Kim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Open AccessR E S E A R C H A R T I C L EResearch articleMidkine, heparin-binding growth factor, blocks kainic acid-induced seizure and neuronal cell death in mouse hippocampusYun B Kim†1,2, Jae K Ryu†1, Hong J Lee1,3, In J Lim4, Dongsun Park2, Min C Lee1,5 and Seung U Kim*1,3AbstractBackground: Midkine (MK), a member of the heparin-binding growth factor family, which includes MK and pleiotrophin, is known to possess neurotrophic and neuroprotective properties in the central nervous system. Previous studies have shown that MK is an effective neuroprotective agent in reducing retinal degeneration caused by excessive light and decreasing hippocampal neuronal death in ischemic gerbil brain. The present study was undertaken to investigate whether MK acts as an anticonvulsant in kainic acid (KA)-induced seizure in mouse and blocks KA-mediated neuronal cell death in hippocampus.Results: Increased expression of MK was found in hippocampus of mouse following seizures induced by intracerebroventricular injection of KA, and MK expression was found in glial fibrillary acidic protein (GFAP)-positive astrocytes. Concurrent injection of MK and KA attenuated KA-induced seizure activity and cell death of hippocampal neurons including pyramidal cells and glutamic acid decarboxylase 67 (GAD67)-positive GABAergic interneurons in the CA3 and hilar area.Conclusion: The results of the present study indicate that MK functions as an anticonvulsant and neuroprotective agent in hippocampus during KA-induced seizures.BackgroundTemporal lobe epilepsy (TLE) is pathologically character-ized by extensive neuronal loss in the CA1, CA3 and hilarregions of hippocampus [1,2]. Previous studies have dem-onstrated that the animal models of TLE generated byintracerebroventricular injection of kainic acid (KA) faith-fully reproduce clinical and pathological features found inhuman TLE [3-7].Previous studies have reported the possible involvementof neurotrophic factors in epilepsy as suggested by the geneexpression of neurotrophic factors such as NGF, BDNF andNT-3 in hippocampus in human TLE as well as in TLE ani-mal models [8,9]. Midkine (MK), one of such neurotrophicfactors, has emerged as an important neuromodulator in thecentral nervous system (CNS). MK, a member of the hepa-rin-binding growth factor family, which includes MK andpleiotrophin, is known to possess neurotrophic and neuro-protective properties [10,11]. MK was originally isolated asthe product of retinoic acid-responsive gene that functionsprimarily in inducing cell differentiation in mouse terato-carcinoma cells [12], and has the ability to influence a vari-ety of neuronal functions including neurite extension [13],neuronal differentiation [14,15] and neuronal survival fol-lowing injury or damage in the CNS [15,16]. During thefetal development of the CNS, MK expression was demon-strated in neuroepithelial/neural progenitor cells followingethylnitrosourea injury [17] indicating that MK might havea role in cellular proliferation [18]. Recent studies furthershowed that MK has been implicated in neurological dis-eases, including Alzheimer's disease [19], cerebral ischemia[20] and Parkinson-dementia complex of Guam (Lytico-bodig disease) [21]. In patients with Alzheimer's disease[19] or Lytico-bodig disease [21], MK immunoreactivitywas found in senile plaques and neurofibrillary tangles. Inaddition, an increased expression of MK was found in* Correspondence: sukim@interchange.ubc.ca1 Division of Neurology, Department of Medicine, UBC Hospital, University of © 2010 Kim et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons At-tribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in anymedium, provided the original work is properly cited.astrocytes in rat models of cerebral ischemia [22]. It is notknown, however, whether the expression of MK in the brainBritish Columbia, Vancouver, Canada† Contributed equallyFull list of author information is available at the end of the articleKim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 2 of 9after the brain injury is a part of an endogenous repair pro-cess to prevent further damage in the CNS.The objectives of the present study