当前位置：文档库 › No-Nature_Janus_et_al[1]

No-Nature_Janus_et_al[1]

14.Chen,K.S.et al.Neurodegenerative Alzheimer-like pathology in PDAPP717V!F transgenic mice.

Prog.Brain Res.117,327±334(1998).

15.Ennaceur,A.&Delacour,J.A new one-trial test for neurobiological studies of memory in rats.1:

Behavioral data.Behav.Brain Res.31,47±59(1988).

16.Tang,Y.P.et al.Genetic enhancement of learning and memory in mice.Nature401,63±69(1999).

17.Chapman,P.F.et al.Impaired synaptic plasticity and learning in aged amyloid precursor protein

transgenic mice.Nature Neurosci.2,271±276(1999).

18.Hsiao,K.et al.Correlative memory de?cits,A b elevation,and amyloid plaques in transgenic mice.

Science274,99±102(1996).

19.Johnson-Wood,K.et al.Amyloid precursor protein processing and A b42deposition in a transgenic

mouse model of Alzheimer disease.Proc.Natl https://www.wendangku.net/doc/1a1001430.html,A94,1550±1555(1997).

20.Dodart,J.-C.,Mathis,C.,Bales,K.R.,Paul,S.M.&Ungerer,A.Early regional cerebral glucose

hypometabolism in transgenic mice overexpressing the V717F b-amyloid precursor protein.Neurosci.

Lett.277,49±52(1999).

21.Dodart,J.-C.et al.Neuroanatomical abnormalities in behaviorally characterised APPV717F Trans-

genic Mice.Neurobiol.Dis.7,71±85(2000).

https://www.wendangku.net/doc/1a1001430.html,rson,J.,Lynch,G.,Games,D.&Seubert,P.Alterations in synaptic transmission from young and

aged PDAPP mice.Brain Res.840,23±35(1999).

23.Clayton,N.S.&Dickinson,A.What,where and when:Episodic-like memory during cache recovery

by scrub jays.Nature395,272±274(1998).

24.O'Keefe,J.&Nadel,L.The Hippocampus as a Cognitive Map(Clarendon,Oxford,1978).

25.Dodart,J.C.et al.Behavioral disturbances in transgenic mice overexpressing the V717F b-amyloid

precursor protein.Behav.Neurosci.113,982±990(1999).

26.Naslund,J.et al.Correlation between elevated levels of amyloid b-peptide in the brain and cognitive

decline.J.Am.Med.Assoc.283,1571±1577(2000).

27.Schenk,D.et al.Immunization with amyloid-b attenuates Alzheimer-disease-like pathology in the

PDAPP mouse.Nature400,173±177(1999).

28.Masliah,E.et https://www.wendangku.net/doc/1a1001430.html,parison of neurodegenerative pathology in transgenic mice overexpressing

V717F b-amyloid precursor protein and Alzheimer's disease.J.Neurosci.16,5795±5811(1996). Acknowledgements

This work was supported by grants from the MRC and the Cunningham Trust.We are indebted to Bill Nailon(CJD Surveillance Unit,Edinburgh)and Karen Khan(Elan Pharmaceuticals)for assistance with immunocytochemistry and image analysis. Correspondence and requests for materials should be addressed to R.G.M.M.

(e-mail:R.G.M.Morris@https://www.wendangku.net/doc/1a1001430.html,).

.................................................................

A b peptide immunization reduces behavioural impairment and plaques in a model of Alzheimer's disease

Christopher Janus*,Jacqueline Pearson*,JoAnne McLaurin*,

Paul M.Mathews2,Ying Jiang2,Stephen D.Schmidt2,M.Azhar Chishti*, Patrick Horne*,Donna Heslin*,Janet French*,Howard T.J.Mount*, Ralph A.Nixon2,Marc Mercken3,Catherine Bergeron*§,Paul E.Fraser*, Peter St George-Hyslop*§&David Westaway*

