Disfunção da sinalização por insulina hipocampal na Doença de Alzheimer · 2016-01-19 · entenderá o que é temer ao Senhor e achará o conhecimento de Deus. Pois o Senhor

Universidade Federal do Rio de Janeiro

Disfunção da sinalização por insulina hipocampal na Doença de Alzheimer

Theresa Rachel Jacinto de Souza Bomfim

Tese submetida ao Programa de Pós-Graduação em Química Biológica do Instituto de Bioquímica Médica da Universidade Federal do Rio de Janeiro, como parte dos requisitos

para a obtenção do grau de Doutor em Ciências (Química Biológica)

Rio de Janeiro 2014

ii



Tese submetida ao Programa de Pós-Graduação em Química Biológica do Instituto de Bioquímica Médica da Universidade Federal do Rio de Janeiro, como parte dos requisitos para a obtenção do grau de Doutor em Ciências (Química Biológica)

Orientadora: Profa. Fernanda G. De Felice

Co-orientador: Prof. Sérgio Teixeira Ferreira

Rio de Janeiro 2014

iii

Ficha Catalográfica

Rachel Jacinto de Souza Bomfim, Theresa. Disfunção da sinalização por insulina hipocampal na Doença de

Alzheimer – Theresa Rachel Jacinto de Souza Bomfim. Rio de Janeiro: UFRJ/IBqM, 2014.

115 fl.: 9 Il. Orientadora: Fernanda G. De Felice Tese (Doutorado) – UFRJ / Instituto de Bioquímica Médica /

Programa de Pós-graduação em Química Biológica, 2014. Referências Bibliográficas: f. 90-115

1. Doença de Alzheimer. 2. Oligômeros do peptídeo beta-amilóide 3. Resistência à insulina. 4. Sinalização por insulina. 5. GLP-1. 6. Inflamação – Tese. I. De Felice, Fernanda Guarino. II. Universidade Federal do Rio de Janeiro, Instituto de Bioquímica Médica, Programa de Pós-graduação em Química Biológica. III. Título.

iv



Rio de Janeiro, 03 de setembro de 2014.

_______________________________________________

Fernanda Guarino De Felice (Orientadora, Professora Adjunta do Instituto de Bioquímica Médica, UFRJ)

_______________________________________________ Sérgio Teixeira Ferreira

(Co-orientador, Professor Titular do Instituto de Bioquímica Médica, UFRJ)

________________________________________________ José Roberto Meyer Fernandes

(Professor Titular do Instituto de Bioquímica Médica, UFRJ)

________________________________________________ José Donato Júnior

(Professor Doutor do Departamento de Fisiologia e Biofísica do ICB/USP)

________________________________________________ Ricardo Augusto de Melo Reis

(Professor Associado III do Instituto de Biofísica Carlos Chagas Filho, UFRJ)

________________________________________________ Antônio Galina Filho

(Revisor, Professor Adjunto do Instituto de Bioquímica Médica, UFRJ)

________________________________________________ Rogério Arena Panizzutti

(Professor Adjunto do Instituto de Ciências Biomédicas, UFRJ)

v

Esta Tese foi realizada no Laboratório de Doenças Neurodegenerativas,

do Instituto de Bioquímica Médica da Universidade Federal do Rio de Janeiro, sob

a orientação da Professora Fernanda Guarino De Felice e co-orientação do

Professor Sérgio T. Ferreira, com auxílios financeiros da Fundação de Amparo à

Pesquisa do Estado do Rio de Janeiro (FAPERJ), do Conselho Nacional do

Desenvolvimento Científico e Tecnológico (CNPq) do Instituto Nacional de

Neurociência Translacional (INNT) e do Human Frontiers Science Program

(HFSP).

vi

“Meu filho, se você aceitar as minhas palavras e guardar no coração os meus mandamentos; se der ouvidos à sabedoria e inclinar o coração para o discernimento; se

clamar por entendimento e por discernimento gritar bem alto, se procurar a sabedoria como se procura a prata e buscá-la como quem busca um tesouro escondido, então você

entenderá o que é temer ao Senhor e achará o conhecimento de Deus. Pois o Senhor é quem dá sabedoria; de sua boca procedem o conhecimento e o discernimento. Ele

reserva a sensatez para o justo; como um escudo protege quem anda com integridade, pois guarda a vereda do justo e protege o caminho de seus fiéis. Então você entenderá o

que é justo, direito e certo, e aprenderá os caminhos do bem. Pois a sabedoria entrará em seu coração, e o conhecimento será agradável à sua alma.” Provérbios 2:1-10

vii

Dedico essa Tese à minha querida mãe e melhor amiga, Vera Jacinto, por ser exemplo

de amor e cuidade em cada detalhe de minha vida! E à minha doce e querida avó, Águeda Jacinto,

por ser exemplo de sabedoria e amor em todas as circunstâncias.

viii

AGRADECIMENTOS

Em primeiro lugar, quero agradecer a Deus, por caminhar comigo todos os

dias nesta jornada acadêmica, sem me deixar faltar nada e por providenciar cada

detalhe necessário para alcançar mais essa etapa. Toda honra e toda glória ao

meu Deus, porque sem Ele, eu não estaria aqui.

Agradeço à minha querida e grande mãe, Vera Jacinto, por todos os muitos

sacrifícios e privações pelos quais passou para permitir que eu pudesse estudar e

morar em outra cidade, por todas as ligações de encorajamento, todos os abraços,

palavras de carinho, conselhos, direcionamentos, por curtir e acompanhar com

grande expectativa cada conquista e vitória no laboratório. Como agradecer por

cada detalhe do seu cuidado incomparável, não tenho palavras para agradecer

por tanto amor e dedicação! Compartilho com você, mãe, essa conquista é nossa!

