Cholinergic forebrain degeneration in the APPswe/PS1ΔE9 transgenic mouse

  • Published on

  • View

  • Download

Embed Size (px)


<ul><li><p>nmo</p><p>in Ag</p><p>l of M</p><p>pathogenic 42-amino-acid -amyloid (A142) peptides (Price and Buttini et al., 2002; German et al., 2003; Aucoin et al., 2005; Bales etal., 2006) compared to APP/PS1 double transgenic mice (Wong etal., 1999; Hernandez et al., 2001; Jaffar et al., 2001). Moreover,choline; ChAT activity; AChE; Stereology; Interneurons; Degeneration</p><p>Introduction</p><p>Alzheimers disease (AD) is characterized by cognitive decline,which is accompanied by beta-amyloid (A) plaque and neuro-fibrillary tangle formation in the diseased brain (Arnold et al.,1991; Braak and Braak, 1991). In early onset familial forms of AD(FAD) and in sporadic AD A-deposits are composed of 4 kDaA-peptides derived from amyloid precursor proteins (APP). FADare autosomal dominant forms of AD caused by the expression ofmutant human genes encoding APP, presenilin 1 (PS1) orpresenilin 2 (PS2), that lead to increased production of highly</p><p>Oddo et al., 2003; Gtz et al., 2004). In terms of cholinergicpathology, most mice over-expressingmutant APP and/or PS1 genesdisplay an age-dependent A deposition, cortical and hippocampalcholinergic fiber degeneration (Wong et al., 1999; Hernandez et al.,2001; Jaffar et al., 2001; Boncristiano et al., 2002; Buttini et al.,2002; German et al., 2003; Aucoin et al., 2005) and memory deficits(see Suh and Checler, 2002). However, these changes are notaccompanied by a loss of cholinergic basal forebrain neurons(Hernandez et al., 2001; Jaffar et al., 2001; Boncristiano et al., 2002;German et al., 2003). Therefore, none of these mutant mice fullyrecapitulates the robust cholinergic neurodegeneration seen as one ofthe hallmarks of human AD. It is interesting to note that the majorityof studies reporting cholinergic pathologies were performed in singleAPP or PS1 mutant mice, which have limited or no A deposition(Hernandez et al., 2001; Jaffar et al., 2001; Boncristiano et al., 2002;Cholinergic forebrain degeneratiotransgenic mouse</p><p>Sylvia E. Perez,a Saleem Dar,a Milos D. IkonoSteven T. DeKosky,b and Elliott J. Mufsona,</p><p>aDepartment of Neurological Sciences, Alla V. and Solomon Jesmer ChairSuite 300, Chicago, IL 60612, USAbDepartments of Neurology and Psychiatry, University of Pittsburgh Schoo</p><p>Received 10 April 2007; revised 5 June 2007; accepted 6 June 2007Available online 27 June 2007</p><p>The impact of A deposition upon cholinergic intrinsic cortical andstriatal, as well as basal forebrain long projection neuronal systemswas qualitatively and quantitatively evaluated in young (26 months)and middle-aged (1016 months) APPswe/PS1E9 transgenic (tg)mice. Cholinergic neuritic swellings occurred as early as 23 months ofage in the cortex and hippocampus and 56 months in the striatum oftg mice. However, cholinergic neuron number or choline acetyltrans-ferase (ChAT) optical density measurements remained unchanged inthe forebrain structures with age in APPswe/PS1E9 tg mice. ChATenzyme activity decreased significantly in the cortex and hippocampusof middle-aged tg mice. These results suggest that A deposition hasage-dependent effects on cortical and hippocampal ChAT fibernetworks and enzyme activity, but does not impact the survival ofcholinergic intrinsic or long projection forebrain neurons in APPswe/PS1E9 tg mice. 2007 Elsevier Inc. All rights reserved.