Neuropeptides 46 (2012) 29–37

Contents lists available at SciVerse ScienceDirect

Neuropeptides

journal homepage: www.elsevier .com/locate /npep

Altered expression of neuropeptides in the primary somatosensory cortexof the Down syndrome model Ts65Dn

Samuel Hernández a, Javier Gilabert-Juan a,b, José Miguel Blasco-Ibáñez a, Carlos Crespo a,Juan Nácher a, Emilio Varea a,⇑a Neurobiology Unit and Program in Basic and Applied Neurosciences, Cell Biology Department, Universitat de València, Spainb Genetics Department, Universitat de València, CIBERSAM, Spain

a r t i c l e i n f o

Article history:Received 20 April 2011Accepted 18 October 2011Available online 10 November 2011

Keywords:Down syndromeTs65DnMurine modelsInhibitory neuronsNeuropeptidesVasoactive intestinal peptideCholecystokininSomatostatinNeuropeptide Y

0143-4179/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.npep.2011.10.002

⇑ Corresponding author. Address: Neurobiology UnUniversitat de València, Dr. Moliner, 50, Burjassot 461354 3241.

E-mail address: emilio.varea@uv.es (E. Varea).

a b s t r a c t

Down syndrome is the most common genetic disorder associated with mental retardation. Subjects andmice models for Down syndrome (such as Ts65Dn) show defects in the formation of neuronal networks inboth the hippocampus and the cerebral cortex. The principal neurons display alterations in the morphol-ogy, density and distribution of dendritic spines in the cortex as well as in the hippocampus. Several evi-dences point to the possibility that the atrophy observed in principal neurons could be mediated bychanges in their inhibitory inputs and, in fact, an imbalance between excitation and inhibition has beenobserved in Ts65Dn mice in these regions, which are crucial for learning and information processing.These animals have an increased density of interneurons in the primary somatosensory cortex, especiallyof those expressing calretinin and calbindin D-28k. Here, we have analysed the expression and distribu-tion of several neuropeptides in the primary somatosensory cortex of Ts65Dn mice in order to investigatewhether these subpopulations of interneurons are affected. We have observed an increase in the totaldensity of somatostatin expressing interneurons and of those expressing VIP in layer IV in Ts65Dn mice.The typology of the somatostatin and VIP interneurons was unaltered as attested by the pattern of co-expression with other markers. Somatostatin immunoreactive neurons co-express mainly D-28k calbin-din and VIP expressing interneurons maintain its pattern of co-expression with calcium binding proteins.These alterations, in case they were also present in subjects with Down syndrome, could be related totheir impairment in cognitive profile and could be involved in the neurological defects observed in thisdisorder.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Down syndrome, with an incidence of one in 800 live births(Roizen and Patterson, 2003), is one of the most common geneticdisorders. The phenotype observed as a consequence of a trisomyon the chromosome 21 may include immune deficiencies, heartdefects, increased risk of leukemia and early development ofAlzheimer’s disease. The principal common feature among all DSindividuals is the presence of mental retardation. The substratefor this retardation has not been fully understood and may includedefects in the formation of neuronal networks and informationprocessing. Alterations in synaptic plasticity have been related toimpaired cognition in different murine models of this geneticalteration (Siarey et al., 2005, 2006).

ll rights reserved.

it, Cell Biology Department,00, Spain. Tel./fax: +34 34 96

Among the murine models available to study this genetic alter-ation, the most widely used is the Ts65Dn mouse, which mimicsmost of the alterations observed in DS. Ts65Dn mice are segmen-tally trisomic for a portion of the murine chromosome 16, whichis orthologous to the long arm of the human chromosome 21. Thissegment contains approximately 140 genes, many of which arehighly conserved between mice and humans (Gardiner et al.,2003). These mice display delays in the acquisition of a numberof sensory and motor tasks (Holtzman et al., 1996; Costa et al.,1999), as well as defects in learning and in the execution of mem-ory tasks mediated by the hippocampus (Reeves et al., 1995;Escorihuela et al., 1995, 1998; Holtzman et al., 1996).

Many studies have shown deficits in the dendritic arborizationof the principal cells of the neocortex and hippocampus of DS sub-jects and murine models for this disorder (Marín-Padilla, 1976;Becker et al., 1986; Vuksic et al., 2002; Takashima et al., 1981,1989; Kaufmann and Moser, 2000; Dierssen et al., 2003). This atro-phy has been related to mental retardation and deficits in cogni-tion (Dierssen and Ramakers, 2006).

30 S. Hernández et al. / Neuropeptides 46 (2012) 29–37

Alterations at the synaptic level have been observed in both DSsubjects and murine models. Studies analyzing the expression ofsynaptophysin (a reliable marker for synapses (Eastwood andHarrison, 2001; Masliah et al., 1990)) have observed that the areaoccupied by synaptophysin is higher in Ts65Dn mice when com-pared with euploid mice, suggesting an increase in the size of thesynapse (Belichenko et al., 2004, 2007, 2009). The increasedexpression of synaptophysin was restricted to superficial layersof cortex (Pérez-Cremades et al., 2010). The detailed study of syn-apse subtypes reflects a reduction in the number of asymmetric(excitatory) synapses per neuron in the hippocampus and in thetemporal cortex of Ts65Dn mice (Kurt et al., 2000). Moreover, theinhibitory contacts are redistributed on the dendrites of these prin-cipal neurons, resulting in an increased density of those contactingspines and in a decrease of those contacting dendritic shafts(Belichenko et al., 2004). There is also an increased density ofinhibitory puncta (observed by immunohistochemistry for GAD-67) in every layer of the cortex (Pérez-Cremades et al., 2010). Alto-gether, these results suggest an unbalanced level of inhibitory andexcitatory inputs in the neocortex and hippocampus in Down syn-drome. Supporting this fact, an overactivation of the inhibitory sys-tem, causing a general inhibition in the brain, has been observed inTs65Dn mice (Fernández et al., 2007). This increased inhibitionmay be responsible, at least to some extent, for the cognitiveimpairment observed in Down syndrome. In fact, some studieshave attempted to reverse the cognitive impairments observed inTs65Dn mice by using GABAA receptor antagonists (Fernández etal., 2007), obtaining promising results.

