UBC Faculty Research and Publications

Brivaracetam, but not ethosuximide, reverses memory impairments in an Alzheimer’s disease mouse model Nygaard, Haakon B; Kaufman, Adam C; Sekine-Konno, Tomoko; Huh, Linda L; Going, Hilary; Feldman, Samantha J; Kostylev, Mikhail A; Strittmatter, Stephen M May 5, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-13195_2015_Article_110.pdf [ 2.42MB ]
JSON: 52383-1.0223646.json
JSON-LD: 52383-1.0223646-ld.json
RDF/XML (Pretty): 52383-1.0223646-rdf.xml
RDF/JSON: 52383-1.0223646-rdf.json
Turtle: 52383-1.0223646-turtle.txt
N-Triples: 52383-1.0223646-rdf-ntriples.txt
Original Record: 52383-1.0223646-source.json
Full Text

Full Text

RESEARCHBrivaracetam, but not ethoaeMAlzheimer’s disease (AD), an effective disease-modifyingintervention has not yet been identified. It is now wellhave seizures [4-6]. These findings have led to the hypoth-esis that amyloid-β (Aβ), the peptide derived from APPNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 DOI 10.1186/s13195-015-0110-9peptide Y. The same drug was recently shown to improveselect hippocampal function in human subjects diagnosedUniversity School of Medicine, 295 Congress Avenue, New Haven, CT 06536, USAFull list of author information is available at the end of the articleestablished that patients with AD have an increased riskof seizures [1]. In sporadic AD, the frequency of seizuresvary considerably between studies, with more recent re-ports estimating an incidence of approximately 4 to 5 per1,000 persons per year [2,3]. Epilepsy is common in famil-ial AD, with an incidence as high as 83% in these patientsand widely believed to play a critical role in AD pathogen-esis, may trigger neuronal hyperexcitability, seizures, andultimately worsen neuronal dysfunction in AD. This hy-pothesis was partly tested in two recent studies wheretransgenic AD mice underwent chronic treatment withthe antiepileptic drug (AED) levetiracetam [7,8]. In the ini-tial report, treatment with levetiracetam was described asstrongly reducing epileptiform discharges (single spikes),ameliorating memory impairments and reversing markersof hyperexcitability, including calbindin D28 and neuro-* Correspondence: haakon.nygaard@ubc.ca; stephen.strittmatter@yale.edu1Department of Neurology, Yale University School of Medicine, 800 HowardAvenue, New Haven, CT 06510, USA2Cellular Neuroscience, Neurodegeneration, and Repair Program (CNNR), YaleIntroduction: Recent studies have shown that several strains of transgenic Alzheimer’s disease (AD) miceoverexpressing the amyloid precursor protein (APP) have cortical hyperexcitability, and their results have suggestedthat this aberrant network activity may be a mechanism by which amyloid-β (Aβ) causes more widespread neuronaldysfunction. Specific anticonvulsant therapy reverses memory impairments in various transgenic mouse strains, butit is not known whether reduction of epileptiform activity might serve as a surrogate marker of drug efficacy formemory improvement in AD mouse models.Methods: Transgenic AD mice (APP/PS1 and 3xTg-AD) were chronically implanted with dural electroencephalographyelectrodes, and epileptiform activity was correlated with spatial memory function and transgene-specific pathology.The antiepileptic drugs ethosuximide and brivaracetam were tested for their ability to suppress epileptiform activityand to reverse memory impairments and synapse loss in APP/PS1 mice.Results: We report that in two transgenic mouse models of AD (APP/PS1 and 3xTg-AD), the presence of spike-wavedischarges (SWDs) correlated with impairments in spatial memory. Both ethosuximide and brivaracetam reduce mouseSWDs, but only brivaracetam reverses memory impairments in APP/PS1 mice.Conclusions: Our data confirm an intriguing therapeutic role of anticonvulsant drugs targeting synaptic vesicle protein2A across AD mouse models. Chronic ethosuximide dosing did not reverse spatial memory impairments in APP/PS1mice, despite reduction of SWDs. Our data indicate that SWDs are not a reliable surrogate marker of appropriate targetengagement for reversal of memory dysfunction in APP/PS1 mice.IntroductionDespite significant advances in the understanding of[1]. Several groups, including ours, have shown that miceoverexpressing the amyloid precursor protein (APP) alsomemory impairments inmouse modelHaakon B Nygaard1,2,3*, Adam C Kaufman2, Tomoko SekinSamantha J Feldman1, Mikhail A Kostylev1,2 and StephenAbstract© 2015 Nygaard et al.; licensee BioMed CentraCommons Attribution License (http://creativecreproduction in any medium, provided the orDedication waiver (http://creativecommons.orunless otherwise stated.Open Accesssuximide, reversesn Alzheimer’s disease-Konno1,2, Linda L Huh1,4, Hilary Going1,Strittmatter1,2*l. This is an Open Access article distributed under the terms of the Creativeommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andiginal work is properly credited. The Creative Commons Public Domaing/publicdomain/zero/1.0/) applies to the data made available in this article,Nygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 2 of 12with amnestic mild cognitive impairment (aMCI) [9],suggesting a potential therapeutic benefit of levetirace-tam in aMCI and possibly AD.The mechanisms underlying the improvements seen inAD mice treated with levetiracetam are presumed to in-volve a reduction in neuronal excitability, and, althoughthis hypothesis has not been directly tested, targeting epi-leptiform discharges has emerged as a potential thera-peutic approach in AD [7,10]. This is supported by recentwork showing that a genetic reduction in either endogen-ous tau protein or cellular prion protein (PrPC), both ofwhich reverse impairment in spatial memory in AD mice,is associated with a reduction in aberrant neuronal activityin rodent models of AD [6,11,12]. These findings wouldsuggest that a reduction in epileptiform discharges canpredict a therapeutic reversal in spatial memory impair-ments, with reduced neuropathology, in transgenic ADmice. This would be important because behavioral testingin mice, still considered an important step in preclinicaldrug development, requires significant time and resourcesthat could be optimized by availability of a reliable surro-gate marker of drug efficacy.Using continuous in vivo electroencephalography (EEG)recording, coupled with spatial memory testing, we stud-ied whether epileptiform discharges in transgenic ADmice could be used as a marker of drug efficacy for mem-ory improvement. We report that in