</p><p>Keywords: Alzheimers disease; Amyloid; Forebrain; Transgenics; Acetyl- Corresponding author. Fax: +1 312 563 3571.E-mail address: (E.J. Mufson).Available online on ScienceDirect (</p><p>0969-9961/$ - see front matter 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.nbd.2007.06.015in the APPswe/PS1E9</p><p>vic,b</p><p>ing, Rush University Medical Center, 1735 W. Harrison Street,</p><p>edicine, Pittsburgh, PA 15261, USA</p><p>Sisodia, 1998). In addition to plaque and tangle formation there is areduction in subcortical cholinergic nucleus basalis neurons as wellas cortical acetylcholine in AD (Davies and Maloney, 1976;Davies, 1979; Richter et al., 1980; Whitehouse et al., 1981, 1985;Mufson et al., 1989a,b; DeKosky et al., 2002). Likewise,cholinergic interneurons and metabolic markers are reduced inthe striatum in the presence of A and neurofibrillary tangles inAD (Davies, 1979; Oyanagi et al., 1989; Braak and Braak, 1990;Selden et al., 1994a,b; Scott et al., 1995; Boissiere et al., 1997;Klunk et al., 2004). Although A aggregates/deposits are thoughtto be neurotoxic (Hartley et al., 1999; Selkoe, 2001), perhaps byspreading from the cortex to subcortical nuclei (Saper et al., 1987;Mufson et al., 1995; Yanker, 1996), the impact A upon centralcholinergic neuron degeneration in AD remains unclear.</p><p>Various types of transgenic mice have been engineered to mimicdifferent aspects of AD neurodegeneration (Suh and Checler, 2002;</p><p> of Disease 28 (2007) 315A142 disrupts normal cholinergic neurotransmission (Wang et al.,2000; Nagele et al., 2002; Chen et al., 2006), further suggesting thatA may precipitate cholinergic system dysfunction in AD.</p></li><li><p>logyThe APPswe/PS1E9 transgenic mouse displays an early andaggressive onset of neuritic A deposition in the cortex, hippo-campus and striatum (Lazarov et al., 2002; Perez et al., 2005)together with memory impairment (Savonenko et al., 2005).Despite the report of alterations in cortical acetylcholinesteraseand ChAT activity seen in 19-month-old APPswe/PS1E9transgenic mice (Savonenko et al., 2005), there are no detailedinvestigations of the impact of A deposition upon cholinergiclocal and long projection neuronal systems in this transgenicmouse model of AD. Therefore, we performed quantitativemorphologic and biochemical analyses of cholinergic cortical,striatal, hippocampal, and nucleus basalis cholinergic systems inyoung (26 months) and middle-aged (1016 months) amyloidover-expressing APPswe/PS1E9 transgenic compared to age-matched non-transgenic mice.</p><p>Materials and methods</p><p>The present study used a total of 40 animals (both genders)consisting of young (26 months of age) and old (1016 monthsof age) heterozygous transgenic (tg) mice harboring FAD-linkedmutant APPswe/PS1E9 [co-expressing presenilin 1 (PS1) and achimerica mousehuman amyloid precursor protein (APP) 695with mutations (K595N, M596L) linked to Swedish FADpedigrees (APPswe) via the mouse prion protein promoter(Borchelt et al., 1996a,b, 1997; Lesuisse et al., 2001)] and age-matched non-transgenic (ntg) littermate mice. At least twofemale mice were included in each group examined. These micewere obtained by crossing single APPswe line C3-3 andPS1E9 line S-9 tg mice and PS1E9 line S-9 tg with ntglittermate mice. The background strains for APPswe are {C3H/HeJC57BL/6J F3}C57BL/6J n1, and PS1E9 are C3H/HeJC57BL/6J F3. Animal care and procedures were con-ducted according to the National Institutes of Health Guide forthe Care and Use of Laboratory Animals. All mice wereanesthetized with an injection of ketamine/xylazine (285 mg/kg/9.5 mg/kg) and perfused transcardially with ice-cold 0.9%sodium chloride (NaCl) solution. Brains were rapidly removedfrom the skull and hemisected in a frozen stainless steel brainblocker. One hemisphere was sectioned into 1 mm coronal slabson wet ice in the brain blocker. Tissue pieces were dissectedfrom the cortex, hippocampus and striatum using fiduciarylandmarks and frozen at 80 C until processed for enzymaticassay. The other hemisphere was immersion-fixed in 4%paraformaldehyde and 0.1% glutaraldehyde in 0.1 M phosphatebuffer for 12 h at 4 C and cryoprotected in 30% sucrose at4 C until the brain sank, and then cut on a freezing slidingknife microtome in the coronal plane at a thickness of 40 m.The sections were stored at 4 C in a cryoprotectant solution(30% glycerol, 30% ethylene glycol in 0.1 M phosphatebuffered saline) prior to use.