*Centre for Research in Neurodegenerative Diseases,Departments of Medicine, Laboratory Medicine and Pathobiology,and Medical Biophysics,University of Toronto,Tanz Neuroscience Building,6Queen's Park Crescent West,Toronto, Ontario M5S3H2,Canada

2Nathan Kline Institute Center for Dementia Research,and New York University School of Medicine,140Old Orangeburg Road,Orangeburg,New York10962, USA

3Janssen Research Foundation,Turnhoutseweg,30,B-2340Beerse,Belgium

§Departments of Medicine(Division of Neurology)and Pathology,Toronto Western Hospital,University Health Network,Toronto,Ontario M5S1A8, Canada .............................................................................................................................................. Much evidence indicates that abnormal processing and extracel-lular deposition of amyloid-b peptide(A b),a proteolytic deriva-tive of the b-amyloid precursor protein(b APP),is central to the pathogenesis of Alzheimer's disease(reviewed in ref.1).In the PDAPP transgenic mouse model of Alzheimer's disease,immuni-zation with A b causes a marked reduction in burden of the brain amyloid2,3.Evidence that A b immunization also reduces cognitive dysfunction in murine models of Alzheimer's disease would support the hypothesis that abnormal A b processing is essential to the pathogenesis of Alzheimer's disease,and would encourage the development of other strategies directed at the`amyloid cascade'.Here we show that A b immunization reduces both deposition of cerebral?brillar A b and cognitive dysfunction in the TgCRND8murine model of Alzheimer's disease without, however,altering total levels of A b in the brain.This implies that either a,50%reduction in dense-cored A b plaques is suf?cient to affect cognition,or that vaccination may modulate the activity/abundance of a small subpopulation of especially toxic A b species.

To explore the behavioural consequences of A b immunization, we used the TgCRND8murine model of Alzheimer's disease that expresses a mutant(K670N/M671L and V717F)human b APP695 transgene under the regulation of the Syrian hamster prion pro-moter on a C3H/B6strain background(M.A.C.et al.,manuscript in preparation).TgCRND8mice have spatial learning de?cits at 3months of age that are accompanied by both increasing levels of SDS-soluble A b and increasing numbers of A b-containing amyloid plaques in the brain.Age-and sex-matched TgCRND8mice and non-Tg littermates in three cohorts were vaccinated at6,8,12,16 and20weeks with either A b42or islet-associated polypeptide (IAPP),which has similar biophysical properties to A b but is associated with a non-central nervous system(CNS)amyloidosis. Both immunogens were in b-pleated-sheet conformation at the

Wavelength (nm)

Figure1A b and IAPP peptide immunogens were predominantly b-structured and induced antibodies recognizing?brillar A b deposits.a,Circular dichroism spectra of A b42 (dotted line)and IAPP(solid line)before immunization are predominantly b-structured. m Degrees,millidegrees.b,Negative-stain electron microscopy(scale150nm)of A b42 peptide immunogen showed varying length?bres;c,IAPP peptide immunogen showed short laterally aggregated?bres.d,f,Serum(1:1,000dilution)from non-immunized TgCRND8mice(d)and from23-week old IAPP-immunized mice(f)did not recognize the A b plaques in adjacent,non-formic-acid treated sections from a non-immunized TgCRND8mouse with abundant A b-positive plaques.e,Sera from A b42-immunized mice strongly labelled dense-cored amyloid plaques but not diffuse A b deposits(which are profusely present in these animals and were labelled with other anti-A b antibodies such as4G8;ref.28).

time of injection (Fig.1)and induced detectable antibody titres in all mice by 13weeks of age (as measured by enzyme-linked immu-noabsorbent assay (ELISA)using ?brillar forms of the respective immunogen).These titres increased by a further ,2±3-fold at 23weeks (A b 42titres in A b 42-immunized mice (mean 6s.e.m.):1:3,6406470at 13weeks;1:7,50061,712at 23weeks;IAPP titres in IAPP-immunized mice:1:3,83361,167at 13weeks;1:11,50063661at 23weeks).The sera from A b -immunized mice intensely decorated extracellular,dense-cored plaque deposits when applied as an immunohistochemical reagent to sections of brain from TgCRND8mice containing abundant amyloid plaques (which predominantly display A b in a b -sheet conformation)or to formic-acid-treated sections (which also display additional non-b -sheet A b -epitopes)(Fig.1).However,these sera reacted only very weakly with diffuse,non-?brillar A b deposits,which can be readily detected in these tissues by anti-A b monoclonal antibodies such as

