Neurobiology of Disease 47 (2012) 1–12

Contents lists available at SciVerse ScienceDirect

Neurobiology of Disease

j ourna l homepage: www.e lsev ie r .com/ locate /ynbd i

Selective degeneration of septal and hippocampal GABAergic neurons in a mousemodel of amyloidosis and tauopathy

Desirée Loreth a, Laurence Ozmen b, Florent G. Revel b, Frédéric Knoflach b, Philine Wetzel a,Michael Frotscher a,c, Friedrich Metzger b, Oliver Kretz a,⁎a Department of Neuroanatomy, University of Freiburg, Germanyb Roche CNS Research and Early Development, Basel, Switzerlandc Center for Molecular Neurobiology Hamburg, University of Hamburg, Germany

⁎ Corresponding author at: Department of NeuroanatUniversity of Freiburg, Albertstrasse 23, D-79104 Freibu

E-mail address: oliver.kretz@anat.uni-freiburg.de (OAvailable online on ScienceDirect (www.scienced

0969-9961/$ – see front matter © 2012 Published by Eldoi:10.1016/j.nbd.2012.03.011

a b s t r a c t

a r t i c l e i n f o

Article history:Received 7 October 2011Revised 30 January 2012Accepted 1 March 2012Available online 9 March 2012

Keywords:Alzheimer's diseaseTauAβGABADisinhibitionMedial septumHippocampus

Alzheimer's disease (AD) is a neurodegenerative disorder characterized by brain accumulation of amyloid-βpeptide and neurofibrillary tangles, which are believed to initiate a pathological cascade that results in pro-gressive impairment of cognitive functions and eventual neuronal death. To obtain a mouse model displayingthe typical AD histopathology of amyloidosis and tauopathy, we generated a triple-transgenic mouse line(TauPS2APP) by overexpressing human mutations of the amyloid precursor protein, presenilin2 and taugenes. Stereological analysis of TauPS2APP mice revealed significant neurodegeneration of GABAergicsepto-hippocampal projection neurons as well as their target cells, the GABAergic hippocampal interneurons.In contrast, the cholinergic medial septum neurons remained unaffected. Moreover, the degeneration of hip-pocampal GABAergic interneurons was dependent on the hippocampal subfield and interneuronal subtypeinvestigated, whereby the dentate gyrus and the NPY-positive interneurons, respectively, were most stronglyaffected. Neurodegeneration was also accompanied by a change in the mRNA expression of markers for in-hibitory interneurons. In line with the loss of inhibitory neurons, we observed functional changes in TauP-S2APP mice relative to WT mice, with strongly enhanced long-term potentiation in the medial-perforantpathway input to the dentate gyrus, and stereotypic hyperactivity. Our data indicate that inhibitory neuronsare the targets of neurodegeneration in a mouse model of amyloidosis and tauopathy, thus pointing to a pos-sible role of the inhibitory network in the pathophysiological and functional cascade of Alzheimer's disease.

© 2012 Published by Elsevier Inc.

Introduction

Alzheimer's disease (AD) is the most common neurodegenerativecondition in adult humans, characterized by a slow-progressing declinein cognitive function that results in dementia and eventually death. ADis clinically divided into an early onset or familial (FAD) form, and a spo-radic or late onset form (SAD). FAD is associatedwith geneticmutationsin amyloid precursor protein (APP) and presenilin (PS) 1 or 2 (Bertramand Tanzi, 2005; Goate, 2006; Sherrington et al., 1995), makingmodelsbased on these human mutations highly relevant for investigations onthe pathophysiology of FAD. The late-onset cases have been ascribedto apolipoprotein E (APOE) gene mutations that are now consideredto be the most important risk factor for the development of SAD(Bertram et al., 2010; Corder et al., 1993; Strittmatter et al., 1993).The histopathological changes occurring in the brains of AD patients in-clude the formation of β-amyloid plaques and tau neurofibrillary

omy, Center for Neurosciences,rg, Germany.. Kretz).irect.com).

sevier Inc.

tangles. Accordingly, transgenic mouse lines expressing mutatedforms of APP, PS or tau provide suitable models for the investigationof AD pathophysiology. However, when mice transgenically expressmutated APP either alone or in combination with PS, there is an in-crease both in Aβ levels and plaque formation, but no development oftau neurofibrillary tangles (Games et al., 1995; Holcomb et al., 1998;Hsiao et al., 1996; Sturchler-Pierrat et al., 1997). While mutations inhuman tau are yet to be reported in AD patients, tau dysfunction hasbeen linked to neurodegeneration in frontotemporal dementia(Hutton et al., 1998). Moreover, mice overexpressing mutated formsof tau show neurofibrillary pathology and neurodegeneration (Götzand Ittner, 2008). Thus, in order to establish a mouse model displayingthe relevant histopathological features of AD, we recently generated atriple-transgenic mouse line (TauPS2APP) expressing mutated formsof tau, PS2 and APP (Grueninger et al., 2010). These mice display anage-dependent accumulation of Aβ, with formation of Aβ-plaques andneurofibrillary tangles in the hippocampus. Notably, accumulation ofAβ in TauPS2APPmice leads to accelerated tau pathology and increasedphosphorylation of tau at serine 422 when compared to tau single-transgenic mice. Moreover, TauPS2APP mice are hyperactive andshow impaired spatial learning in the Morris water maze. However,

2 D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

stereological analysis of pyramidal neurons within the hippocampusdid not reveal any signs of neuronal loss (Grueninger et al., 2010). Inthe present study,we provide a comprehensive analysis of neurodegen-eration in TauPS2APP mice in the basal forebrain, an important sourceof GABAergic and cholinergic input for the hippocampus, and in thehippocampus itself. In addition, we performed electrophysiologicalcharacterization of TauPS2APP mice, as well as an age-dependent eval-uation of their behavioral activity. Our results show that GABAergicneurons undergo neurodegeneration in a triple transgenic mousemodel for amyloidosis and tauopathy, thus suggesting a role for the in-hibitory network in the pathophysiology of AD.

Material and methods

Transgenic mice

The transgenic APP (line 147.72H) is homozygous for the humanAPP Swedish FAD transgene, the transgenic PS2 (line PS2.30H) is ho-mozygous for human PS2 N141I and the transgenic PS2APP (lineB6.152H) is homozygous for both human PS2 N141I and APP SwedishFAD transgenes. The double transgenic mouse line was created by co-injection of both transgenes into C57BL/6 zygotes, as previously de-scribed (Ozmen et al., 2009). The transgenic TauP301L mouse (linepR5) was generated on a mixed C57Bl/6, DBA/2 background (Gotzet al., 2001a, 2001b) and backcrossed for seven generations intoC57Bl/6 before generation of the triple transgenic (TauPS2APP)mice. TauPS2APP mice were obtained as recently described(Grueninger et al., 2010). Mice were screened by PCR using oligonu-cleotide pairs specific for the transgenes that overlap promoter andcoding regions, resulting in the amplification of fragments that areonly present in the transgenic mice. Age-matched C57Bl/6 micewere used as wild-type (WT) controls. Mice were housed under a12-h light/12-h dark cycle (lights on at 6:00 a.m., unless otherwisestated) at 20–22 °C, with ad libitum access to food and water. All pro-cedures were conducted in strict adherence to the Swiss and Germanfederal regulations for animal protection and to the rules of the Asso-ciation for Assessment and Accreditation of Laboratory Animal Care,with the explicit approval of the local veterinary authority.

Histology and immunohistochemistry

All animals assigned to morphological analyses were anesthetizedwith a mixture of ketamine, xylazine and acepromazine (0.3 ml/20 gbody weight) and transcardially perfused with 4% paraformaldehydein PBS. Coronal brain sections (50 μm) were cut on a vibratome andthe sections stored in 0.1 M PB. Histological examination of fornixmorphology was performed on routine Nissl-stained vibratome sec-tions at different rostro-caudal levels. For immunohistochemistry,free-floating sections were treated with 3% H2O2 in 0.1 M PB for15 min, followed by pre-incubation in a solution containing 5–10% nor-mal serum of the species in which the secondary antibody was raised(see secondary antibodies below), 2% BSA, and 0.1% Triton X in0.1 M PB for 1 h at room temperature (RT). The following primary anti-bodies were used: polyclonal anti-choline acetyltransferase (ChAT,1:100 dilution; Millipore); monoclonal anti-GAD67 (1:20,000 dilution;Millipore); polyclonal anti-parvalbumin (PV, 1:5000 dilution; Swant);polyclonal anti-Somatostatin (SOM, 1:5000 dilution; Bachem); poly-clonal anti-calretinin (CALRET, 1:5000 dilution; Swant) and polyclonalanti-neuropeptide Y (NPY, 1:1500 dilution; ImmunoStar Inc., Hudson,WI, USA). After incubation with primary antibodies the sections weretransferred to solutions containing biotinylated goat anti-rabbit IgG(for parvalbumin, somatostatin, calretinin and NPY), biotinylated rabbitanti-mouse IgG (for GAD-67), or biotinylated rat anti-goat IgG (forChAT; all secondary antibodies: Vector Laboratories, Burlingame, CA),diluted 1:250 in PB for 2 h at RT. Subsequent visualization of thelabeling by an avidin–biotin complex (1:250; Elite Vectastain ABC kit,

Vector Laboratories, Burlingame, CA) was performed using DAB as asubstrate. Immunofluorescence stainings were used for double labelingof sections using ptau (PS422, 1:500) or Aβ (BAP-2, 1:1000) antibodiesand GAD67 antibody, respectively. These sections were labeled withprimary antibodies and visualized with cy3- (dilution 1:800; JacksonImmunoResearch Laboratories INC., West Grove, PA, USA) or Alexa488-labeled (dilution 1:250; Invitrogen, Karlsruhe, Germany) second-ary antibodies.

The cholinergic fiber density in the hippocampus was examinedusing acetylcholine esterase (AChE)-histochemistry according to pre-vious protocols (Eckenstein and Sofroniew, 1983), with acetylcholineas the substrate and ethopropazine as the inhibitor of non-acetylcholinesterase.