are to determinewhether intracerebroventricularly injected MK acts as ananticonvulsant and blocks KA-mediated neuronal cell deathin hippocampus.ResultsMK expression after seizuresWe first examined MK expression immunohistochemicallyin mouse hippocampus following KA injection. Injection ofKA (0.2 μg/mouse) to mice induced severe epileptiformseizures (mean score 4.2/maximum score 5.0). Basal levelof MK immunoreactivity was found in hippocampal pyra-midal neurons in control mouse brain injected with vehicle[phosphate-buffered saline (PBS)] (Figure 1A, top leftpanel), while in mouse injected with KA decreased MKexpression was detected in pyramidal neurons (Figure 1A,bottom left panel; see arrows); Nissl staining of the adjacentsections confirmed that the cellular area of decreased MKimmunoreactivity was associated with damaged pyramidalneurons (Figure 1A, bottom right panel; see arrows). Nisslstaining in control animals receiving PBS injection showedno evident neuronal damage (Figure 1A, top right panel).Interestingly, we have found the number of MK-positivecells markedly increased in stratum lacunosum moleculare(Slm) area of CA3 at 24-hr post-KA injection relative toPBS controls. Representative MK immunostaining in CA3Slm is shown in Figure 1B (box denotes area of high mag-nification, top panel). In PBS-injected brain, only smallnumber of MK-positive cells was observed (Figure 1B,middle panel). Following injection of KA, CA3 Slm areademonstrated an increase in the number of MK-positivecells (Figure 1B, bottom panel). Double-labeling immuno-fluorescence staining was then used to investigate MKexpression in glial cells in KA-injected hippocampus. Rep-resentative double staining for MK (green staining, Figure1C top panel) and GFAP (red staining, Figure 1C middlepanel) showed that MK was expressed in GFAP-positiveastrocytes; there was no MK expression in anti-complementreceptor 3 (OX-42)-positive microglia (data not shown).Quantification of MK expression data (expressed as areadensity of MK-positive cells, Figure 2) showed that KAinjection significantly decreased MK immunoreactivity by44% in SP area relative to PBS-injected control. However,the MK immunoreactivity in animals with KA injectionmarkedly increased by 651% in Slm area as compared toPBS-injected control.Anticonvulsant effect of MKTo further investigate the role of MK in KA-injected hip- tion, which lasted for mean 1,200 sec (Table 1). TheFigure 1 MK expression in the hippocampus after KA injection. (A) Representative immunofluorescence images of MK immunoreac-tivity in hippocampal CA3 pyramidal neurons 24 hr after PBS injection (top left panel) or KA (0.2 μg/mouse, bottom left panel). Adjacent hip-pocampal sections stained with Nissl staining (right panels) are shown here. Following KA treatment, cell death in CA3 pyramidal neurons is clearly visible (arrows). Scale bars: 20 μm (top panels), 30 μm (bottom panels). (B) Representative immunofluorescence images of MK-posi-tive cells in CA3 Slm (striatum lacunosum moleculare) (box in top pan-el indicates region of interest) after injection with PBS (middle panel) or KA (0.2 μg/mouse, bottom panel). (C) Double-labeling immunofluores-cence staining of MK (green, top panel) with GFAP (red, middle panel) showing MK immunoreactivity is co-localized on GFAP-positive astro-cytes in CA3 Slm (bottom panel). Scale bar: 20 μm.pocampus, we examined the potential protective effects ofMK against KA-induced seizures and neurotoxicity. Sei-zure onset was approximately 170 sec following KA injec-intensity of seizures reached mean score 4.2 (maximumscore 5.0) when the seizure activity was highest (5-10 minafter KA challenge). Interestingly, seizure duration andKim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 3 of 9intensity were markedly reduced by co-administration ofMK in a dose-dependent manner, although the time of onsettime slightly delayed at a high dose (0.4 μg/mouse). Seizureduration and intensity were shortened and attenuated to51.4 - 26.5% and 59.5 - 40.5% of the control levels by treat-ment with KA (0.1 - 0.4 μg/mouse), respectively.Neuroprotective effect of MKRepresentative Nissl staining 24 hr following KA injectionshowed considerable neuronal loss in