À minha doce irmã e melhor amiga! Obrigada por sempre ser minha

companheira nesta batalha, por me apoiar em todos os momentos de todas as

formas! Obriagda por todas as vezes que confortou meu coração, me aconselhou,

pela comidinha quentinha e gostosa que fazia. Essa conquista também é nossa

por tudo que passamos juntas!

Ao meu Pai, Valdir Bomfim, por ser amigo, me apoiar em diversos

momentos, nas mudanças, nas lutas e batalhas vividas aqui no Rio. Seu apoio foi

fundamental

A toda minha família, que é minha base, meu suporte, por todas as vezes

que se alegrou comigo em cada conquista, por torcer em todos os momentos.

Agradeço por todo o apoio durante essa fase de estudos, principalmente aos

meus tios Nestor e Maria Célia, e meu avô Antônio Jacinto, que sem exitar me

ajudaram de forma significativa.

Agradeço aos meus Orientadores, Fernanda De Felice e Sérgio Ferreira por

toda confiança, ensinamentos, conselhos, por tantas vezes que se preocuparam

comigo como uma filha! Agradeço por toda paciência e dedicação. Mais do que

Professores, para mim são como amigos queridos que tanto me ajudaram!

ix

Aos meus queridos amigos do LDN, são tantos durante essa longa

caminhada. Mas não podia me esquecer destes queridos que desde o início me

ensinaram tanto e para não cometer nenhuma falta ou injustiça, não mencionarei

nomes. São todos muito imporante na minha formação como aluna e como

pessoa.

Ainda do LDN, não poderia deixar de agradecer às três guerreiras que

sempre estão dispostas a nos ajudar, e mesmo nos casos em que eu chegava aos

45 do segundo tempo! Minhas queridas, Claudinha, Maíra e Mariangela, vocês

conseguem de forma completa deixar nossos dias de tarabalho mais leves!

Agradeço também a uma mulher muito amável e generosa que me iniciou

com tanto carinho na caminha científica, e se fez muito presente para minha

formação. Agradeço a ela, Profª. Martha M. Sorenson, por todas as conversas,

apoio, suporte e ajuda. Nunca me esquecerei de tudo que fez por mim! A Profª

Verônica Salerno, por ter me recebido com tanto carinho, por me apoiar e motivar

sempre a crescer!

Ao Prof. Antônio Galina, que sempre foi mais do que um Revisor, foi uma

grande amigo e Mestre que me ensinou que um Professor, além de ensinar o

conteúdo, também transforma a visão do aluno em relação ao conhecimento.

Nunca me esquecerei de nossas conversas, conselhos e risadas!

Aos membros da Banca, Prof. José Roberto Meyer, Prof. José Donato

Júnior, Prof. Ricardo Reis e Prof. Rogério Panizzutti, muito obrigada por aceitar ao

convite, e tornar esse momento mais rico com as suas contribuições!

x

RESUMO BOMFIM, Theresa Rachel Jacinto de Souza. Disfunção da sinalização por insulina hipocampal na Doença de Alzheimer. Rio de Janeiro, 2014. Tese (Doutorado em Química Biológica) Instituto de Bioquímica Médica, Universidade Federal do Rio de Janeiro, Rio de Janeiro, 2014 Estudos recentes sugerem que existe uma correlação entre a doença de Alzheimer (DA) e a diabetes melitus tipo 2, uma vez que a disfunção na sinalização de insulina ocorre no hipocampo de pacientes com DA. Acredita-se que essa disfunção possa contribuir para o declinio cognitivo na DA. Embora essa conexão entre a DA e o diabetes seja sugerida, o mecanismo pelo qual a inibição da via de sinalização de insulina ocorre no cérebro ainda é desconhecido. No presente trabalho, procuramos investigar os mecanismos que levam a sinalização deficiente de insulina em neurônios hipocampais. Amostras de tecido de cérebro humano com DA apresentaram níveis elevados de IRS-1pSer e JNK ativada, similar ao que ocorre nos tecidos periféricos em pacientes com diabetes. Nossos resultados demonstraram que oligômeros do peptíeo β amiloide (Aβ), sinaptotoxinas que acumulam no cérebro de pacientes DA, ativam a via de sinalização TNF-α/JNK, induzindo a fosforilação do IRS-1 em múltiplos resíduos de serina, e inibem a fosforilação fisiológica IRS-1pTyr em culturas maduras de neurônios hipocampais. A inibição do IRS-1 também foi observada no hipocampo de camundongos transgênicos para a doença de Alzheimer. Interessantemente, a injeção intracerebroventricular de oligômeros de Aβ induziu a ativação da JNK e consequente inibição do IRS-1 em macacos cinomolgos. As patologias neuronais induzidas pelos oligômeros de Aβ, incluindo disfunção no transporte axonal, foram prevenidas pela exposição à exendina-4 (exenatida), um agente anti-diabético. Em camundongos transgênicos, a exendina-4 diminuiu os níveis de IRS-1pSer e JNK ativada no hipocampo e contribuiu para melhores resultados em testes comportamentais de memória. Dados adicionais indicaram que a citocina IL-1β tem uma importante participação tanto na inibição do IRS-1, quanto no prejuízo cognitivo induzido pelos oligômeros. Por estabelecer um link molecular entre a desregulação da sinalização por insulina na doença de Alzheimer e o diabetes, nossos resultados abrem caminhos para investigação de novas abordagens terapêuticas em Alzheimer.