Stereological cell counts

Stereological cell counts (Stereo Investigator software:MicroBright-Field, Inc., Colchester, VT; version 4.31) were performed to determinethe number of ChAT- and PV-immunoreactive (ir) neurons in themedi-al septum-vertical limb of the diagonal band complex, and the respec-tive number of GAD-67-, PV-, SOM-, CALRET- and NPY-ir interneuronsin the hippocampus. Every second serial section (50 μm) of the medialseptum/vertical limb of the diagonal band and every third serial sectionof the hippocampus was used for stereological analysis. The optical dis-ector/fractionator method (OF) was applied to each region as previous-ly described (Guijarro et al., 2006; Naumann et al., 2002; West et al.,1991). Briefly, sections were visualized on a computer screen attachedto an Olympus BX60 microscope F5 (Olympus Optical Co. Ltd, Düssel-dorf, Germany) with a computer-controlled stepper motor stage andfocus assembly that allow movement in the x-, y- and z-axes. Cellcounts were performed using Stereo Investigator software (version3.0; MicroBrightField, Inc., Colchester, USA). The region of interestwas first outlined in each section at low magnification (4× objective,nA: 0.10) and the following parameters were implemented: a countingframe of 50×30 μm; a guard zone of 2 μm; and a counting depth of8 μm. The view was then switched to high magnification (100× oilobjective, NA: 1.35,), where immuno-positive cells that fulfilled thecriteria of the unbiased counting rules (e.g. presence of a recognizablesoma within the counting frame, somata showing a distinct nucleus;cf. Coggeshall and Lekan, 1996) were marked and added to the proberun list. Total cell numbers estimated by the OF were statisticallyanalyzed by a two way-analysis of variance (ANOVA; for details seeNaumann T et al., 2002). Statistical significance was analyzed for thecorresponding classifiers and classes (WT, TauPS2APP).

Quantification of AChE staining

The density of cholinergic fibres in subregions of the dorsal andventral hippocampus (i.e. CA1, CA3, dentate gyrus) was determinedby optical densitometry measurements (n=6 per genotype). Afterdelineation of the relevant region of interest on digital images themean optical density (OD) of AChE-ir was measured using imageanalysis software (Olympus SIS, Stuttgart, Germany). The mean ODin the region of the corpus callosum in each section was consideredas ‘background’ and subtracted from all mean ODs measured in theregions of interest.

Ultrastructural analysis

For electron microscopy (EM), 12-month-old WT and TauPS2APPmice (n=3 for each genotype) were deeply anesthetized as de-scribed above and transcardially perfused with 0.9% saline followedby a fixative containing 4% PFA and 1% glutaraldehyde in 0.1 M PB,pH 7.4. Brains were removed and postfixed in the same fixative for12 h. Tissue blocks were washed with PB and horizontal sections(50 μm) were cut on a vibratome. Sections were treated with OsO4,

3D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

stained with uranyl acetate, dehydrated, and flat-embedded in epoxyresin (Durcupan ACM; Fluka, Gillingham, UK). Ultrathin sections werecut and examined in a Philips CM 100 electron microscope.

RNA purification and quantitative RT-PCR

Total RNA of hippocampi from 6-, 12- and 18-month-old TauPS2APPand WT controls (n=6) was purified using the RNaqaeous 4PCR kit(Applied Biosystems, Rotkreuz, Switzerland) according to themanufac-turer's protocol. Prior to reverse transcription, the sampleswere treatedwith DNAse to remove DNA. RNA concentration was determined bymeasuring absorbance at 260/280 nm per NanoDrop ND-1000, whileRNA integrity was confirmed by electrophoresis using BioRad Experion.Quantitative RT-PCR was performed with the quantitect™ probeRT-PCR kit (Qiagen AG, Basel, Switzerland) and Taqman probesfollowing manufacturer's protocol. All primers used to quantify murineGAPDH, Gad67, neuropeptide-Y (NPY), neuropeptide-Y receptor 1(Y1R), NMDAR1, mGluR1, RyR3 and Gabra1 were purchased fromApplied Biosystems. RT-PCR experiments were performed in triplicateand each sample normalized to the housekeeping gene GAPDH.

Electrophysiology

WT and TauPS2APP mice were anaesthetized in a mixture of 3.5%halothane/ 96.5% oxygen and decapitated in accordance with thelocal animal welfare committee guidelines. The brains were quicklysubmerged in saline and the hippocampi dissected free. Transverseslices (400 μm) of the dorsal hippocampus were cut with a Sorvall tis-sue chopper, maintained in a submerged chamber and perfused at RTin a simple salt solution gassed with 95% O2/5% CO2 and containing(in mM): 124 NaCl, 2.5 KCl, 2 MgSO4, 2.5 CaCl2, 1.25 KH2PO4, 26NaHCO3, 10 glucose, and 4 sucrose, pH 7.4. Slices were left to recoverfor 1 h.

For electrophysiological experiments, slices were perfused at35 °C and glass micropipettes (2–4 MΩ) containing 2 M NaCl wereplaced in the mid-molecular layer of the dentate gyrus (DG) and con-nected to a preamplifier (Cyberamp 380, Molecular Devices). The per-forant path input to the mid-molecular layer was stimulated withbipolar tungsten electrodes (50 μm diameter) connected to a stimu-lus generator (STG1008, Multi Channels Systems GmbH, Reutlingen,Germany). Stimulation strength was adjusted so as to evoke fieldEPSPs (fEPSPs) equal to 30–40% of the relative maximum amplitudeswithout superimposed population spikes. Field potentials were digi-tized at 20 kHz and low-pass filtered at 4.0 kHz using the PC-connected Digidata 1322A interface (Molecular devices). Test pulses(100 μs duration) were delivered every 30 s. To induce LTP, a stimulusparadigm comprising 8 bursts of 8 pulses at 200 Hz (80% maximalfEPSP amplitude; 200 μsec duration) was delivered to the stimulatingelectrodes in the presence of the competitive GABAA receptor antag-onist picrotoxin (100 μM). Slopes of fEPSPs were determined by line-ar regression to at least 30% of the rising phase between 30 and 80% ofthe peak amplitude and normalized to the control period 10 min be-fore delivering the LTP stimulus paradigm.

Behavioral recordings

Locomotor activity and stereotypic activity were evaluated usingan automated Omnitech Digiscan system (Omnitech Electronics,Colombus, OH) placed in a sound-proof room, as described previously(Revel et al., 2011). Each activity monitor consisted of a Plexiglas box(41×41×30.5 cm) that contained sawdust bedding and was sur-rounded by invisible horizontal and vertical infrared sensor beams.Water was provided in a bottle fixed in one corner of the box, andfood pellets were dispersed on the bedding. Data were obtained si-multaneously from 16 monitors connected to a Digiscan Analyzerthat constantly collected the beam status information. Locomotor

activity was defined as the total distance travelled (in cm) as deter-mined by the number of interruptions by the photocell beams. If amouse repeatedly broke the same beam (or set of beams), then themonitor considered the animal to be exhibiting stereotypy (e.g., dur-ing grooming). Stereotypy counts were measured as the number ofbeam breaks occurring during this period of stereotypic activity. Lo-comotor activity and stereotypy counts for individual animals weretaken every 10 min.

At 6 months of age (±4 days), individually-housed male TauP-S2APP and WT (n=8/group) mice were transferred to the experi-mental room and housed under a 12/12 light/dark cycle (lightonset at 10:00 p.m.). Following a defined period of habituation, be-havioral recordings were performed successively at 7-, 10-, 12-,15- and 17-months of age using the same cohort of animals. Foreach recording session, the mice were weighed and placed intothe monitor chamber shortly before the beginning of the dark (ac-tive) phase, and behavioral recordings were started at 10:00 a.m.(dark onset) simultaneously for all animals. The mice were moni-tored for 48 h, and then returned to their home cage. The miceremained in the experimental room between recording sessionsand were inspected daily at random time points.

Statistical analysis

Differences between the experimental groups were determinedby a two-way ANOVA (cell counts, real time PCR), a two-wayANOVA with repeated measures (behavior), or by a two-tailed un-paired Student's t-test (LTP) with confidence intervals of 95%. For be-havioral data, post-hoc comparisons were performed using a two-tailed unpaired Student's t-test (comparison of genotypes at eachage) and Dunnett's test (comparison of various ages versus 7 monthold mice for each genotype). All calculations were performed withGraphPad Prism for Microsoft Windows (GraphPad Software, SanDiego, CA, USA). In all cases, pb0.05 was accepted as statisticallysignificant.

Results

GABAergic but not cholinergic septo-hippocampal projection neuronsundergo neurodegeneration

In 18-month-old TauPS2APP mice, the diameter of the fornix at thelevel of the columna (Figs. 1a–c) and at the level of the fimbria fornicis(Figs. 1d–f) was reduced by ~50% and ~20%, respectively, when com-pared to WT controls. Since the fimbria-fornix largely comprises theaxons of septo-hippocampal projection neurons, the age-dependent re-duction in fornix diameter led us to the assumption that these projec-tion neurons undergo progressive degeneration in TauPS2APP mice.We thus performed stereological analysis of both cholinergic andGABAergic septo-hippocampal projection neurons within the medialseptum of 12- and 18-month-old TauPS2APP and WT mice. Whereasthe number of ChAT-positive (cholinergic) neurons within the medialseptum decreased significantly with age, this decrease was comparablebetween TauPS2APP and WT mice (Figs. 2a–c). We then investigatedthe cholinergic septo-hippocampal projection further by quantitativeanalysis of AChE staining; again, no significant difference in cholinergicinnervation of the hippocampus by AChE-positive neurons was ob-served in 18-month-old TauPS2APP mice compared to WT controls(Figs. 2d–i). In contrast, stereological analysis of PV-positive, GABAergicsepto-hippocampal projection neurons in the medial septum complexrevealed a significant loss of this population in TauPS2APP mice,whichwas already evident at 12 months of age.While the total numberof PV-positive septal neurons significantly decreased with age in bothgenotypes, this cell loss was more pronounced in 18-month-old TauP-S2APP mice compared to age-matched WT controls (Figs. 2j–l). Taken

Fig. 1. Fornix pathology in TauPS2APP mice. At the age of 18 months quantitative analysis of Nissl-stained serial sections revealed that fornix diameter is reduced by about 50% inTauPS2APP mice at the level of the column of the fornix (arrowheads, bregma 0.02; b,c) and by about 20% in the region of the fimbria fornicis (arrows, bregma -0.82; e,f), as com-pared to age matched WT (a,c,d,f). Scale bar: 500 µm in a, b, d and e. Data represent mean ± SEM (n= 7 per genotype).