4G8(data not shown).The A b -immune sera did not stain normal neurons,indicating limited crossreactivity with b APP holoprotein.In contrast,sera from non-immunized or IAPP-immunized mice did not stain any structures.Together,these data indicate that in this strain of mice immunization with A b 42proto?brillar assemblies induced antibodies directed primarily towards A b in a b -sheet conformation.

The mice were tested longitudinally in a reference memory version of the Morris water maze test at 11,15,19and 23weeks (Fig.2).At each age of testing,the hidden platform was placed in a different quadrant of the pool.These data were analysed for the entire test period using a mixed model analysis of variance (ANOVA),with immunogen (A b 42versus IAPP)and genotype (TgCRND8versus non-Tg)as a between-subject factor,and age-of-testing (11,15,19and 23weeks)as a within-subject factor.This longitudinal design and mode of analysis,which simulates long-itudinal human clinical trials 4,revealed that A b 42-immunized TgCRND8mice performed signi?cantly better than IAPP-immu-nized TgCRND8mice (P ,0.05),with 31%of the performance variance being due to the effects of the immunogen.However,the improvement was partialDthe A b 42-immunized TgCRND8mice did not perform as well as their non-Tg littermates (P ,0.01).Because the experimental design involved testing of naive mice at 11weeks of age,followed by a series of reversal tests at 15,19and 23weeks,an additional analysis carried out on the reversal tests con?rmed improved performance for the A b -immunized Tg mice (P ,0.02)during this phase of testing as well.The overall conclu-sion that A b 42immunization ameliorates the cognitive de?cit of TgCRND8mice was robust,regardless of whether the analysis assessed latency to reach the hidden platform or swim path length (a measure that is less sensitive to swim speed and ˉoating 5).

The improved performance of A b 42-immunized TgCRND8mice was not due to a nonspeci?c effect of immunization or to an effect on other behavioural,motor,or perceptual systems.Control studies

L a t e n c y (s )

L a t e n c y (s

)

L a t e n c y (s )

Session

L a t e n c y (s

)

L a t e n c y (s

)

L a t e n c y (s )

Session

L a t e n c y (s e c )

Figure 2Reference memory version of Morris water maze test in TgCRND8mice.At 11weeks of age,non-immunized TgCRND8mice (n =5)show cognitive impairment relative to non-Tg controls (n =8)(a ),which is similar to that of IAPP-immunized

TgCRND8mice (n =12)(b ,left),whereas the performance of A b 42-immunized TgCRND8mice (n =9)(b ,right)approaches that of non-Tg littermates (n =19).At 15(c )and 19(d )weeks of age,the IAPP-immunized TgCRND8mice (n =6)(left)were impaired compared with non-Tg littermates (P ,0.01,q 2<36%;n <15),but were not signi?cantly different from the A b 42-immunized TgCRND8mice (n <6)(right).At

23weeks of age,the IAPP-immunized TgCRND8mice (n =6)(e ,left)were signi?cantly impaired relative to both non-Tg littermates (n =15;P ,0.001,q 2=65%)and A b 42-immunized TgCRND8mice (n =6;P ,0.01)(e ,right).Vertical bars represent s.e.m.See text and Supplementary Information for statistical analyses.

c d

T o t a l p l a q u e a r e a H I P P (μm 2)