xi

ABSTRACT

BOMFIM, Theresa Rachel Jacinto de Souza. Disfunção da sinalização por insulina hipocampal na Doença de Alzheimer. Rio de Janeiro, 2014. Tese (Doutorado em Química Biológica) Instituto de Bioquímica Médica, Universidade Federal do Rio de Janeiro, Rio de Janeiro, 2014 Defective brain insulin signaling has been suggested to contribute to the cognitive deficits in patients with Alzheimer’s disease (AD). Although a connection between AD and diabetes has been suggested, a major unknown is the mechanism(s) by which insulin resistance in the brain arises in individuals with AD. Here, we show that serine phosphorylation of IRS-1 (IRS-1pSer) is common to both diseases. Brain tissue from humans with AD had elevated levels of IRS-1pSer and activated JNK, analogous to what occurs in peripheral tissue in patients with diabetes. We found that amyloid-β peptide (Aβ) oligomers, synaptotoxins that accumulate in the brains of AD patients, activated the JNK/TNF-α pathway, induced IRS-1 phosphorylation at multiple serine residues, and inhibited physiological IRS-1pTyr in mature cultured hippocampal neurons. Additional data suggest that IL-1β also contributs to IRS-1 inhibition and cognition disfunction induced by Aβ oligomers. Impaired IRS-1 signaling was also present in the hippocampi of Tg mice with a brain condition that models AD. Importantly, intracerebroventricular injection of Aβ oligomers triggered hippocampal IRS-1pSer and JNK activation in cynomolgus monkeys. The oligomer-induced neuronal pathologies observed in vitro, including impaired axonal transport, were prevented by exposure to exendin-4 (exenatide), an anti-diabetes agent. In Tg mice, exendin-4 decreased levels of hippocampal IRS-1pSer and activated JNK and improved behavioral measures of cognition. By establishing molecular links between the dysregulated insulin signaling in AD and diabetes, our results open avenues for the investigation of new therapeutics in AD.

xii

SUMÁRIO

1) Introdução ........................................................................................................ 14 1.1) Doença de Alzheimer ..................................................................................... 14 1.2) Peptídeo β-amilóide (Aβ) ............................................................................... 19 1.3) Oligômeros de Aβ .......................................................................................... 22 1.4) Sinalização por Insulina ................................................................................. 28

1.4.1) Mecanismos Moleculares ........................................................................ 29 1.4.2) Resistência a insulina ............................................................................. 34 1.4.3) Sinalização por Insulina no Sistema Nervoso Central ............................ 36

1.5) Doença de Alzheimer e resistencia a insulina ................................................ 37 1.6) Doença de Alzheimer e Inflamação ............................................................... 41 1.7)Transporte Axonal ........................................................................................... 43

1.7.1) Transporte Axonal, envelhecimento e doença de Alzheimer .................. 47 1.8) Uso da exendina-4 na estimulação da sinalização por insulina ..................... 49 2) Objetivos .......................................................................................................... 52 3) Metodologia ..................................................................................................... 53 3.1) Materiais ......................................................................................................... 53 3.2) Preparo e caracterização dos oligômeros de Aβ ............................................ 54

3.2.1) Gel-filtração por HPLC ............................................................................ 55 3.2.2)Western blot para Oligômeros de Aβ (ADDLs) ........................................ 56

3.3) Culturas primárias de hipocampo de rato....................................................... 58 3.4) Neuropatologia de cérebro humano ............................................................... 59 3.5) Injeção intracerebroventricular de Aβ (i.c.v.) em camundongos..................... 61 3.6) Injeção de Aβ no cérebro de macacos e estudos neuropatológicos ............. 62 3.7) Tarefa de Reconhecimento de Objetos .......................................................... 63 3.8) Tratamento com os oligômeros de Aβ e agentes farmacológicos .................. 65 3.9) Animais transgênicos ..................................................................................... 65 3.10) Administração exendina-4 em animais transgênicos ................................... 66 3.11) Imunocitoquímica ......................................................................................... 67 3.12) Análise por western blot das amostras de hipocampo dos animais transgênicos e das culturas de neurônios hipocampais ........................................ 68 3.13) Análise dos dados ........................................................................................ 69 3.14) Expressão dos plasmídeos .......................................................................... 70 4) Resultados ....................................................................................................... 72 4.1) Agentes antidiabéticos protegem o cérebro de camundongos de uma disfunção na sinalização por insulina causada pelos oligômeros de Aβ associados a doença de Alzheimer. ......................................................................................... 72 4.2) Resultados Adicionais I .................................................................................. 73

xiii

4.3) Resultados Adicionais II ................................................................................. 77 5) Discussão ........................................................................................................ 80 6) Conclusões ...................................................................................................... 89 7) Referências ...................................................................................................... 90 Anexos ............................................................................................................... 115

14

1) INTRODUÇÃO

1.1) Doença de Alzheimer

Poucos diagnósticos na medicina moderna evocan uma apreensão tão

profunda no paciente e em sua família. A possibilidade de desenvolver doenças

cardiovasculares, câncer ou doenças metabólicas é bastante preocupante, mas

existem algumas indicações de que as pessoas tem mais medo de desenvolver

Alzheimer. Isso porque esta doença nos rouba de nossas qualidades mais

humanas: raciocínio, memória, abstração, lingagem, controle emocional

(ALZHEIMER e cols., 1995; WALSH e SELKOE, 2004). E ainda é uma doença

cujo tratamento está muito além de nosso alcance. A doença de Alzheimer (DA) é

o tipo de demência mais comum que atinge o homem durante o envelhecimento.