4 D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

together, these results suggest a selective degeneration of GABAergicsepto-hippocampal projection neurons and their corresponding axons.

Region- and subtype-specific loss of hippocampal interneurons

Given that PV-positive septo-hippocampal projection neurons se-lectively terminate on GABAergic interneurons in the hippocampus,we next performed stereological analysis of this latter population.To determine whether hippocampal interneurons also undergo neu-rodegeneration we next performed a stereological analysis of thispopulation. We used anti-Gad67 as a global marker for inhibitory in-terneurons, as well as antibodies against NPY, somatostatin, calretininand parvalbumin to identify interneuronal subpopulations. Stereolog-ical cells counts were performed in the CA1-3 subregion of the hippo-campus as well as in the whole DG and in the hilar region of 12 and18 months old TauPS2APP mice and age matched WT controls. Al-though the number of Gad67-positive interneurons did not differ be-tween WT and TauPS2APP mice at 12 months of age (Fig. 3a), weobserved in 18-month-old TauPS2APP mice a significant loss ofGad67-positive neurons in all subregions investigated, with cell lossmost pronounced in the DG and hilus (Fig. 3b). The NPY-positive in-terneurons were also strongly affected since they already showed asignificant reduction in number in 12-month-old TauPS2APP micein all subregions analyzed; this cell loss persisted in 18-month-oldTauPS2APP mice, with the DG once again harboring the most signifi-cant loss (Figs. 3a, b). In contrast, no significant reduction in the num-ber of PV- and CALRET-positive interneurons at 12 months of age wasobserved. However, both markers were significantly reduced in theDG of aged TauPS2APP mice compared to WT controls (Figs. 3a, b). Fi-nally, the numbers of SOM-positive interneurons in TauPS2APP vs.WT mice were comparable at each time point and in each subregioninvestigated (Figs. 3a, b). In summary, the detailed stereological anal-ysis revealed that the extent of interneuronal loss within the hippo-campus of TauPS2APP mice is highly dependent on the subregionand neuronal types investigated.

Because a stereological reduction in cell number does not discrim-inate between cell death and loss of marker immunoreactivity, we

next performed an ultrastructural analysis of the medial septum anddentate hilar region of 12-month-old TauPS2APP vs. WT mice to de-termine whether the observed differences could be attributed to celldeath. In TauPS2APP mice, neurons with multiple intracellular inclu-sions were frequently observed both in the septum (Fig. 4a) and thehippocampus (Figs. 4b, c). Moreover, many neurons within both re-gions displayed signs of neurodegeneration, i.e. disaggregation ofthe nuclear and cellular membranes, loss of integrity of cell organellesor apoptotic blebs (Figs. 4a–c).

Downregulation of markers for inhibitory transmission

In an earlier study it was shown that principal neurons of the hip-pocampus are not affected by neurodegeneration (Grueninger et al.,2010). To determine whether the selective loss of GABAergic neuronswas also represented at the mRNA level we next performed a quanti-tative RT-PCR analysis of excitatory and inhibitory markers of hippo-campal neurons in TauPS2APP mice and WT controls at the ages of 6,12 and 18 months. We found that the expression of NPY was alreadysignificantly reduced in TauPS2APP mice at 6 months of age and fur-ther decreased at 18 months of age to less than 50% of WT levels(Fig. 5a). The expression levels of the global interneuron markerGad67 (data not shown) and the NPY receptor Y1R( Fig. 5b) wereonly significantly reduced in 18-month-old TauPS2APP mice. In con-trast, expression levels of GABA receptor alpha1 (Gabra1) were unal-tered at all time points investigated (Fig. 5c). To confirm that thechanges were specific to interneuron subtypes and not related tothe loss of excitatory input, we analyzed NMDA receptor subunit 1,mGluR1 and the endoplasmatic reticulum calcium channel ryanodinereceptor type 3 (Ryr3). None of these excitatory neuron markers dis-played altered expression at 6 or 12 months of age; however, at18 months all 3 genes showed reduced expression in TauPS2APPmice compared to WT controls (Figs. 5d–f). These data underlinethe absence of a general loss of inhibitory neurons, but rather pointto a specific deficit of NPY signaling in younger animals before severeamyloid and tau pathologies manifest.

Fig. 2. Stereological cell counts of septo-hippocampal projection neurons. (a–c) TauPS2APP mice versus WT displayed a significant, age-dependent reduction in cholinergic septo-hippocampal projection neurons in both groups, however, no significant difference between the genotypes could be observed either at 12 or 18 months of age, respectively (a). Repre-sentative staining for ChAT in the medial septum ofWT (b) and TauPS2APP mice (c) at the age of 18 months. (d, e) AChE staining for cholinergic fibers in hippocampi of 18 month-oldWT (d) and TauPS2APPmice (e). High power views ofWT (f, h), TauPS2APP (h, i), dentate gyrus (f, g) and CA3 subregions (h, i) reveal no obviousmorphological alterations to theAChE-positive fibers in TauPS2APP mice compared to age-matched controls. Moreover, a quantitative analysis of this staining did not reveal any significant difference between genotypes incholinergic fiber density (data not shown). (j–l) However, the number of PV-positive septo-hippocampal projection neurons is significantly reduced in 12-month-old TauPS2APP mice,and further decreases with age in comparison toWT (j). Representative staining for parvalbumin in themedial septum ofWT (k) and TauPS2APPmice (l) at the age of 18 months. Scalebars: 250 μm in b–e, k, l and 50 μm in f–i . ***, Pb0.001, *, Pb0.05 vs WT. Data represent mean±SEM (n=6 per staining, time point and genotype).

5D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

TauPS2APP mice display increased locomotor activity and enhanced LTPin the dentate gyrus

We hypothesized that the loss of inhibitory hippocampal inter-neurons in the DG would alter the inhibitory regulation of synaptictransmission and thereby possibly modulate LTP. To this end, we per-formed electrophysiological recordings in the perforant path/DG ofTauPS2APP and WT mice and observed that 13-month-old TauP-S2APPmice displayed strongly enhanced LTP in the presence of picro-toxin (Fig. 6). This effect was of a similar size to that observedrecently in the PS2APP founder line at 3 and 10 months of age

(Poirier et al., 2010). During our behavioral observations, seizureswere only occasionally observed, which was not unexpected, giventhe altered inhibitory regulation. In contrast, no epileptic dischargeswere seen during electrophysiological recordings in brain slices.However, while this would not exclude the presence of electrophysi-ological seizure activity in the EEG, such measurements were beyondthe scope of this study. Nonetheless, TauPS2APP mice were hyperac-tive in a stereotypic manner, as evaluated from the daily profile re-cordings of locomotor activity and stereotypies in TauPS2APP andWT mice at several ages (Fig. 7). Both genotypes showed elevated lo-comotion and stereotypic activity following introduction to the

Fig. 3. Stereological cell counts of hippocampal interneurons. Shown are stereological data of Gad67-, Parvalbumin- (PV), CALRET-, NPY- and SOM-ir cells in the hippocampus of (a) 12-and (b) 18-month-oldWT and TauPS2APP (3TG) mice. The reduced number of interneurons in TauPS2APPmice varies according to the hippocampal subregion and interneuronal sub-type, with themost pronounced cell loss seenwithin the DG and hilus. Moreover, the earliest and strongest effect is observed for the NPY-positive interneuron population. ***, Pb0.001,**, Pb0.01, *, Pb0.05 vs WT. Data represent mean±SEM (n=6 per staining, time point and genotype). DG = total dentate gyrus; hilus = hilar region of the dentate gyrus.

6 D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

monitoring chamber, thus likely responding to increased exploratoryactivity to the new environment. Such exploratory activity was higherat both 7 and 17 months of age in TauPS2APP mice compared to WT(Figs. 7a, c). On the subsequent day of recording (habituated ani-mals), locomotor activity and stereotypy counts were still significant-ly elevated in aged TauPS2APP mice (12-, 15-, 17-months old), butnot in 7- and 10-month-old mice (Figs. 7b, d). This resulted from anage-dependent reduction in activity in WT, whereas the TauPS2APPmice maintained throughout life elevated locomotor activity after

habituation (Fig. 7b). As a consequence, age-dependent body weightgain was reduced in TauPS2APP mice (data not shown), and reachedsignificance at 17 months of age (TauPS2APP, 37.60±1.75 g; WT,45.35±1.81 g; Pb0.01). Importantly, the TauPS2APP mice did notshow any abnormalities in their daily activity profile (Figs. 7a, c), con-firming a general elevation in activity during the active (dark) phase.Taken together, these data show that unlike WT mice, TauPS2APPmice do not experience an age-related reduction in locomotor activi-ty, but rather maintain hyperactivity and stereotypic behavior. How

Fig. 4. Ultrastructural features of neurodegeneration in TauPS2APP mice. At the ultrastructural level, several neurons show signs of neurodegeneration in 12-month-old TauPS2APPmice, but not in WT mice. Both in the medial septum (a) and hilus of the DG (b, c), neurons display numerous intracellular inclusions (a, arrows), loss of nuclear and cell membraneintegrity as well as of cell organelles (b, c). Scale bar 10 μm in a–c.