IAPP-immunized TgCRND8A β42-immunized TgCRND8

Figure 3At 25weeks of age,in TgCRND8mice immunized with A b 42peptides,dense-cored A b plaque burden is reduced in the hippocampus (HIPP)(a )and in the cerebral cortex (CTX)(b ).The A b 42-immunized TgCRND8mice had 50%fewer plaques than IAPP-immunized TgCRND8controls (71.4610.8per area counted versus 119.7614.6in the cortex,P ,0.05;and 11.661.6per area counted versus 20.961.7in hippocampus,P ,0.01).Representative pictures of the distribution of A b plaques labelled by the Dako 6F/3D anti-A b monoclonal antibody in hippocampus of IAPP-and A b 42-immunized TgCRND8mice (c and d ,respectively).Vertical bars represent s.e.m.Askerisk,P ,0.05;two asterisks,P ,0.01.Scale bars,100m m.

in non-Tg mice established that immunization with Freund's Adjuvant plus phosphate buffered saline,Freund's plus A b42, Freund's plus IAPP,IAPP alone,or A b42alone had no effect on performance in the Morris water maze test(P.0.1).During non-spatial pre-training at11weeks,the latency of random search for a hidden platform,as well as swim speed,and swim path length to a visible platform were not affected by genotype or immunization (P.0.05).Similarly,at23weeks,performance during a visible platform test and spontaneous exploration of the open?eld were not signi?cantly inˉuenced by immunization(P.0.05).Gender effects were not signi?cant between the TgCRND8and IAPP-or A b42-immunized non-Tg littermates(P.0.05).

Analysis of performance in the probe trials with the platform removed(annulus crossing index,passes over platform site or dwelling in the target quadrant)revealed no signi?cant differences owing to immunization or genotype,or interactions between these two factors(each P.0.05).This probably arises both because the probe trials were conducted within30min of the?nal training trial on day5(by which point most mice had learned),and because repeated administration of probe trials(at15,19and23weeks) makes them a less effective measure of spatial memory6. Because the main analysis revealed signi?cant immunogen′genotype(P,0.01)and immunogen′genotype′age(P,0.05) interactions,as well as signi?cant main factor effects for immuno-gen,genotype and age of testing(P,0.01for all),post hoc analyses were carried out for each age of testing.These showed markedly improved performance in A b42-immunized TgCRND8mice at11 and23weeks,with A b42-immunization accounting for large por-tions of the variance(q2)(19%at11weeks;42%at23weeks). Analyses at15and19weeks did not show statistically signi?cant differentiation of the two immunogens.However,this probably reˉects the effects of previous test experience(`carrying-over' effect7).Reduction of impairment during re-testing in the water maze has been reported in another study of a murine model of Alzheimer's disease8,and similar re-test effects are commonly seen in placebo-treated patients in human trials9.The subsequent re-emer-gence of impaired cognition in the IAPP-immunized TgCRND8mice at23weeks probably reˉects the advance of Alzheimer's-disease-related phenotypes,which are clearly progressive over this epoch in TgCRND8mice.Thus,increased A b levels are?rst detectable by western blot or ELISA analysis in brain homogenates at8.5±17and 10weeks,respectively,but thereafter rise progressively.Similarly, cortical dense-cored A b-plaques are?rst detectable immuno-histochemically at7±10weeks,but increase by a further18-fold at 19±27weeks(M.A.C.et al.,manuscript in preparation).

In agreement with previous reports2,3,A b42immunization caused ,50%reduction in the number and size of A b-positive dense-cored plaques(Fig.3)without affecting steady-state levels of b APP holoprotein,amino-terminal secreted fragments(b APP S),or car-boxy-terminal fragments in the brain(see Supplementary Informa-tion).However,in contrast to the previous reports,A b42 immunization had no signi?cant effect on the levels of formic-acid-extractable A b in brain at13weeks(P.0.1)or25weeks (IAPP immunized(n=7):A b40=11,47561,567fmol per mg protein,A b42=39,98364,387;A b42immunized(n=6):A b40= 12,27662,386;and A b42=50,45768552,mean6s.e.m.,P<0.27 two-tailed t-test).Sandwich ELISAs using two other detection antibodies(directed at A b residues1±4and17±24)yielded similar results,arguing against a selective effect of immunization on A b species differing in their amino terminus identity or in post-translational modi?cation.