Uma em cada oito pessoas com 65 anos ou mais é acometida pela DA, com uma

incidência de 13% (Plassman e cols., 2007). Entretanto, esse número aumenta

para 33% para indivíduos com 85 anos ou mais (SESHADRI e cols., 2006). Um

relatório publicado pela Associação Americana de Alzheimer (Alzheimer’

Association) mostrou que aproximadamente 35,6 milhões de pessoas no mundo

foram acometidas pela doença de Alzheimer em 2010 (World Alzheimer Report,

2009). Com o aumento da expectativa de vida, estima-se que este número deverá

dobrar para 65,7 milhões em 2030, avançando ainda mais em 2050 podendo

atingir um número de 115,4 milhões de pessoas (FERRI e cols., 2005;

BROOKMEYER e cols., 2007; KALARIA e cols., 2008).

15

A doença de Alzheimer é uma demência predominantemente esporádica, já

que mais de 95% dos casos não apresentam uma causa genética definida e as

possíveis causas ambientais tem sido amplamente debatidas (BALLARD e cols.,

2011). Atualmente, sabe-se que o envelhecimento é o maior fator de risco para o

desenvolvimento da DA esporádica, embora um desequilibrio no metabolismo

corporal ao longo da vida seja recentemente considerado como um importante

fator que parece contribuir para o seu estabelecimento (DAVIGLUS e cols., 2011;

MATTSON, 2012; DE FELICE, 2013). Entretanto, não existem evindências

consistentemente suficientes para concluir e assumir a associação de algum fator

modificável com o risco para desenvolver Alzheimer (DAVIGLUS e cols., 2011).

O diagnóstico definitivo é realizado após análise histopatológica post-

mortem, o que dificulta a detecção precoce dessa doença. Existem, no entanto,

diversos testes neuropsicológicos específicos com escalas propedêuticas que

fornecem indicações funcionais do estado cognitivo dos pacientes. A partir da

aplicação destas escalas, sugere-se um estado cognitivo normal ou demente, em

diversos níveis (NOWRANGI e cols., 2011).

O avanço da doença de Alzheimer ocorre de forma lenta e progressiva. Nos

casos esporádicos, os primeiros sintomas cognitivos geralmente aparecem após

os 65 anos, sendo na maioria das vezes confundidos com alterações relacionadas

ao envelhecimento. Nos casos genéticos, os sintomas aparecem mais cedo,

dependendo da severidade das mutações apresentadas (BERTRAM e cols.,

2010).

Inicialmente, verifica-se a incapacidade de evocar lembranças recentes,

enquanto que acontecimentos mais antigos conseguem ser recuperados. Em

16

estágios mais avançados, o paciente desenvolve déficit cognitivo severo, confusão

mental, alteração de humor, deficiências na fala, inabilidade motora e,

eventualmente, é levado a óbito (GREENE e cols., 1995).

É importante observar que grande parte dos pacientes que desenvolverão

demência passa por um estágio prodrômico conhecido como disfunção cognitiva

leve (MCI, do inglês mild cognitive impairment). Embora já seja possível ter

indícios de demência nesta etapa, não há garantias de que esta progredirá para a

DA (PETERSEN e cols., 2009). Ainda assim, o MCI é um fenômeno clínico

bastante relevante, pois causa deficiências efetivas no paciente e aumenta

consideravelmente o risco de evolução para a doença de Alzheimer.

Morfologicamente, o cérebro DA, em um estágio bastante avançado, é

caracterizado pela redução da formação hipocampal e do cótex cerebral, além do

alargamento dos ventrículos laterais e estreitamento dos giros (Figura 1 A e B).

As primeiras descrições clínicas e histopatológicas da DA foram feitas, pelo

médico alemão Alois Alzheimer, a partir do estudo do cérebro de sua paciente

Auguste D em 1906 (para uma tradução em inglês do artigo original de Alois

Alzheimer, ver ALZHEIMER e cols., 1995). Nas análises histopatológicas do córtex

dessa paciente o Dr. Alzheimer identificou dois marcadores que são considerados

como lesões bastante importantes para a caracterização desta doença (Figura 1 C

e D).

Uma das caracteríscticas é a presença de inclusões intracelulares que

demarcam o corpo celular e parte dos processos neuronais, hoje conhecidas

como emaranhados neurofibrilares (Figura 1C). Estes agregados são constituídos

de proteína tau hiperfosforilada. Nesta condição, a tau perde sua função fisiológica

17

de associar-se aos microtúbulos e estabilizá-los, formando assim, filamentos

helicoidais pareados que agregam no interior das células neuronais. Esses

emaranhados são observados principalmente no hipocampo, no córtex entorrinal e

na amígdala, principais regiões afetadas na DA (THAL e cols., 2006).

Outros marcadores importantes presentes em todo o córtex cerebral são as

placas senis (Figura 1 D). Alguns anos mais tarde, percebeu-se que essas lesões

eram positivamente marcadas com o corante vermelho-de-Congo, que possui

afinidade por estruturas amilóides. Assim, surgiram os primeiros indícios de que os

componentes destas placas teriam esta natureza, o que permitiu a criação do

termo “placas amilóides”. A caracterização completa destas placas somente viria a

acontecer várias décadas após sua descrição inicial, quando se verificou que o

peptídeo beta-amilóide (Aβ) é o principal componente das placas amilóides

(GLENNER e WONG, 1984; MASTERS e cols., 1985), embora várias outras

proteínas também estejam associadas às placas (WELANDER e cols., 2009).