7D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

this behavior relates to the increased LTP measured in the dentategyrus remains to be shown.

Discussion

The present study provides a morphological and functional charac-terization of a recently-generated triple-transgenic mouse model foramyloidosis and tauopathy (TauPS2APP), whereby the major pheno-typical finding was selective neurodegeneration of GABAergic neuronsboth in the hippocampus andmedial septum complex. The selective af-fection of the GABAergic septo-hippocampal population but relative

Fig. 5. Quantitative RT-PCR expression analysis. Expression of gene markers for inhibitory anand TauPS2APP mice. NPY expression is constantly decreased at 6, 12 and 18 months of age (is not affected (c). The mRNA levels of excitatory markers such as NMDAR1 (d), mGluR1 or RNPY expression is already significantly reduced at the age of 6 months. ***, Pb0.001, **, Pb

sparing of the cholinergic system within the basal forebrain is some-what surprising. Three decades ago, publications describing the degen-eration of cholinergic neurons (for review see Auld et al., 2002;Mesulam, 2004) and the downregulation of cholinergic markers withinthe basal forebrain of AD patients (Perry et al., 1977) led to the ‘cholin-ergic hypothesis’ (Bartus et al., 1982; Whitehouse et al., 1982), whichsuggests that the degenerating cholinergic system plays a crucial rolein the pathomechanisms of AD. However, recent findings have chal-lenged this theory, since more reliable analysis techniques haveshown that cholinergic degeneration in the basal forebrain is detectableneither in the brains of early-stage AD patients, nor in those suffering

d excitatory synaptic transmission in the hippocampus of 6-, 12- and 18-month-old WTa). In contrast, expression of Y1R only declines at 18 months of age (b), whereas Gabra1YR3 (f) only show a significant decrease in 18-month-old TauPS2APP mice. In contrast,0.01, *Pb0.05 vs WT. Data represent the mean±SEM (n=6 per group).

Fig. 6. Electrophysiological alterations in the hippocampus of TauPS2APP mice. Highfrequency stimulation of the medial perforant pathway was delivered to hippocampalslices from WT (grey traces; n=5 slices from 3 animals) and TauPS2APP mice (blacktraces; n=6 slices from 3 animals). LTP was induced by tetanic stimulation, whichwas delivered at the time indicated (arrow) in the presence of picrotoxin (100 μM,bar). LTP is markedly enhanced in TauPS2APP mice compared to WT controls. Dataare expressed as mean±SEM of the field EPSP slope normalized to baseline valuesobtained 10 min before the tetanic stimulation. Differences between both groupswere assessed with an unpaired Student's t-test; **, Pb0.01 with the data obtained60 min after the tetanic stimulation.

8 D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

from mild cognitive impairment; in contrast, some of these patientseven exhibited increased ChAT activity in the hippocampus in compar-ison to age-matched, non-demented controls (Davis et al., 1999;DeKosky et al., 2002; Gilmor et al., 1999; Tiraboschi et al., 2000). More-over, a critical review of the literature shows that in the majority ofmousemodels for AD, the cholinergic basal forebrain was either not in-vestigated or found to be unchanged (Cassel et al., 2008). Thus, evi-dence based on some AD animal models suggests that any potentialdeficits within the cholinergic system do not make a primary contribu-tion to the pathophysiology of AD (Buttini et al., 2002; Contestabile etal., 2008; German et al., 2003). Early cholinergic neurodegeneration inthe basal forebrain was nonetheless described in a recent study of thetransgenicmouse line Thy-tau22 (Belarbi et al., 2009). Although the au-thors did not specifically identify the degenerating neuronal popula-tion, the reduced number of retrogradely fluorogold-labelled medialseptum neurons in this mouse model was attributed to a loss of cholin-ergic neurons. However, our observations of the TauPS2APPmouse linerather suggest that it is the GABAergic septo-hippocampal projectionthat undergoes selective neurodegeneration. Ascertainment ofwhetherthis feature is unique to our triple transgenic model system would ne-cessitate a careful analysis of both populations of medial septal projec-tion neurons in other AD mouse models.

The GABAergic septo-hippocampal projection plays a crucial rolein the generation of hippocampal theta rhythm activity (Hangya etal., 2009), an electrophysiological characteristic indispensible forhippocampal-dependent learning and memory tasks (Düzel et al.,2010). Thus, it is not unlikely that the degeneration of GABAergicneurons in the medial septum also contributes to the cognitive im-pairment observed both in our mouse model system and in AD. TheGABAergic septal projection terminates exclusively on inhibitoryGABAergic interneurons in the hippocampus (Freund and Antal,1988), which were observed to have undergone progressive degener-ation in our TauPS2APP model. Here, the magnitude of neuronal loss

was dependent on the hippocampal subregion and subpopulation ofinterneurons investigated, whereby the DG and the NPY positive in-terneurons, respectively, showed the most pronounced and earliestsigns of cell loss. This observation expands on our recent findingthat the total number of principal excitatory neurons is unchangedin the hippocampus of TauPS2APP mice, even in aged animals(Grueninger et al., 2010). It remains unclear whether hippocampaland septal neurodegeneration occur independently or whether hip-pocampal cell loss is secondary to the loss of septal input. Althoughthere are no studies on the effect of reduced GABAergic input ontohippocampal interneurons e.g. after fimbria-fornix transection, a neu-rotrophic effect of septal innervation cannot be excluded since, for ex-ample, immunolesion of cholinergic basal forebrain neurons leads tothe degeneration of NPY-positive hippocampal interneurons (Milneret al., 1999). On the other hand, it has been shown that hippocampalhyperexcitability can predominantly lead to the degeneration ofGABAergic septal neurons (Garrido Sanabria et al., 2006). Thus, inTauPS2APP mice, impaired inhibition within the hippocampus dueto the regression of GABAergic interneurons might be the reason forconsecutive cell loss in the medial septum.

TauPS2APP mice displayed enhanced LTP in the perforant path/DGat 13 months of age. This alteration in LTP might be the direct conse-quence of inhibitory neuron loss in the DG. However, increased LTP inthe DG has been described in young, single-transgenic tau-P301Lmice (Boekhoorn et al., 2006) and in 3-month-old B6.152H mice,which comprise the founder line of our mouse model, (Poirier et al.,2010). Therefore, the increased LTP observed in both models was reg-istered at ages preceding hallmark neuropathological changes such ashyperphosphorylation, tauopathy and amyloid plaque deposition. Onthe other hand, several mouse models of AD that harbor mutant var-iants of APP display impaired LTP in the hippocampus that is relatedto the presence of brain pathology (Yamin, 2009). Thus, tau andAPP, as well as the disturbance of the inhibitory network, have bothoverlapping and opposite effects on hippocampal LTP in our triple-transgenic mouse model. Further electrophysiological investigationsin TauPS2APP mice, as well as in their single-transgenic founderlines, are needed to provide clarity about the nature of the unexpect-ed increase in LTP.

Up to 22% of the patients suffering from AD display unprovokedseizures (Mendez and Lim, 2003), with seizure frequency even higherin early-onset AD (Amatniek et al., 2006). Moreover, non-convulsiveseizures and progressive epilepsy have also been described in differ-ent APP transgenic mouse lines (Minkeviciene et al., 2009; Palop etal., 2007), in which exaggerated depolarization of primary neuronsby fibrillar Aβ1-42 peptide has been assumed to be the primary epi-leptogenic event (Minkeviciene et al., 2009). However, the alterationsin the inhibitory network recorded in these studies resemble thoseseen after excitotoxic stimuli and might rather compensate for theoverexcitability of the corticolimbic network (Palop et al., 2007).Since occasional signs of convulsion were observed in TauPS2APPmice, it would be pertinent to perform systematic electroencephalo-graphic recordings to evaluate the occurrence of asymptomatic sei-zures. Abolishment of hippocampal theta activity, which is assumedto be part of a seizure resistant state, lowers seizure thresholds(Miller et al., 1994). Moreover, even a modest impairment of the in-hibitory hippocampal network in mouse models has been shown toresult in reduced seizure threshold and activity, a feature likely toplay a role in human patients suffering from spontaneous seizures(Dudek and Sutula, 2007). Thus, the degeneration in TauPS2APPmice of GABAergic septo-hippocampal neurons that normally syn-chronize theta activity, coupled with the disinhibition arising fromsignificant loss of hippocampal interneurons, might offer a new ratio-nale for understanding seizure formation in AD. The results of our be-havioral studies revealed a significant, age-dependent increase inlocomotor activity and stereotypic behavior in TauPS2APP mice com-pared to age-matched controls. This is in line with a recent study in

Fig. 7. Recording of locomotor and stereotypic activities in aging TauPS2APP mice. Locomotor (a, b) and stereotypic (c, d) activities were recorded for 48 h in aging TauPS2APP miceand WT controls. (a, c) Daily profiles of locomotor (a) and stereotypic (c) activities of the mice at 7- and 17-months of age. During the first day of the recordings, animals displayedhigher activity levels, illustrating the exploratory period of the new environment. The locomotor activity (b) and stereotypy counts (d) of the habituated mice were then accumu-lated over the second day of recording and compared at 7, 10, 12, 15 and 17 months of age. Whereas locomotor activity decreased with age in WT mice, that of TauPS2APP miceremained elevated (b). Similarly, TauPS2APP mice showed increased stereotypies with age (d). The daily activity profile of aged TauPS2APP mice is similar to that of WT controls(a), with increased locomotor activity (a) and stereotypies (c) occurring during the active (dark) phase. Time 0 corresponds to ‘lights off’ under the 12 h light/12 h dark cycle. Datarepresent mean±SEM, with n=8 mice per group. **Pb0.01, *Pb0.05 vs WT (same age); $$$Pb0.001, $$$Pb0.05 vs 7 months of age group (same genotype).