Changes in inhibitory puncta density and distribution could berelated to changes in the number and types of interneurons in theaffected regions. In fact, we have observed an increase in the totalnumber of inhibitory neurons in the primary somatosensory cortexof Ts65Dn mice. The phenotypical characterization revealed thatamong the increased subpopulations of inhibitory neurons werethe calretinin and the calbindin D-28k expressing neurons(Pérez-Cremades et al., 2010).

The aim of this study is to deepen into the changes in interneu-ron populations in a specific region of the neocortex of Ts65Dnmice, the primary somatosensory cortex. We have chosen this re-gion because previous reports have shown atrophy in principalneurons (Dierssen et al., 2003) and increased density of inhibitoryneurons (Pérez-Cremades et al., 2010) We analysed the distributionand density of interneurons expressing four different neuropep-tides: cholecystokinin (CCK); somatostatin (SST), neuropeptide Y(NPY) and vasoactive intestinal peptide (VIP) in this cortical regionof Ts65Dn mice. The analysis of these neuropeptides allowed us todiscriminate the different subtypes of interneurons that were notpreviously studied in this area. In this way to aim to complete ourprevious study where we studied calcium binding protein interneu-rons and observed an increased density of interneurons expressingcalretinin (Pérez-Cremades et al., 2010).

2. Experimental procedures

Experimental mice were generated by repeated backcrossing ofTs65Dn females to C57/6Ei 9 C3H/HeSnJ (B6EiC3) F1 hybrid males.The parental generation was purchased from the research colony ofThe Jackson Laboratory (Ben Harbor, Maine, USA) Euploidlittermates of Ts65Dn mice served as controls. The genotypic char-acterization was established by qRT-PCR using SYBR Green PCRmaster mix (Applied Biosystems) from genomic DNA extracted ofmice tails by mean of the phenol–chloroform method. The relativeamount of each gene was quantified by the ABI PRISM 7700Sequence Detection System (Applied Biosystems). The genes ana-lysed where APP (3 copies) and Apo-B (2 copies). The primers used

where for APP (APP-F 50-TGT TCG GCT GTG TGA TCC TGT GAC-30;APP-R 50-AGA AAC GAG CGG CGA AGG GC-30) and for Apo-B(Apo-B-F 50-TGC CAG GCT TGT GCT GCT GT-30; Apo-B-R 50-GGGTGC TGC CTT TCT CTT GGG G-30).

For this study we used 4–5 month-old male mice (9 trisomic; 12euploid). All animal experimentation was conducted in accordancewith the European Communities Council Directive of 24 November1986 (86/609/EEC) on the protection of animals used for scientificpurposes and approved by the Committee on Bioethics of the Uni-versidad Miguel Hernández of Elche.

Animals were transcardially perfused using a solution contain-ing 4% paraformaldehyde in PB (0.1 M, pH 7.4). Brains were re-moved and cryoprotected using 30% sucrose. Fifty micronsections (10 subseries for each brain) were obtained using a slidingfreezing microtome.

2.1. Immunohistochemical procedure

Tissue was processed ‘‘free-floating’’ for immunohistochemistryas follows. Briefly, sections were incubated with 10% methanol, 3%hydrogen peroxide in phosphate-buffered saline (PBS, pH 7.4) for10 min to block endogenous peroxidase activity. After this, sectionswere treated for 1 h with 5% normal donkey serum (NDS) (JacksonImmunoResearch Laboratories, West Grove, PA, USA) in PBS with0.2% Triton- X100 (Sigma–Aldrich, St Louis, MO, USA) and wereincubated overnight at room temperature in either monoclonalmouse anti-CCK (Cat No. 9303, Cure (Gastroenteric Biology Center),University of California, USA, 1:1000), polyclonal rabbit anti-VIP(VA 1285, Affiniti, UK, 1:1000); polyclonal rabbit anti-NPY (kindlyprovided by Dr. T.J. Görcs, 1:1000) (Csiffary et al., 1990) or poly-clonal rabbit anti-SST (Cat. No 20067, Diasorin, Stillwater, USA,1:1000). After washing, sections were incubated for 2 h with don-key anti-mouse IgG or donkey anti-rabbit IgG biotinylated antibod-ies (Jackson ImmunoResearch Laboratories, West Grove, PA, USA,1:250), followed by avidin–biotin–peroxidase complex (ABC; Vec-tor Laboratories, Peterborough, UK. 1:200) for 30 min. PBS contain-ing 0.2% Triton-X-100 and 3% NDS was used for primary andsecondary antibody dilutions. Color development was achievedby incubating in 0.05% 3,3-diaminobenzidine tetrahydrochloride(Sigma–Aldrich), 0.033% hydrogen peroxide in PB for 4 min.