18

Figura 1: Marcadores histopatológicos da doença de Alzheimer. A) Cérebro acometido pela doença (à direita), exibindo considerável alargamento dos sulcos e estreitamento dos giros, além de notável redução do volume cerebral em comparação a um cérebro de um indivíduo não demenciado (à esquerda). B) Cortes coronais de um cérebro de um paciente com Alzheimer (à direita) revelam o aumento do volume ventricular e a perda de massa cerebral quando comparado com um cérebro de um indivíduo não demenciado (à esquerda). C) Emaranhados neurofibrilares (intracelulares) corados por prata. D) Placas senis (extracelulares) coradas por prata. Adaptado de www.alzheimer.sk.ca.

19

1.2) Peptídeo β-amilóide (Aβ)

Quase um século após a primeira descrição das placas amilódes pelo Dr.

Alois Alzheimer, no meio da década de 80, bioquímicos se esforçaram para

identificar seu principal componente. Foi então que a partir do isolamento de

placas amilóides associadas à vasculatura cerebral de pacientes com Alzheimer

que Glenner e Wong indetificaram um fragmente de aproximadamente 4kDa, o

qual eles nomeram como proteína β-amilóide (GLENNER e WONG, 1984). No ano

seguinte Masters e colaboradores confirmaram a presença do peptídeo β amiloide

nas placas senis isoladas de córtices de pacientes com Alzheimer (MASTERS e

cols., 1985). Estes trabalhos foram fundamentais para dar início a diversos

estudos com o propósito de melhor entender a participação deste peptídeo na

fisiopatologia da DA.

O peptídeo Aβ é originado a partir da clivagem proteolítica da proteína

precursora amilóide (APP do inglês amyloid precursor protein). A APP é uma

proteína integral de membrana codificada pelo gene app, localizado no

cromossomo 21. A APP é expressa em diversos tipos celulares e pode sofrer

diferentes regulações pós-transcricionais e pós-traducionais de acordo com a

sinalização celular. A isoforma mais expressa em neurônios é a APP695, que está

presente nas sinapses e parece exercer uma importante função em regular a

adesão, migração e sinalização neuronal (GRALLE e FERREIRA, 2007).

O resultado do processamento da APP por diferentes complexos

enzimáticos multiproteicos (α, β e γ secretases) é a geração de fragmentos

peptídicos bioativos, dentre os quais o peptídeo beta-amilóide (Aβ). A formação

20

deste peptídeo ocorre a partir da clivagem inicial da APP na membrana

plasmática, primeiramente pela β-secretase (BACE), sendo formado um fragmento

solúvel (sAPPβ) e outro que ainda transpassa a membrana (C99). Posteriormente

o complexo enzimático γ-secretase cliva o peptídeo C99 originando o fragmento

AICD e o peptídeo beta-amilóide na porção extracelular da membrana (ZHANG e

cols., 2012). Dependendo do ponto exato desta última clivagem pela γ-secretase a

geração do peptídeo Aβ pode apresentar um número variável de aminoácidos

(entre 39 e 43) (GU e cols., 2004). Este processamento da APP é conhecido como

amiloidogênico devido à produção de monômeros de Aβ.

Por outro lado, a APP pode ser clivada inicialmente pelo complexo α-

secretase, gerando os fragmentos sAPPα e C83. A posterior clivagem destes

fragmentos não gera o peptídeo beta-amilóide, o que torna este processamento

não-amiloidogênico (Figura 2).

A quantidade relativa de Aβ1-42 formado é particularmente notável, devido

ao fato desta forma mais longa do Aβ ter uma altíssima capacidade de agregar em

comparação a forma mais abundantemente gerada, o peptídeo Aβ1-40 (BURDICK

e cols., 1992; JARRETT e cols., 1993). A produção do peptíeo Aβ é um processo

normal, mas em um pequeno número de indivíduos essa condição biológica

parece estar alterada. O Aβ produzido é liberado no meio extracelular, e

dependendo da concentração e condições físico-químicas tende a agregar

anormalmente. O tamanho destes agregados pode variar desde oligômeros

solúveis até fibras que se depositam nas placas amilóides (SNYDER e cols.,

1994). Embora seja pouco conhecido sobre a regulação da formação destes

agregados, sabe-se que estes podem se interconverter e que podem ser desfeitos

21

através da ação enzimática de duas proteases extracelulares: neprilisina e enzima

degradante de insulina (IDE, do inglês insulin-degrading enzyme) (MINERS e

cols., 2011).

Figura 2: Processamento da proteína precursora amilóide (APP). Processamento amiloidogênico e não amiloidogênico da APP na membrana plasmática. Na via não amiloidogênica, a clivagem da APP pela α-secretase libera um ectodomínio solúvel (SAPPα) para o meio extracelular e gera um fragmento residual de 83 aminoácidos C-terminal (C-83) na membrana. O C-83 é clivado pela γ-secretase liberando o peptídeo p3. Na via amiloidogênica a APP é clivada pela β-secretase, o que resulta na secreção da molécula levemente truncada sAPPβ e na retenção de um fragmento de 99 resíduos C-terminal (C-99). O C-99 é clivado pela γ-secretase gerando o peptídeo Aβ (destacado em verde). A clivagem de ambos os fragmentos, C83 e C99, libera o fragmento intracelular AICD. Figura produzida e gentilmente cedida por Marcelo Nunes Vieira.

22

1.3) Oligômeros de Aβ

A descoberta do Aβ como principal componente das placas amilóides

motivou o interesse pela potencial toxicidade dessa molécula. Em poucos anos,

um grande conjunto de evidências demonstrou que agregados deste peptídeo são

neurotóxicos, o que poderia explicar a neurodegeneração observada em estágios

avançados da doença de Alzheimer (GEULA e cols., 1998; WANG e cols., 2004;

KLEIN, 2006; FERREIRA e cols., 2007). Essas evidências são resultados de

estudos genéticos de casos de uma forma familiar (hereditária) da doença de

Alzheimer. Estes, ocorrem precocemente, em torno da quarta ou quinta decada de

vida, com uma incidência aproximada de 10% dos casos (SELKOE, 2004;

BERTRAM e TANZI, 2005). Os achados desses estudos genéticos apontam

fortemente para o gene da APP como um importante sítio de mutações

causadoras da doença.