9D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

which mice with a combined lesion of the cholinergic and GABAergicsepto-hippocampal projection neurons displayed increased locomo-tor activity both at baseline and after amphetamine administration(Lecourtier et al., 2010). Moreover, impairment of the GABAergic sys-tem in a mouse model carrying a mutation in the MECP2 gene was re-cently reported to result in neuropsychiatric phenotypes such asstereotypic behavior (Chao et al., 2010). On the other hand, hyperac-tivity (Aalten et al., 2007, 2008) and stereotypies (Helvink andHolroyd, 2006; Nyatsanza et al., 2003) are common clinical featuresin patients suffering from dementia-related diseases including AD.Given that increased exploratory behavior and stereotypies havealso been described in mouse models of AD (Filali et al., 2011;Sterniczuk et al., 2010), it would be interesting to investigate whetherthe behavioral disturbances observed in these models are also linkedto pathological changes in the GABAergic network.

The progressive decline in cognitive function is another clinicalhallmark of AD. There is growing evidence that the inhibitoryGABAergic network plays a crucial role in learning and memory pro-cesses; for example, conditional learning paradigms in mice lead tothe formation of inhibitory synapses (Jasinska et al., 2010) and inducean enhancement in synaptic inhibition in olfactory-discriminationlearning (Brosh and Barkai, 2009). GABA release from hippocampalinterneurons is increased after spatial learning tasks (Cui et al.,2008), whereas decreased GABA levels resulting from accelerated re-uptake impair associative learning ability and new object recognitionretention in mice (Hu et al., 2004). Such mechanisms might help toexplain why the progressive loss of GABAergic hippocampal neuronswas associated with an impairment in spatial learning in the Morriswater maze in 4- and 12-month-old TauPS2APP mice (Grueningeret al., 2010). In addition, altered inhibition might also play a role in

the cognitive decline of AD patients. In line with this, GABA levels inthe brain and CSF have been shown to decrease both with normalaging and in AD patients (Bareggi et al., 1982; Seidl et al., 2001;Zimmer et al., 1984). Similarly, the numbers of PV-positive interneu-rons and NOS-positive small inhibitory interneurons in the hippo-campus and cortex, respectively, are reduced in post mortem ADbrains (Brady and Mufson, 1997; Koliatsos et al., 2006). In contrast,other studies did not find any differences in the number of CALRET-or PV-positive interneurons in the temporal cortex of AD patients(Fonseca and Soriano, 1995; Fonseca et al., 1993). However, in trans-genic mouse lines expressing human APP and/or PS1 mutations as amodel for amyloidosis, a depletion in CALRET- and PV immunoreac-tivity was observed in the hippocampus, as determined by densitom-etry (Popović et al., 2008) and stereology (Baglietto-Vargas et al.,2010; Takahashi et al., 2010), respectively.

Already several years ago another triple transgenic mouse modelfor AD (3xTg-AD) was published. In this mouse line the human Swed-ish double mutation for APP, the tauP301L and Presenilin1M146V tran-genes are expressed under the control of mouse Thy1.2 regulatoryelements (Oddo et al., 2003). Despite differing fromTauPS2APP micein terms of promoter and trangene expression, both lines sharesome obvious phenotypic similarities, including a comparable timecourse and pattern of amyloid plaque formation in the neocortexand the limbic system. Moreover, in both lines, plaque formation pre-cedes tangle formation, and tau-pathology is aggravated by the si-multaneous expression of the APP transgene (Grueninger et al.,2010; Oddo et al., 2003). However, while we observed increasedLTP in the DG of TauPS2APP mice, the 3xTg-AD line showed reducedLTP in the CA1 subregion (Oddo et al., 2003). As neuronal cell deathwas not reported in 3xTg-AD mice, it would be interesting to

10 D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

investigate whether the different molecular approach used in thegeneration of 3xTg-ADmice also leads to the ultimate loss of GABAer-gic neurons in the basal forebrain and hippocampus, as observed inour TauPS2APP model.

Although the pathomechanisms underlying the selective loss ofGABAergic neurons in TauPS2APP mice remains to be defined, recentstudies have suggested a possible role for apolipoprotein E4 (apoE4)and tau (Herz, 2009). The apoE4 isoform has been identified as an im-portant risk factor for the onset of dementia and SAD in humans(Herz, 2009) and knock-in mouse lines carrying human apoE4 displaya significant loss of Gad67-positive interneurons in the hilar region ofthe DG (Andrews-Zwilling et al., 2010).Depletion of tau was sufficientto rescue this phenotype, suggesting a tau-dependent mechanism(Brecht et al., 2004; Harris et al., 2004; Small and Duff, 2008). In TauP-S2APP mice, tau transgene expression and neurofibrillary tangle for-mation are also observed in GABAergic neurons of the basalforebrain and hippocampus. Thus, ptau-dependent neurodegenera-tion seems likely to occur in TauPS2APP mice. Aβ peptides act up-stream of tau and further stimulate hyperphosphorylation andtangle formation (Gotz et al., 2001b). In line with this, we recentlyshowed that expression of the Aβ transgene enhanced tau phosphor-ylation (Grueninger et al., 2010). To what extent the greater aggrava-tion of tau pathology in TauPS2APP mice over tau single-transgenicmice influences the interneuronal phenotype, and why GABAergicneurons are particularly linked to tau-associated alterations, remainsto be elucidated. The effects of Aβ peptides and presenilin on neuro-nal calcium signaling might offer another explanation for the selec-tive loss of GABAergic interneurons in TauPS2APP mice.Perturbation of calcium signaling has been described in AD(Demuro et al., 2005), and represents a causal factor for excitotoxicityand cell death (Lei et al., 1992). Mutant PS increases calcium releasefrom intracellular stores via ryanodine or IP3-receptor-mediatedmechanisms (Cheung et al., 2008; Stutzmann et al., 2006). In addi-tion, administration of Aβ to cultured cells increases calcium perme-ability of the plasma membrane (Demuro et al., 2005; Deshpande etal., 2006), and elicits calcium influx, thereby enhancing NMDA recep-tor currents in cultured neurons (Demuro et al., 2010). GABAergicneurons are vulnerable to excitotoxicity, as observed in mousemodels of epilepsy, brain trauma or ischemia (Johansen, 1993; Liuet al., 2009; Monnerie and Le Roux, 2007; Shetty et al., 2009). Thus,deleterious intracellular signaling cascades induced by perturbationsin calcium regulation might also contribute to the loss of GABAergicneurons in TauPS2APP mice. Future investigations that include elec-trophysiological recordings of identified interneurons and calciumimaging should reveal whether altered glutamatergic synaptic trans-mission and calcium overload are early events in TauPS2APP micethat lead to the excitotoxic cell death of interneurons. In line withthis, electron microscopic analysis of the expression level and distri-bution of glutamatergic receptors at excitatory input synapses ofGABAergic interneurons together with single cell RT-PCR would beuseful, since an alteration in NMDA-receptor distribution could acti-vate pro-apoptotic pathways in interneurons (Hardingham et al.,2002). A recent report described in the forebrain of TauPS2APP micedisturbances in mitochondrial protein expression and related func-tional alterations that were likely evoked by the synergistic effectsof Aβ and tau (Rhein et al., 2009). It would therefore be interestingto see whether mitochondrial dysfunction is more pronounced inGABAergic interneurons, or if these neurons are selectively vulnera-ble to such alterations, as discussed in the context of other neurode-generative diseases (Kumar et al.; Oliveira, 2010).

In summary, while further experiments are needed to elucidatethe precise pathomechanisms of GABAergic neurodegeneration inTauPS2APP mice, our findings indicate that the control of inhibitoryactivity plays an important role in learning and memory formationas well as in cognition and seizure protection, both of which are keyissues in AD patients. The alterations to the inhibitory network in

the DG of TauPS2APP mice thus point to a possible role of GABAergicdegeneration in AD patients.

Acknowledgments

The authors thank S. Nestel, M. Haenggi, J. Messer and P. Biry forskilful technical assistance, J.-L. Moreau and T. Ballard for their valu-able guidance as well as S. Dieni for proofreading the manuscript.MF is Senior Research Professor of the Hertie Foundation.

Appendix A. Supplementary data

Supplementary data to this article can be found online at doi:10.1016/j.nbd.2012.03.011.

References

Aalten, P., Verhey, F.R.J., Boziki, M., Bullock, R., Byrne, E.J., Camus, V., Caputo, M., Collins,D., de Deyn, P.P., Elina, K., Frisoni, G., Girtler, N., Holmes, C., Hurt, C., Marriott, A.,Mecocci, P., Nobili, F., Ousset, P.J., Reynish, E., Salmon, E., Tsolaki, M., Vellas, B.,Robert, P.H., 2007. Neuropsychiatric syndromes in dementia. Results from the Eu-ropean Alzheimer Disease Consortium: Part I. Dement. Geriatr. Cogn. Disord. 24(6), 457–463. doi:10.1159/000110738.

Aalten, P., Verhey, F.R.J., Boziki, M., Brugnolo, A., Bullock, R., Byrne, E.J., Camus, V.,Caputo, M., Collins, D., de Deyn, P.P., Elina, K., Frisoni, G., Holmes, C., Hurt, C.,Marriott, A., Mecocci, P., Nobili, F., Ousset, P.J., Reynish, E., Salmon, E., Tsolaki, M.,Vellas, B., Robert, P.H., 2008. Consistency of neuropsychiatric syndromes across de-mentias: results from the European Alzheimer Disease Consortium. Part II.Dement. Geriatr. Cogn. Disord. 25 (1), 1–8. doi:10.1159/111082.

Amatniek, J.C., Hauser, W.A., DelCastillo-Castaneda, C., Jacobs, D.M., Marder, K., Bell, K.,Albert, M., Brandt, J., Stern, Y., 2006. Incidence and predictors of seizures in pa-tients with Alzheimer's disease. Epilepsia 47 (5), 867–872. doi:10.1111/j.1528-1167.2006.00554.x.