One explanation for our results is that dense-cored cerebral amyloid plaques are the toxic moiety,and the,50%reduction caused by A b42vaccination is suf?cient to prevent or reverse the behavioural de?cits.Although this cannot be excluded,the density of such plaques correlates poorly with the ante-mortem severity of dementia in most human studies10.Furthermore,a dissociation between plaque deposition and cognitive and/or neuronal dysfunc-tion has been described in other TgAPP mice11,12.An alternative explanation is that immunization affects A b either in a particular conformation(for example,b-sheet forms in proto?brils)or in a restricted compartment.The former is more likely because we used oligomeric assemblies of A b in b-sheets(`proto?brils')as an immunogen,and the resultant antisera preferentially recognized b-sheet forms of A b.This is signi?cant because monoclonal antibodies raised to A b epitopes that initiate?bril aggregation inhibit assembly of synthetic A b oligomeric proto?brils in vitro13.It is possible,therefore,that the antibodies induced in the current strain of mice may bind to b-sheet oligomeric aggregates and inhibit further assembly.This A b species is especially neurotoxic,a critical intermediary in?brillogenesis and an accurate predictor of neuro-degeneration14±17.Consequently,small amounts of such antibodies that cross the blood±brain barrier(0.1%of serum levels3)might be suf?cient to attenuate both the behavioural de?cits caused by this neurotoxic form of A b and the further aggregation of these species into?brillar A b in dense-cored plaques.Because this pool of A b is small,and because antibodies to this form of A b might need only inhibit assembly of A b?brils to have a functional effect,these antibodies need not necessarily cause large changes in total cerebral A b.It is also possible that the antibodies redistribute A b from dense-cored plaques to diffuse A b deposits.Either might explain the divergent effects on levels of dense-cored plaques and A b measured biochemically(although high levels of cerebral A b in the presence of relatively few plaques are seen in humans18and in transgenic mouse models19).

It is conceivable,however,that immunization might modulate A b metabolism through several distinct mechanisms,including destruction of A b by microglial phagocytosis3.Such different effects,which might reˉect variations in antigen presentation or in strain-speci?c immune response,might also explain the disparity between the effects of A b immunization on total A b levels in this and the previous studies.If correct,such variations could compli-cate the use of active immunization in humans.In addition to clinical caveats raised previously20,it is also important to emphasize that,although A b42immunization had a strong effect on behaviour and neuropathology in this mouse model,it did not fully reverse these features.This might reˉect either inef?cient ingress of anti-bodies to the CNS and/or the possibility that other b APP proteo-lytic fragments may be involved in the pathogenesis of Alzheimer's disease21.These issues will need to be addressed directly by future studies.Nonetheless,our data support the hypotheses that A b plays a central role in Alzheimer's disease and that procedures that modulate its production,assembly and/or removal might be used as treatments.M

Methods

Mice

The TgCRND8mice(M.A.C.et al.,manuscript in preparation)were maintained in a hybrid C3H/B6background.Experimental groups derived from crosses to B6mice were matched for gender and weight,and transgene identity was unknown to experimenters at all stages of the study.Sixty-eight6-week-old mice were immunized(TgCRND8,n=28, non-Tg,n=40);60entered the behavioural testing at11weeks(A b immunized: TgCRND8,n=12;non-Tg,n=20;and IAPP immunized:TgCRND8,n=9;non-Tg, n=19).Thirteen mice were killed at13weeks,and their brains analysed(A b immunized: TgCRND8,n=3;non-Tg,n=4;IAPP immunized:TgCRND8,n=2,non-Tg=4).All remaining mice were longitudinally tested until23weeks of age(A b immunized: TgCRND8,n=9;non-Tg,n=16;IAPP immunized:TgCRND8,n=7;non-Tg,n=15). Thirteen non-immunized mice(TgCRND8,n=5;non-Tg,n=8)were included as controls in the test conducted at11weeks of age.