two transgenic ADmodels, APP/PS1 [13] and 3xTg-AD [14], the presence ofspike-wave discharges (SWDs) correlate with impairmentsin spatial memory, although a weaker correlation was seenin 3xTg-AD mice. Biochemical and immunohistochemicalanalyses indicated that these epileptiform dischargeswere not associated with changes in Aβ metabolism ordeposition. We further demonstrate that SWDs can besuppressed by the AEDs ethosuximide and brivaracetam,with no effect seen when phenytoin was used. Interestingly,brivaracetam, but not ethosuximide, reversed memorydeficits in APP/PS1 mice, despite both drugs causing astrong reduction in epileptiform discharges. Our dataindicate that SWDs are associated with poor cognitiveperformance in APP/PS1 mice, but that the reductionof this abnormal network activity does not reliably pre-dict therapeutic reversal of age-associated impairmentsin spatial memory in this mouse model. We confirmthat targeting synaptic vesicle protein 2A (SV2A), whichresults in broad-spectrum anticonvulsant action, reversesmemory impairments in the APP/PS1 model of AD.MethodsMiceThe use of mice in this study was approved by the YaleAnimal Resources Center according to internationally rec-ognized guidelines. All mice were housed with a 12-hourlight/12-hour dark cycle and fed ad libitum. Coinjectedcongenic APPswe/PSEN1dE9 transgenic mice [13] on apure C57BL/6J background were obtained from TheJackson Laboratory (Bar Harbor, ME, USA). 3xTg-ADmice were a gift from Dr Frank LaFerla (UC Irvine,CA, USA) and were obtained via Dr Paul Lombroso(Yale University). They express the mutated knockingene PS1M146V, as well as APPswe and tauP301L, atthe same locus, both under control of the mouse Thy1.2regulatory element [14]. 3xTg-AD mice were on a mixedC57BL/6J × 129/Sv background as described elsewhere[14]. For chronic drug experiments, mouse groups weresex-matched, with 40% to 60% of each sex in differentcohorts.Behavioral studiesFor experiments correlating epileptiform activity, ani-mals underwent EEG monitoring prior to behavioraltesting. Animals tested for behavior during treatmentwith AEDs did not have EEG electrodes implanted.Animals were randomized, and the experimenter wasblinded to genotype for the duration of behavioral testing.Morris water maze testing [15] based on previously de-scribed methods [12] was performed over the course of3 days. Each swim was performed at room temperature inan open-water pool approximately 1.3 m in diameter, util-izing a submerged, nonvisible escape platform located inthe center of one of the pool’s four quadrants. This loca-tion remained constant for the 3 days of testing. Over thecourse of each testing day, an animal swam a total of eighttimes—four times in the morning, constituting one“block” of swims, and four times in an afternoon block.The interval between blocks was approximately 2 hours.For each block, the mice would begin their swim in one offour distinct locations around the wall of the pool andwere timed for its latency and path length to reach theescape platform for a maximum time of 1 minute. Ifthe mouse did not find the submerged platform by1 minute, it was placed on the platform for approxi-mately 10 seconds before being removed from the pool.The water maze probe trial was performed 48 hoursfollowing the third and last day of the memory acquisi-tion phase and in the same 1.3-m pool describedabove. For the purposes of the probe trial, the platformwas removed from the pool. All mice were startedfrom a location opposite to the platform location andallowed to swim for 1 minute. To ensure that all micewere equal in terms of swim speed, motivation and visualacuity, a block of five swims to a visible platform was con-ducted after the probe trial. Mice were excluded from thestudy if the latency to the visible platform exceeded 3standard deviations above the average latency for controlmice, as previously described [16]. By this criterion, oneAPP/PS1 mouse with SWDs was excluded. Latency to theplatform, swim speed, path length and resting time wereNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 3 of 12automatically recorded using Panlab SMART videotracking and analysis program, v2.5 (Panlab, Cornellàde Llobregat, Spain).Brain tissue collectionMice were deeply anesthetized with isoflurane and imme-diately perfused with ice-cold 0.9% NaCl for 2 minutes.Their brains were then dissected out and placed in ice-cold 0.9% NaCl. For biochemical analysis, the righthemibrain was weighed and immediately frozen in liquidnitrogen, followed by storage at −80°C. To extract thesoluble cytosolic fraction, the brains were homogenizedin 3 volumes (w/v) of 50 mM Tris-HCl, 150 mM NaCl,pH 7.6 (TBS), containing a protease inhibitor cocktail(cOmplete Protease Inhibitor Cocktail, catalog number10745000; Roche Diagnostics, Mannheim, Germany),1 mM sodium orthovanadate and 50 mM sodium fluoride.Tissue was homogenized using an ultrasonic cell disruptor(Branson Ultrasonics Corporation, Danbury, CT, USA) andultracentrifuged at 100,000 × g for 20 minutes at 4°C.The pellet was then resuspended to the same volumeas the original homogenate in TBS with 2% TritonX-100 (AmericanBio, Natick, MA, USA), 0.1% SDS(AmericanBio), a protease inhibitor cocktail (cOmpleteProtease Inhibitor Cocktail), 1 mM sodium orthovanadateand 50 mM sodium fluoride. Tissue was homogenized andultracentrifuged at 100,000 × g for 20 minutes. The super-natant was mixed with 4× SDS-PAGE loading buffer, boiledfor 5 minutes and stored for subsequent analysis.ImmunohistochemistryOne hemibrain was immersed in fresh 4% paraformalde-hyde (PFA) overnight. After the brains were fixed, theywere embedded in 10% gelatin and placed in 4% PFA for20 hours at 4°C. Parasagittal sections (30 μm) were thencut using a Leica VT1000 S vibratome (Leica Biosystems,Buffalo Grove, IL, USA). For immunohistochemistry,sections were blocked in 10% donkey serum for 1 hour,followed by incubation with primary antibody over-night at room temperature. Primary antibodies werediluted in phosphate-buffered saline (PBS) with 0.2%Triton X-100 (AmericanBio). The following antibodieswere used: Aβ antibody (catalog number 2454, CellSignaling Technology, Danvers, MA, USA; and clone6E10, monoclonal antibody 1560, EMD Millipore, Billerica,MA, USA: both diluted 1:250), rabbit anti-PSD-95 poly-clonal antibody (1:250 dilution, catalog number 51-6900;Invitrogen, Camarillo, CA, USA) and anti-calbindin D28antibody (1:1,000 dilution; Swant, Marly, Switzerland).Following incubation, the sections