The antibodies had been previously tested in their laboratory oforigin; additionally, they showed a regional and cellular immu-nolabelling similar to previous descriptions of these antigens andthe distribution of neurons is in accordance with previous studiesusing these antibodies and others specific for these neuropeptides(Chronwall et al., 1985; Forloni et al., 1990; Meziane et al., 1997;Sims et al., 1980). In order to confirm that some of the immuno-staining was not produced by the secondary antibodies or by theimmunohistochemical protocol itself, we omitted primary anti-bodies or substituted them by normal donkey serum. Moreover,we have pre-incubated the primary antibodies with the corre-sponding antigenic peptide. All these controls resulted in a com-plete absence of immunostaining in every case.

2.2. Density of neuropeptide expressing interneurons in the primarysomatosensory cortex of Ts65Dn mice

We have analysed changes in the distribution of specific sub-populations of interneurons (using immunohistochemistry againstCCK, SST, NPY and VIP) in the primary somatosensorial cortex ofTs65Dn mice and in their euploid littermates. We counted thenumber of immunoreactive cells found in 500-lm wide strips(20 strips per group) running perpendicular to the pial surfaceincluding all the layers of the primary somatosensory cortex(Berbel et al., 1996; Pérez-Cremades et al., 2010). Our method con-sisted in counting all the cells in a strip of somatosensory cortex

S. Hernández et al. / Neuropeptides 46 (2012) 29–37 31

containing all layers. Layers were delimited using consecutive Nisslstained series. A cell was counted if its nucleus was detected on thesection using a 63� oil objective. Of the nuclei cut on the surface ofthe section, only those in the upper side were considered to avoiddouble counting. For the same reason, only nuclei cut by the leftside of the boxed area were considered. This method is, in fact, amodification of the dissector method. We have expressed the re-sults as a cellular density in every layer and in the total primarysomatosensory cortex. Means were determined for each experi-mental group and data were statistically analysed using the SPSSsoftware package (version 15). The difference between groupswas analysed with unpaired t-student test.

According to previous reports, there were no differences inbrain weight and cortex volume in Ts65Dn mice (Belichenkoet al., 2004; Roper et al., 2006).

2.3. Double immunofluorescence

In order to investigate the phenotype of the subpopulations ofneuropeptide expressing interneurons affected in Ts65Dn mice,we performed double immunofluorescence for calbindin D-28k

Fig. 1. Density and distribution of interneurons expressing the neuropeptide cholecystoshowing the density of CCK IR interneurons in the primary somatosensory cortex of euplointerneurons in each layer of the primary somatosensory cortex in euploid and Ts65Dn m(D), showing the distribution of CCK IR interneurons. Scale bar: 200 lm.

and calretinin against the neuropeptides somatostatin and VIP.Sections were processed as described above, but the endogenousperoxidase block was omitted. The sections were incubated over-night at room temperature with monoclonal mouse anti-calbindinD-28k (SWANT, 1:1000) or monoclonal mouse anti-calretinin(SWANT, 1:1000) on one hand, and polyclonal rabbit anti-SST(1:1000) or polyclonal rabbit anti-VIP (1:1000) antibodies on theother. After washing, sections were incubated with donkey anti-rabbit IgG and donkey anti-mouse IgG secondary antibodies conju-gated with Alexafluor 488 or Alexafluor 555 (Molecular Probes,1:200) respectively. For antibody dilutions, PBS containing 0.2%Triton X-100 and 3% NDS was used.

2.4. Confocal imaging

All sections processed for immunofluorescence were mounted onslides and coverslipped using Permafluor mounting medium(Immunon/Shandon, Pittsburgh, PA, USA). Then, the sections wereobserved under a confocal microscope (Leica, Wein, Austria; TCS-SPE). Z-series of optical sections (1 lm apart) were obtained usingsequential scanning mode. These stacks were processed with LSM

kinin (CCK IR) in the primary somatosensory cortex of the Ts65Dn mice. (A) Graphid (black bar) and Ts65Dn mice (white bar). (B) Graph showing the density of CCK IRice. Field survey of the somatosensory cortex of a control (C) and a Ts65Dn mouse

Table 1Densities of the different subpopulations of interneurons (cells/mm3) in the wholeprimary somatosensory cortex and in each individual layer of euploid and trisomicmice. Bold numbers represent parameters statistically significant.

CCK NPY VIP Somatostatin

I Euploid 227 ± 72 454 ± 105 630 ± 118 303 ± 93Trisomic 228 ± 93 227 ± 72 998 ± 170 304 ± 76

II Euploid 495 ± 103 774 ± 145 2830 ± 335 845 ± 101Trisomic 619 ± 184 460 ± 71 3007 ± 244 914 ± 124

III Euploid 257 ± 19 1880 ± 182 2419 ± 266 3200 ± 296Trisomic 321 ± 36 1905 ± 343 3235 ± 442 3616 ± 314

IV Euploid 233 ± 72 468 ± 65 435 ± 69* 2368 ± 299Trisomic 250 ± 34 466 ± 73 699 ± 97 3066 ± 184

V Euploid 393 ± 61 964 ± 70 958 ± 107 5197 ± 574⁄

Trisomic 378 ± 44 847 ± 103 1180 ± 194 6936 ± 466VI Euploid 552 ± 76 1189 ± 36⁄ 720 ± 110 2466 ± 285⁄

Trisomic 655 ± 118 800 ± 141 717 ± 110 3424 ± 286TOTAL Euploid 338 ± 34 1055 ± 72 1372 ± 119 2628 ± 197⁄

Trisomic 386 ± 28 932 ± 119 1741 ± 125 3291 ± 123

* p < 0.05.