the hippocampal sub-regions, CA3 and hilus of the dentate gyrus as compared toPBS control (Figure 3A, left and middle panels). MK treat-ment was effective in reducing neuronal loss in KA-injected hippocampus as shown in a representative findingin a high dose (0.4 μg/mouse) group (Figure 3A, right pan-els). The extent of degeneration of hippocampal neuronswas quantified in hippocampal subregions, CA1, CA3, andhilus (Figure 3B). The number of hippocampal neurons wassignificantly reduced in CA3 (-81%) and hilus (-85%) byKA exposure, whereas CA1 region (-19%) was relativelyspared when compared with PBS control. Co-application ofMK increased survival of neurons in KA-injected hip-pocampus in a dose-dependent manner, leading to signifi-cant improvements at KA doses of 0.2 and 0.4 μg/mouse.Next we investigated the efficacy of MK to block theKA-induced cell death of GABAergic interneurons(GAD67-positive neurons). A previous study has reportedthat GAD67-positive interneurons are lost in the hippocam-pus in KA-injected animal model of excitotoxicity [23].Animals receiving KA injection showed significant reduc-tion in the number of GAD67-positive interneurons in sub-fields of CA1 (strata oriens -80%, strata pyramidale -64%,strata radiatum -33%), CA3 (strata oriens -91%, strata pyra-midale -78%, strata radiatum -62%) and layers of dentategyrus (molecular layer -37%, granule cell layer -69%, den-tate hilus -77%) as compared to PBS control (Figures 4 and5). Administration of MK (0.4 μg/mouse) reduced the cellloss in GAD67-positive neurons caused by KA injectionand this neuroprotective effect of MK was evident in thesubfields of CA3 and layers of dentate gyrus (Figure 4).Application of MK markedly increased survival of GAD67-positive neurons in the strata pyramidale (68% improve-ment), radiatum (45% improvement) of CA3 and molecularlayer (27% improvement), granule cell layer (29%improvement), and dentate hilus (33% improvement) ofdentate gyrus, whereas neurons in subfields of CA1 was noteffectively preserved (Figure 5).DiscussionThe main finding of this study is that intracerebroventricu-lar administration of MK conferred neuroprotection againstKA-induced excitotoxic cell death of hippocampal neurons.Our results also demonstrated that MK was effective inattenuating KA-induced seizures and degeneration ofGABAergic interneurons in hippocampus. The results fromimmunohistochemical staining showed that damaged pyra-midal neurons in hippocampus are correlated with thedecreased level of MK expression (Figure 1A). At present,it is not known whether this reduced level of MK immuno-reactivity in neurons is causative factor for neuronal degen-Table 1: Effects of MK on KA-induced seizure activitiesTreatment(μg/mouse)Latency to onset(sec)Seizure duration(sec)Seizure intensity(maximum score 5.0)Vehicle 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0KA (0.2) alone 169.9 ± 38.8* 1,200.6 ± 372.3* 4.2 ± 0.1*+MK (0.1) 160.0 ± 28.5 616.7 ± 50.3# 2.5 ± 0.3#Figure 2 Site-selective MK expression in the KA-damaged hip-pocampus. Bar graph shows the quantification of MK immunoreactiv-ity in SP (stratum pyramidale) and Slm (stratum lacunosum moleculare) of hippocampal CA3 region 24 hr after PBS or KA (0.2 μg/mouse) injection. Data are presented as mean ± SEM. * p < 0.05, as compared with PBS. One-way ANOVA, Student-Newman-Keuls multi-ple comparison test.+MK (0.2) 172.8 ± 33.8 422.4 ± 71.5# 2.2 ± 0.2#+MK (0.4) 188.8 ± 47.0 318.6 ± 31.7# 1.7 ± 0.3#Data are presented as mean ± SEM. * p < 0.05, as compared with PBS. # p < 0.05, as compared with KA. Student's t-test.Kim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 4 of 9Figure 3 Effect of MK on hippocampal neuronal damage induced by KA injection. (A) Representative Nissl-stained sections from animals 24 hr after PBS (left panels), KA (0.2 μg/mouse, middle panels) or KA plus MK (0.4 μg/mouse, right panels) injection. Top panels shows at low magnification of the hippocampus. Scale bar: 400 μm. Middle and bottom panels show at high magnification of CA3 and dentate gyrus (DG) subregions of hip-pocampus. Scale bars: 200 μm. Note that the damaged areas induced by KA are indicated by absence of Nissl-positive neurons. (B) Quantification of