Andrews-Zwilling, Y., Bien-Ly, N., Xu, Q., Li, G., Bernardo, A., Yoon, S.Y., Zwilling, D., Yan,T.X., Chen, L., Huang, Y., 2010. Apolipoprotein E4 causes age- and Tau-dependentimpairment of GABAergic interneurons, leading to learning and memory deficitsin mice. J. Neurosci. 30 (41), 13707–13717. doi:10.1523/JNEUROSCI.4040-10.2010.

Auld, D.S., Kornecook, T.J., Bastianetto, S., Quirion, R., 2002. Alzheimer's disease and thebasal forebrain cholinergic system: relations to beta-amyloid peptides, cognition,and treatment strategies. Prog. Neurobiol. 68 (3), 209–245.

Baglietto-Vargas, D., Moreno-Gonzalez, I., Sanchez-Varo, R., Jimenez, S., Trujillo-Estrada, L., Sanchez-Mejias, E., Torres, M., Romero-Acebal, M., Ruano, D., Vizuete,M., Vitorica, J., Gutierrez, A., 2010. Calretinin interneurons are early targets of ex-tracellular amyloid-beta pathology in PS1/AbetaPP Alzheimer mice hippocampus.J. Alzheimers Dis. 21 (1), 119–132. doi:10.3233/JAD-2010-100066.

Bareggi, S.R., Franceschi, M., Bonini, L., Zecca, L., Smirne, S., 1982. Decreased CSF con-centrations of homovanillic acid and gamma-aminobutyric acid in Alzheimer's dis-ease. Age- or disease-related modifications? Arch. Neurol. 39 (11), 709–712.

Bartus, R.T., Dean, R.L., Beer, B., Lippa, A.S., 1982. The cholinergic hypothesis of geriatricmemory dysfunction. Science 217 (4558), 408–414.

Belarbi, K., Schindowski, K., Burnouf, S., Caillierez, R., Grosjean, M.-E., Demeyer, D.,Hamdane, M., Sergeant, N., Blum, D., Buée, L., 2009. Early Tau pathology involvingthe septo-hippocampal pathway in a Tau transgenic model: relevance to Alzhei-mer's disease. Curr. Alzheimer Res. 6 (2), 152–157.

Bertram, L., Tanzi, R.E., 2005. The genetic epidemiology of neurodegenerative disease. J.Clin. Invest. 115 (6), 1449–1457. doi:10.1172/JCI24761.

Bertram, L., Lill, C.M., Tanzi, R.E., 2010. The genetics of Alzheimer disease: back to thefuture. Neuron 68 (2), 270–281. doi:10.1016/j.neuron.2010.10.013.

Boekhoorn, K., Terwel, D., Biemans, B., Borghgraef, P., Wiegert, O., Ramakers, G.J.A., deVos, K., Krugers, H., Tomiyama, T., Mori, H., Joels, M., van Leuven, F., Lucassen,P.J., 2006. Improved long-term potentiation and memory in young tau-P301Ltransgenic mice before onset of hyperphosphorylation and tauopathy. J. Neurosci.26 (13), 3514–3523. doi:10.1523/JNEUROSCI.5425-05.2006.

Brady, D.R., Mufson, E.J., 1997. Parvalbumin-immunoreactive neurons in the hippo-campal formation of Alzheimer's diseased brain. Neuroscience 80 (4), 1113–1125.

Brecht, W.J., Harris, F.M., Chang, S., Tesseur, I., Yu, G.-Q., Xu, Q., Dee Fish, J., Wyss-Coray,T., Buttini, M., Mucke, L., Mahley, R.W., Huang, Y., 2004. Neuron-specific apolipo-protein e4 proteolysis is associated with increased tau phosphorylation in brainsof transgenic mice. J. Neurosci. 24 (10), 2527–2534. doi:10.1523/JNEUR-OSCI.4315-03.2004.

Brosh, I., Barkai, E., 2009. Learning-induced enhancement of feedback inhibitory synap-tic transmission. Learn. Mem. 16 (7), 413–416. doi:10.1101/lm.1430809.

Buttini, M., Yu, G.-Q., Shockley, K., Huang, Y., Jones, B., Masliah, E., Mallory, M., Yeo, T.,Longo, F.M., Mucke, L., 2002. Modulation of Alzheimer-like synaptic and choliner-gic deficits in transgenic mice by human apolipoprotein E depends on isoform,aging, and overexpression of amyloid beta peptides but not on plaque formation.J. Neurosci. 22 (24), 10539–10548.

Cassel, J.-C., Mathis, C., Majchrzak, M., Moreau, P.-H., Dalrymple-Alford, J.C., 2008.Coexisting cholinergic and parahippocampal degeneration: a key to memory lossin dementia and a challenge for transgenic models? Neurodegener. Dis. 5 (5),304–317. doi:10.1159/000135615.

11D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

Chao, H.-T., Chen, H., Samaco, R.C., Xue, M., Chahrour, M., Yoo, J., Neul, J.L., Gong, S., Lu,H.-C., Heintz, N., Ekker, M., Rubenstein, J.L.R., Noebels, J.L., Rosenmund, C., Zoghbi,H.Y., 2010. Dysfunction in GABA signalling mediates autism-like stereotypies andRett syndrome phenotypes. Nature 468 (7321), 263–269. doi:10.1038/nature09582.

Cheung, K.-H., Shineman, D., Müller, M., Cárdenas, C., Mei, L., Yang, J., Tomita, T.,Iwatsubo, T., Lee, V.M.-Y., Foskett, J.K., 2008. Mechanism of Ca2+ disruption in Alz-heimer's disease by presenilin regulation of InsP3 receptor channel gating. Neuron58 (6), 871–883. doi:10.1016/j.neuron.2008.04.015.

Coggeshall, R.E., Lekan, H.A., 1996. Methods for determining numbers of cells and syn-apses: a case for more uniform standards of review. J. Comp. Neurol. 364 (1), 6–15.doi:10.1002/(SICI)1096-9861(19960101)364:1b6:AID-CNE2>3.0.CO;2-9.

Contestabile, A., Ciani, E., Contestabile, A., 2008. The place of choline acetyltransferaseactivity measurement in the “cholinergic hypothesis” of neurodegenerative dis-eases. Neurochem. Res. 33 (2), 318–327. doi:10.1007/s11064-007-9497-4.

Corder, E.H., Saunders, A.M., Strittmatter, W.J., Schmechel, D.E., Gaskell, P.C., Small,G.W., Roses, A.D., Haines, J.L., Pericak-Vance, M.A., 1993. Gene dose of apolipopro-tein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science261 (5123), 921–923.

Cui, Y., Costa, R.M., Murphy, G.G., Elgersma, Y., Zhu, Y., Gutmann, D.H., Parada, L.F.,Mody, I., Silva, A.J., 2008. Neurofibromin regulation of ERK signaling modulatesGABA release and learning. Cell 135 (3), 549–560. doi:10.1016/j.cell.2008.09.060.

Davis, K.L., Mohs, R.C., Marin, D., Purohit, D.P., Perl, D.P., Lantz, M., Austin, G.,Haroutunian, V., 1999. Cholinergic markers in elderly patients with early signs ofAlzheimer disease. JAMA 281 (15), 1401–1406.

DeKosky, S.T., Ikonomovic, M.D., Styren, S.D., Beckett, L., Wisniewski, S., Bennett, D.A.,Cochran, E.J., Kordower, J.H., Mufson, E.J., 2002. Upregulation of choline acetyl-transferase activity in hippocampus and frontal cortex of elderly subjects withmild cognitive impairment. Ann. Neurol. 51 (2), 145–155.

Demuro, A., Mina, E., Kayed, R., Milton, S.C., Parker, I., Glabe, C.G., 2005. Calcium dysre-gulation and membrane disruption as a ubiquitous neurotoxic mechanism of solu-ble amyloid oligomers. J. Biol. Chem. 280 (17), 17294–17300. doi:10.1074/jbc.M500997200.

Demuro, A., Parker, I., Stutzmann, G.E., 2010. Calcium signaling and amyloid toxicity inAlzheimer disease. J. Biol. Chem. 285 (17), 12463–12468. doi:10.1074/jbc.R109.080895.

Deshpande, A., Mina, E., Glabe, C., Busciglio, J., 2006. Different conformations of amy-loid beta induce neurotoxicity by distinct mechanisms in human cortical neurons.J. Neurosci. 26 (22), 6011–6018. doi:10.1523/JNEUROSCI.1189-06.2006.

Dudek, F.E., Sutula, T.P., 2007. Epileptogenesis in the dentate gyrus: a critical perspec-tive. Prog. Brain Res. 163, 755–773. doi:10.1016/S0079-6123(07), 63041-6.

Düzel, E., Penny, W.D., Burgess, N., 2010. Brain oscillations and memory. Curr. Opin.Neurobiol. 20 (2), 143–149. doi:10.1016/j.conb.2010.01.004.

Eckenstein, F., Sofroniew, M.V., 1983. Identification of central cholinergic neurons con-taining both choline acetyltransferase and acetylcholinesterase and of central neu-rons containing only acetylcholinesterase. J. Neurosci. 3 (11), 2286–2291.

Filali, M., Lalonde, R., Rivest, S., 2011. Subchronic memantine administration on spatiallearning, exploratory activity, and nest-building in an APP/PS1 mouse model ofAlzheimer's disease. Neuropharmacology 60 (6), 930–936. doi:10.1016/j.neuropharm.2011.01.035.

Fonseca, M., Soriano, E., 1995. Calretinin-immunoreactive neurons in the normalhuman temporal cortex and in Alzheimer's disease. Brain Res. 691 (1–2),83–91.

Fonseca, M., Soriano, E., Ferrer, I., Martinez, A., Tuñon, T., 1993. Chandelier cell axonsidentified by parvalbumin-immunoreactivity in the normal human temporal cor-tex and in Alzheimer's disease. Neuroscience 55 (4), 1107–1116.

Freund, T.F., Antal, M., 1988. GABA-containing neurons in the septum control inhibito-ry interneurons in the hippocampus. Nature 336 (6195), 170–173. doi:10.1038/336170a0.