were washed threetimes with PBS and incubated in Alexa Fluor fluorescentsecondary antibody (donkey anti-rabbit or anti-mouse,all at 1:500 dilution; Invitrogen) for 2 hours at roomtemperature. The slices were then washed three timesand transferred to PBS. Sections were also stained withsecondary antibody alone to rule out nonspecific staining.Each free-floating section was mounted on a microscopeslide (Fisherbrand Superfrost Plus; Fisher Scientific,Pittsburgh, PA, USA) and coverslipped using VECTA-SHIELD mounting medium (H-1000; Vector Laboratories,Burlingame, CA, USA).Imaging and analysisAll images and analyses were generated by personnelwho had no knowledge of the mouse genotype. Aβ imageswere obtained using a Zeiss Axio Imager Z1 fluorescencemicroscope (Carl Zeiss Microscopy, Jena, Germany) witha 10× lens objective. Mosaic images of the entire cortexand hippocampus of each animal were obtained and ana-lyzed, and plaque burden was calculated using ImageJsoftware (National Institutes of Health, Bethesda, MD,USA). This was done by isolating the cortex or hippocam-pus, thresholding to a standard value and calculatingthe area occupied. Using an UltraVIEW VoX spinningdisc confocal microscope (PerkinElmer, Waltham, MA,USA), hippocampal PSD-95 immunoreactive punctawere imaged with a 60× lens objective and digitallymagnified to × 100. Two images were obtained in themolecular layer of the dentate gyrus with two slicesfrom each mouse analyzed. Puncta from the dentategyrus were analyzed and counted using ImageJ, exclud-ing cell somata. Hippocampal calbindin D28 imageswere obtained using a Zeiss Axio Imager Z1 fluores-cence microscope with a 20× lens objective. Mosaicimages of the entire hippocampus of each animal wereobtained and analyzed. All histologic analyses weredone using ImageJ and analyzed statistically by Student’st-test, or by analysis of variance (ANOVA) with post hoccomparisons as indicated, using SPSS software (IBM,Armonk, NY, USA).Immunoblotting and enzyme-linked immunoassayexperimentsPrecast 10% Tris-glycine or 10–20% Tris-tricine gelswere used (Bio-Rad Laboratories, Hercules, CA, USA).After transfer, the polyvinylidene fluoride membranes(catalog number 162-0174; Bio-Rad Laboratories) wereincubated in blocking buffer for 1 hour at room temperature(catalog number 927-4000, Odyssey blocking buffer;LI-COR Biosciences, Lincoln, NE, USA). Membraneswere then washed five times in a mixture of Tris-bufferedsaline and Tween 20 (TBST) and incubated overnight inprimary antibodies. The following antibodies were used:clone 6E10 (MAB1560, 1:1,000 dilution; EMD Millipore),clone 22C11 (MAB348, 1:100 dilution; EMD Millipore)and actin (catalog number sc-1616, 1:200 dilution; SantaCruz Biotechnology, Santa Cruz, CA, USA). All antibodieswere diluted in Odyssey blocking buffer, and membranesNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 4 of 12were incubated overnight at 4°C. Following primaryantibody incubation, the membranes were washed fivetimes with TBST, and secondary antibodies were ap-plied for 1 hour at room temperature (Odyssey donkeyanti-mouse or anti-goat IRDye (LI-COR Biosciences)at 680 or 800 nm). Membranes were then washed, andproteins were visualized using a LI-COR Odyssey In-frared imaging system. Blots were analyzed using Ima-geJ and normalized to actin optical density. Total Aβenzyme-linked immunosorbent assay (ELISA) experi-ments on TBS-soluble mouse brain lysates were per-formed according to the manufacturer’s instructions(Invitrogen).Continuous electroencephalography video monitoringFor dural electrode implantation, the mice anesthetizedand maintained with inhaled isoflurane and mounted ina stereotaxic frame (David Kopf Instruments, Tujunga,CA, USA). A midline incision was made, and two bilat-eral burr holes were manually drilled anterolateral andposterolateral to the bregma. Four presoldered intracra-nial screw electrodes (catalog number 8403; PinnacleTechnology, Lawrence, KS, USA) or a prefabricatedheadmount (catalog number 8201; Pinnacle Technology)was inserted and secured with a layer of dental cement(catalog number 526000; A-M Systems, Sequim, WA,USA). In the case of presoldered screw electrodes, theelectrode wires were soldered onto a six-pin surfacemount connector (catalog number 8235-SM; PinnacleTechnology) and secured by a final layer of dental cement.All mice were allowed to recover for 7 days prior tochronic EEG recordings.Mice were video-recorded using an in vivo EEG videomonitoring system (8200-K1-SE3, 8236; Pinnacle Systems).EEGs were sampled at 400 Hz with 100× preamplifier gainand filtered at 30 Hz. Each mouse underwent 72 hours ofcontinuous EEG video recording and was maintained on aregular 12-hour light/12-hour dark cycle with full access tofood and water. EEG traces were scored manually by an in-vestigator, blinded to genotype, using Pinnacle Technologysoftware. A convulsive seizure was defined as an abruptonset of evolving SWDs lasting >30 seconds, associatedwith tonic-clonic activity by synchronized video analysis,and followed by postictal attenuation of cerebral EEGrhythms. A SWD was defined as a burst of sharp-wavedischarges, with an amplitude of at least twice the back-ground amplitude and 1 to 2 seconds in duration. A singlespike was defined as a sharp discharge at least twice thebackground amplitude and <100 milliseconds in duration.SWDs were manually correlated with synchronized videoanalysis and scored as with or without behavioral arrest.Twenty-four-hour EEGs were manually scored for singlespikes. A full 72 hours of EEG were manually screened forSWDs for each mouse, comprising 24 hours before drugdelivery and 48 hours afterward. Epileptiform dischargeswere analyzed using Student’s t-test.Drug administrationEach mouse received a single intraperitoneal (IP) injectionof drug as indicated. All drugs were dissolved in normalsaline. Each mouse underwent a 1-week washout withverification of a return to baseline SWD frequency priorto subsequent drug injection. Each mouse first received anIP injection of levetiracetam, followed by ethosuximide,phenytoin and brivaracetam. For long-term drug delivery,ethosuximide was delivered in the drinking water at a con-centration of 30 mg/ml. Brivaracetam was continuouslyadministered IP for 28 days using an osmotic minipump(ALZET Osmotic Pumps, Cupertino, CA, USA) at a rateof 8.5 mg/kg/day. Owing to the short half-life of ethosuxi-mide in mice (1 hour) [17], periodic injections were notfeasible. Minipump infusions could not be used, becausethe