O primeiro caso de mutação foi descoberto em uma família com amiloidose

cerebrovascular com múltiplas hemorragias (LEVY e cols., 1990). Nos anos

seguintes, uma distinta mutação na APP foi identificada em uma família que

desenvolvia precocemente a doença de Alzheimer (GOATE e cols., 1991), no

mesmo período foram detectadas mutações adicionais em outras famílias

(CHARTIER-HARLIN e cols., 1991; MULLAN e cols., 1992). Estes e outros

estudos levaram a proposta de que estas mutações na APP, ou nas enzimas

responsáveis pelo seu processamento proteolítico resultaria na geração

exacerbada de Aβ (BERTRAM e cols., 2010; SELKOE, 2000; DE e cols., 2010).

Entretanto, a maioria dos casos da DA é esporádica, não tendo sido ainda

identificado nenhum fator genético determinante desta condição (SELKOE, 2004;

23

ROBERSON e MUCKE, 2006). Nesses casos, a doença está bastante associada

ao envelhecimento. É importante ressaltar que o acúmulo de Aβ é o ponto comum

a todos os casos de DA, esporádicos e hereditários. No entanto, o que promove o

acúmulo de Aβ nos casos esporádicos da DA é ainda um grande mistério.

Pouco tempo depois da associação entre o peptídeo Aβ e a doença de

Alzheimer, ainda não havia uma proposta de mecanismo que explicasse como a

presença e agregação deste peptídeo poderiam desencadear o quadro de

demência observado. Neste sentido, os pesquisadores John Hardy e Gerald

Higgins publicaram, em 1992, um pequeno trabalho onde propunham, pela

primeira vez, uma sequência de eventos que serviu de paradigma para o

entendimento do processo que rege a DA (HARDY e HIGGINS, 1992).

Esta proposta, que ficou conhecida como hipótese da cascata amilóide,

postulava que o evento patogênico inicial seria um aumento nos níveis de Aβ no

cérebro do paciente, iniciando uma cascata de sinalização tóxica que leva à morte

neuronal, resultando em um estado de demência.

No entanto, apesar do sólido embasamento da hipótese amilóide proposta,

estudos seguintes forneceram evidências de que havia falhas neste mecanismo.

Por exemplo, a existência de deposição de placas amilóides em indivíduos não-

demenciados (MOCHIZUKI e cols., 1996) e a falta de correlação entre a

intensidade de deposição das placas e a severidade da disfunção cognitiva

(SLOANE e cols., 1997; RICHARDSON e cols., 2003) comprometeram as

predições desta proposta. Com as contribuições destes e de outros estudos que

sucederam a proposta inicial da cascata amilóide, uma nova versão foi elaborada

por Hardy e Selkoe (HARDY e SELKOE, 2002). Essa hipótese revisada apresenta

24

algumas modificações importantes, como incluir o impacto neurotóxico do

peptídeo beta amilóide, especialmente na sua forma oligomérica (Figura 3).

Desde a década de 90, muitos estudos têm focado seus esforços sobre a

forma oligomérica do peptídeo Aβ, em vista das crescentes evidências de que

estas formas são neurotóxicas, mesmo na ausência dos depósitos amilóides

(MUCKE e cols., 2000). Um estudo pioneiro demonstrou que os oligômeros de Aβ

(também conhecidos como ADDLs, do inglês Aβ-Derived Diffusible Ligands),

prejudicam mecanismos de plasticidade sináptica por inibir a potenciação de longa

duração (LTP, do inglês long-term potentiation) no hipocampo, evento fisiológico

crucial para a aquisição e formação de memória (LAMBERT e cols., 1998).

Adicionalmente, os oligômeros inibem a reversão da depressão de longa duração

(LTD, do inglês long-term depression) (WALSH e cols., 2002), que é um processo

inicial de depressão sináptica que culmina com a eliminação da espinha (HSIEH e

cols., 2006) e estão presentes no cérebro de pacientes com DA (GONG e cols.,

2003; FERREIRA e cols., 2007). De fato, estudos identificaram os oligômeros de

Aβ em concentrações significativamente mais altas em extratos de cérebros com

DA comparados com indivíduos controles não demenciados, e revelaram ainda,

que as características físico-químicas destes oligômeros endógenos são muito

similares a dos sintetizados in vitro (GONG e cols., 2003, DE FELICE e cols.,

2008). Esses dados reforçam e validam a relevância clínica do uso dos oligômeros

do peptídeo β-amilóide como modelo experimental e ainda como alvo terapêutico

para a Doença de Alzheimer.

Uma dramática redução do número de espinhas dendríticas é observada

após períodos prolongados de incubação com os oligômeros (Hsieh e cols., 2006;

25

LACOR e cols., 2007; De Felice e cols., 2009). Como as espinhas são protrusões

especializadas dos dendritos nas quais ocorrem as sinapses excitatórias, o

impacto dos oligomeros nas sinapses sugere a base para a perda da

conectividade neuronal e conseqüentemente para o déficit cognitivo observado

nos pacientes com a DA. Estes efeitos deletérios sobre as sinapses são

provavelmente resultado da capacidade dos oligômeros de se ligarem com grande

especificidade a sítios sinápticos excitatórios (LACOR e cols., 2004), evento este,

de particular relevância para o mecanismo da perda de memória.