Games, D., Adams, D., Alessandrini, R., Barbour, R., Berthelette, P., Blackwell, C., Carr, T.,Clemens, J., Donaldson, T., Gillespie, F., 1995. Alzheimer-type neuropathology intransgenic mice overexpressing V717F beta-amyloid precursor protein. Nature373 (6514), 523–527. doi:10.1038/373523a0.

Garrido Sanabria, E.R., Castañeda, M.T., Banuelos, C., Perez-Cordova, M.G., Hernandez,S., Colom, L.V., 2006. Septal GABAergic neurons are selectively vulnerable topilocarpine-induced status epilepticus and chronic spontaneous seizures. Neuro-science 142 (3), 871–883. doi:10.1016/j.neuroscience.2006.06.057.

German, D.C., Yazdani, U., Speciale, S.G., Pasbakhsh, P., Games, D., Liang, C.-L., 2003.Cholinergic neuropathology in a mouse model of Alzheimer's disease. J. Comp.Neurol. 462 (4), 371–381. doi:10.1002/cne.10737.

Gilmor, M.L., Erickson, J.D., Varoqui, H., Hersh, L.B., Bennett, D.A., Cochran, E.J., Mufson,E.J., Levey, A.I., 1999. Preservation of nucleus basalis neurons containing cholineacetyltransferase and the vesicular acetylcholine transporter in the elderly withmild cognitive impairment and early Alzheimer's disease. J. Comp. Neurol. 411(4), 693–704.

Goate, A., 2006. Segregation of a missense mutation in the amyloid beta-protein pre-cursor gene with familial Alzheimer's disease. J. Alzheimers Dis. 9 (3 Suppl.),341–347.

Götz, J., Ittner, L.M., 2008. Animal models of Alzheimer's disease and frontotemporaldementia. Nat. Rev. Neurosci. 9 (7), 532–544. doi:10.1038/nrn2420.

Gotz, J., Chen, F., Barmettler, R., Nitsch, R.M., 2001a. Tau filament formation in trans-genic mice expressing P301L tau. J. Biol. Chem. 276 (1), 529–534. doi:10.1074/jbc.M006531200.

Gotz, J., Chen, F., van Dorpe, J., Nitsch, R.M., 2001b. Formation of neurofibrillary tanglesin P301l tau transgenic mice induced by Abeta 42 fibrils. Science 293 (5534),1491–1495. doi:10.1126/science.1062097.

Grueninger, F., Bohrmann, B., Czech, C., Ballard, T.M., Frey, J.R., Weidensteiner, C., vonKienlin, M., Ozmen, L., 2010. Phosphorylation of Tau at S422 is enhanced byAbeta in TauPS2APP triple transgenic mice. Neurobiol. Dis. 37 (2), 294–306.doi:10.1016/j.nbd.2009.09.004.

Guijarro, C., Rutz, S., Rothmaier, K., Turiault, M., Zhi, Q., Naumann, T., Frotscher, M.,Tronche, F., Jackisch, R., Kretz, O., 2006. Maturation and maintenance of cholinergicmedial septum neurons require glucocorticoid receptor signaling. J. Neurochem. 97(3), 747–758. doi:10.1111/j.1471-4159.2006.03728.x.

Hangya, B., Borhegyi, Z., Szilágyi, N., Freund, T.F., Varga, V., 2009. GABAergic neurons ofthe medial septum lead the hippocampal network during theta activity. J. Neurosci.29 (25), 8094–8102. doi:10.1523/JNEUROSCI.5665-08.2009.

Hardingham, G.E., Fukunaga, Y., Bading, H., 2002. Extrasynaptic NMDARs oppose syn-aptic NMDARs by triggering CREB shut-off and cell death pathways. Nat. Neurosci.5 (5), 405–414. doi:10.1038/nn835.

Harris, F.M., Brecht, W.J., Xu, Q., Mahley, R.W., Huang, Y., 2004. Increased tau phosphor-ylation in apolipoprotein E4 transgenic mice is associated with activation of extra-cellular signal-regulated kinase: modulation by zinc. J. Biol. Chem. 279 (43),44795–44801. doi:10.1074/jbc.M408127200.

Helvink, B., Holroyd, S., 2006. Buspirone for stereotypic movements in elderly withcognitive impairment. J. Neuropsychiatry Clin. Neurosci. 18 (2), 242–244.doi:10.1176/appi.neuropsych.18.2.242.

Herz, J., 2009. Apolipoprotein E receptors in the nervous system. Curr. Opin. Lipidol. 20(3), 190–196. doi:10.1097/MOL.0b013e32832d3a10.

Holcomb, L., Gordon, M.N., McGowan, E., Yu, X., Benkovic, S., Jantzen, P., Wright, K.,Saad, I., Mueller, R., Morgan, D., Sanders, S., Zehr, C., O'Campo, K., Hardy, J., Prada,C.M., Eckman, C., Younkin, S., Hsiao, K., Duff, K., 1998. Accelerated Alzheimer-type phenotype in transgenic mice carrying both mutant amyloid precursor pro-tein and presenilin 1 transgenes. Nat. Med. 4 (1), 97–100.

Hsiao, K., Chapman, P., Nilsen, S., Eckman, C., Harigaya, Y., Younkin, S., Yang, F., Cole, G.,1996. Correlative memory deficits, Abeta elevation, and amyloid plaques in trans-genic mice. Science 274 (5284), 99–102.

Hu, J.-H., Ma, Y.-H., Jiang, J., Yang, N., Duan, S.-h., Jiang, Z.-H., Mei, Z.-T., Fei, J., Guo, L.-H.,2004. Cognitive impairment in mice over-expressing gamma-aminobutyric acidtransporter 1 (GAT1). Neuroreport 15 (1), 9–12.

Hutton, M., Lendon, C.L., Rizzu, P., Baker, M., Froelich, S., Houlden, H., Pickering-Brown,S., Chakraverty, S., Isaacs, A., Grover, A., Hackett, J., Adamson, J., Lincoln, S., Dickson,D., Davies, P., Petersen, R.C., Stevens, M., de Graaff, E., Wauters, E., van Baren, J.,Hillebrand, M., Joosse, M., Kwon, J.M., Nowotny, P., Che, L.K., Norton, J., Morris,J.C., Reed, L.A., Trojanowski, J., Basun, H., Lannfelt, L., Neystat, M., Fahn, S., Dark,F., Tannenberg, T., Dodd, P.R., Hayward, N., Kwok, J.B., Schofield, P.R., Andreadis,A., Snowden, J., Craufurd, D., Neary, D., Owen, F., Oostra, B.A., Hardy, J., Goate, A.,van Swieten, J., Mann, D., Lynch, T., Heutink, P., 1998. Association of missenseand 5′-splice-site mutations in tau with the inherited dementia FTDP-17. Nature393 (6686), 702–705. doi:10.1038/31508.

Jasinska, M., Siucinska, E., Cybulska-Klosowicz, A., Pyza, E., Furness, D.N., Kossut, M.,Glazewski, S., 2010. Rapid, learning-induced inhibitory synaptogenesis in murinebarrel field. J. Neurosci. 30 (3), 1176–1184. doi:10.1523/JNEUROSCI.2970-09.2010.

Johansen, F.F., 1993. Interneurons in rat hippocampus after cerebral ischemia. Morpho-metric, functional, and therapeutic investigations. Acta Neurol. Scand. (Suppl.150), 1–32.

Koliatsos, V.E., Kecojevic, A., Troncoso, J.C., Gastard, M.C., Bennett, D.A., Schneider, J.A.,2006. Early involvement of small inhibitory cortical interneurons in Alzheimer'sdisease. Acta Neuropathol. 112 (2), 147–162. doi:10.1007/s00401-006-0068-6.

Kumar, P., Kalonia, H., Kumar, A. Huntington's disease: pathogenesis to animal models.Pharmacol Rep 62 (1), 1–14.

Lecourtier, L., de Vasconcelos, A.P., Cosquer, B., Cassel, J.-C., 2010. Combined lesions ofGABAergic and cholinergic septal neurons increase locomotor activity and potenti-ate the locomotor response to amphetamine. Behav. Brain Res. 213 (2), 175–182.doi:10.1016/j.bbr.2010.04.050.

Lei, S.Z., Zhang, D., Abele, A.E., Lipton, S.A., 1992. Blockade of NMDA receptor-mediatedmobilization of intracellular Ca2+ prevents neurotoxicity. Brain Res. 598 (1–2),196–202.

Liu, Y.-H., Wang, L., Wei, L.-C., Huang, Y.-G., Chen, L.-W., 2009. Up-regulation of D-serine might induce GABAergic neuronal degeneration in the cerebral cortex andhippocampus in the mouse pilocarpine model of epilepsy. Neurochem. Res. 34(7), 1209–1218. doi:10.1007/s11064-008-9897-0.

Mendez, M., Lim, G., 2003. Seizures in elderly patients with dementia: epidemiologyand management. Drugs Aging 20 (11), 791–803.

Mesulam, M., 2004. The cholinergic lesion of Alzheimer's disease: pivotal factor or sideshow? Learn. Mem. 11 (1), 43–49. doi:10.1101/lm.69204.

Miller, J.W., Turner, G.M., Gray, B.C., 1994. Anticonvulsant effects of the experimentalinduction of hippocampal theta activity. Epilepsy Res. 18 (3), 195–204.

Milner, T.A., Hammel, J.R., Ghorbani, T.T., Wiley, R.G., Pierce, J.P., 1999. Septal choliner-gic deafferentation of the dentate gyrus results in a loss of a subset of neuropeptideY somata and an increase in synaptic area on remaining neuropeptide Y dendrites.Brain Res. 831 (1–2), 322–336.

Minkeviciene, R., Rheims, S., Dobszay, M.B., Zilberter, M., Hartikainen, J., Fülöp, L.,Penke, B., Zilberter, Y., Harkany, T., Pitkänen, A., Tanila, H., 2009. Amyloid beta-induced neuronal hyperexcitability triggers progressive epilepsy. J. Neurosci. 29(11), 3453–3462. doi:10.1523/JNEUROSCI.5215-08.2009.

