Igh

JCa

b

c

d

e

f

h

•••

a

ARA

KADCT

1

n

mno

sB

0h

Neuroscience Letters 552 (2013) 129– 134

Contents lists available at ScienceDirect

Neuroscience Letters

jou rn al hom epage: www.elsev ier .com/ locate /neule t

ncreased densities of resting and activated microglia in the dentateyrus follow senile plaque formation in the CA1 subfield of theippocampus in the triple transgenic model of Alzheimer’s disease�

.J. Rodrígueza,b,∗, H.N. Noristanic, T. Hilditchd, M. Olabarriad,.Y. Yehd, J. Wittone,f, A. Verkhratskya,b,d

Ikerbasque, Basque Foundation for Science, 48011 Bilbao, SpainDepartment of Neurosciences, University of the Basque Country UPV/EHU, 48940 Leioa, SpainINSERM U1051, Institute for Neurosciences of Montpellier, 80, Av. Augustin Fliche, F-34091 Montpellier Cedex 05, FranceFaculty of Life Sciences, The University of Manchester, Manchester, UKSchool of Physiology and Pharmacology, University of Bristol, University Walk, Bristol BS8 1TD, UKPfizer Applied Neurophysiology Group, University of Bristol, University Walk, Bristol BS8 1TD, UK

i g h l i g h t s

Nobel studies of microglia in CA1 of hippocampus in the 3xTgAD.Microglial remodelling during AD progression.Study of resting and activated microglia in CA1.

r t i c l e i n f o

rticle history:eceived 10 May 2013ccepted 20 June 2013

eywords:ctivated microgliaentate gyrusA1 hippocampus

a b s t r a c t

Alzheimer’s disease (AD) is an irreversible neurodegenerative disease that is characterised by the pres-ence of �-amyloid (A�) plaques, neurofibrillary tangles (NFTs) and synaptic loss specifically in brainregions involved in learning and memory such as the neocortex and the hippocampus. A� depositions inthe form of neuritic plaques trigger activation of microglia that is believed to be a common neuropatho-logical feature of AD brains. As an integral part of the hippocampus, the dentate gyrus (DG) plays animportant role in cognitive function. Although post-mortem studies suggest later involvement of the DGinto the AD progression, changes in microglia have not been studied in this subfield of the hippocam-

3

riple transgenic AD pus. In the present study the numerical density (Nv, #/mm ) of both resting (identified by tomato lectin

staining) and activated (identified by Mac-1 immunoreactivity) microglia was analysed in the molecularlayer (ML) of the DG in the triple transgenic (3xTg-AD) mouse model of AD at different ages (9, 12 and 18months). The 3xTg-AD mouse model of AD showed a significant increase in the Nv of resting (by 75%) andactivated (by 67%) at 18 months of age compared to non-Tg controls. These results indicate a complexmicroglial remodelling during AD progression.

©

. Introduction

Microglial cells are the immuno-competent cells of the centralervous system (CNS) which mount an array of defensive responses

� This is an open-access article distributed under the terms of the Creative Com-ons Attribution-NonCommercial-No Derivative Works License, which permits

on-commercial use, distribution, and reproduction in any medium, provided theriginal author and source are credited.∗ Corresponding author at: Ikerbasque, Department of Neuroscience, The Univer-

ity of the Basque Country UPV/EHU, Technological Park, Bldg. 205, Floor-1, Laidaidea, 48170 Zamudio, Vizkaia, Spain. Tel.: +34 946018305; fax: +34 946018289.

E-mail address: j.rodriguez-arellano@ikerbasque.org (J.J. Rodríguez).

304-3940/$ – see front matter © 2013 The Authors. Published by Elsevier Ireland Ltd. Alttp://dx.doi.org/10.1016/j.neulet.2013.06.036

2013 The Authors. Published by Elsevier Ireland Ltd. All rights reserved.

to any threat to the CNS [6]. Originally described by Del Rio Hortegain 1919, microglial cells have myelomonocytic origin and invadethe CNS during early embryonic and postnatal development [6].Following entry into the CNS, microglia disseminate through thebrain parenchyma and acquire ramified phenotype characterisedby small cell bodies and several very thin processes. The pro-cesses of ramified microglial cells constantly scan their territorialdomains [6]. Insults to the CNS trigger the multi-stage programmeof microglial activation [6].

