduced in pathological brain conditionssuch as ischemia (Lin et al., 2001; Lu et al., 2008), traumaticinjury (Lin et al., 2001) and stroke (Gursoy-Ozdemir et al., 2004),suggesting impairment of the BBB. In the present study, weemployed EBA and laminin immunolabeling and observed amarked reduction in the intensity of EBA immunolabeling inmicrovessels in the area where neurodegeneration occurs andwhere laminin expression is upregulated. This result not onlyconfirms the disruption of the BBB but also reveals theexpression of laminin glycoprotein in the area of the brainwhere degenerating neurons were prevalent following the KAinsult.

Microglia cells are the resident macrophages of the CNS,and their activation plays a critical role in inflammatoryreactions associated with many brain disorders, includingischemia, Alzheimer's and Parkinson's diseases, multiplesclerosis and certain toxic insults (Vesce et al., 2007; Wyss-Coray and Mucke, 2002). Microglia and astrocytes representthe major source of pro-inflammatory molecules, and theyplay a critical role in determining the spatial and temporalextent of brain inflammation as well as its impact on neuronalfunction and survival (Haynes et al., 2006; Vesce et al., 2007;Viviani et al., 2007;Wyss-Coray andMucke, 2002). Accordingly,microglia and astrocytes become reactive after status epilep-ticus. In the case of microglia, this activation has been welldocumented in terms of changes in morphology and expres-sion of specific markers (Turrin and Rivest, 2004; Vezzani andGranata, 2005). However, activation of microglial cells is acomplex process that includes changes in pharmacologicaland electrophysiological properties, as well as the migration,proliferation and release of a variety of mediators (Farber andKettenmann, 2005; Hanisch and Kettenmann, 2007). In thepresent study, CD11b was used as a marker for microglialactivation to further characterize the effect of a KA insult onthe brain. A marked increase in CD11b labeling was observed

245B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

in the thalamus, hippocampus, amygdala and cortex ascompared to the control animals in which reduced CD11blabeling was apparent. This correlates with the same areas ofneurodegeneration as previously reported.

This study also performed CD11b labeling in the sectionsobtained from 3-NPA treated animals. CD11b immunolabelingwas typically observed in the striatum and occasionally in thehippocampus but not in the adjacent non-injured areas. Tothe best of our knowledge, this is the first study to showactivation of microglia following a 3-NPA insult. Severalunderlying mechanisms are possible. It is possible that 3-NPA initiates the cascade of events which triggers theproduction of pro-inflammatory cytokines, free radicals, andnitrous oxide whichmay be deleterious for neurons and resultin activated inflammatory responses.We have also performedEBA (SMI71) and laminin double immunolabeling in the brainsof animals treated with 3-NPA. Most of the EBA containingmicrovessels are present in the non-injured part of thestriatum, in contrast to laminin containing microvessels thatare apparent in the striatum and sometimes extend to thehippocampus, depending on the severity of the pathologyobserved following the insult. Several angiogenic factors anddownstream signaling pathways have been shown to beinvolved in brain angiogenesis (Greenberg and Jin, 2005).Probably, the most important is vascular endothelial growthfactor (VEGF) which is expressed in the brain during differentneuropathological conditions such as ischemia, brain tumorand hypoxia. There are reports suggesting that microglia andastrocytes can secrete neurotrophic and angiogenic factors,whichmay in turn, enhance vascular remodeling as evidencedby increased vascular endothelial growth factor expression inendothelial cells (Greenberg and Jin, 2005).

This study also examined the localization of laminin andFluoro-Jade C positive degenerating neurons in our neurotox-icity model. We found that laminin expression was increasedin the cortex, thalamus and hippocampus and correlated withthe regions containing Fluoro-Jade C positive neurons. Thismay imply that laminin is likely to be protective or reparative,although this hypothesis needs to be examined further toelucidate the exact role of extra cellular matrix proteins suchas laminin.

Our findings suggest that two mechanistically diverseneurotoxicants, KA or 3-NPA, not only induce neurodegenera-tion but also cause inflammatory responses from reactivemicroglia which could induce angiogenesis. Treatments tothat block laminin receptor activation or block the productionof one of these ECM proteins (laminin) could serve astherapeutic strategies to reduce tumor angiogenesis in cancerpatients. One follow up study would be a systematic exami-nation of the time course of laminin expression. This mayfacilitate the elucidation of its function.