Fig. 2. Density and distribution of interneurons expressing the neuropeptide Y (NPY IRdensity of NPY IR interneurons in the primary somatosensory cortex of euploid (blainterneurons in each layer of the primary somatosensory cortex in euploid and Ts65Dn m(D), showing the distribution of NPY IR interneurons. Scale bar: 200 lm (⁄p < 0.05).

32 S. Hernández et al. / Neuropeptides 46 (2012) 29–37

5 Image Browser software. VIP or SST immunoreactive cells werefirst identified using conventional fluorescence microscopy. FiftyVIP (from layer IV) or SST (from layer V–VI) immunoreactive neuronsin the primary somatosensory cortex were analyzed in each case todetermine the co-expression with calretinin and calbindin D-28k.

3. Results

The analysis of volume per layer and total in the primarysomatosensory cortex of Ts65Dn showed not significant changeswhen compared with their euploid littermates, in accordance withprevious results (Belichenko et al., 2004).

3.1. Densities of peptidergic subpopulations in the primarysomatosensory cortex of Ts65Dn mice

CCK immunoreactive (CCK IR) neurons were scarce in theprimary somatosensory cortex in both trisomic mice and euploid

) in the primary somatosensory cortex of the Ts65Dn mice. (A) Graph showing theck bar) and Ts65Dn mice (white bar). (B) Graph showing the density of NPY IR

ice. Field survey of the somatosensory cortex of a control (C) and a Ts65Dn mouse

S. Hernández et al. / Neuropeptides 46 (2012) 29–37 33

littermates, being more abundant in layers II and VI (Fig. 1). Theanalysis of CCK expressing interneuron density (Table 1) revealedthat this population is maintained in trisomic mice(386 ± 28 cells/mm3 vs. 338 ± 34 cells/mm3, p = 0.68) (Fig. 1A).The analysis by cortical layers did not reveal any significant differ-ence (Fig. 1B).

NPY immunoreactive (NPY IR) neurons were more abundantthan those expressing CCK. They were found in every layer of theprimary somatosensory cortex, but presented a higher density inlayer III (Fig. 2). The density of NPY expressing neurons in this cor-tical region (Table 1) remained unaltered in trisomic mice(932 ± 119 cells/mm3 vs. 1055 ± 72 cells/mm3, n.s.) (Fig. 2A). Theanalysis by cortical layers revealed a decrease in the density ofinterneurons expressing NPY in layer VI (799 ± 140 cells/mm3 vs.1189 ± 37 cells/mm3, p < 0.05) (Fig. 2B).

VIP immunoreactive (VIP IR) interneurons were present inevery layer, although they were more abundant in layers II andIII (Fig. 3). The analysis of the density of VIP expressing interneu-rons in the primary somatosensory cortex of Ts65Dn (Table 1)reflected a general non significant increment in this subpopulation

Fig. 3. Density and distribution of interneurons expressing the vasoactive intestinal peshowing the total density of VIP IR interneurons in the primary somatosensory cortex ofVIP IR interneurons in each layer of the primary somatosensory cortex in euploid and Tsmouse (D), showing the distribution of VIP IR interneurons. Scale bar: 200 lm (⁄p < 0.05

of interneurons (1741 ± 125 cells/mm3 vs. 1372 ± 119 cells/mm3,p = 0.07) (Fig. 3A). The detailed study of the expression by corticallayers reflected that only their density in layer IV (699 ± 97 cells/mm3 vs. 435 ± 69 cells/mm3, p < 0.05) was increased significantlyin the trisomic model (Fig. 3B).

Somatostatin immunoreactive (SST IR) interneurons were themost abundant of all the analysed subpopulations. They were scarcein layers I–II but abounded in layers V–VI (Fig. 4). The analysis of thedensity of SST IR interneurons in the primary somatosensory cortexof Ts65Dn mice (Table 1) showed an increment in the density of thissubpopulation (3292 ± 123 cells/mm3 vs. 2628 ± 196 cells/mm3,p < 0.05) (Fig. 4A). The detailed study by cortical layers reflected thatthe increment in density of SST IR neurons is confined to layer V(6935 ± 466 cells/mm3 vs. 5196 ± 574 cells/mm3, p < 0.05) and VI(3424 ± 285 cells/mm3 vs. 2466 ± 282 cells/mm3, p < 0.05).

We have analysed whether the increase in the density of VIP-and SST-IR cells was accompanied by changes in their neurochem-ical phenotype (attending to their calcium binding protein expres-sion) in the layers where the significant alteration in density werefound. For that reason, we performed double immunolabelling for

ptide (VIP IR) in the primary somatosensory cortex of the Ts65Dn mice. (A) Grapheuploid (black bar) and Ts65Dn mice (white bar). (B) Graph showing the density of65Dn mice. Field survey of the somatosensory cortex of a control (C) and a Ts65Dn).

Fig. 4. Density and distribution of Somatostatin immunoreactive interneurons (SST IR) in the somatosensory cortex of Ts65Dn mice. (A) Graph showing the total density ofSST IR interneurons in the primary somatosensory cortex of euploid (black bar) and Ts65Dn mice (white bar). (B) Graph showing the density of SST IR interneurons in eachlayer of the primary somatosensory cortex in euploid and Ts65Dn mice. Field survey of the somatosensory cortex of a control (C) and a Ts65Dn mouse (D), showing thedistribution of SST IR interneurons. Scale bar: 200 lm (⁄p < 0.05).