Due to their defensive role, microglial cells are involved in alltypes of CNS pathologies including Alzheimer’s disease (AD) [7,3]the most common form of dementia in the elderly that causes asevere and irreparable decline in cognition including learning and

l rights reserved.

1 ience L

mropw[

tmmho[retooups[

pot[atopApta(t

2

tuAtor

2

m[(tTcm1akPcsf

30 J.J. Rodríguez et al. / Neurosc

emory [2,3]. Histopathological hallmarks of AD brains are rep-esented by extracellular �-amyloid (A�) depositions in the formf neuritic plaques and appearance of intraneuronal hyperphos-horylated tau protein in the form of neurofibrillary tangles (NFTs)hich are accompanied by severe synaptic loss and neuronal death

2].AD-associated increase in microglia activation and prolifera-

ion was initially reported by McGeer et al., who found activatedicroglia in the hippocampus of AD brains concomitant with for-ation of A� plaques [9]. Increased number of activated microglia

as been repeatedly reported in the vicinity of A� plaques in vari-us transgenic mouse models of the disease including the PS1xAPP5] and the APPswe/PS1d9xYFP mice [10], although changes in theesting microglia were studied neither in AD nor in transgenic mod-ls mentioned earlier. Previously, we had reported that the tripleransgenic (3xTg-AD) mouse model of AD display increased numberf not only activated but also ramified microglia in the CA1 subfieldf the hippocampus [14]. This increase in the resting microglia pop-lation precedes the increase in activated microglia cells and occursrior to formation of extracellular A� plaque deposition in the CA1ubfield of the hippocampus in the 3xTg-AD mouse mode of AD14].

As an integral part of the hippocampus, the dentate gyrus (DG)lays an important role in memory [1]. The molecular layer (ML)f the DG regulates the flow of cortical input to the hippocampushrough the perforant path, which is critical for cognitive function1]. Despite pronounced synaptic loss in the ML of the DG [15], AD-ssociated neuronal loss is more prominent in the CA1 subfield ofhe hippocampus [17], which may be due to increased presencef A� plaques and NFT neuropathology in the CA1 subfield com-ared to the DG of the hippocampus [2]. Post-mortem analysis ofD brains have shown that the DG of the hippocampus resists A�laque deposition, NFT formation and associated neurodegenera-ion until the late stages of the disease (for review see [11]). Theim of the present study was to quantify the numerical density#/mm3) of both resting and activated microglial cells in the DG ofhe hippocampus in the 3xTg-AD mouse model of AD.

. Materials and methods

All animal procedures were carried out in accordance withhe United Kingdom Animals (Scientific Procedures) Act of 1986nder the license from the Home Office and were approved by thenimal Care and Ethical Committee at The University of Manches-

er (PCD 50/2506). All efforts were made to reduce the numberf animals used by following the 3Rs (reduction, refinement andeplacement).

.1. Mice

The procedure for generating the triple transgenic (3xTg-AD)ouse model of AD has been described in details previously

13]. In brief, the human APP cDNA harbouring the SwedishKM670/671NL, APPSwe) and human four repeat tau, harbouringhe P301L (tauP301L) mutations (both under control of the mousehy1.2 regulatory element) were co-microinjected into a single-ell embryo of a homozygous presenilin 1 (PS1M146V) knock-inouse [13]. The background of the PS1 knock-in mice is a hybrid

29/C57BL6. The non-transgenic (non-Tg) control mice used werelso from the same strain and genetic background as the PS1nock-in mice, but they harbour the endogenous wild-type mouse

S1 gene. All 3xTg-AD and non-Tg control mice were obtained byrossing homozygous breeders. The animals were housed in theame-sex cage, kept in 12 h light–dark cycles with free access toood and water.

etters 552 (2013) 129– 134

2.2. Fixation and tissue processing

Male 3xTg-AD and their respective non-Tg control mice ofdifferent ages (9, 12 and 18 months of age, n = 4–5) were anaes-thetised with an intraperitoneal injection of sodium pentobarbital(50 mg/kg). All mice were perfused through the aortic archwith 3.75% acrolein (TAAB, Berkshire, UK) in a solution of 2%paraformaldehyde (PFA, Sigma, UK) and 0.1 M phosphate buffer(PB) pH 7.4, followed by 2% PFA. Brains were cut into 4–5 mm coro-nal slabs of tissue, post-fixed in 2% PFA for 24 h and kept in 0.1 MPB, pH 7.4. Coronal brain sections were cut into 40–50 �m thick-ness using a vibrating microtome (VT1000S, Leica, Milton Keynes,UK). Free floating brain sections in 0.1 M PB, pH 7.4 were collectedand stored in cryoprotectant solution in 0.05 M PB at pH 7.4. Coronalbrain sections at levels −1.70 mm/−2.18 mm (hippocampus) poste-rior to bregma were selected for immunohistochemistry accordingto the mouse brain atlas of Paxinos and Franklin [12].