amount required exceeded the solubility of ethosuxi-mide. Therefore, drinking water was chosen as the routeof administration for ethosuximide. For brivaracetam, anosmotic minipump is the most reliable route of adminis-tration for continuous dosing.ResultsTransgenic Alzheimer’s disease mice have frequentepileptiform dischargesRecently, authors have reported various types of epilepti-form discharges in transgenic AD mice overexpressingAPP. In the J20 model (APPswe,Ind), both generalizedseizures and single spikes were reported [4], and APP/PS1 mice were found to have single spikes, clusters ofSWDs and generalized seizures, when compared withnontransgenic littermate controls [5,18]. We have previ-ously shown that 40% of aged APP/PS1 mice have convul-sive seizures when recorded for 72 hours by continuousEEG video monitoring (Figure 1A) [6]. To furthercharacterize epileptiform activity in transgenic ADmice, we assessed nineteen 8- to 10-month-old APP/PS1mice using long-term in vivo EEG video monitoring. Inaddition to convulsive seizures, 9 (47%) of 19 APP/PS1mice had frequent clusters of SWDs, compared with 0 of8 of their wild-type littermates (Figure 1B,D). Using syn-chronized video analysis, a total of 240 hours of EEG wereanalyzed, and freezing behavior during the SWDs thatmight interfere with memory testing was quantified.Overall, only 82 SWDs were associated with brief behav-ioral arrest for the duration of the SWD, for a rate of lessthan one arrest per hour. In contrast with previous au-thors, we show that the frequency of single spikes isnot transgene-dependent and that APP/PS1 mice donot differ from their wild-type littermates (Figure 1C,E).To assess EEG characteristics in a second AD mousemodel, a limited cohort of ten 3xTg-AD mice aged 8 toNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 5 of 1210 months underwent continuous in vivo EEG recordingfor 24 hours as described for APP/PS1 mice. None of the3xTg-AD mice had convulsive seizures during the record-ing period. Four (40%) of ten of the 3xTg-AD mice hadSWDs over a 12-hour period (11 ± 6 SWDs/hr). On thebasis of EEG morphology, no differences in SWDs wereobserved between mouse strains, and the frequency ofSWDs was not found to be significantly different betweenthe two mouse models (APP/PS1 mean: 5 ± 1 SWDs/hrversus 3xTg-AD mean: 11 ± 6 SWDs/hr; P= 0.2 by Student’st-test). Owing to the high frequency and concordancebetween transgenic lines, we focused on SWDs as theprimary manifestation of epileptiform activity.Spike-wave discharges correlate with impairments inspatial memory in APP/PS1 and 3xTg-AD miceAlthough several groups have reported epileptiform dis-charges in AD mice, it is not known to what extent thesedischarges affect the phenotypic manifestations in transgenicFigure 1 APP/PS1 mice have frequent epileptiform discharges. (A) Sp(SWD) (arrows) and (C) single spike in a 10-month-old APP/PS1 mouse (arrofollowing single spike. (D) Quantification of SWDs in APP/PS1 mice compareddoes not show a difference between APP/PS1 mice and WT littermates (n = 8and three males; APP/PS1 mice with SWDs, four males and five females;SWD = 200 μV/0.5 s; spike = 200 μV/1 s.models. We correlated the presence of SWDs in APP/PS1and 3xTg-AD mice with spatial memory function as mea-sured with the Morris water maze test. In APP/PS1 mice,the presence of SWDs measured prior to memory testingwas associated with worsened performance in the acquisi-tion phase of the Morris water maze (Figure 2A). Analysisof swim path length corroborated these findings (Additionalfile 1: Figure S1A). After 48 hours, the mice were tested forlong-term memory in the probe trial. A modest inverse rela-tionship between the number of entries in the correct targetarea and the number of SWDs was seen (Figure 2B). Similarfindings were seen in 3xTg-AD mice, with the presenceof SWDs correlating with spatial memory performance(Figure 2C). However, we did not see a correlation inthe delayed probe trial between SWDs and the numberof correct target entries in 3xTg-AD mice (Figure 2D),and path length analysis did not indicate a differencein memory performance of 3xTg-AD mice with SWDscompared with mice without them (Additional file 1:ontaneous generalized convulsive seizure, (B) spike-wave dischargews). The arrowheads indicate normal electroencephalogram backgroundwith their wild-type (WT) littermates. (E) Quantification of single spikesfor WT, 19 for APP/PS1). *P < 0.05 by Student’s t-test. WT, five femalesAPP/PS1 mice without SWDs, six females and four males. Calibration:Nygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 6 of 12Figure S1D). Thus, the correlation between SWDs andspatial memory performance was less robust in 3xTg-ADtransgenic mice compared with APP/PS1 mice. Althoughrare, SWDs can be associated with behavioral arrest thatcould interfere with the results of the Morris water mazetest. To assess whether reduced latency to the platformobserved in the transgenic AD mice with SWDs was dueto excessive freezing or reduced swim speed, we measuredswim speed and average resting time in addition to plat-form latency. The swim speed and rest times did not differbetween groups (Additional file 1: Figure S1).Spike-wave discharges do not affect amyloid-β metabolismof plaque deposition in APP/PS1 miceHaving shown that the presence of SWDs correlates withimpairments in spatial memory, we assessed whetherFigure 2 Presence of spike-wave discharges correlates with impairmeof transgenic mice underwent Morris water maze testing immediately afterpresence of more than one spike-wave discharge (SWD) worsened the perof the Morris water maze (A), and the same was true of 3xTg-AD mice (C)with post hoc analysis). (B) and (D) A 48-hour probe trial showed an inversarea, defined as the platform area, during a 1-minute trial in APP/PS1 micePearson correlation coefficient). WT: n = 8; APP/PS1 with SWDs: n = 8; APPfour males. v1 through v4 indicate visible platform swim trials.SWDs also impact biochemical and histologic mea-sures in transgenic AD brain, including APP metabo-lites and Aβ levels. Immunoblotting of soluble anddetergent extracts of brain homogenates revealed nocorrelation between SWDs and levels of soluble APP-α,β-C-terminal fragment or Aβ in APP/PS1 or 3xTg-ADmice (Figure 3A,D,F,I). In addition, neither cortical norhippocampal deposits of insoluble Aβ plaque differedwith regard to the presence of SWDs in APP/PS1 mice(Figure 3B,C,E). Although no cortical plaques wereseen in 3xTg mice at 8 to 10 months of age, the pres-ence or absence of SWDs was not associated with hip-pocampal plaque density in these mice (Figure 3G,H,J).Hippocampal calbindin D-28K was first reported to bereduced in human AD several decades ago [19] andplays a role in normal hippocampal physiology as annts in spatial memory in APP/PS1 and 3xTg-AD mice. Both strainscontinuous in vivo electroencephalography recording. (A) and (C) Theformance of 8- to 10-month-old APP/PS1 mice in the acquisition phase(*P = 0.039 and **P = 0.002 by repeated-measures analysis of variancee relationship between frequency of SWDs and entries into the target(B), but not in 3xTg-AD mice (D) (P = 0.005 and 0.39, respectively;/PS1 without SWDs: n = 9. 