34 S. Hernández et al. / Neuropeptides 46 (2012) 29–37

somatostatin or VIP and the calcium binding proteins calbindinD-28k or calretinin (Fig. 5). The analysis of colocalization betweenthese markers revealed that these subpopulations did not changetheir phenotype. In the case of SST-IR interneurons, the majorityof them also expressed the calcium binding protein calbindinD-28k in both groups of animals (68.1 ± 9.7% in trisomic mice vs.77.0 ± 9.5% in euploid mice p = 0.53), whereas only a few SST-IRneurons also expressed calretinin (2.3 ± 1.4% in trisomic mice vs.5.2 ± 2.2% in euploid mice p = 0.38). For VIP-IR interneurons, weobserved that they did not express calbindin D-28k, and that theexpression of calretinin did not change between groups (33.6 ±8.8% in trisomic mice vs. 36.8 ± 7.5% in euploid mice, p = 0.79).

4. Discussion

In this study we have analysed the changes in the subpopula-tions of interneurons expressing neuropeptides in the primarysomatosensory cortex of Ts65Dn mice. Trisomic mice displayed ahigher density of interneurons expressing somatostatin and VIP,

and a lower density of interneurons expressing NPY than their eu-ploid littermates. The differences were confined to specific layers,V and VI for somatostatin, IV for VIP, and VI for NPY. The studyof the phenotype of the subpopulations that increased their densi-ties (somatostatin and VIP) revealed that these interneurons main-tain its phenotype in trisomic mice and therefore the increment isnot likely to come from a change in neuropeptide expression byother interneurons. Somatostatin immunoreactive interneuronswere mainly calbindin D-28k positive as in euploids. The percent-age of VIP immunoreactive neurons expressing calretinin remainedunaltered.

There are previous reports analysing possible alterations in NPYexpression in animal models of DS. These studies have been donein a different mice model, a mice model carrying a completetrisomy of murine chromosome 16. Since whole chromosome 16trisomic foetuses die in utero, studies using this model must bedone in primary cultures obtained from the foetuses (Caserta,1994). Under these conditions, the authors have found an increasein the number of NPY immunoreactive neurons and they suggestthat this finding is due to a difference in the trophic environment

Fig. 5. Characterization of VIP and somatostatin expressing interneurons in primary somatosensory cortex of Ts65Dn mice using calcium binding proteins. (A) Colocalizationof VIP with calretinin in layer IV of the primary somatosensory cortex of Ts65Dn mice. (B) Colocalization of somatostatin with calbindin D-28k in layer V–VI in the sameregion. (C) Colocalization of somatostatin with calretinin in layer V–VI in the same region. Scale bar: 30 lm. Images correspond to focal planes.

S. Hernández et al. / Neuropeptides 46 (2012) 29–37 35

in the cortex, mainly related to glial cells (Caserta, 1994). In ourexperiment, when analysing the expression of NPY, we have founda decrease in the density of NPY IR neurons in layer VI on Ts65Dnmice, which are only trisomic for a region of chromosome 16 thatcontain genes homologous to the human chromosome 21. More-over the model for complete trisomy in chromosome 16 displaysover-expression of a wide quantity of genes that are not presentin human chromosome 21, for this reason Ts65Dn mice modelreproduces more closely the phenotype observed in DS subjects.There are two possible explanations for this discrepancy: first,the genes influencing NPY expressing interneurons (or other inhib-itory neurons) may be located in a region of the chromosome 16that is not trisomic in Ts65Dn; second and more plausible, thatthe increment observed during prenatal development is not main-tained during adulthood.

Using cell cultures from foetuses of mice trisomic for the wholechromosome 16, Caserta and collaborators found an increment inthe number of somatostatin immunoreactive neurons in culturesfrom foetuses of trisomic chromosome 16 mice. In this case, thepossible explanation that the authors provide for that increase isthe presence of the gene pre-prosomatostatin in the murinechromosome 16 (Caserta et al., 1990). In the Ts65Dn model theincrement is maintained during adulthood. Our study revealed ahigher density of somatostatin immunoreactive interneuronsmainly in layers V and VI of the primary somatosensory cortex ofTs65Dn mice. Recently, (Chakrabarti et al. (2010)) have observedalterations in the density of somatostatin immunoreactive neuronsin the neocortex of Ts65Dn. Some of their observations are inaccordance with the present work; however they have also foundincreased density of calbindin and parvalbumin immunoreactiveneurons. In a previous report by our group (Pérez-Cremadeset al., 2010) we found increased density of calretinin immunoreac-tive neurons, but only a slight increase in calbindin and parvalbu-min immunoreactive neurons. One possible explanation for these

different results could be that Chakrabarti et al. used mice between0–30 days-old, but our results have been obtained in 4–5 monthsold mice. On one hand young interneurons can change neurochem-ical phenotype, on the other, there is a progressive atrophy of thepyramidal morphology during age and perhaps that could producechanges in the subpopulations of interneurons.