2.3. Antibodies

The ramified (resting) microglia cells were identified usingbiotinylated tomato lectin (Sigma–Aldrich, Poole, UK), as previ-ously described [14]. Tomato lectin is a sugar-binding glycoproteinthat has a specific affinity for poly-N-acetyl lactosamine residuesoccurring on the surface membranes of ramified microglia cells andendothelial cells. A polyclonal affinity-purified rat antiserum raisedagainst CD11b (Mac-1, Serotec, Kidlington, UK) was used for thedetermination of amoeboid (activated) microglia cells. Mac-1 anti-body has a specific affinity for the microglial surface membranereceptor CD11b, which is highly expressed by amoeboid microgliaand is commonly used as a specific marker for amoeboid microgliacells. For the identification of A� plaques a monoclonal mouseantibody against the amino acid residues 1–16 of beta amyloid(Covance, Emeryville, CA, USA) was used. The specificity of theseantibodies has been reported previously using immunohistochem-istry [13,14]. To assess for non-specific background labelling orcross-reactivity between antibodies derived from different hostspecies, a series of control experiments were performed. Omis-sion of primary and/or secondary antibodies from the incubationsolutions resulted in a total absence of target labelling (data notshown).

2.4. Immunohistochemistry

To optimise the detection of Mac-1 and tomato lectin (TL)microglia, brain sections of all 3xTg-AD and non-Tg control ani-mals were processed by using the highly sensitive avidin–biotinperoxidise complex (ABC) method, as previously described [14].Brain sections were treated with 1% sodium borohydride in 0.1 MPB for 30 min and were then washed with 0.1 M PB followed by0.1 M trizma base saline (TS) for further 10 min before incuba-tion in 0.5% bovine serum albumin (BSA, Sigma, Gillingham, UK)in 0.1 M TS and 0.25% Triton (Sigma, Gillingham, UK, ×100) for30 min. All brain sections were incubated for 48 h at room tem-perature in primary antibody solution (biotinylated tomato lectin,1:500, Sigma–Aldrich, Poole, UK or rat anti-Mac-1, 1:1000, Sigma,Saint Louise, MO, USA). For tomato lectin labelling, brain sectionswere directly incubated in the ABC (Elite kit, Vector LaboratoriesLtd., Peterborough, UK) following their incubation in the primaryantibody as described previously [14]. For Mac-1 staining, brainsections were incubated in 1:200 dilutions of secondary antibodies(biotinylated donkey anti-rat IgG, Jackson Immunoresearch, Strat-

ech Scientific Ltd., Soham, UK) for 1 h at room temperature followedby incubation in ABC solution for further 30 min. The peroxidasereaction product was visualised for both TL and Mac-1 labelling, byincubating the brain sections in a chromogen solution containing

ence L

00

rrt�CUlodg

2

im3ps[iaTaeDwamaTwrtb

2

Ai3fwaP

3

lai3s

tacm(t

J.J. Rodríguez et al. / Neurosci

.022% of 3,3′-diaminobenzidine (DAB, Aldrich, Gilligham, UK) and

.003% H2O2 for 6 min, as previously described [13,14].For the detection and determination of microglia cells and their

elationship with A� neuritic plaques dual indirect immunofluo-escence was used. The sections were incubated for 48 h at roomemperature in a primary antibody cocktail containing mouse anti-

amyloid monoclonal antibody (A�, 1:2000, Covance, Emeryville,A, USA) and rat anti-Mac-1 (1:30,000, Sigma, Saint Louise, MO,SA) simultaneously. For the dual indirect immunofluorescence

abelling, A� and Mac-1 were detected in a sequential mannern the same sections by incubation with 1:200 dilutions of rho-amine (TRITC)-conjugated goat anti-mouse and FITC-conjugatedoat anti-rat (Invitrogen, Paisley, UK) IgG, respectively.