In conclusion, this study demonstrates an increasedlaminin immunoreactivity in the areas where degeneratingneurons are also located, suggesting that neovascularizationoccurs in response to neurotoxic insults. Increased CD11bexpression in the KA or 3-NPA treated animals suggests thepossible association of microglial activation, angiogenesis andneurodegeneration. Therefore, laminin immunolabeling canbe used as a marker to identify the effect of neurotoxicsubstances on the vascular elements in the brain.

4. Experimental procedures

4.1. Animals

Experiments were performed on adult male Sprague–Dawleyrats (NCTR breeding colony) weighing 400–450 g. The animalswere housed under standard environmental conditions (lightbetween 06.00 and 18.00 h, temperature 22±1 °C, with adlibitum access to water and rat chow). All experimentalprotocols were approved by NCTR IACUC.

4.2. Administration of kainic acid and 3-NPA

We selected one dose of kainic acid for this study. In order tomaximize theneurodegeneration, adoseof 10mg/kgkainic acid(Schmued et al., 1997) and 3 subcutaneous doses of 20mg/kg 3-NPAwas selected (Schmued et al., 1997). Twentymale rats wereused for the present study. Animals were allotted into fourgroups (n=5). The first group received kainic acid (KA; Sigma;10 mg/kg; i.p.), the second group received 3-NPA (Sigma; 3 dosesof 20mg/kg; s.c. one day interval) and the third and fourthgroups received only saline. Kainic acid treated animals wereperfused intra-cardially with 10% formalin two days after drugadministration along with saline treated controls, whereas 3-NPA treated animals were sacrificed intra-cardially with thesame fixative (10% formalin) six days after the first dose of thedrug, along with saline treated controls. Brains were sectionedon a sliding freezing microtome at 25 μm thickness andcollected in 0.1 M neutral phosphate buffer (PB) and processedfor histochemistry.

4.3. Measurement of body temperature

Body temperaturewas recorded using rectal probe once in every30min for 8 h. Generally, KA injected rats showed hyperther-mia (>40 °C) whereas, 3-NPA treated rats showed hypothermia(35–36 °C).

4.4. Immunohistochemical controls

The antibody employed against EBA was raised in mouse andits specificity has been characterized previously (Sternbergerand Sternberger, 1987). Specificity of the laminin antibody(Sigma— Cat# L 9393) was established by (a) incubation of 1 mlof diluted (1:1000) antibody with the laminin (Sigma — Cat# L2020 from Engelbreth–Holm–Swarm murine sarcoma base-ment membrane) at 10−5 M for 24 h at 4°C, using it in thecomplete staining protocol; or (b) omitting the primaryantiserum (laminin). Either procedure resulted in a completeabsence to detectable laminin immunoreactivity in the brainsections of saline, KA and 3-NPA injected rats.

4.5. Tissue preparation and Fluoro-Jade C labeling

To identify and validate lesions after neurotoxic insults,Fluoro-Jade C (Histo-Chem, Jefferson, AR) labeling was per-formed on 25 μm thick coronal sections as described previ-ously by Schmued et al. (2005). Briefly, sections were mountedon gelatin coated slides and air dried for one hour, rinsed inbasic alcohol for 3 min, followed by a 2 min rinse in 70%

246 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

alcohol. Sectionswere then briefly rinsed in distilledwater andincubated in a 0.06% solution of KMnO4 for 10 min. Sectionswere briefly rinsed in distilled water to remove the excessKMnO4. Sectionswere incubated in 0.0001%Fluoro-Jade C stainin 0.1% acetic acid for 10 min. Following Fluoro-Jade C labeling,sectionswere rinsed three times in distilledwater, air dried for10 min and cleared in xylene and coverslipped with DPX.