</p><p>Immunohistochemistry, immunofluorescence and histochemistry</p><p>Fixed tissue was processed as free-floating sections andsingly labeled with a polyclonal antiserum raised against cholineacetyltransferase (ChAT) from human placenta (1:1000 dilution,Chemicon, CA, USA). The specificity of the antibody has beendescribed extensively (Mesulam et al., 1983; Mufson et al.,</p><p>4 S.E. Perez et al. / Neurobio1989a,b). Sections were washed several times in Tris-bufferedsaline (TBS) before incubation with 0.1 M sodium periodate (toinhibit endogenous peroxidase activity) in a TBS solution for20 min. After several rinses in a solution containing 0.25%Triton X-100 in TBS, the tissue was placed in a blockingsolution containing TBS with 0.25% Triton X-100 and 3%horse serum for 1 h. Sections were subsequently incubated ingoat anti-human ChAT for 48 h in a solution containing TBS,1% Triton X-100 and 1% horse normal serum. Followingwashes with 1% horse normal serum in TBS, sections wereincubated with biotinylated horse anti-goat secondary antibodies(Vector Laboratories, CA, USA) for 1 h. After several washes inTBS the tissue was incubated for 60 min with an avidinbiotincomplex (1:500; Elite Kit, Vector Laboratories, CA, USA).Tissue was rinsed in 0.2 M sodium acetate, 1.0 M imidazolbuffer (pH 7.4), and developed in an acetateimidazol buffercontaining 2.5% nickel sulfate, 0.05% 3,3-diaminobenzidinetetrahydrochochloride (DAB; Sigma-Aldrich, St. Louis, MO,USA) and 0.0015% H2O2. The reaction was terminated using anacetateimidazol buffer solution. Finally, sections were mountedon glass slides, dehydrated in graded alcohols, cleared inxylenes and cover slipped with DPX (Biochemica Fluka,Switzerland). All sections were processed at the same timeusing the same chemical reagents to avoid batch-to-batchvariation during immunostaining. Additional sections weredual-labeled for ChAT and A using the 10D5 monoclonalantibody raised against amino acids 116 of the human beta-amyloid protein (1:10,000 dilution, gift of Elan Pharmaceutics,San Francisco, CA, USA). For double immunostaining, ChATwas developed using a nickel chromagen followed by incubationwith the monoclonal 10D5 antibody and visualized using a novared substrate kit (Vector Laboratories, CA, USA). This dual-staining method results in a two-colored profile: dark-blue ChATpositive profiles, and red A containing plaques. Immunofluor-escence was also used for A visualization: these sections werestained with 10D5 antiserum (1:1000 dilution) and developedwith Cy3-conjugated donkey anti-mouse antibodies (1:300;Jackson ImmunoResearch Labs, West Grove, PA, USA). Inaddition sections were also double stained with 10D5 (nova redsubstrate kit) and/or histochemically reacted for acetylcholines-terase (AChE), the degrading enzyme for acetylcholine, usingthe Karnovsky and Roots method (1964) followed by silverintensification (Emre et al., 1993).</p><p>Radioenzymatic assay</p><p>Brain regions including cortex, hippocampus and striatumfrom 26 months old and 1016 months old APPswe/PS1E9 tgand age-matched ntg littermate mice were processed fordetermination of ChAT enzyme activity using a modification ofthe Fonnum method (Fonnum, 1975; DeKosky et al., 1985).Frozen tissue was homogenized using high frequency sonicationin a solution containing 0.5% Triton X-100 and 10 mM EDTA.Briefly, 5 l of tissue homogenate was combined with C-14labeled acetyl Co-A (New England Nuclear, Boston, MA, USA),incubation buffer (100 mM sodium phosphate, 600 mM NaCl,20 mM choline chloride, 10 mM