.5. Cell counting

The numerical density (Nv, #/mm3) of tomato lectin-mmunoreactive (TL-IR) and Mac-1-immunoreactive (Mac-1-IR)

icroglia were determined at 9, 12 and 18 months of age in bothxTg-AD and non-Tg mice in the dentate gyrus (DG) of the hip-ocampus. For this, 3–4 representative non-consecutive coronalections throughout the dorsal hippocampus at levels −1.70/−2.1812] were quantified accounting for an analysed volume of approx-mately 6,000,000 �m3 in the DG of the hippocampus. The specificnalysed areas were the molecular layer (ML) of the DG. AllL- and Mac-1-IR microglia cells were clearly labelled against

light background making them easy to be identified with anqual chance of being counted. The surface area of the ML of theG was measured using a Nikon Eclipse 80i microscope coupledith a 8001 MicroFIRE camera and a computer-assisted imaging

nalysis (ImageJ 1.32j, NIH, USA) software. All TL and Mac-1-IRicroglia cells located in the ML of the DG were counted using

10 mm × 10 mm graticule under a Nikon Eclipse 80i microscope.he boundaries of areas in which TL- and Mac-1-IR microglia cellsere to be counted were clearly delineated, thus, counts were

eproducible. All counting of TL- and Mac-1-IR microglia cells inhe DG were conducted by a single investigator (T. H) who waslind to the subject number and group assignment.

.6. Statistical analysis

Data are expressed as mean ± standard error of the mean (SEM).nalysis of variance (one-way) was used to examine differences

n the mean number of TL-IR and Mac-1-IR microglia cells in thexTg-AD and non-Tg control mice during ageing followed by a Bon-erroni’s post hoc test where appropriate. Whenever two groupsere compared, an unpaired t-test was applied. Significance was

ccepted at p ≤ 0.05. All data were analysed using the graphPadrism 4.0 (GraphPad Software, Inc., La Jolla, CA, USA) program.

. Results

The populations of resting and activated microglia were ana-ysed in the ML of the DG in the 3xTg-AD and non-Tg control micet three different ages namely: 9, 12 and 18 months. Previous stud-es have shown that formation of A� plaques in the hippocampus ofxTg-AD mice begins between 9 and 12 months of age and becomesubstantial by 18 months [13].

In the ML of the DG of both non-Tg control and 3xTg-AD mice,wo different phenotypes of microglial cells were observed (Fig. 1Bnd D). The resting tomato-lectin immunoreactive (TL-IR) microglia

ells were characterised by a small cell body equipped with thin to-edium ramified processes extending to the surrounding neuropil

Fig. 1B). The other type of microglial cells, which were imunoreac-ive for Mac-1, showed enlarged cell bodies from which processes

etters 552 (2013) 129– 134 131

with an enlarged and thicker appearance emanated (Fig. 1D and E),thus being reflective of an activated phenotype.

3.1. Resting microglia in AD

No differences were observed between the morphological char-acteristics of TL-IR ramified microglia in either 3xTg-AD or non-Tgcontrol mice, irrespective of their age (Fig. 1B). When comparingthe age-effect on the Nv of resting microglia, the 3xTg-AD mousemodel of AD showed a significant increase at 18 months of age com-pared to 3xTg-AD mice at both 9 and 12 months of age (F5,21 = 9.994,56%, p < 0.05 compared to 9 months and 60%, p < 0.01 compared to12 months, respectively; Fig. 1A). No age-associated changes wereobserved in the Nv of resting microglia in the ML of the DG in thenon-Tg control group between 9, 12 and 18 months of age. Compari-son between the two genotypes revealed a non-significant increasein the 3xTg-AD mouse model of AD compared to age-matched non-Tg control group at 9 (10%, p = 0.3609) and 12 (12%, p = 0.4413)months of age (Fig. 1A). Significant increase in the Nv of restingmicroglia was observed at 18 months of age in the 3xTg-AD mousemodel of AD compared to non-Tg control group (75%, p = 0.0063,Fig. 1A).

3.2. Activated microglia in AD

Similarly to resting microglia, no changes were observed inthe typical morphology of Mac-1-IR activated microglia in all agegroups between 3xTg-AD and non-Tg control mice. When compar-ing the age-effect on the Nv of activated microglia, the 3xTg-ADmouse model of AD showed a significant increase at 18 months ofage compared to 3xTg-AD mice at 9 (F5,20 = 4.934, 64%, p < 0.01). Anon-significant increase in the Nv of activated microglia was alsoobserved in 3xTg-AD mice between 12 and 18 months of age (25%,p > 0.05; Fig. 1C).