3xTg-AD mice included six females andNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 7 of 12intracellular calcium buffer [20]. Its link to epilepsycomes from the finding that patients with epilepsyhave a loss of calbindin D28 in several areas of thehippocampus, and these changes have been proposedto affect the plasticity changes associated with themaintenance of the epileptic phenotype [21]. In studiesin the J20 model of AD, researchers have reported a de-crease in hippocampal calbindin D28 in the hippocampus,which is thought to reflect neuronal hyperexcitability [4].In contrast to studies in J20 mice, in our present study wedid not detect a reduction of calbindin D28 in APP/PS1mice compared with WT littermates (data not shown).Ethosuximide and brivaracetam reduce spike-wave dischargesin Alzheimer’s disease miceBecause epileptiform discharges are relatively frequentin APP/PS1 mice and correlated with spatial memoryperformance, we hypothesized that the reduction ofthese discharges might predict the ability of anticonvul-sant drugs to reverse spatial memory deficits in thismouse model. In a similar approach to that reportedby Sanchez et al. [7], we screened several AEDs fortheir ability to reduce SWDs, including phenytoin,Figure 3 The presence of spike wave discharges is not associated witlevels. In APP/PS1 mice, the presence of spike-wave discharges (SWDs) didmarker), β-C-terminal fragment (β-CTF) (15 kDa marker) or amyloid-β (Aβ) mcontrol is indicated by the 37 kDa marker. In the same strain, Aβ plaque de3xTg-AD mice, the presence of SWDs did not alter levels of sAPP-α or β-CTplaque (G, H, J). Data were analyzed by two-tailed Student’s t-test. APP/PSno SWDs. HC, Hippocampus; C, Cortex; PSEN, Presenelin; WT, Wild type.levetiracetam, brivaracetam and ethosuximide. Thesedrugs were chosen because both phenytoin and levetirac-etam have been used in AD mice previously [7,8,18] andbecause the SWDs reported here show similarities toethosuximide-sensitive SWDs seen in the C3H/He mousemodel of absence seizures [22]. APP/PS1 and 3xTg-ADmice at 8 to 10 months of age underwent continuousin vivo EEG recording for 72 hours. After a 24-hourbaseline EEG, mice were given a single IP injection ofdrug, followed by quantification of SWDs before andafter injection. In contrast to a previous report [18],phenytoin (20 mg/kg) did not acutely decrease SWDs ineither APP/PS1 mice or 3xTg-AD mice (Figure 4A,E).Ethosuximide (200 mg/kg) showed the strongest reduc-tion of SWDs in both APP/PS1 (93% ± 4) and 3xTg-AD(83% ± 5) mice, with almost complete elimination ofSWDs in the first several hours after the loading dose(Figure 4B,F). Levetiracetam (20 mg/kg) reduced SWDsby 45% ± 12 in APP/PS1 mice and by 61% ±12 in3xTg-AD mice (Figure 4C,G). Brivaracetam (10 mg/kg)reduced SWDs by 41% ± 7 in APP/PS1 mice, with a trendtoward reduced frequency even at 24 hours postdose(Figure 4D).h changes in amyloid precursor protein metabolism or amyloid-βnot alter levels of soluble amyloid precursor protein (sAPP)-α (100 kDaonomers (5 kDa marker) by Western blot analysis (A, D). Actin loadingposits were not altered by the presence of SWDs (B, C, E). Similarly, inF as analyzed by Western blotting (F, I), nor did it alter levels of Aβ1, n = 11; 3xTg-AD, n = 11. +Indicates the presence of SWDs, and− indicatesrg(PHb1: −(B,P/PPNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 8 of 12Figure 4 Ethosuximide and brivaracetam reduce spike-wave dischadisease (AD) mice were given a single injection of 20 mg/kg phenytoinlevetiracetam (LVT) (C, G) or 10 mg/kg brivaracetam (Briva) (D), followedno reduction in SWDs at 0 to 5 hours were seen in either strain (A, E) (APP/PSP= 0.23). Ethosuximide strongly reduced SWDs at 0 to 5 hours in both strainsLevetiracetam reduced SWDs at 0 to 5 hours in both strains (C, G) (APreduced SWDs in APP/PS1 mice by 41%± 7, *P < 0.01, with an R2 of 0.95. AStudent’s t-test. n.s., Not significant.Brivaracetam, but not ethosuximide, reversesimpairments in spatial memory in APP/PS1 miceHaving demonstrated that both brivaracetam and etho-suximide significantly reduce SWDs, we tested whetherthis reduction in epileptiform activity could accuratelypredict therapeutic reversal of impairments in spatialmemory in aged APP/PS1 mice. To assess the role ofbrivaracetam in APP/PS1 mouse phenotypes, we treated13-month-old mice chronically, measuring the effect ofdrug therapy on spatial memory, Aβ levels, synapse lossand hippocampal calbindin D28 immunoreactivity. ALZETosmotic minipumps were implanted into the APP/PS1 andWT mice, and the mice received continuous IP dosing of8.5 mg/kg/day of brivaracetam versus saline. After 28 days,mice were tested in the Morris water maze while drugdelivery was continued. Chronic brivaracetam therapyfully reversed memory impairments in APP/PS1 mice(Figure 5A,B; Additional file 2: Figure S2), but it didnot change the brain concentration of soluble Aβ orinsoluble plaque (Figure 6A,B). Despite the improvedmemory performance with brivaracetam, synapse dens-ity was not recovered (Figure 6C). Treatment with bri-varacetam did not affect hippocampal calbindin D28immunoreactivity (data not shown).To assess the role of ethosuximide in APP/PS1mouse phenotypes, we treated 16-month-old mice chron-ically (45 days), followed by measurements of spatialmemory, APP metabolism and Aβ levels, synapse loss,es in transgenic Alzheimer’s disease mice. Transgenic Alzheimer’sT) (A, E), 200 mg/kg ethosuximide (ETX) (B, F), 20 mg/kgy hourly quantification of spike-wave discharges (SWDs). For phenytoin,6%±17 (increased SWDs), P= 0.75; 3xTg-AD: −61%±76 (increased SWDs),F) (APP/PS1: 93% ± 4, ***P < 0.0001; 3xTg-AD: 83% ± 5, ***P < 0.0001).PS1: 45% ± 12, *P < 0.01; 3xTg-AD: 61% ± 24, **P = 0.002). Brivaracetam/PS1, n = 4; 3xTg-AD, n = 4. P-values were calculated by