Immunization procedures

Synthetic A b42and IAPP peptides were isolated by reverse phase high performance liquid chromatography on a C18m bondapak column,with purity determined using mass spectrometry and amino-acid analyses.Before preparation of the vaccine,the secondary structure and?bre morphology of the peptides were assessed using circular dichroism spectroscopy and by electron microscopy22.The immunization protocol and schedule have been described2.Antibody titres were assayed in triplicate by ELISA(see

Supplementary Information for more details)in serum samples(200m l of blood) collected at13and25weeks.

Behavioural tests and data analysis

The water maze apparatus,mouse handling and general testing procedures have been described23.Before the?rst spatial learning test at11weeks,all mice underwent non-spatial pre-training(NSP)to assess swimming abilities and to accustom mice to the test24,25 (see Supplementary Information).Two days after the NSP phase,all mice underwent a reference memory training with a hidden platform placed in the centre of one quadrant of the pool for5days,with four trials per day.After the last trial of day5,the platform was removed from the pool and each mouse received one60-s swim probe trial.Escape latency (s),length of swim path(cm),swim speed(cm s-1),%ofˉoating(speed less then5cm s-1), %of time in outer zone(near the pool wall),and%of time and path in each quadrant of the pool were recorded using an on-line HVS image video tracking system23(see Supplementary Information).

For the probe trials,an annulus-crossing index was calculated that represents the number of passes over the platform site,minus the mean of passes over alternative sites in other quadrants.The index expresses the spatial place preference and controls for alternative search strategies without place preferences,such as circular search paths26,27.All mice were re-tested at15,19and23weeks of age,one week before the next immunization. At each re-testing,the platform was placed in the centre of a different,semi-randomly chosen pool quadrant for all?ve sessions of training.At the end of the experiment,all mice were given a cue(visual platform)learning test.This was followed by the open-?eld test to investigate spontaneous locomotor exploration.Behavioural data was analysed using a mixed model of factorial ANOV A.Degrees of freedom were adjusted by Greenhouse±Geisser epsilon correction for heterogeneity of variance.A Bonferroni Inequality correc-tion was applied for multiple comparisons.Omega squared(q2)was used as a measure of effect size caused by different factors.

Analysis of b APP and amyloid burden in brain

Three5-m m sections at25-m m intervals from one cerebral hemisphere were immunos-tained with Dako6F/3D anti-A b monoclonal antibody to residues8±17(which is primarily reactive against dense-cored plaques)with4G8(ref.28),or with sera from immunized mice,and counterstained with haematoxylin and resin mounted as described (M.A.C.et al.,manuscript in preparation).For some samples the formic-acid treatment step was omitted.End products were visualized with diaminobenzidine.Amyloid plaque burden was assessed using Leco IA-3001image analysis software interfaced with a Leica microscope and a Hitachi KP-M1U CCD video camera.The quantitative analysis was performed at of′25magni?cation,and the image frame and guard size was set to

0,0,639,479(307,200m m2)for each slide.The brain area(cortex or hippocampus)was outlined using the edit plane function,and the area and number of plaques in the outlined structure were recorded.Data were pooled for all three sections.

Cerebral A b levels were assayed from formic-acid-extracted29,hemi-brain sucrose homogenates using an ELISA method(see Supplementary Information)in which A b was trapped with either monoclonal antibody to A b40(JRF/cAb40/10)or A b42(JRF/cAb42/ 26)and then detected with horseradish peroxidase(HRP)-conjugated JRF/Abtot/17.The dilution of JRF/Abtot/17and samples were optimized to detect A b in the range of50to 800fmol ml-1.ELISA signals are reported as the mean6s.e.m.of four replica wells in fmol A b per mg total protein(determined with the BioRad DC protein assay),based on standard curves using synthetic A b1±40and A b1±42peptide standards(American Peptide Co.Sunnyvale,CA).Cerebral b APPs levels were analysed in supernatant of brain as described30.

Received19July;accepted16November2000.