Monnerie, H., Le Roux, P.D., 2007. Reduced dendrite growth and altered glutamic aciddecarboxylase (GAD) 65- and 67-kDa isoform protein expression from mouse cor-tical GABAergic neurons following excitotoxic injury in vitro. Exp. Neurol. 205 (2),367–382. doi:10.1016/j.expneurol.2007.02.007.

Naumann, T., Casademunt, E., Hollerbach, E., Hofmann, J., Dechant, G., Frotscher, M.,Barde, Y.A., 2002. Complete deletion of the neurotrophin receptor p75NTR leads

12 D. Loreth et al. / Neurobiology of Disease 47 (2012) 1–12

to long-lasting increases in the number of basal forebrain cholinergic neurons. J.Neurosci. 22 (7), 2409–2418.

Nyatsanza, S., Shetty, T., Gregory, C., Lough, S., Dawson, K., Hodges, J.R., 2003. A study ofstereotypic behaviours in Alzheimer's disease and frontal and temporal variantfrontotemporal dementia. J. Neurol. Neurosurg. Psychiatry 74 (10), 1398–1402.

Oddo, S., Caccamo, A., Shepherd, J.D., Murphy, M.P., Golde, T.E., Kayed, R., Metherate, R.,Mattson, M.P., Akbari, Y., Laferla, F.M., 2003. Triple-transgenic model of Alzhei-mer's disease with plaques and tangles: intracellular Abeta and synaptic dysfunc-tion. Neuron 39 (3), 409–421.

Oliveira, J.M.A., 2010. Nature and cause of mitochondrial dysfunction in Huntington'sdisease: focusing on huntingtin and the striatum. J. Neurochem. 114 (1), 1–12.doi:10.1111/j.1471-4159.2010.06741.x.

Ozmen, L., Albientz, A., Czech, C., Jacobsen, H., 2009. Expression of transgenic APPmRNA is the key determinant for beta-amyloid deposition in PS2APP transgenicmice. Neurodegener. Dis. 6 (1–2), 29–36. doi:10.1159/000170884.

Palop, J.J., Chin, J., Roberson, E.D., Wang, J., Thwin, M.T., Bien-Ly, N., Yoo, J., Ho, K.O., Yu,G.-Q., Kreitzer, A., Finkbeiner, S., Noebels, J.L., Mucke, L., 2007. Aberrant excitatoryneuronal activity and compensatory remodeling of inhibitory hippocampal circuitsin mouse models of Alzheimer's disease. Neuron 55 (5), 697–711. doi:10.1016/j.neuron.2007.07.025.

Perry, E.K., Gibson, P.H., Blessed, G., Perry, R.H., Tomlinson, B.E., 1977. Neurotransmitterenzyme abnormalities in senile dementia. Choline acetyltransferase and glutamicacid decarboxylase activities in necropsy brain tissue. J. Neurol. Sci. 34 (2),247–265.

Poirier, R., Veltman, I., Pflimlin, M.C., Knoflach, F., Metzger, F., 2010. Enhanced dentategyrus synaptic plasticity but reduced neurogenesis in a mouse model of amyloid-osis. Neurobiol. Dis. 40 (2), 386–393. doi:10.1016/j.nbd.2010.06.014.

Popović, M., Caballero-Bleda, M., Kadish, I., van Groen, T., 2008. Subfield and layer-specific depletion in calbindin-D28K, calretinin and parvalbumin immunoreactivi-ty in the dentate gyrus of amyloid precursor protein/presenilin 1 transgenic mice.Neuroscience 155 (1), 182–191. doi:10.1016/j.neuroscience.2008.05.023.

Revel, F.G., Moreau, J.-L., Gainetdinov, R.R., Bradaia, A., Sotnikova, T.D., Mory, R., Durkin,S., Zbinden, K.G., Norcross, R., Meyer, C.A., Metzler, V., Chaboz, S., Ozmen, L., Trube,G., Pouzet, B., Bettler, B., Caron, M.G., Wettstein, J.G., Hoener, M.C., 2011. TAAR1 ac-tivation modulates monoaminergic neurotransmission, preventing hyperdopami-nergic and hypoglutamatergic activity. Proc. Natl. Acad. Sci. U. S. A. 108 (20),8485–8490. doi:10.1073/pnas.1103029108.

Rhein, V., Song, X., Wiesner, A., Ittner, L.M., Baysang, G., Meier, F., Ozmen, L.,Bluethmann, H., Drose, S., Brandt, U., Savaskan, E., Czech, C., Gotz, J., Eckert, A.,2009. Amyloid- and tau synergistically impair the oxidative phosphorylation sys-tem in triple transgenic Alzheimer's disease mice. Proc. Natl. Acad. Sci.doi:10.1073/pnas.0905529106

Seidl, R., Cairns, N., Singewald, N., Kaehler, S.T., Lubec, G., 2001. Differences betweenGABA levels in Alzheimer's disease and Down syndrome with Alzheimer-like neu-ropathology. Naunyn Schmiedebergs Arch. Pharmacol. 363 (2), 139–145.

Sherrington, R., Rogaev, E.I., Liang, Y., Rogaeva, E.A., Levesque, G., Ikeda, M., Chi, H., Lin,C., Li, G., Holman, K., Tsuda, T., Mar, L., Foncin, J.F., Bruni, A.C., Montesi, M.P., Sorbi,

S., Rainero, I., Pinessi, L., Nee, L., Chumakov, I., Pollen, D., Brookes, A., Sanseau, P.,Polinsky, R.J., Wasco, W., Da Silva, H.A., Haines, J.L., Perkicak-Vance, M.A., Tanzi,R.E., Roses, A.D., Fraser, P.E., Rommens, J.M., St George-Hyslop, P.H., 1995. Cloningof a gene bearing missense mutations in early-onset familial Alzheimer's disease.Nature 375 (6534), 754–760. doi:10.1038/375754a0.

Shetty, A.K., Hattiangady, B., Rao, M.S., 2009. Vulnerability of hippocampal GABA-ergicinterneurons to kainate-induced excitotoxic injury during old age. J. Cell. Mol. Med.13 (8B), 2408–2423. doi:10.1111/j.1582-4934.2009.00675.x.

Small, S.A., Duff, K., 2008. Linking Abeta and tau in late-onset Alzheimer's disease: a dualpathway hypothesis. Neuron 60 (4), 534–542. doi:10.1016/j.neuron.2008.11.007.

Sterniczuk, R., Dyck, R.H., Laferla, F.M., Antle, M.C., 2010. Characterization of the 3xTg-ADmouse model of Alzheimer's disease: Part 1. Circadian changes. Brain Res. 1348,139–148. doi:10.1016/j.brainres.2010.05.013.

Strittmatter, W.J., Weisgraber, K.H., Huang, D.Y., Dong, L.M., Salvesen, G.S., Pericak-Vance, M., Schmechel, D., Saunders, A.M., Goldgaber, D., Roses, A.D., 1993. Bindingof human apolipoprotein E to synthetic amyloid beta peptide: isoform-specific ef-fects and implications for late-onset Alzheimer disease. Proc. Natl. Acad. Sci. U. S. A.90 (17), 8098–8102.

Sturchler-Pierrat, C., Abramowski, D., Duke, M., Wiederhold, K.H., Mistl, C., Rothacher,S., Ledermann, B., Bürki, K., Frey, P., Paganetti, P.A., Waridel, C., Calhoun, M.E.,Jucker, M., Probst, A., Staufenbiel, M., Sommer, B., 1997. Two amyloid precursorprotein transgenic mouse models with Alzheimer disease-like pathology. Proc.Natl. Acad. Sci. U. S. A. 94 (24), 13287–13292.

Stutzmann, G.E., Smith, I., Caccamo, A., Oddo, S., Laferla, F.M., Parker, I., 2006. Enhancedryanodine receptor recruitment contributes to Ca2+ disruptions in young, adult,and aged Alzheimer's disease mice. J. Neurosci. 26 (19), 5180–5189. doi:10.1523/JNEUROSCI.0739-06.2006.

Takahashi, H., Brasnjevic, I., Rutten, B.P.F., van der Kolk, N., Perl, D.P., Bouras, C.,Steinbusch, H.W.M., Schmitz, C., Hof, P.R., Dickstein, D.L., 2010. Hippocampal inter-neuron loss in an APP/PS1 double mutant mouse and in Alzheimer's disease. BrainStruct. Funct. 214 (2–3), 145–160. doi:10.1007/s00429-010-0242-4.

Tiraboschi, P., Hansen, L.A., Alford, M., Masliah, E., Thal, L.J., Corey-Bloom, J., 2000. Thedecline in synapses and cholinergic activity is asynchronous in Alzheimer's disease.Neurology 55 (9), 1278–1283.

West, M.J., Slomianka, L., Gundersen, H.J., 1991. Unbiased stereological estimation ofthe total number of neurons in the subdivisions of the rat hippocampus usingthe optical fractionator. Anat. Rec. 231 (4), 482–497. doi:10.1002/ar.1092310411.

Whitehouse, P.J., Price, D.L., Struble, R.G., Clark, A.W., Coyle, J.T., Delon, M.R., 1982. Alz-heimer's disease and senile dementia: loss of neurons in the basal forebrain. Sci-ence 215 (4537), 1237–1239.

Yamin, G., 2009. NMDA receptor-dependent signaling pathways that underlie amyloidbeta-protein disruption of LTP in the hippocampus. J. Neurosci. Res. 87 (8),1729–1736. doi:10.1002/jnr.21998.

Zimmer, R., Teelken, A.W., Trieling, W.B., Weber, W., Weihmayr, T., Lauter, H., 1984.Gamma-aminobutyric acid and homovanillic acid concentration in the CSF ofpatients with senile dementia of Alzheimer's type. Arch. Neurol. 41 (6),602–604.