The 3xTg-AD mouse model of AD showed a significant increaseat 18 months of age compared to 3xTg-AD mice at 9 but not at 12months of age (F5,20 = 4.934, 64%, p < 0.01 compared to 9 monthsand 25%, p > 0.05 compared to 12 months, respectively; Fig. 1C).No age-associated changes were observed in the Nv of activatedmicroglia in the ML of the DG in non-Tg control group between 9,12 and 18 months of age. Comparison between the two genotypesshowed a non-significant decrease in the 3xTg-AD mouse model ofAD compared to age-matched non-Tg control group at 9 months(−12%, p = 0.2538) and a non-significant increase at 12 months ofage (26% p = 0.1103, Fig. 1C). However, the 3xTg-AD mouse model ofAD displayed a significant increase in the Nv of activated microgliain the ML of DG at 18 months of age compared to age-matchednon-Tg control group (67%, p = 0.0228, Fig. 1C). The Mac-1-positiveactivated microglial cells were accumulated around the A� plaquesand adjacent to the vascular elements that accumulated A�, whichis typical of AD-associated angiopathy (Fig. 1F and G).

4. Discussion

The main finding of the present study is that the 3xTg-AD mousemodel of AD displays age-dependent increase in the numerical den-sity (Nv, #/mm3) of both resting and activated microglial cells in thedentate gyrus (DG) of the hippocampus. Increased Nv of ramifiedand activated microglia cells were observed at 18 months of ageand occurred in the absence of A� plaque formation in the DG ofthe hippocampus in the 3xTg-AD mouse model of AD.

These findings are in agreement with our previously reported

increase in both resting and activated microglial cells in the CA1subfield of the hippocampus in the 3xTg-AD mouse model of AD[14]. Similarly, Janelsins et al. had reported increased number ofF4/80-positive microglia/macrophages in the entorhinal cortex in

132 J.J. Rodríguez et al. / Neuroscience Letters 552 (2013) 129– 134

Fig. 1. Visualisation and quantification of ramified and amoeboid microglia in the dentate gyrus. (A) Bar graph showing the age-effect on the numerical density (Nv, #/mm3)of ramified microglia in the dentate gyrus of the hippocampus in 3xTg-AD and non-Tg control mice at 9, 12 and 18 months of age. (B) Brightfield micrograph showing thecharacteristic morphology of ramified tomato lectin-immunoreactive (TL-IR) microglia with small cell body equipped with thin to medium ramified processes extending tothe surrounding neuropil in the dentate gyrus subfield of the hippocampus. (C) Bar graph showing the age-effect on the Nv of amoeboid microglia in the dentate gyrus of thehippocampus in 3xTg-AD mice and non-Tg control mice. (D) and (E) Brightfield micrograph (D) and confocal image (E) showing the characteristic morphology of amoeboidMAC-1-IR microglia with enlarged cell bodies from which a greater number of numerous processes emanated, but with an enlarged and thicker appearance. (F) Confocalimage showing the recruitment of MAC-1-IR microglia (green) in the vicinity of A� plaques (red) in the CA1 subfield of the hippocampus of an 18-month-old 3xTg-AD mouse.(G) Confocal image of MAC-1-IR microglia (green) adjacent to a blood vessel (BV) with A� deposits. Bars represent mean ± SEM (n = 4–5). *, **p < 0.05, p < 0.01 compared toage-matched non-Tg, #p < 0.05 compared to 9 months 3xTg-AD and •p < 0.05 compared to 12 months 3xTg-AD group. Scale bars: (B) = 5 �m, (D) = 2 �m, (E) and (G) = 10 �ma he rea

tdrc1tmctCma

nd (F) 20 �m. (For interpretation of the references to colour in this figure legend, t

he 3xTg-AD mouse model of AD [4]. However, there is a temporalifference in the increased number of microglia in different brainegions in the 3xTg-AD mice. Specifically, the increase in microgliaells were evident at 6 months of age in the entorhinal cortex, at2 months in the CA1 subfield [14] and at 18 months of age inhe DG of the hippocampus (current study) in the 3xTg-AD mouse

odel of AD (Fig. 2). Such specific spatial increase in microglia cellslosely mimics the evolution of AD-related neuropathology, with

he earliest appearance in the entorhinal cortex followed by theA1 and DG of the hippocampus not only in the 3xTg-AD mouseodel of AD [7] but also in post-mortem AD brains [2]. Incidentally,

n increase in microglial density was also reported in normal aged

der is referred to the web version of this article.)

rodent brains at 24 months of age [16]. In this the AD pathologyaccelerates this age-dependent rise in numbers of microglial cells.