1.Steiner,H.,Capell,A.,Leimer,U.&Haass,C.Genes and mechanisms involved in beta-amyloid

generation and Alzheimer's disease.Eur.Arch.Psychiat.Clin.Neurosci.249,266±270(1999).

2.Schenk,D.et al.Immunization with A b attenuates Alzheimer's disease-like pathology in the PDAPP

mouse.Nature400,173±177(1999).

3.Bard,F.et al.Peripherally administered antibodies against amyloid beta-peptide enter the central

nervous system and reduce pathology in a mouse model of Alzheimer disease.Nature Med.6,916±919 (2000).

4.Whitehouse,P.J.et al.Clinical trial designs for demonstrating disease-course-altering effects in

dementia.Alzheimer Dis.Assoc.Disord.12,281±294(1998).

5.Morris,R.Developments of a water-maze procedure for studying spatial learning in the rat.J.Neurosci.

Methods11,47±60(1984).

6.Markowska,A.L.,Long,J.M.,Johnson,C.T.&Olton,D.S.Variable-interval probe test as a tool for

repeated measurements of spatial memory in the water maze.Behav.Neurosci.107,627±632(1993).

7.Martin,P.&Bateson,P.Measuring Behaviour(Cambridge Univ.Press,Cambridge,1996).

8.Hsiao,K.Correlative memory de?cits,A b elevation,and amyloid plaques in transgenic mice.Science

274,99±102(1996).

9.Rogers,S.L.,Farlow,M.R.,Doody,R.S.,Mohs,R.&Friedhoff,L.T.A24-week,double-blind,

placebo-controlled trial of donepezil in patients with Alzheimer's disease.Donepezil study group.

Neurology50,136±145(1998).

10.Terry,R.D.,Masliah,E.&Hansen,L.A.in Alzheimer Disease2nd edn(eds Terry,R.D.,Katzman,R.,

Bick,K.C.&Sisodia,S.S.)(Lippincott Williams Wilkins,Philadelphia,1999).

11.Moechars,D.et al.Early phenotypic changes in transgenic mice that over-express different mutants of

amyloid precursor protein.J.Biol.Chem.274,6483±6492(1999).

12.Hsia,A.Y.Plaque-independent disruption of neural circuits in Alzheimer's disease mouse models.

Proc.Natl https://www.wendangku.net/doc/1a1001430.html,A96,3228±3233(1999).

13.Solomon,B.,Koppel,R.,Hanan,E.&Katzav,T.Monoclonal antibodies inhibit in vitro?brillar

aggregation of the Alzheimer b-amyloid peptide.Proc.Natl https://www.wendangku.net/doc/1a1001430.html,A93,452±455(1996). 14.Lorenzo,A.&Yanker,B.A.b-amyloid neurotoxicity requires?bril formation and is inhibited by

Congo red.Proc.Natl https://www.wendangku.net/doc/1a1001430.html,A91,12243±12247(1994).

15.Walsh,D.M.et al.Amyloid b-protein?brillogenesis.Structure and biological activity of proto?brillar

intermediates.J.Biol.Chem.274,25945±25952(1999).

16.Hartley,D.M.Proto?brillar intermediates of amyloid b-protein induce acute electrophysiological

changes and progressive neurotoxicity in cortical neurons.J.Neurosci.19,8876±8884(1999). 17.McLean,C.A.et al.Soluble pool of A b amyloid as a determinant of severity of neurodegeneration in

Alzheimer's disease.Ann.Neurol.46,860±866(1999).

18.Naslund,J.et al.Correlation between elevated levels of amyloid b-peptide in the brain and cognitive

decline.J.Am.Med.Assoc.283,1571±1577(2000).

19.Mucke,L.et al.High-level neuronal expression of A b1±42in wild-type human amyloid protein

precursor transgenic mice:synaptotoxicity without plaque formation.J.Neurosci.20,4050±4058 (2000).

20.St George-Hyslop,P.&Westaway,D.Alzheimer's disease:antibody clears senile plaques.Nature400,

116±117(1999).