Recentes estudos têm investigado os mecanismos utilizados pelos

oligômeros para danificar a plasticidade sináptica. Um destes, a transmissão

glutamatérgica, é considerada a base para os estímulos excitatórios associados

com aprendizado e memória (BENNETT, 2000). Os receptores de glutamato do

tipo AMPA e NMDA realizam um papel central na transmissão glutamatérgica e

são requeridos para indução de LTP e fortalecimento sináptico (DERKACH e cols.,

2007; LAU e ZUKIN, 2007). Trabalhos adicionais demonstraram uma maciça

redução dos níveis de receptores AMPA e NMDA da membrana neuronal

plasmática (Lacor e cols., 2007; Snyder e cols., 2005; Almeida e cols., 2005;

ROSELLI e cols., 2005; HSIEH e cols., 2006; GOTO e cols., 2006).

De forma significativa, a densidade pós-sináptica 95 (PSD-95, do inglês

post-density-95), uma proteína que forma um arcabouço fundamental para o

ancoramento e estabilização dos receptores AMPA e NMDA, também é reduzida

após tratamento com os oligômeros de Aβ (Roselli e cols., 2005). Dentre os

receptores acima citados, podemos ainda destacar que os receptores de insulina,

que também desempenham um papel crucial para a plasticidade neuronal,

26

encontram-se reduzidos de forma significativa na membrana plasmática após

exposição neuronal aos oligômeros (DE FELICE e cols., 2009; Zhao e cols.,

2008).

É importante ressaltar que, a perda patológica de espinhas e suas

moléculas associadas estão bem descrita para os casos de doença de Alzheimer

(SCHEIBEL e cols., 1975; SHIM e LUBEC, 2002; SCHEFF e PRICE, 2003) e

modelos de camundongos trangênicos para DA (LANZ e cols., 2003; CALON e

cols., 2004; MOOLMAN e cols., 2004; SPIRES e cols., 2005; JACOBSEN e cols.,

2006). Adicionalmente, outros trabalhos demonstraram que tanto pacientes com

DA (GONG e cols., 2003; Lacor e cols., 2004) quanto modelos de animais

transgênicos para DA (Chang e cols., 2003; Oddo e cols., 2006; Ohno e cols.,

2006) acumulam níveis substanciais de oligômeros de Aβ.

Eventos adicionais causados pelos oligômeros, que também interferem com

processos sinápticos importantes, incluem a fosforilação aberrante de tau em

resíduos que característicamente estão hiperfosforilados em cérebros de

pacientes com DA (De Felice e cols., 2008), além de estresse oxidativo e um

influxo neuronal excessivo de cálcio, possivelmente causado pela disregulação

das funções dos receptores NMDA (De Felice e cols., 2007). A reunião destes

achados sugere que as sinapses são o principal alvo dos oligômeros, e que esta

interação leva a mudanças morfológicas e na composição de receptores

resultando na disfunção sináptica, provavelmente responsável pelo declínio

cognitivo inicial que acomete indivíduos com DA (KLEIN, 2006; FERREIRA e

KLEIN, 2011).

27

Figura 3: Hipótese da cascata amilóde revisada. Esquema representativo da provável sequência de eventos que devem levar à doença de Alzheimer. As setas curvas indicam que os oligômeros de Aβ podem diretamente afetar as sinapses. Adaptado de (HARDY e SELKOE, 2002).

28

1.4) Sinalização por Insulina

O tecido muscular esquelético, o tecido adiposo, e o fígado são

classicamente descritos como órgãos sensíveis a insulina, e que desempenham

um papel central, através da sinalização insulinérgica, na regulação periférica do

metabolismo de carboidratos, lipídeos e proteínas. Por outro lado, desde a

primeira descrição da insulina (BANTING e cols., 1922), o cérebro já era

considerado como um órgão insensível à insulina. Entretanto, 30 anos mais tarde,

surgiram as primeiras evidências da ação da insulina sobre o cérebro (WOODS e

cols., 1979), e anos mais tarde, foi demonstrado que seus receptores encontram-

se amplamente distribuídos pelo encéfalo (ZHAO e cols., 2004). Descreveremos

aqui e nos subtópicos seguintes, a sinalização por insulina periférica e central,

respectivamente.

Um dilema interessante que precisa ser enfrentado por todos os animais é

o ciclo de alimentação, seguido pelo jejum que se dá entre as refeições. Os

mamíferos resolvem bem este problema através da estocagem de nutrientes, de

forma que esse estoque possa ser utilizado como fonte de energia durante os

períodos longos de jejum.

Um dos importantes agentes que regulam este processo é a insulina, um

hormônio peptídico de 5,8 kDa, majoritariamente produzido pelas células beta

pancreáticas e liberado na corrente sanguínea. Esta proteína desempenha uma

função bastante importante em regular o ciclo jejum-alimentação, cujo bom

funcionamento é essencial para o suprimento das demandas energéticas e para o

gerenciamento do metabolismo corporal.

29

O aumento da concentração plasmática de glicose, devido a ingestão e

metabolismo de alimentos, estimula a liberação de insulina, a qual induz a

captação de glicose por tecidos-alvo, promove a atividade de processos

anabólicos e inibe a quebra de macromoléculas energéticas no fígado e no tecido

adiposo. Conjuntamente, suas funções antagonizam a ação do glucagon, o qual

apresenta uma atividade máxima em situações de jejum. Neste sentido, os

organismos conseguem atender à demanda corporal por energia e estocá-la

quando em excesso (CARVALHEIRA e cols., 2002).