undamaged Nissl-positive neurons for the hippocampal subregions CA1 and CA3, and hilus of dentate gyrus with different MK doses (0.1 -- 0.4 μg/mouse). Data are presented as mean ± SEM. * p < 0.05, as compared with PBS. # p < 0.05, as compared with KA. One-way ANOVA, Student-Newman-Keuls multiple comparison test.Kim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 5 of 9eration induced by KA injection. However, previous in vivostudies have demonstrated that MK is an effective neuro-protective agent in reducing retinal degeneration caused byexcessive light [24] and decreasing hippocampal neuronaldeath in ischemic gerbil brain [25]. Furthermore, MKknock-out mice displayed altered expression of calcium-binding protein in hippocampus and defective workingmemory [26]. In view of these observations, it is possiblethat MK is involved in the regulation of endogenous neuro-protective process against externally applied injury or dis-ease.We have found that the level of MK immunoreactivitymarkedly increased in astrocytes after KA injection (Figure1C). In a previous report, we have demonstrated rapid acti-vation of astrocytes in hippocampal alveus and fimbria,observations suggest that the activated astrocytes serve asan important source of MK in response to excitotoxicity.Important findings in the present study are therapeutic andneuroprotective effects of MK on KA-induced neuronalinjury in CA3 and dentate hilus of hippocampus (Figure 3).Our results are in good agreement with a previous study ofneuroprotective effect of MK in transient forebrain isch-emia when given immediately before middle cerebral arteryocclusion [25]. Previous studies have demonstrated thatKA-injected brain show dramatic reduction in the numberof GAD-positive interneurons [7,23,28], even in CA1statum oriens and alveus where pyramidal neurons are rela-tively spared [Figures 3-5][29], which can cause abnormalfunctional inhibition of neuronal circuitry leading to hip-pocampal hyperexcitability [2,7]. Thus, it is noteworthyFigure 4 Protective effect of MK on KA-induced cell loss of GAD67-positive GABAergic neurons. Hippocampal coronal sections from animals were immunostained with glutamic acid decarboxylase 67 (GAD67) antibody 24 hr after PBS (left panels), KA (0.2 μg/mouse, middle panels), KA plus MK (0.4 μg/mouse, right panels) injection. High magnification of representative GAD67-stained images of hippocampal subregions CA1 (top panels), CA3 (middle panels), and DG (bottom panels) from different treatments are shown here. Scale bar: 200 μm. Note the paucity of GAD67-positve cells in KA-treated CA3 and dentate gyrus subregions.strata oriens and lacunosum moleculare as well as dentatehilus as early as 1 hr following diisopropyl fluorophosphate(DFP)-induced seizures and neuronal injury [27]. Thesethat icv application of MK blocks the degeneration ofGAD67-positive interneurons, especially in the stratumpyramidale and radiatum of CA3, and the molecular layer,Kim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 6 of 9Figure 5 Protective effect of MK on KA-induced GABAergic neuronal cell death. Numbers of GAD67+ neurons were counted in subregions of hippocampus from PBS, KA (0.2 μg/mouse), and KA plus MK (0.4 μg/mouse)-injected mice; strata oriens (SO), pyramidale (SP) and radiatum (SR) of CA1 (A) and CA3 (B) and all layers of dentate gyrus (C) (ML; molecular layer, GCL; granule cell layer, DH; dentate hilus). Data are presented as mean ± SEM. * p < 0.05, as compared with PBS. # p < 0.05, as compared with KA. One-way ANOVA, Student-Newman-Keuls multiple comparison test.Kim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 7 of 9granule cell layer and hilus of dentate gyrus as caused byKA injection (Figures 4 and 5).The majority of the animals experienced stage 4/5 seizureseverity after KA injection. A previous study has demon-strated in KA-injected rat epilepsy model that dizocilpine(MK-801) inhibited hippocampal neuronal loss withoutblocking seizure development [30]. In addition, severalstudies in animal model of TLE have also shown that sig-nificant neuronal loss is not necessarily a prerequisite forthe development of seizures [31,32]. Such results imply thatthe mechanism of MK action in TLE