21.Lu,D.C.A second cytotoxic proteolytic peptide derived from amyloid b-protein precursor.Nature

Med.6,397±404(2000).

22.McLaurin,J.,Franklin,T.,Chakrabartty,A.&Fraser,P.E.Phosphatidylinositol and inositol

involvement in Alzheimer amyloid-beta?bril growth and arrest.J.Mol.Biol.278,183±194(1998).

23.Janus,C.Spatial learning in transgenic mice expressing human presenilin1(PS1)transgenes.

Neurobiology of Aging21,541±549(2000).

24.Cain,D.P.,Beiko,J.&Boon,F.Navigation in the water maze:the role of proximal and visual cues,

path intergration,and magnetic?eld information.Psychobiology25,286±293(1997).

25.Dudchenko,P.A.,Goodridge,J.P.,Seiterle,D.A.&Taube,J.S.Effects of repeated disorientation on

the acquisition of spatial tasks in rats:dissociation between the appetetive radial arm maze and aversive water maze.J.Exp.Psychol.23,194±210(1997).

26.Gass,P.et al.De?cits in memory tasks of mice with CREB mutations depend on gene dosage.Learn

Mem.5,274±288(1998).

27.Wehner,J.M.,Sleight,S.&Upchurch,M.Hippocampal protein kinase C activity is reduced in poor

spatial learners.Brain Res.523,181±187(1990).

28.Wisniewski,H.M.,Wen,G.Y.&Kim,https://www.wendangku.net/doc/1a1001430.html,parison of four staining methods on the detection of

neuritic plaques.Acta Neuropathol.78,22±27(1989).

29.Susuki,N.et al.An increased percentage of long amyloid B protein secreted by Familial Amyloid b

Protein Precursor(BAPP717)mutants.Science264,1336±1340(1994).

30.Citron,M.et al.Mutant presenilins of Alzheimer's Disease increase production of42residue amyloid

b-protein in both transfected cells and transgenic mice.Nature Med.3,67±72(1997). Supplementary information is available on Nature's World-Wide Web site

(https://www.wendangku.net/doc/1a1001430.html,)or as paper copy from the London editorial of?ce of Nature. Acknowledgements

This work was supported by the Medical Research Council of Canada,Howard Hughes Medical Research Foundation,Alzheimer Society of Ontario,The W.Gar?eld Weston Foundation and the US National Institute of Aging.We thank G.Carlson for useful discussions,and R.Renlund,K.Parisien,J.Haight and J.Cowieson for help during mice immunization.None of the authors has a?nancial or other relationship with Elan Pharmaceuticals Inc.

Correspondence and requests for materials should be addressed to P.H.StG.-H.

(e-mail:p.hyslop@utoronto.ca). .................................................................

A b peptide vaccination prevents memory loss in an animal model

of Alzheimer's disease

Dave Morgan*,David M.Diamond23,Paul E.Gottschall*,

Kenneth E.Ugen§,Chad Dickey§,John Hardy k,Karen Duff?,

Paul Jantzen*,Giovanni DiCarlo*,Donna Wilcock*,Karen Connor*, Jaime Hatcher#,Caroline Hope#,Marcia Gordon*&Gary W.Arendash#

*Alzheimer Research Laboratory,Department of Pharmacology,2Department of Psychology,§Department of Medical Microbiology and Immunology,#Alzheimer Research Laboratory,Department of Biology,University of South Florida,Tampa, Florida33612,USA

3James A.Haley VA Center,Tampa,Florida33612,USA

k Department of Pharmacology,Mayo Clinic,Jacksonville,Florida32224,USA ?Nathan Kline Institute,Orangeburg,New York10962,USA .............................................................................................................................................. Vaccinations with amyloid-b peptide(AB)can dramatically reduce amyloid deposition in a transgenic mouse model of Alzheimer's disease1.To determine if the vaccinations had dele-terious or bene?cial functional consequences,we tested eight months of A b vaccination in a different transgenic model for