It has been described that somatostatin immunoreactive neu-rons located in layers V and VI of the neocortex, the ones affectedin the primary somatosensory cortex of the Ts65Dn model, areburst spiking (Wang et al., 2004). Inhibitory neurons located inthese layers displaying co-localization with D-28k calbindin havebeen characterized as ‘‘Martinotti cells’’ (Martinotti, 1889; Ramóny Cajal, 1891; Wang et al., 2004). These neurons have a fundamen-tal role in the information processing and signaling of principalneurons within and between neocortical layers and columns (Cauliet al., 1997; Markram et al., 2004). The impairment in cognitiveprofile displayed in Ts65Dn mice could be related to the disregula-tion of this interneuron subpopulation.

Previous studies using Ts65Dn mice have shown an increase inthe presence of VIP and in the expression its receptor 1 (VPAC-1,present in astrocytes) (Sahir et al., 2006). Astrocytes from trisomicmice are less sensitive to VIP stimulation and they do not release asmuch survival promoting substances after VIP stimulation, as ithappens in control animals (Sahir et al., 2006). Analysing theTs65Dn model, Hill et al. (2003) found that they have significantlymore VIP binding sites than euploid littermates in all regions. Theyhave a higher number of VIP immunoreactive neurons in some cor-tical regions (Hill et al., 2003). Our results are in agreement withthis report, but our study reflects that the increment is restrictedto specific layers of the primary somatosensory cortex (mainlyIV). Previous studies have shown co-localization between VIP andcalretinin (Rogers, 1992). In a previous study of our laboratory(Pérez-Cremades et al., 2010), we have found an increase in thedensity of calretinin immunoreactive neurons in the primary

36 S. Hernández et al. / Neuropeptides 46 (2012) 29–37

somatosensory cortex. Other pathologies carrying alterations incortical formation and increasing the frequency of epilepsy, suchas the cerebrocortical microdysgenesis, showed an increment inthe number of VIP interneurons (Sherman et al., 1990). Interest-ingly, several studies have shown a prevalence of epilepsy in DSsubjects (Pueschel et al., 1991; Stafstrom, 1993; Prasher, 1995;Johannsen et al., 1996; Goldberg-Stern et al., 2001 ). Concretly,Johannsen and col. found a prevalence of epilepsy observing that17% of DS subjects presented epilepsy, this percentage is higherin elderly subjects reaching 24% (Johannsen et al., 1996). Anotherstudy from Prasher analysing 201 adults with Down syndrome ob-served that 32 of them developed epilepsy (15.9%) (Prasher, 1995).This propensity could be related paradoxically to an increased den-sity of inhibitory neurons. The increase in interneurons and inhibi-tion could be an adaptative phenomenon intending to cope withthis increment in excitation that has been described in DS. Butthe price to pay for this increase inhibition can be an impairmentin cognitive functions. In this context, the treatment with antago-nists of GABAa receptors that has been shown to improve the cog-nitive profile (Fernández et al., 2007) could work blocking thisincrement in inhibition. However, these authors did not check forthe appearance of epilepsy in the mice model.

Further studies must be performed in order to confirm thatthese alterations observed in murine models are present in DS sub-jects. The alterations in the densities of specific subpopulations ofinterneurons can influence the normal function of inhibitory cir-cuits and may underlie the cognitive deficits observed in DS. Infact, changes in the relative proportions of different interneuronsubtypes can cause significant alterations in the properties of amodeled neuronal network (Foldy et al., 2004). The study of thechanges in interneuron subpopulations and types of inhibition inDS can shed light on the processes leading to neurological deficitsand can help to design novel strategies to contain the mental retar-dation associated with this disorder.

Conflict of interest

The authors declare no conflict of interest.

Acknowledgments

This study has been founded by Spanish Ministry of Educationand Science (BFU2007-64130/BFI); Foundation Jerome Lejeuneand The Spanish Ministry of Science and Innovation BFU2009-12284/BFI and PIM2010ERN-00577/NEUCONNECT in the frame ofERA-NET NEURON.

References

Becker, L.E., Armstrong, D.L., Chan, F., 1986. Dendritic atrophy in children withDown’s syndrome. Ann. Neurol. 20, 520–526.

Belichenko, P.V., Masliah, E., Kleschevnikov, A.M., Villar, A.J., Epstein, C.J., Salehi, A.,Mobley, W.C., 2004. Synaptic structural abnormalities in the Ts65Dn mousemodel of Down syndrome. J. Comp. Neurol. 480 (3), 281–298.

Belichenko, P.V., Kleschevnikov, A.M., Salehi, A., Epstein, C.J., Mobley, W.C., 2007.Synaptic and cognitive abnormalities in mouse models of Down syndrome:exploring genotype-phenotype relationship. J. Comp. Neurol. 504, 329–345.

Belichenko, P.V., Kleschevnikov, A.M., Masliah, E., Wu, C., Takimoto-Kimura, R.,Salehi, A., Mobley, W., 2009. Excitatory-inhibitory relationship in the fasciadentate in the Ts65Dn mouse model of Down syndrome. J. Comp. Neurol. 512,453–466.

Berbel, P., Marco, P., Cerezo, J.R., DeFelipe, J., 1996. Distribution of parvalbuminimmunoreactivity in the neocortex of hypothyroid adult rats. Neurosci. Lett.204, 65–68.

Caserta, M.T., 1994. Neuropeptide Y immunoreactive neurons in murine trisomy 16cortical cultures. Plasticity of expression and differentiation. Mol. Chem.Neuropathol. 22 (3), 197–210.