model is differentfrom that of MK-801, an NMDA antagonist. Our resultsshow that MK-induced neuroprotection against KA toxicityis primarily associated with moderation of seizure activity(Table 1). It is believed that MK plays a role as an anticon-vulsant directly (Table 1) or indirectly by preserving aninhibitory amino acid (GABA) system including GAD67-positive interneurons (Figures 4 and 5). Furthermore, it ispossible to speculate that MK could be a neurotrophic fac-tor, especially for GAD67-positive interneurons or astro-cytes expressing high levels of MK (Figure 1) which maycontribute to early cessation of seizure and control of itsrecurrence.The exact mechanism of the neuroprotective activity ofMK remains to be further clarified but activation of geneproduct(s) associated with apoptosis by MK may providesome answers. For instance, previous studies have shownthat MK inhibits apoptotic process by up-regulation of bcl-2 expression [33] and by inhibition of caspase-3 activation[34]. Thus, it is possible that the MK-mediated neuropro-tective mechanism involves activation of signal transduc-tion pathways involved in regulation of apoptotic celldeath.ConclusionsThe results of the present study demonstrate that the admin-istration of MK produces a significant neuroprotectiveeffect against KA-induced neuronal loss in mouse model ofepilepsy. Additional studies of MK to elucidate their neuro-protective activity in animal models of brain injury andneurodegeneration such as Parkinson's disease, Hunting-ton's disease, amyotropic lateral sclerosis, stroke or spinalcord injury should prove MK as a member of neurotrophicfactors that are valuable in providing effective treatment forpatients with various neurological disorders.MethodsTreatment and seizure monitoringMale C57BL/6 mice (n = 8/group) weighing 25 - 30 g wereused for the experiments. The animals were housed in atemperature- and humidity-controlled room that was kepton Animal Care Committee of the University of BritishColumbia.Animals were anesthetized with intraperitoneal injectionof chloral hydrate (7%, 0.1 mL/kg) and then mounted in astereotaxic apparatus (David Kopf Instruments, Tujunga,CA). KA (0.2 μg/0.4 μL PBS/mouse; Sigma, St. Louis,MO) was injected into the right lateral cerebroventricle atthe coordinate (AP, -2.0; ML, -2.9; DV, -3.8) using a 10 μLHamilton syringe fitted with 26G needle at a rate of 0.1 μL/min. The needle was left in place for 5 min. MK (0.1, 0.2 or0.4 μg/0.4 μL PBS; LG Biotech, Daejon, Korea) was unilat-erally co-injected with KA into the right lateral cerebroven-tricle. For sham-operated animals, the same volume of 0.1M PBS was stereotaxically injected into the same coordi-nates described above. Wounds were sutured, and animalswere allowed to recover and then returned to their cages.After KA injection, each animal was placed in a Plexiglascylinder and their seizure behaviors - latency to onset, dura-tion, and intensity - were recorded for a period of 120 minin a blind manner. Seizure intensity was scored with a slightmodification from a previous scoring system [35] as fol-lowed: stage 1, immobilization and staring; stage 2, headnodding; stage 3, rearing accompanied by forelimb clonusand wet dog shakes; stage 4, falling and wobbling; stage 5,jumping, circling, or rolling. The starting time of head nod-ding was considered onset time of seizures, because thestarting of immobilization and staring behaviors (stage 1)was not clear.ImmunohistochemistryThe animals were anesthetized and transcardially perfusedwith 50 mL cold saline followed by 100 mL of 4% para-formaldehyde in 0.1 M phosphate buffer (pH 7.4) 24 hrafter KA administration. The brains were removed from theskull and post-fixed in 4% paraformaldehyde for 24 hr, fol-lowed with cryoprotection in 30% sucrose in phosphatebuffer for 2 days. Serial coronal sections at 30 μm were pre-pared on a cryostat (CM 1900; Leica, Heerbrugg, Switzer-land). Free-floating sections were prepared from the brainsof PBS, KA, and KA plus MK-injected mice. For singleimmunofluorescence staining, brain sections were incu-bated in PBS containing 5% normal goat serum and 0.2%Triton X-100 for 30 min at room temperature (RT), andthen incubated overnight with