paired two-tailedand hippocampal calbindin D28 immunoreactivity. Drugwas delivered by dissolving ethosuximide in drinkingwater (30 mg/ml). This dose yields a chronic plasma drugconcentration of 55 μg/ml after 1 week of dosing in a sep-arate dose range test. In the 1-week dose range test, SWDsdetected by EEG were reduced from 25/hr to 6/hr, a 76%reduction. Chronic therapy with this dosage over 45 daysdid not reverse impairments in spatial memory in APP/PS1 mice (Figure 5C,D; Additional file 2: Figure S2), nordid this treatment affect soluble Aβ levels or Aβ plaque(Figure 6D,E). Ethosuximide did not reverse the loss ofhippocampal PSD-95-positive puncta (Figure 6F) orhippocampal calbindin D28 immunoreactivity (data notshown). Thus, although both brivaracetam and etho-suximide significantly reduced SWDs in APP/PS1 mice,only brivaracetam reversed memory impairments in thismodel.DiscussionSeizures and epileptiform discharges have been observedin several strains of AD mice, including J20 and APP/PS1 transgenic models [4,6]. In the former model, sei-zures and single spikes have been reported, whereas thelatter model also displays longer runs of epileptiformdischarges resembling SWDs seen in models of absenceepilepsy [18,23]. It is widely believed that seizures andepileptiform discharges play a role in the pathophysi-ology of AD. Chronic treatment with the anticonvulsantsNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 9 of 12levetiracetam and topiramate reverses impairments inspatial memory in J20 and APP/PS1 AD mice and mayaffect the dynamics of both Aβ and tau protein [7,8].Moreover, genetic knockouts found to reverse the patho-logic phenotypes in AD mice also eliminate cortical hy-perexcitability, including the reduction of tau proteinand removal of PrPC [6,11]. A low dose of levetiracetamwas recently shown to reduce hippocampal hyperactivityduring encoding processes in patients with aMCI [9].This reduction showed slight improvements in selecthippocampal function, suggesting that neuronal hyper-activity in aMCI may be a pathologic rather than com-pensatory response to neurodegeneration and reducedconnectivity. Thus, accumulating indirect evidence sug-gests that cortical hyperactivity may play an importantrole in the pathophysiology of AD, making chronic EEGrecordings a promising marker of target engagementand efficacy for new drugs for AD.We found that SWDs constitute the most frequent ep-ileptiform discharges in APP/PS1 mice, in contrast tothe J20 mice, in which single spikes seem to predomin-ate [4]. SWDs correlate with worsened memory per-formance in APP/PS1 mice, which we considered as aFigure 5 Brivaracetam, but not ethosuximide, reverses impairments iadministered brivaracetam by osmotic minipump or continuous delivery obrivaracetam (Briva) fully reversed memory impairments in APP/PS1 mice inadministration of ethosuximide did not improve performance in the MBriva cohort, wild-type (WT) + vehicle: n = 11; WT + Briva: n = 11, APP/PSWT + vehicle: n = 7, WT + ETX: n = 7, APP + vehicle: n = 6, APP + ETX: n = 7post hoc comparisons. V1 through V4 indicate visible platform swim trialpromising feature for a possible surrogate marker ofboth disease and drug efficacy. However, the pharmaco-logic elimination of SWDs does not consistently predictimprovements in spatial memory. Indeed, although bothethosuximide and brivaracetam significantly reducedSWDs in APP/PS1 mice, only the latter reversed impair-ments in spatial memory performance in these mice.These findings suggest that a reduction in SWDs does notrepresent a robust surrogate marker of drug efficacy inAPP/PS1 mice. Our data further emphasize the role ofdrugs targeting SV2A, such as levetiracetam and brivara-cetam, in reversing spatial memory impairments acrossseveral AD mouse strains [7]. Further, our conclusion thatchronic ethosuximide administration does not reversememory impairments in APP/PS1 mice is important as,apart from its antiepileptic effects, ethosuximide has pre-viously been shown to have neuroprotective propertiesand thus is seen as a candidate to alleviate aging and age-related disease. In a screen of compounds that affect lon-gevity in the Caenorhabditis elegans model, ethosuximidewas found to extend lifespan by an average of 17% [24].The underlying mechanism was later found to be modula-tion of sensory perception by ethosuximide with reducedn spatial memory in APP/PS1 mice. Aged APP/PS1 mice weref ethosuximide (ETX) via drinking water. Four-week administration ofthe Morris water maze (A) and Probe Trial (B). Chronic (7-week)orris water maze or probe trial in APP/PS1 mice (C, D). For the1 + vehicle: n = 15, APP/PS1 + Briva: n = 16. For chronic ETX therapy,. *P < 0.05, ***P < 0.001, repeated-measures analysis of variance withs. + Indicates drug therapy; - indicates vehicle.Nygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 10 of 12sensorineural activity [25]. Ethosuximide has also beenshown to prevent cochlear injury in a mouse model ofsensorineural hearing loss, again linked to reducing neur-onal activity [26]. Despite these interesting findings, etho-suximide does not appear to have a therapeutic effect inthe APP/PS1 model of AD.Several limitations of our data must be considered.We focused on SWDs as these are the most frequent ep-ileptiform discharges unique to APP/PS1 mice comparedwith their nontransgenic littermates. However, it is notyet known which, if any, epileptiform discharges pre-dominate in patients with AD. We note that Sanchezet al. [7] reported single spikes as the predominant epilep-tiform activity in the J20 mouse model of AD. Although itis likely these findings represent strain differences, theirimportance in AD pathophysiology is unclear. Seizureshave been studied for decades in humans with AD, but thepresence of epileptiform discharges, which requires EEGFigure 6 Chronic brivaracetam or ethosuximide do not alter amyloid pThirteen-month old APP/PS1 mice were treated with brivaracetam (Briva) (soluble amyloid-β (Aβ) levels (A, D), Aβ plaque density (B, E) or dentate gyrassay showed no effect of brivaracetam (A) or ethosuximide (D) on total Adeposits of Aβ plaque were not different between treatment groups (Studenthe dentate gyrus by immunohistochemistry. Drug therapy did not reverse sypost hoc comparisons). + Indicates drug therapy; - indicates vehicle. n.s., Not srecordings, are not well characterized. In the largest studyto date, researchers examined routine EEG recordings from1,674 patients with various cognitive