Caserta, M.T., Corsi, P., Oster-Granite, M.L., Coyle, J.T., 1990. Increased number ofsomatostatin-immunoreactive neurons in primary cultures of trisomy 15mouse neocortex. Brain Res. Mol. Brain Res. 7 (3), 269–272.

Cauli, B., Audinat, E., Lambolez, B., Angulo, M.C., Ropert, N., Tsuzuki, K., Hestrin, S.,Rossier, J., 1997. Molecular and physiological diversity of cortical nonpyramidalcells. J. Neurosci. 17 (10), 3894–3906.

Chakrabarti, L., Best, T.K., Cramer, N.P., Carney, R.S.E., Isaac, J.T.R., Gadzicki, Z.,Haydar, T.F., 2010. Olig1 and Olig2 triplication causes developmental braindefects in Down syndrome. Nat. Neurosci. 13, 927–934.

Chronwall, B., DiMaggio, D.A., Massari, V.J., Pickel, V.M., Ruggiero, D.A., O’Donohue,T.A., 1985. The anatomy of neuropeptide-Y-containing neurons in the rat brain.Neuroscience 15, 1159–1181.

Costa, A.C., Walsh, K., Davisson, M.T., 1999. Motor dysfunction in a mouse model forDown syndrome. Physiol. Behav. 68, 211–220.

Csiffary, A., Gorcs, T.J., Palkovits, M., 1990. Neuropeptide Y innervations of ACTH-immunoreactive neurons in the arcuate nucleus of rats: a correlated light andelectron microscopic double immunolabeling study. Brain Res. 506, 215–222.

Dierssen, M., Benavides-Piccione, R., Martínez-Cué, C., Estivill, X., Flórez, J., Elston,G.N., DeFelipe, J., 2003. Alterations of neocortical pyramidal cell phenotype inthe Ts65Dn mouse model of Down syndrome: effects of environmentalenrichment. Cereb. Cortex 13, 758–764.

Dierssen, M., Ramakers, G.J., 2006. Dendritic pathology in mental retardation frommolecular genetics to neurobiology. Genes Brain Behav. 5 (Suppl 2), 48–60.

Eastwood, S.L., Harrison, P.J., 2001. Synaptic pathology in the anterior cingulatedcortex in schizophrenia and mood disorders. A review and a Western blot studyof synaptophysin, Gap-43 and the complexins. Brain Res. Bull. 55, 569–578.

Escorihuela, R.M., Fernández-Teruel, A., Vallina, I.F., Baamonde, C., Lumbreras, M.A.,Dierssen, M., Tobeña, A., Flórez, J., 1995. A behavioural assessment of Ts65Dnmice: a putative Down syndrome model. Neurosci. Lett. 199, 143–146.

Escorihuela, R.M., Vallina, I.F., Martínez-Cué, C., Baamonde, C., Dierssen, M., Tobeña,A., Flórez, J., Fernández-Teruel, A., 1998. Impaired short- and long-term memoryin Ts65Dn mice, a model for Down syndrome. Neurosci. Lett. 247, 171–174.

Fernández, F., Morishita, W., Zuniga, E., Nguyen, J., Blank, M., Malenka, R.C., Garner,C.C., 2007. Pharmacotherapy for cognitive impairment in a mouse model ofDown syndrome. Nat. Neurosci. 10, 411–413.

Foldy, C., Aradi, I., Howard, A., Soltesz, I., 2004. Diversity beyond variance:modulation of firing rates and network coherence by GABAergicsubpopulations. Eur. J. Neurosci. 19, 119–130.

Forloni, G., Hohmann, C., Coyle, J.T., 1990. Developmental expression ofsomatostatin in mouse brain. I. Immunocytochemical studies. Brain Res. Dev.Brain Res. 53, 6–25.

Gardiner, K., Fortna, A., Bechtel, L., Davisson, M.T., 2003. Mouse models of Downsyndrome: how useful can they be? Comparison of the gene content of humanchromosome 21 with orthologous mouse genomic regions. Gene 318, 137–147.

Goldberg-stern, H., Strawsburg, R.H., Patterson, B., Hickey, F., Bare, M., Gadoth, N.,Degrauw, T.J., 2001. Seizure frequencey and characteristics in children withDown syndrome. Brain Dev. 23, 375–378.

Hill, J.M., Ades, A.M., McCune, S.K., Sahir, N., Moody, E.M., Abebe, D.T., Crnic, L.S.,Brenneman, D.E., 2003. Vasoactive intestinal peptide in the brain of a mousemodel for Down syndrome. Exp. Neurol. 183, 56–65.

Holtzman, D.M., Santucci, D., Kilbridge, J., Chua-Couzens, J., Fontana, D.J., Daniels,S.E., Johnson, R.M., Chen, K., Sun, Y., Carlson, E., Alleva, E., Epstein, C.J., Mobley,W.C., 1996. Developmental abnormalities and age-related neurodegeneration ina mouse model of Down syndrome. Proc. Natl. Acad. Sci. USA 96, 13333–13338.

Johannsen, P., Crhistensen, J.E.J., Goldstein, H., Nielsen, V.K., Mai, J., 1996. Epilepsy inDown Syndrome- prevalence in three age groups. Seizure 5, 121–125.

Kaufmann, W.E., Moser, H.W., 2000. Dendritic anomalies in disorders associatedwith mental retardation. Cereb. Cortex 10, 981–991.

Kurt, M.A., Davies, D.C., Kidd, M., Dierssen, M., Florez, J., 2000. Synaptic deficit in thetemporal cortex of partial trisomy 16 (Ts65Dn) mice. Brain Res. 858, 191–197.