rabbit anti-MK antibody(1:500, kindly provided by Dr. T. Muramatsu, Nagoya Uni-versity, Japan). Sections were then incubated with AlexaFluor 488-conjugated goat anti-rabbit IgG (1:200; Molecu-lar Probes, Eugene, OR) at RT for 2 hr in the dark. Sectionswere washed in phosphate buffer, mounted on slides. Nisslstaining was also performed on slide-mounted brain sec-tions with 0.1% cresyl violet (Sigma) for the evaluation ofon an alternating 2-hr light/dark schedule. Food and waterwere available ad libitum throughout the experiments. Allanimal experiments were conducted in accordance with thehippocampal neuronal loss.For GAD67 immunohistochemical staining, all incuba-tion solutions did not contain Triton X-100. The brain sec-Kim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 8 of 9tions were briefly quenched with 3% H2O2 in PBS for 10min and incubated with 5% normal goat serum for 30 min.The sections were incubated overnight at 4°C with rabbitanti-GAD67 (1:1000; Chemicon, Temecula, CA). The sec-tions were then incubated for 1 hr with biotinylated anti-rabbit IgG (1:200; Vector, Burlingame, CA), followed byincubation with avidin-biotin complex (1:200, Vector) for 1hr and then visualized with 0.05% 3,3'-diaminobenzidine(Sigma) and 0.003% H2O2. Negative control sections wereprepared for immunohistochemical staining in an identicalmanner except the primary antibodies were omitted.Double-labeling immunofluorescence microscopyFree floating sections were incubated in PBS containing 3%normal goat serum and 0.3% Triton X-100 for 30 min atroom temperature (RT). Brain sections were incubated for48 hr at 4°C in a mixture of two primary antibodies: MK(1:100) in combination with mouse anti-GFAP (1:500;Sigma) or mouse OX-42 (1:200; Serotec, Oxford, UK).Sections were then incubated in a mixture of Alexa Fluor488-conjugated goat anti-rabbit IgG (1:200; MolecularProbes) and Alexa Fluor 594-conjugated goat anti-mouseIgG (1:200; Molecular Probes) at RT for 2 hr in the dark.Processed sections were mounted on gelatin-coated slides,coversliped and examined under a Zeiss Axioplan-2 micro-scope.Quantitative analysisFive coronal hippocampal sections (at the level of the injec-tion site and spaced 60 μm from each other) were used forimmunohistochemical analysis. All quantitative analyseswere performed in a blind manner. To ensure consistency intissue sampling, matched hippocampal sections (based onanatomical landmark) were always processed throughoutthe experiments. Digitized images of stained sections wereacquired using a Zeiss Axioplan-2 microscope equippedwith a DVC camera (Diagnostic Instruments, SterlingHeights, MI). These images were then analyzed usingNorthern Eclipse software (Empix Imaging, Mississauga,ON, Canada). Three microscopic fields within SP (stratumpyramidale) and Slm (stratum lacunosum moleculare) areasin CA3 were selected (magnification of ×40) in MK-stainedsections. MK immunoreactivity was then measured andexpressed as area density of MK (the fraction of the totalgiven area occupied by MK-positive cells). Three micro-scopic fields (magnification at ×40) in CA1, CA3 and hilusregion were counted for Nissl-stained undamaged neuronsin each coronal brain section. Neurons are identified bydark nucleoli within lightly stained nuclei. GAD67-positiveneurons were quantified in CA1 and CA3 subfields (strataoriens, radiatum, and pyramidale) and dentate gyrus layersitive neurons was expressed as a percentage of the PBS-injected control sections.Statistical analysisData are presented as means SEM. The statistical signifi-cance was determined by one-way ANOVA and Student-Newman-Keuls test or Student's t-test using the StatViewprogram (Abacus Concepts, Berkeley, CA). p values < 0.05were considered to be statistically significant.Authors' contributionsYBK and JKR designed the study, carried out animal experiments and draftedthe manuscript. HJL, IJL, and DP performed histology and histochemistry, andMCL participated in the design of the study and histological analysis. SUK con-ceived of the study, participated in its design and coordination, and drafted themanuscript. All authors read and approved the final manuscript.AcknowledgementsThis work was supported by grants from the