4.6. Combined Fluoro-Jade C labeling andlaminin immunohistochemistry

A series of 25-μmcoronal sectionswere cut on a sliding freezingmicrotome at 25 μm from each brain, starting at the genu of thecorpus callosum and extending caudally to the posterior midbrain, collected in 0.1 M PB (pH 7.4) and stored at −20 °C in acryoprotectant solution (mixture of 25% glycerol, 30% ethyleneglycol and 25% 0.1 Mphosphate buffer, 20% distiiledwater) untilsubsequent processing. Briefly, the sections were washed in0.01 M PBS three times and then incubated in 0.5% Triton X-100in PBS for 30min to improve antibody penetration. To reducenon-specific binding of the primary antibody, the sections weretreated with 10% normal horse serum (Sigma, MO, USA) in PBSfor 20min and then incubated in polyclonal rabbit antiseraagainst laminin (Sigma, Cat# L9393) at 1:1000 dilution for 2 daysat 4 °C on a shaker. The sections were then washed in PBS andincubated in biotinylated donkey anti-rabbit IgG (1:200, JacksonImmunoresearch, PA, USA) for 4 h at room temperature. Afterrinsing in PBS three times, sections were incubated in TRITCstreptavidin solution at 1:250 dilution (Jackson Immunore-search, PA, USA) for 3 h at room temperature. To avoid loss ofimmunolabeling, pretreatment in basic alcohol was omittedand the sectionswere incubated inKMnO4only for 2 min, brieflyrinsed in distilled water, incubated in Fluoro-Jade C for 10minand then rinsed in 3 changes of distilled water. Sections werethen air dried, cleared in xylene and cover slipped in DPXmountant. The tissue was examined with a Nikon epifluores-cent microscope using the following set of filters: for TRITCstreptavidin, excitation of 540–590 nm, bandpass of 596 nmandemission of 600–660 nm; for Fluoro-Jade C, excitation of 460–500 nm, band pass of 505 nm and emission of 510–560 nm.Combinations of these filters resulted in negligible cross-talkbetween the signals from individual fluorochromes, such thattherewasnoobservablebleed-throughbetweenanyof the threesignals. Thus, neuronspositive for Fluoro-JadeCexhibited greenfluorescence while laminin containing endothelial cells exhib-ited red fluorescence.

4.7. EBA and laminin double immunolabeling

Another set of sections was taken for EBA and laminin doubleimmunolabeling. The first aim was to see whether KA or 3-NPAtreatments cause blood brain barrier disruption by using EBAimmunolabeling and the second aimwas to seewhether laminincontainingmicrovessels are present in the same area where EBAis being down regulated following neurotoxic insult. Twodifferent fluorochromes were employed to visualize these twoantigens in every fifth section from the brain of all rats. Briefly,the sections werewashed in PBS three times and then incubatedin 0.5% Triton X-100 in PBS for 30min to improve antibodypenetration. To reduce non-specific binding of the primary

antibody, the sections were treated with 10% normal horseserum (Sigma, MO, USA) in PBS for 20 min prior to beingincubated in a mixture of mouse monoclonal antiserum againstEBA (SMI71; Sternberger Monoclonals, Inc. Cat# SMI71) at 1:2000dilution and rabbit antiserum against laminin (Sigma, Cat#L9393) at 1:2000 dilution for 3 days at 4 °C on a shaker. Thesections were then washed in PBS and incubated in amixture ofthe two respective secondary antibodies: Alexa Fluor 594 chickenanti-mouse IgG (Invitrogen, USA) at 1:50 dilution and Alexa Fluor488 chicken anti-rabbit IgG (Invitrogen, USA) at 1:50 dilution for4 h at room temperature. Sections were then washed in PBS fourtimes to remove excess secondary antibody. Sections were thenair dried and cleared in xylene and cover slipped with DPXmountant and examined with a Nikon epifluorescent micro-scope using two sets of filters as described earlier. Thus, laminincontaining endothelial cells exhibited green fluorescence where-as EBA containing microvessels exhibited red fluorescence.