These results suggest that the DG is affected relatively late dur-ing the progression of the disease in the 3xTg-AD mouse model ofAD, which is consistent with post-mortem analysis of AD brainsthat also reported no evident A� plaque deposition, NFT formationand associated neurodegeneration in the DG of the hippocampusuntil the late stages [11]. Taken together, these results suggest that

AD-associated microglia proliferation/activation display spatio-temporal differences that may explain why certain brain regionsare more susceptible to AD-related neurodegeneration during theprogression of the disease.

J.J. Rodríguez et al. / Neuroscience L

Fig. 2. AD-related activation/proliferation of microglia in different subfields of thehippocampus. At the early (plaque-free stages of AD), there is an increase in thenumber of ramified microglia in the CA1 subfield, whilst no changes are observedin the dentate gyrus of the hippocampus. Consolidation of A� plaques in the CA1subfield of the hippocampus is associated with an increase in the density of bothramified and amoeboid microglia cells [14]. At the late stage of the disease, increasedmg

pimGim

erioeim

pd3

[

[

[Elesvier Academic Press, San Diego, 2004.

icroglial density is observed both in plaque-rich (CA1) and plaque-free (dentateyrus) subfields of the hippocampus.

Although AD-associated increase in microglia activation androliferation had been reported in post-mortem AD brains [9] and

n transgenic mouse models of the disease [5,10], no attempts wereade to differentiate between resting and activated microglia cells.iven the functional diversity between these phenotypes [6], it

s critical to distinguish AD-associated changes between the twoicroglia cells.Despite the lack of evident A� deposition, microglial cells were

xtensively distributed in the DG of the hippocampus, being mainlyepresented by resting microglia with typical ramified morphologyncluding small cell bodies and large/thin process (Fig. 1B). Thesebservations suggest that the increased Nv of microglia is not onlyvident in the plaque-rich (CA1) regions of the hippocampus, butt also occurs in the plaque-free (DG) region of the hippocampus at

ore advanced stage of the disease progression (Fig. 2).The underlying mechanism responsible for triggering microglial

roliferation and activation remains unclear. Given the lack of evi-ent A� plaque deposition in the DG at 18 months of age in thexTg-AD mouse model of AD, our result suggest that microglial

[

etters 552 (2013) 129– 134 133

activation/may occur independently of A� plaques, although onecannot exclude the role of smaller A� aggregates. In this sense, wehad previously reported increased activated microglia surround-ing not only extracellular A� neuritic plaques but also smaller A�aggregates in the hippocampus [14]. In addition, previous study inPS/APP transgenic mouse model of AD reported increased microgliaactivation following both fibrillar and diffuse A� deposits [8].

In summary, our results show increased microglia activa-tion/proliferation in the DG of the hippocampus at advanced stagesof the disease in the 3xTg-AD mouse model of AD. This increasein microglia response occurs independently of extracellular A�plaque formation in this brain region suggesting different suscep-tibility of the DG to AD-like pathology compared with other brainregions.

Acknowledgements

This study was supported by an Alzheimer’s Research Trust Pro-gramme Grant [ART/PG2004A/1] to JJR and AV, grants from theGrant Agency of the Czech Republic to JJR ([GACR 309/09/1696],to ES and JJR [GACR 304/11/0184] and to AV [GACR 305/08/1381,GACR 305/08/1384]). Support from the Spanish Government, PlanNacional de I+D+I 2008-2011 and ISCIII- Subdirección General deEvaluación y Fomento de la Investigación [PI10/02738] co-financedby FEDER to JJR and AV and the Government of the Basque Country[AE-2010-1-28, AEGV10/16, GV-2011111020] gratefully acknowl-edged.

References

[1] D.G. Amaral, H.E. Scharfman, P. Lavenex, The dentate gyrus: fundamental neu-roanatomical organization (dentate gyrus for dummies), Prog. Brain Res. 163(2007) 3–22.

[2] H. Braak, E. Braak, Neuropathological stageing of Alzheimer-related changes,Acta Neuropathol. 82 (1991) 239–259.