disorders, including510 with AD and 225 with MCI [27]. Of the former, 2%had epileptiform discharges on routine EEG, the same asthe percentage seen among patients with “subjective com-plaints.” There was no correlation between the presence ofepileptiform discharges and performance on bedside neuro-psychological testing, and it was concluded that routineEEG could not be recommended as part of routine clinicalworkup in AD. In another study, the investigators reporteda frequency of epileptiform discharges of 16% in patientswith AD [2]. In a more recent report, authors showed a fre-quency of 62% in aMCI and AD patients known to haveepilepsy and 6% in patients without seizures [28]. The pres-ence of epileptiform discharges also predicted earlier cogni-tive decline [28]. None of the studies published to datehave differentiated various types of epileptiform discharges,recursor protein metabolism or reduce synapse loss in APP/PS1 mice.A–C) or ethosuximide (ETX) (D–F) as in Figure 5 and then analyzed forus PSD-95 synaptic area (C, F). An Aβ enzyme-linked immunosorbentβ monomers (P > 0.05 by Student’s t-test), and cortical and hippocampalt’s t-test) (B, E). Synaptic puncta were quantified in the molecular layer ofnapse loss seen in APP/PS1 mice (P > 0.05 by analysis of variance withignificant; WT, Wild type.Arch Neurol. 2009;66:435–40.Nygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 11 of 12thus limiting the correlation to SWDs reported here. Thecurrent evidence would suggest that epileptiform activity isless prominent in sporadic AD than in mouse models ofautosomal dominant disease. However, prospective studieswith long-term EEG monitoring are needed to furthercharacterize cortical hyperexcitability in AD, the relation-ship of EEG profiles to AD pathophysiology, and whetherthe presence and reduction of epileptiform discharges mayrepresent a marker of drug efficacy.Our primary objective was to establish whether epilep-tiform discharges could be used as a marker for overalldrug efficacy in improving memory function in an ADmouse model. Thus, we did not test whether a subgroupof mice, all displaying SWDs, would respond better toethosuximide therapy compared with a mixed groupwith varying SWD frequencies. Our findings do suggestthat a reduction in SWDs is not sufficient to reversememory impairments in APP/PS1 mice, but future stud-ies using a different experimental design are required toextend the generalizability of this finding. We also notethat both J20 and APP/PS1 mice have prominent epi-lepsy, and our divergent findings with ethosuximide andbrivaracetam with respect to reversal of impairments inspatial memory may be explained by the efficacy oftreating partial versus generalized seizures. Ethosuximideis exclusively used for absence seizures in humans, witha narrow antiepileptic spectrum, whereas brivaracetamhas broad antiepileptic action.ConclusionsOur study is the first to demonstrate efficacy of brivara-cetam in treating impairments in spatial memory in ADmice. Brivaracetam interacts with SV2A, and is closelyrelated to the widely used anticonvulsant levetiracetam.As noted, two previous studies in J20 and APP/PS1 micehave shown clear benefits of levetiracetam in reversingmemory impairments in this model, suggesting that tar-geting SV2A alleviates AD symptoms across AD models.We also show that, despite some promise as a neuropro-tective agent in other model systems, chronic ethosuxi-mide treatment does not reverse impairments in spatialmemory in APP/PS1 mice. Moreover, whereas SWDs inAPP/PS1 mice correlate with impairments in spatialmemory, the reduction of these discharges is not a reli-able surrogate marker of preclinical drug efficacy in theAPP/PS1 AD mouse model.Additional filesAdditional file 1: Figure S1. Extended Morris water maze analysis ofAPP/PS1 and 3xTg-AD mice. Path length, average swim speed and timespent resting were analyzed as described for platform latency in Figure 2.The presence of >1 SWDs worsened performance of 8- to 10-month-oldAPP/PS1 mice in the acquisition phase of the Morris water maze using pathlength analysis (A) (*P < 0.05 by repeated-measures ANOVA with leastsignificant difference post hoc analysis). Path lengths were not differentin 3xTg-AD mice with or without SWDs (D). Swim speed and time spentresting were even across mouse cohorts (B, C, E, F).Additional file 2: Figure S2. Extended Morris water maze analysis ofAPP/PS1-treated mice with chronic brivaracetam or ethosuximide. Pathlength, average swim speed and time spent resting were analyzed asdescribed for platform latency in Figure 5. Path length was significantlyshorter in APP/PS1 mice treated with brivaracetam compared withAPP/PS1 mice that were on vehicle therapy (A) (P < 0.001 by repeated-measures ANOVA with post hoc comparisons). Chronic ethosuximidetreatment did not alter path length in the Morris water maze (D). Neitherbrivaracetam nor ethosuximide treatment affected average swim speed oraverage resting time across mouse cohorts (B, C, E, F). + Indicates drugtherapy; - indicates vehicle.AbbreviationsAβ: Amyloid-β; AD: Alzheimer’s disease; AED: Antiepileptic drug;aMCI: Amnestic mild cognitive impairment; ANOVA: Analysis of variance;APP: Amyloid precursor protein; β-CTF: β-C-terminal fragment;EEG: Electroencephalography; ELISA: Enzyme-linked immunosorbent assay;ETX: Ethosuximide; IP: Intraperitoneal; PBS: Phosphate-buffered saline;PFA: Paraformaldehyde; PHT: Phenytoin; PrPC: Cellular prion protein;SWD: Spike-wave discharge; SV2A: Synaptic vesicle protein 2A; TBS: 50 mMTris-HCl, 150 mM NaCl; TBST: Tris-buffered saline and Tween 20 mixture.Competing interestsSMS is a cofounder of Axerion Therapeutics, which seeks to develop Nogoreceptor (NgR)- and PrP-based therapeutics. This work was supported in partby a Sponsored Research Agreement (to SMS) from UCB Pharma, the companywhich holds patent rights to brivaracetam. The remaining authors declare thatthey have no competing interests.Authors’ contributionsHBN conceived of the study, participated in its design and coordination,helped draft the manuscript, performed behavioral experiments, EEG analysis,and tissue preparation and histology. ACK performed behavioral experimentsand tissue preparation and histology. TSK performed tissue preparation andhistology. LH, HG, and SF performed EEG analysis. MK performed ELISAmeasurements. SMS conceived of the study, participated in its design andcoordination, and helped draft the manuscript. All authors read andapproved the final