4.8. Laminin and CD11b double immunolabeling

Another set of sections was taken for CD11b and laminin doubleimmunolabeling study. The first aimwas to seewhetherCD11b, amarker for microglia, is being activated upon neurotoxic insult.The second aim was to see whether laminin containing micro-vessels are present in the same area where CD11b is beingupregulated following neurotoxic insult. As previously described,two fluorochromes were employed to visualize these twoantigens respectively in every fifth section fromall brains. Briefly,the sections were washed in PBS three times and then incubatedin 0.5% Triton X-100 in PBS for 30 min to improve antibodypenetration. To reduce non-specific binding of the primaryantibody, the sections were treated with 10% normal horseserum(Sigma,MO,USA) inPBS for 20minand then incubated inamixture ofmousemonoclonal antiserum against CD11b (Chemi-con,Cat#MAB1405)at1:100dilutionandrabbit antiserumagainstlaminin (Sigma, Cat# L9393) at 1:2000 dilution for 3 days at 4 °C ona shaker. The sectionswere thenwashed in PBS and incubated ina mixture of two secondary antibodies: Alexa Fluor 594 chickenanti-mouse IgG (Invitrogen, USA) at 1:50 dilution and Alexa Fluor488 chicken anti-rabbit IgG (Invitrogen, USA) at 1:50 dilution for4 h at room temperature. Sections were then washed in PBS fourtimes to remove excess secondary antibody. Sections were thenair dried, cleared in xylene, cover slippedwithDPXmountant andexaminedwith aNikon epifluorescentmicroscopeusing two setsof filters as described earlier. Thus, laminin containing endothe-lial cells exhibited green fluorescence whereas CD11b containingmicroglia exhibited red fluorescence.

Acknowledgment

This work was supported by FDA protocol E 7312.

R E F E R E N C E S

Akiyama, H., Itagaki, S., McGeer, P.L., 1988. Majorhistocompatibility complex antigen expression on ratmicroglia following epidural kainic acid lesions. J. Neurosci.Res. 20, 147–157.

247B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 2 3 9 – 2 4 7

Beck, D.W., Hart, M.N., Cancilla, P.A., 1983. The role of themacrophage in microvascular regeneration following braininjury. J. Neuropathol. Exp. Neurol. 42, 601–614.

Bohling, T., Paetau, A., Ekblom, P., Haltia, M., 1983. Distribution ofendothelial and basement membrane markers in angiogenictumors of the nervous system. Acta Neuropathol. 62, 67–72.

Carbonetto, S., Evans, D., Cochard, P., 1987. Nerve fiber growth inculture on tissue substrata from central and peripheralnervous systems. J. Neurosci. 7, 610–620.

Colognato, H., Yurchenco, P.D., 2000. Form and function: thelaminin family of heterotrimers. Dev. Dyn. 218, 213–234.

deVries, H.E., Kuiper, J., de Boer, A.G., VanBerkel, T.J., Breimer, D.D.,1997. The blood-brain barrier in neuroinflammatory diseases.Pharmacol. Rev. 49, 143–155.

Duran-Vilaregut, J., del Valle, J., Camins, A., Pallas, M., Pelegri, C.,Vilaplana, J., 2009. Blood-brain barrier disruption in thestriatum of rats treated with 3-nitropropionic acidNeurotoxicology 30, 136–143.

Durbeej, M., 2009. Laminins. Cell Tissue Res. 339, 259–268.Ekblom, P., 1981. Formation of basement membranes in the

embryonic kidney: an immunohistological study. J. Cell Biol.91, 1–10.

Farber, K., Kettenmann, H., 2005. Physiology of microglial cells.Brain Res. Brain Res. Rev. 48, 133–143.

Foidart, J.M., Bere Jr., E.W., Yaar, M., Rennard, S.I., Gullino, M.,Martin, G.R., Katz, S.I., 1980. Distribution and immunoelectronmicroscopic localization of laminin, a noncollagenousbasement membrane glycoprotein. Lab Invest. 42, 336–342.

Form, D.M., Pratt, B.M., Madri, J.A., 1986. Endothelial cellproliferation during angiogenesis. In vitro modulation bybasement membrane components. Lab Invest. 55, 521–530.

Ghabriel, M.N., Zhu, C., Hermanis, G., Allt, G., 2000. Immunologicaltargeting of the endothelial barrier antigen (EBA) in vivo leads toopening of the blood-brain barrier. Brain Res. 878, 127–135.

Giordana, M.T., Germano, I., Giaccone, G., Mauro, A., Migheli, A.,Schiffer, D., 1985. The distribution of laminin in human braintumors: an immunohistochemical study. Acta Neuropathol.67, 51–57.

Greenberg, D.A., Jin, K., 2005. From angiogenesis toneuropathology. Nature 438, 954–959.

Gursoy-Ozdemir, Y., Qiu, J., Matsuoka, N., Bolay, H., Bermpohl, D.,Jin, H., Wang, X., Rosenberg, G.A., Lo, E.H., Moskowitz, M.A.,2004. Cortical spreading depression activates and upregulatesMMP-9. J. Clin. Invest. 113, 1447–1455.