Marín-Padilla, M., 1976. Pyramidal cell abnormalities in the motor cortex of a childwith Down’s syndrome. A Golgi study. J. Comp. Neurol. 167, 63–81.

Markram, H., Toledo-Rodríguez, M., Wang, Y., Gupta, A., Silberberg, G., Wu, C., 2004.Interneurons of the neocortical inhibitory system. Nat. Rev. Neurosci. 5, 793–807.

Martinotti, C., 1889. Contributo allo studio della corteccia cerebral, ed all’originecentral dei nervi. Ann. Freniatr Sci. affini 1, 14–381.

Masliah, E., Terry, R.D., Alford, M., DeTeresa, R., 1990. Quantitativeimmunohistochemistry of synaptophysin in human neocortex: an alternativemethod to estimate density of presynaptic terminals in paraffin sections. J.Histochem. Cytochem. 38, 837–844.

Meziane, H., Devigne, C., Tramu, G., Soumireu-Mourat, B., 1997. Distribution ofcholecystokinin immunoreactivity in the BALB/c mouse forebrain: animmunocytochemical study. J. Chem. Neuroanat. 12, 191–209.

Pérez-Cremades, D., Hernández, S., Blasco-Ibáñez, J.M., Crespo, C., Nácher, J., Varea,E., 2010. Alteration of inhibitory circuits in the somatosensory cortex of Ts65Dnmice, a model for Down’s síndrome. J. Neural Transm. 117, 445–455.

Prasher, V.P., 1995. Epilepsy and associated effects on adaptive behaviour in adultswith Down syndrome. Seizure 4, 53–56.

Pueschel, S.M., Louis, S., Mcknight, P., 1991. Seizure disorders in Down syndrome.Arch. Neurol. 48, 318–320.

Ramón y Cajal, S., 1891. Sur la structure de l’ecorce cerebrale de quelquesmammiferes. Cellule 7, 3–54.

Reeves, R.H., Irving, N.G., Moran, T.H., Wohn, A., Kitt, C., Sisodia, S.S., Schmidt, C.,Bronson, R.T., Davisson, M.T., 1995. A mouse model for Down syndrome exhibitslearning and behaviour deficits. Nat. Genet. 11, 177–184.

Rogers, J.H., 1992. Immunohistochemical markers in the rat cortex: co-localizationof calretinin and calbindin-D28k with neuropeptides and GABA. Brain Res. 587,147–157.

Roizen, N.J., Patterson, D., 2003. Down’s syndrome. Lancet 361, 1281–1289.

S. Hernández et al. / Neuropeptides 46 (2012) 29–37 37

Roper, R.J., John St, H.K., Philip, J., Lawler, A., Reeves, R.H., 2006. Perinatal loss ofTs65Dn Down syndrome mice. Genetics 172 (1), 437–443.

Sahir, N., Brenneman, D.E., Hill, J.M., 2006. Neonatal mice of the down syndromemodel, Ts65Dn exhibit upregulated VIP measures and reduced responsivenessof cortical astrocytes to VIP stimulation. J. Mol. Neurosci. 30 (3), 329–340.

Sherman, G.F., Stone, J.S., Rosen, G.D., Galaburda, A.M., 1990. Neocortical VIPneurons are increased in the hemisphere containing focal cerebrocorticalmicrodysgenesis in New Zealand Black mice. Brain Res. 532, 232–236.

Siarey, R.J., Villar, A.J., Epstein, C.J., Galdzicki, Z., 2005. Abnormal synaptic plasticityin the Ts1Cje segmental trisomy 16 mouse model of Down syndrome.Neuropharmacology 49 (1), 122–128.

Siarey, R.J., Kline-Burgess, A., Cho, M., Balbo, A., Best, T.K., Harashima, C., Klann, E.,Galdzicki, Z., 2006. Altered signaling pathways underlying abnormalhippocampal synaptic plasticity in the Ts65Dn mouse model of Downsyndrome. J. Neurochem. 98 (4), 1266–1277.

Sims, K.B., Hoffman, D.L., Said, S.I., Zimmerman, E.A., 1980. Vasoactive intestinalpolypeptide (VIP) in mouse and rat brain: an immunocytochemical study. BrainRes. 186, 165–183.

Stafstrom, C.E., 1993. Epilepys in Down syndrome. Clinical aspects and possiblemechanisms. Am J. Ment. Retard. 98 Suppl, 12–26.

Takashima, S., Becker, L.E., Armstrong, D.L., Chan, F., 1981. Abnormal neuronaldevelopment in the visual cortex of the human fetus and infant with down’ssyndrome. A quantitative and qualitative Golgi study. Brain Res. 225, 1–21.

Takashima, S., Ieshima, A., Nakamura, H., Becker, L.E., 1989. Dendrites, dementiaand the Down syndrome. Brain Dev. 11, 131–133.

Vuksic, M., Petanjek, Z., Rasin, M.R., Kostovic, I., 2002. Perinatal growth of prefrontallayer III pyramids in Down syndrome. Pediatr. Neurol. 27, 36–38.

Wang, Y., Toledo-Rodríguez, M., Gupta, A., Wu, C., Silberberg, G., Luo, J., Markram, H.,2004. Anatomical, physiological and molecular properties of Martinotti cells inthe somatosensory cortex of the juvenile rat. J. Physiol. 561.1, 65–90.