[3] C. Giaume, F. Kirchhoff, C. Matute, A. Reichenbach, A. Verkhratsky,Glia: the fulcrum of brain diseases, Cell Death Differ. 14 (2007)1324–1335.

[4] M.C. Janelsins, M.A. Mastrangelo, S. Oddo, F.M. LaFerla, H.J. Federoff, W.J. Bow-ers, Early correlation of microglial activation with enhanced tumor necrosisfactor-alpha and monocyte chemoattractant protein-1 expression specificallywithin the entorhinal cortex of triple transgenic Alzheimer’s disease mice, J.Neuroinflammation 2 (2005) 23.

[5] S. Jimenez, D. Baglietto-Vargas, C. Caballero, I. Moreno-Gonzalez, M.Torres, R. Sanchez-Varo, D. Ruano, M. Vizuete, A. Gutierrez, J. Vitor-ica, Inflammatory response in the hippocampus of PS1M146L/APP751SLmouse model of Alzheimer’s disease: age-dependent switch in themicroglial phenotype from alternative to classic, J. Neurosci. 28 (2008)11650–11661.

[6] H. Kettenmann, U.K. Hanisch, M. Noda, A. Verkhratsky, Physiology of microglia,Physiol. Rev. 91 (2011) 461–553.

[7] M.A. Mastrangelo, W.J. Bowers, Detailed immunohistochemical characteri-zation of temporal and spatial progression of Alzheimer’s disease-relatedpathologies in male triple-transgenic mice, BMC Neurosci. 9 (2008)81.

[8] Y. Matsuoka, M. Picciano, B. Malester, J. LaFrancois, C. Zehr, J.M.Daeschner, J.A. Olschowka, M.I. Fonseca, M.K. O‘Banion, A.J. Tenner,C.A. Lemere, K. Duff, Inflammatory responses to amyloidosis in a trans-genic mouse model of Alzheimer’s disease, Am. J. Pathol. 158 (2001)1345–1354.

[9] P.L. McGeer, S. Itagaki, H. Tago, E.G. McGeer, Reactive microglia in patients withsenile dementia of the Alzheimer type are positive for the histocompatibilityglycoprotein HLA-DR, Neurosci. Lett. 79 (1987) 195–200.

10] M. Meyer-Luehmann, T.L. Spires-Jones, C. Prada, M. Garcia-Alloza, A. deCalignon, A. Rozkalne, J. Koenigsknecht-Talboo, D.M. Holtzman, B.J. Bac-skai, B.T. Hyman, Rapid appearance and local toxicity of amyloid-betaplaques in a mouse model of Alzheimer’s disease, Nature 451 (2008)720–724.

11] T.G. Ohm, The dentate gyrus in Alzheimer’s disease, Prog. Brain Res. 163 (2007)723–740.

12] G. Paxinos, K.B.J. Franklin, The Mouse Brain in Stereotaxic Coordinates, 2nd ed.,

13] J.J. Rodríguez, V.C. Jones, M. Tabuchi, S.M. Allan, E.M. Knight, F.M. LaFerla, S.Oddo, A. Verkhratsky, Impaired adult neurogenesis in the dentate gyrus ofa triple transgenic mouse model of Alzheimer’s disease, PLoS ONE 3 (2008)e2935.

1 ience L

[

[

[and sensory loss on glial cells in mouse visual and auditory cortices, Glia 60

34 J.J. Rodríguez et al. / Neurosc

14] J.J. Rodríguez, J. Witton, M. Olabarria, H.N. Noristani, A. Verkhratsky, Increasein the density of resting microglia precedes neuritic plaque formation and

microglial activation in a transgenic model of Alzheimer’s disease, Cell DeathDis. 1 (2010) e1.

15] S.W. Scheff, D.A. Price, Synaptic density in the inner molecular layer of thehippocampal dentate gyrus in Alzheimer disease, J. Neuropathol. Exp. Neurol.57 (1998) 1146–1153.

[

etters 552 (2013) 129– 134

16] M.E. Tremblay, M.L. Zettel, J.R. Ison, P.D. Allen, A.K. Majewska, Effects of aging

(2012) 541–558.17] M.J. West, C.H. Kawas, L.J. Martin, J.C. Troncoso, The CA1 region of the human

hippocampus is a hot spot in Alzheimer’s disease, Ann. N.Y. Acad. Sci. 908 (2000)255–259.