Hanisch, U.K., Kettenmann, H., 2007. Microglia: active sensor andversatile effector cells in the normal and pathologic brain. Nat.Neurosci. 10, 1387–1394.

Haynes, S.E., Hollopeter, G., Yang, G., Kurpius, D., Dailey, M.E., Gan,W.B., Julius, D., 2006. The P2Y12 receptor regulates microglialactivation by extracellular nucleotides. Nat. Neurosci. 9,1512–1519.

Lin, B., Ginsberg, M.D., Zhao, W., Alonso, O.F., Belayev, L., Busto, R.,2001. Quantitative analysis of microvascular alterations intraumatic brain injury by endothelial barrier antigenimmunohistochemistry. J. Neurotrauma 18, 389–397.

Lu, A., Clark, J.F., Broderick, J.P., Pyne-Geithman, G.J., Wagner, K.R.,Ran, R., Khatri, P., Tomsick, T., Sharp, F.R., 2008. Reperfusionactivates metalloproteinases that contribute to neurovascularinjury. Exp. Neurol. 210, 549–559.

McGeer, P.L., Itagaki, S., Boyes, B.E., McGeer, E.G., 1988. Reactivemicroglia are positive for HLA-DR in the substantia nigra ofParkinson's and Alzheimer's disease brains. Neurology 38,1285–1291.

Miner, J.H., 2008. Laminins and their roles in mammals. Microsc.Res. Tech. 71, 349–356.

Nukada, H., Dyck, P.J., 1986. Neovascularization after ischemicnerve injury. Exp. Neurol. 92, 391–397.

Patton, B.L., 2000. Laminins of the neuromuscular system.Microsc.Res. Tech. 51, 247–261.

Racine, R.J., 1972. Modification of seizure activity by electricalstimulation. II. Motor seizure. Electroencephalogr. Clin.Neurophysiol. 32, 281–294.

Schmued, L.C., Albertson,C., Slikker Jr.,W., 1997. Fluoro-Jade: anovelfluorochrome for the sensitive and reliable histochemicallocalization of neuronal degeneration. Brain Res. 751, 37–46.

Schmued, L.C., Hopkins, K.J., 2000. Fluoro-Jade B: a high affinityfluorescent marker for the localization of neuronaldegeneration. Brain Res. 874, 123–130.

Schmued, L.C., Stowers, C.C., Scallet, A.C., Xu, L., 2005. Fluoro-JadeC results in ultra high resolution and contrast labeling ofdegenerating neurons. Brain Res. 1035, 24–31.

Shigematsu, K., Kamo, H., Akiguchi, I., Kameyama, M., Kimura, H.,1989a. Neovascularization of transplanted central nervoustissue suspensions: an immunohistochemical study withlaminin. Neurosci. Lett. 99, 18–23.

Shigematsu, K., Kamo, H., Akiguchi, I., Kimura, J., Kameyama, M.,Kimura, H., 1989b. Neovascularization in kainic acid-inducedlesions of rat striatum. An immunohistochemical study withlaminin. Brain Res. 501, 215–222.

Sternberger, N.H., Sternberger, L.A., 1987. Blood-brain barrierprotein recognized by monoclonal antibody. Proc. Natl Acad.Sci. USA 84, 8169–8173.

Sternberger, N.H., Sternberger, L.A., Kies, M.W., Shear, C.R., 1989.Cell surface endothelial proteins altered in experimentalallergic encephalomyelitis. J. Neuroimmunol. 21, 241–248.

Turrin, N.P., Rivest, S., 2004. Innate immune reaction in responseto seizures: implications for the neuropathology associatedwith epilepsy. Neurobiol. Dis. 16, 321–334.

Vesce, S., Rossi, D., Brambilla, L., Volterra, A., 2007. Glutamaterelease from astrocytes in physiological conditions and inneurodegenerative disorders characterized byneuroinflammation. Int. Rev. Neurobiol. 82, 57–71.

Vezzani, A., Granata, T., 2005. Brain inflammation in epilepsy:experimental and clinical evidence. Epilepsia 46, 1724–1743.

Viviani, B., Gardoni, F., Marinovich, M., 2007. Cytokines andneuronal ion channels in health and disease. Int. Rev.Neurobiol. 82, 247–263.

Wyss-Coray, T., Mucke, L., 2002. Inflammation inneurodegenerative disease — a double-edged sword. Neuron35, 419–432.