National R&D Program for Cancer Control, Korean Ministry of Health and Welfare and the Canadian Myelin Research Initiative.Author Details1Division of Neurology, Department of Medicine, UBC Hospital, University of British Columbia, Vancouver, Canada, 2College of Veterinary Medicine, Chungbuk National University, Cheongju, Korea, 3Medical Research Institute, Chung-Ang University College of Medicine, Seoul, Korea, 4Department of Physiology, Chung-Ang University College of Medicine, Seoul, Korea and 5Department of Pathology, Chonnam National University Medical School, Gwangju, KoreaReferences1. Bruton CJ: The neuropathology of temporal lobe epilepsy. New York, Oxford University Press; 1988. 2. Sloviter RS: The functional organization of the hippocampal dentate gyrus and its relevance to the pathogenesis of temporal lobe epilepsy. Ann Neurol 1994, 35:640-654.3. Nadler JV, Perry BW, Cotman CW: Intraventricular kainic acid preferentially destroys hippocampal pyramidal cells. Nature 1978, 271:676-677.4. Franck JE: Dynamic alterations in hippocampal morphology following intra-ventricular kainic acid. Acta Neuropathol 1984, 62:242-253.5. Ben-Ari Y: Limbic seizures and brain damage produced by kainic acid: Mechanisms and relevance to human temporal lobe epilepsy. Neuroscience 1985, 14:375-403.6. Sperk G: Kainic acid seizures in the rat. Prog Neurobiol 1994, 42:1-32.7. Shetty AK, Turner DA: Fetal hippocampal grafts containing CA3 cells restore host hippocampal glutamate decarboxylase-positive interneuron numbers in a rat model of temporal lobe epilepsy. J Neurosci 2000, 20:8788-8801.8. Gall CM, Isackson PJ: Limbic seizures increase neuronal production of messenger RNA for nerve growth factor. Science 1989, 245:758-761.9. Mathern GW, Babb TL, Micevych PE, Blanco CE, Pretorius JK: Granule cell mRNA levels for BDNF, NGF, and NT-3 correlate with neuron losses or supragranular mossy fiber sprouting in the chronically damaged and epileptic human hippocampus. Mol Chem Neuropathol 1997, 30:53-76.10. Rauvala H: An 18-kd heparin-binding protein of developing brain that is distince tfrom fibroblast growth factors. EMBO J 1989, 8:2933-2941.11. Muramatsu H, Shirahama H, Yonezawa S, Maruta H, Muramatsu T: Midkine, a retinoic acid-inducible growth/differentiation factor: immunochemical evidence for the function and distribution. Dev Biol 1993, 159:392-402.Received: 31 August 2009 Accepted: 26 March 2010 Published: 26 March 2010This article is available from: http://www.biomedcentral.com/1471-2202/11/42© 2010 Kim et al; icensee BioMed Central Ltd. is an Open Ac ss article distributed un er th terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.BMC Neuroscience 2010, 11:42(dentate hilus, granule cell layer, and molecular layer).These regions were established according to the atlas ofPaxinos & Watson [36]. Number of Nissl- and GAD67-pos-12. Kadomatsu K, Tomomura M, Muramatsu T: cDNA cloning and sequencing of a new gene intensely expressed in early differentiation stages of embryonal carcinoma cells and in midgestation period of Kim et al. BMC Neuroscience 2010, 11:42http://www.biomedcentral.com/1471-2202/11/42Page 9 of 9mouse embryogenesis. Biochem Biophys Res Commun 1988, 151:1312-1318.13. Muramatsu H, Muramatsu T: Purification of recombinant midkine and examination of its biological activities: functional comparison of new heparin binding factors. Biochem Biophys Res Commun 1991, 177:652-658.14. Michikawa M, Kikuchi S, Muramatsu H, Muramatsu T, Kim SU: Retinoic acid responsive gene product, midkine, has neurotrphic functions in mouse spinal cord and dorsal root ganglion neurons in culture. J Neurosci Res 1993, 35:530-539.15. Satoh J, Muramatsu H, Moretto G, Chang HJ, Kim ST, Cho JM, Kim SU: Midkine that promotes survival of fetal human neurons is produced by fetal astrocytes in culture. Brain Res Dev Brain Res 1993, 75:201-205.16. Kikuchi S, Muramatsu H, Muramatsu T, Kim SU: Midkine, a novel neurotrophic factor, promotes survival of mesencephalic neurons in culture. Neurosci Lett 1993, 160:9-12.17. Kikuchi-Horie K, Kawakami E, Kamata M, Wada M, Hu JG, Ohara K, Watabe K, Oyanagi K: Distinctive expression of midkine in the repair period of rat brain during neurogenesis: immunohistochemical