manuscript.AcknowledgementsThis work was supported in part by a Sponsored Research Agreement toSMS from UCB Pharma, the company which holds patent rights tobrivaracetam. The work was also supported by grants to SMS from theNational Institutes of Health, the Falk Medical Research Trust, the Alzheimer’sAssociation and the BrightFocus Foundation.Author details1Department of Neurology, Yale University School of Medicine, 800 HowardAvenue, New Haven, CT 06510, USA. 2Cellular Neuroscience,Neurodegeneration, and Repair Program (CNNR), Yale University School ofMedicine, 295 Congress Avenue, New Haven, CT 06536, USA. 3Division ofNeurology, The University of British Columbia, Djavad Mowafaghian Centrefor Brain Health, 2215 Wesbrook Mall, Vancouver, BC V6T 1Z3, Canada.4Division of Pediatric Neurology, The University of British Columbia, BritishColumbia Children’s Hospital, 4480 Oak Street, Vancouver, BC V6H 3V4,Canada.Received: 11 September 2014 Accepted: 19 February 2015References1. Palop JJ, Mucke L. Epilepsy and cognitive impairments in Alzheimer disease.2. Scarmeas N, Honig LS, Choi H, Cantero J, Brandt J, Blacker D, et al.Seizures in Alzheimer disease: who, when, and how common? ArchNeurol. 2009;66:992–7.Submit your next manuscript to BioMed CentralNygaard et al. Alzheimer's Research & Therapy  (2015) 7:25 Page 12 of 123. Irizarry MC, Jin S, He F, Emond JA, Raman R, Thomas RG, et al. Incidence ofnew-onset seizures in mild to moderate Alzheimer disease. Arch Neurol.2012;69:368–72.4. Palop J, Chin J, Roberson E, Wang J, Thwin M, Bien-Ly N, et al. Aberrant excitatoryneuronal activity and compensatory remodeling of inhibitory hippocampalcircuits in mouse models of Alzheimer’s disease. Neuron. 2007;55:697–711.5. Minkeviciene R, Rheims S, Dobszay MB, Zilberter M, Hartikainen J, Fülöp L, et al.Amyloid β-induced neuronal hyperexcitability triggers progressive epilepsy.J Neurosci. 2009;29:3453–62.6. Um JW, Nygaard HB, Heiss JK, Kostylev MA, Stagi M, Vortmeyer A, et al.Alzheimer amyloid-β oligomer bound to postsynaptic prion protein activatesFyn to impair neurons. Nat Neurosci. 2012;15:1227–35.7. Sanchez PE, Zhu L, Verret L, Vossel KA, Orr AG, Cirrito JR, et al. Levetiracetamsuppresses neuronal network dysfunction and reverses synaptic andcognitive deficits in an Alzheimer’s disease model. Proc Natl Acad Sci U S A.2012;109:E2895–903.8. Shi JQ, Wang BR, Tian YY, Xu J, Gao L, Zhao SL, et al. Antiepilepticstopiramate and levetiracetam alleviate behavioral deficits and reduceneuropathology in APPswe/PS1dE9 transgenic mice. CNS Neurosci Ther.2013;19:871–81.9. Bakker A, Krauss GL, Albert MS, Speck CL, Jones LR, Stark CE, et al. Reduction ofhippocampal hyperactivity improves cognition in amnestic mild cognitiveimpairment. Neuron. 2012;74:467–74.10. Bomben V, Holth J, Reed J, Cramer P, Landreth G, Noebels J. Bexarotene reducesnetwork excitability in models of Alzheimer’s disease and epilepsy. NeurobiolAging. 2014;35:2091–5.11. Roberson ED, Halabisky B, Yoo JW, Yao J, Chin J, Yan F, et al. Amyloid-β/Fyn-inducedsynaptic, network, and cognitive impairments depend on tau levels in multiplemouse models of Alzheimer’s disease. J Neurosci. 2011;31:700–11.12. Gimbel DA, Nygaard HB, Coffey EE, Gunther EC, Laurén J, Gimbel ZA, et al.Memory impairment in transgenic Alzheimer mice requires cellular prionprotein. J Neurosci. 2010;30:6367–74.13. Jankowsky J, Fadale D, Anderson J, Xu G, Gonzales V, Jenkins N, et al.Mutant presenilins specifically elevate the levels of the 42 residue β-amyloidpeptide in vivo: evidence for augmentation of a 42-specific γ secretase.Hum Mol Genet. 2004;13:159–70.14. Oddo S, Caccamo A, Shepherd J, Murphy M, Golde T, Kayed R, et al.Triple-transgenic model of Alzheimer’s disease with plaques and tangles:intracellular Aβ and synaptic dysfunction. Neuron. 2003;39:409–21.15. Morris RGM. Spatial localization does not require the presence of local cues.Learn Motiv. 1981;12:239–60.16. Kashiwaya Y, Bergman C, Lee JH, Wan R, King MT, Mughal MR, et al. A ketoneester diet exhibits anxiolytic and cognition-sparing properties, and lessensamyloid and tau pathologies in a mouse model of Alzheimer’s disease.Neurobiol Aging. 2013;34:1530–9.17. el Sayed MA, Löscher W, Frey HH. Pharmacokinetics of ethosuximide in thedog. Arch Int Pharmacodyn Ther. 1978;234:180–92.18. Ziyatdinova S, Gurevicius K, Kutchiashvili N, Bolkvadze T, Nissinen J, Tanila H,et al. Spontaneous epileptiform discharges in a mouse model of Alzheimer’sdisease are suppressed by antiepileptic drugs that block sodium channels.Epilepsy Res. 2011;94:75–85.19. Ichimiya Y, Emson PC, Mountjoy CQ, Lawson DE, Heizmann CW. Loss ofcalbindin-28 K immunoreactive neurones from the cortex in Alzheimer-typedementia. Brain Res. 1988;475:156–9.20. Dutar P, Potier B, Lamour Y, Emson PC, Senut MC. Loss of calbindin-28 Kimmunoreactivity in hippocampal slices from aged rats: a role for calcium?Eur J Neurosci. 1991;3:839–49.21. Carter DS, Harrison AJ, Falenski KW, Blair RE, DeLorenzo RJ. Long-term decreasein calbindin-D28K expression in the hippocampus of epileptic rats followingpilocarpine-induced status epilepticus. Epilepsy Res. 2008;79:213–23.22. Frankel WN, Beyer B, Maxwell CR, Pretel S, Letts VA, Siegel SJ. Developmentof a new genetic model for absence epilepsy: spike-wave seizures in C3H/Heand backcross mice. J Neurosci. 2005;25:3452–8.23. Ellens DJ, Hong E, Giblin K, Singleton MJ, Bashyal C, Englot DJ, et al.Development of spike-wave seizures in C3H/HeJ mice. Epilepsy Res.2009;85:53–9.24. Evason K, Huang C, Yamben I, Covey DF, Kornfeld K. Anticonvulsantmedications extend worm life-span. Science. 2005;307:258–62.25. Collins JJ, Evason K, Pickett CL, Schneider DL, Kornfeld K. The anticonvulsantethosuximide disrupts sensory function to extend C. elegans lifespan. PLoSGenet. 2008;4:e1000230.and take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistribution26. Shen H, Zhang B, Shin JH, Lei D, Du Y, Gao X, et al. Prophylactic andtherapeutic functions of T-type calcium blockers against noise-inducedhearing loss. Hear Res. 2007;226:52–60.27. Liedorp M, Stam CJ, van der Flier WM, Pijnenburg YA, Scheltens P.Prevalence and clinical significance of epileptiform EEG discharges in a largememory clinic cohort. Dement Geriatr Cogn Disord. 2010;29:432–7.28. Vossel KA, Beagle AJ, Rabinovici GD, Shu HD, Lee SE, Naasan G, et al.Seizures and epileptiform activity in the early stages of Alzheimer disease.JAMA Neurol. 2013;70:1158–66.Submit your manuscript at www.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items