disodium EDTA, pH7.4), andphysostigmine (20 mM, Sigma-Aldrich, St. Louis, MO, USA).After 30 min of incubation at 37 C in a water bath, the reactionwas stopped by the addition of 4 ml of 10 mM phosphate buffer(pH 7.4). Subsequently, 1.6 ml of acetonitrile/tetrephenalboron</p><p>of Disease 28 (2007) 315mixture and 8 ml of scintillation fluid were added to cause phaseseparation. After the samples stabilized for 24 h, they were</p></li><li><p>counted in a scintillation counter. Protein content of the sampleswas determined using BCA protein assay kits (Pierce, Rockford,IL, USA). ChAT activity was expressed as mol/h/g protein.Samples were coded, and all assays performed in triplicate by atechnician blinded to experimental groups.</p><p>Stereologic analysis</p><p>The optical dissector method was used to determine thenumber of ChAT immunoreactive (-ir) neurons in the cerebralcortex (motor, cingulate and sensory cortices), striatum andnucleus basalis in 3- to 6-month-old and 10- to 16-month-oldAPPswe/PS1E9 tg as well as age-matched ntg mice aspreviously described (Jaffar et al., 2001; Perez et al., 2005). Theregions were manually outlined under low magnification andsystematically analyzed using a random sampling design. Theestimated numbers of ChAT-ir neurons were performed usingMicroBrightField stereological software (Williston, VT) in anOlympus BX-60 microscope coupled with LEP MAC5000(BioVision Technologies; Exton, PA, USA). The coefficients oferror were calculated according to Gundersen et al. (1988) andvalues b0.10 were accepted (West, 1993). The ChAT-ir neurons inthe nucleus basalis were counted immediately caudal to thedecussation of the anterior commissure until the end of the globuspallidus (Figs. 1BD). These neurons were located ventral to andwithin the globus pallidus as well as intermingled within theinternal capsule tracts. The ChAT-ir neurons in the sensory cortexwere counted until the appearance of cholinergic immunoreactivityin the interpeduncular nucleus (Figs. 1BE), whereas in the</p><p>counted throughout their entire extension (Fig. 1). Since hemi-secting a brain can result in asymmetrical midline cuts,quantification of ChAT-ir neurons in the medial septum andvertical limb of the diagonal band of Broca was not performed. Allof the ChAT-ir neuron counts were performed by an observerblinded to age and genotype.</p><p>Optical density and area measurements</p><p>Quantification of the relative optical density (OD) of neuronalChAT immunoreactivity in the cingulate, motor and sensorycortices, striatum and nucleus basalis from young (36 months)and middle-aged (1016 months) APPswe/PS1E9 tg and ntgmice was performed using a densitometry software program(Image 1.60, Scion 1.6) as previously described (Mufson et al.,1997; Ma et al., 1999; Jaffar et al., 2001; Perez et al., 2005). TheChAT-ir neurons of the different regions were outlined manually,and the OD and area measurements were automatically analyzedin gray-scale images. Background tissue levels were measured andthe average subtracted from the OD measurements of ChAT-ir.These measurements were performed by an observed blinded toage and genotype.</p><p>Statistical analysis</p><p>Data obtained from the ChAT enzymatic assay, stereologic andOD measurements were evaluated using the MannWhitney rank-sum test, a non-parametric test (SigmaStat 3.0; Aspire SoftwareInternational, Leesburg, VA, USA). This test is more powerful in</p><p>mitatissed fx; DGnal baact; LV</p><p>5S.E. Perez et al. / Neurobiology of Disease 28 (2007) 315cingulate and motor cortex as well as in the striatum they were</p><p>Fig. 1. (AE) Rostral to caudal diagrams showing transverse sections delicingulate and sensory cortices as well as the striatum and nucleus basalis procenucleus basalis; BST, bed nucleus of the stria ter...</p></li></ul>


View more >