and immunoelectron microscopic observations. J Neurosci Res 2004, 75:678-687.18. Ratovitski EA, Kotzbauer PT, Milbrandt J, Lowenstein CJ, Burrow CR: Midkine induces tumor cell proliferation and binds to a high affinity signaling receptor associated with JAK tyrosine kinases. J Biol Chem 1998, 273:3654-3660.19. Yasuhara O, Muramatsu H, Kim SU, Muramatsu T, Maruta H, McGeer PL: Midkine, a novel neurotrophic factor, is present in senile plaques of Alzheimer disease. Brain Res Dev Brain Res 1993, 192:246-251.20. Yoshida Y, Goto M, Tsutsui J: Midkine is present in the early stage of cerebral infarct. Brain Res Dev Brain Res 1995, 85:25-30.21. Yasuhara O, Schwab C, Matsuo A, Kim SU, Steele JC, Akiguchi I, Kimura H, McGeer EG, McGeer PL: Midkine-like immunoreactivity in extracellular neurofibrillary tangles in brains of patients with parkinsonism-dementia complex of Guam. Neurosci Lett 1996, 205:107-110.22. Mochizuki R, Takeda A, Sato N, Kimpara T, Onodera H, Itoyama Y, Muramatsu T: Induction of midkine expression in reactive astrocytes following rat transient forebrain ischemia. Exp Neurol 1998, 149:73-78.23. Buckmaster PS, Jongen-Relo AL: Highly specific neuron loss preserves lateral inhibitory circuits in the dentate gyrus of kainate-induced epileptic rats. J Neurosci 1999, 19:9519-9529.24. Unoki K, Ohba N, Arimura H, Muramatsu H, Muramatsu T: Rescue of photoreceptors from the damaging effects of constant light by midkine, a retinoic acid-responsive gene product. Invest Ophthal Vis Sci 1994, 35:4063-4068.25. Yoshida Y, Ikematsu S, Moritoyo T, Goto M, Tsutsuji J, Sakuma S, Osame M, Muramatsu T: Intraventricular administration of the neurotrophic factor midkine ameliorates hippocampal delayed neuronal death following transient forebrain ischemia in gerbils. Brain Res 2001, 894:46-55.26. Nakamura E, Kadomatsu K, Yuasa S, Muramatsu H, Nabeshima T, Fan QW, Ishiguro K, Igakura T, Matsubara S, Kaname T, Horiba M, Saaito H, Muramatsu T: Disruption of the midkine gene (Mdk) resulted in altered expression of a calcium binding protein in the hippocampus of infant mice and their abnormal behavior. Genes Cells 1998, 12:811-822.27. Kim Y-B, Hur G-H: Seizure-related encephalopathy in rats intoxicated with diisopropylfluorophosphate. J Toxicol Pub Health 2001, 17:73-82.28. Houser CR, Esclapez M: Vulnerability and plasticity of the GABA system in the pilocarpine model of spontaneous recurrent seizures. Epilepsy Res 1996, 26:207-218.29. Morin F, Beaulieu C, Lacaille J-C: Selective loss of GABA neurons in area CA1 of the rat hippocampus after intraventricular kainite. Epilepsy Res 1998, 32:363-369.30. Brandt C, Potschka H, Löscher W, Ebert U: N-methyl-D-aspartate receptor blockade after status epilepticus protects against limbic brain damage but not against epilepsy in the kainate model of temporal lobe epilepsy. Neuroscience 2003, 118:727-740.31. Bertram EH, Scott C: The pathological substrate of limbic epilepsy: neuronal loss in the medial dorsal thalamic nucleus as the consistent change. Epilepsia 2000, 41(Suppl 6):S3-S8.33. Qi M, Ikematsu S, Ichihara-Tanaka K, Sakuma S, Muramatsu T, Kadomatsu K: Midkine rescues Wilms' tumor cells from cisplatin-induced apoptosis: regulation of Bcl-2 expression by midkine. J Biochem 2000, 127:269-277.34. Owada K, Sanjo N, Kobayashi T, Mizusawa H, Muramatsu H, Muramatsu T, Michikawa M: Midkine inhibits caspase-dependent apoptosis via the activation of mitogen-activated protein kinase and phosphatidylinositol 3-kinase in cultured neurons. J Neurochem 1999, 73:2084-2092.35. Benkovic SA, O'Callaghan JP, Miller DB: Sensitive indicators of injury reveal hippocampal damage in C57BL/6J mice treated with kainic acid in the absence of tonic-clonic seizures. Brain Res 2004, 1024:59-76.36. Paxinos G, Watson C: The rat brain in stereotaxic coordinates. 5th edition. Elsevier Academic Press; 2005. doi: 10.1186/1471-2202-11-42Cite this article as: Kim et al., Midkine, heparin-binding growth factor, blocks kainic acid-induced seizure and neuronal cell death in mouse hippocampus BMC Neuroscience 2010, 11:4232. Raol YS, Budreck EC, Brooks-Kayal AR: Epilepsy after early-life seizures can be independent of hippocampal injury. Ann Neurol 2003, 53:503-511.