Complementando às suas funções metabólicas clássicas, a insulina, como

um hormônio anabólico, desempenha um importante papel durante o

desenvolvimento e crescimento, regulando a proliferação e sobrevivência celular.

Inclusive no organismo maduro a insulina é fundamental na modulação dos

processos de envelhecimento e expectativa de vida (TAGUCHI e cols., 2007).

1.4.1) Mecanismos Moleculares

O esclarecimento mais profundo dos mecanismos moleculares da

sinalização por insulina sempre foi de grande interesse de muitos pesquisadores,

diante da complexa ação apresentada por este hormônio (WHITE, 2003). A

cascata de sinalização intracelular da insulina começa com a sua ligação a um

receptor específico de membrana, que pertence a uma subfamília de receptores

tirosina quinase. O receptor de insulina (RI) é uma proteína heterotetramérica

constituído por dímeros intrínsecos ligados por pontes disulfeto, estes, são

compostos por domínios extracelulares (subunidades α) que controlam a atividade

30

tirosina quinase dos domínios transmembrana (subunidades β) (PATTI e KAHN,

1998; Ebina e cols., 1985; ULLRICH e cols., 1985; CARVALHEIRA e cols., 2002).

Após a ligação da insulina na subunidade α, ocorre um aumento da

atividade tirosina quinase das subunidades β, resultando na autofosforilação dos

resíduos de tirosina desta subunidade do receptor de insulina. A fosforilação de

tirosina realiza um papel importante em diversos processos fisiológicos modulando

a atividade de receptores e/ou enzimas presentes em passos iniciais de algumas

cascatas de sinalização, e ainda por coordenar o ancoramento de

multicomponentes de um complexo de sinalização próximo aos receptores

ativados (PAWSON, 1995). É sabido que grande parte dos sinais intracelulares da

sinalização de insulina é gerada através de complexos de sinalização formados

por substratos proteicos que se ancoram próximo aos resíduos de fosfotirosina

após a ativação do receptor de insulina (CARVALHEIRA e cols., 2002; TAGUCHI

e WHITE, 2008).

Atualmente, dez substratos intracelulares do RI tem sido identificados,

dentre os quais, quatro pertencem à família dos substratos do receptor de insulina

(IRS, do inglês insulin receptor substrate protein) (WHITE e YENUSH, 1998;

WHITE, 2006). Outros substratos incluem as proteínas adaptadoras Shc, Gab-1,

p60dok,Cbl, JAK2 e APS (BAUMANN e cols., 2000; PAWSON e SCOTT, 1997;

WHITE e YENUSH, 1998). Embora o papel de cada uma destas proteínas mereça

atenção, estudos sugerem que a maioria das respostas a insulina são amplamente

mediadas através de duas isoformas do substrato do receptor de insulina, IRS-1 e

IRS-2 (WHITE, 2003; TAGUCHI e WHITE, 2008).

31

As proteínas IRS não apresentam uma atividade catalítica intrínseca, sendo

compostas por múltiplos domínios de interação e sítios de fosforilação. No mínimo,

três isoformas de IRS estão presentes em humanos e camundongos, sendo estas

IRS-1 e IRS-2 que são amplamente expressos, e ainda IRS-4, o qual tem sua

expressão limitada ao timo, cérebro, rim e possivelmente a células β pancreáticas.

Roedores também expressam IRS-3, o qual é bastante restrito ao tecido adiposo e

demonstra atividades similares ao IRS-1 (UCHIDA e cols., 2000). Todas as

proteínas IRS apresentam domínios protéicos em sua estrutura denominados a

partir de sua homologia com proteínas já descritas. Dentre os domínios que

permitem o reconhecimento dos IRSs pelo RI e por outras proteínas da via,

podemos citar os domínios PH (do inglês pleckstrin homology), o PTB (phospho-

tyrosine binding) e o SH (Src homology, podendo ser denominados SH2 e SH3),

que são os mais bem descritos e de maior importância (LIZCANO e ALESSI,

2002;Yenush e cols., 1998).

As proteínas IRS são suscetíveis a diversas modificações pós-traducionais

que regulam seus estados de ativação. A fosforilação em resíduos de tirosina, por

exemplo, confere ao IRS a capacidade de ligação a outras proteínas adaptadoras

com domínios SH. Esta propriedade do IRS é fundamental para a amplificação

intracelular do sinal da insulina (ZHAO e ALKON, 2001).

Dentre as moléculas com domínio SH, o alvo mais conhecido do IRS é a

PI3K (do inglês, phosphatidylinositol-3 kinase). Esta enzima, quando ativada pelo

IRS, regula positivamente outros alvos subjacentes que coletivamente propagam

sinais de suma importância para a homeostase corporal, como por exemplo o

crescimento e proliferação celular, além de estimular a exposição dos

32

transportadores de glicose na superfície das células (LIZCANO e ALESSI, 2002;

SALTIEL e KAHN, 2001; WHITE, 2003; PIROLA e cols., 2004).

O balanço da fosforilação do IRS-1 é um dos principais fatores envolvidos

na modulação da amplitude e duração da sinalização por insulina nos órgãos

periféricos, como no fígado e no músculo esquelético. Entretanto, esta via de

sinalização pode sofrer intervenções diversas, como pelo cruzamento de vias

independentes, retroalimentação negativa ou até mesmo pelo balanço energético

celular (TANTI e JAGER, 2009). Portanto a sinalização efetora clássica mediada

pela insulina envolve a ativação de quinases bastante conhecidas, como por

exemplo, a PI3K e Akt/PKB, dentre outras que são apresentados com mais

detalhe abaixo na Figura 6.

33

Figura 4: Sinalização pela insulina e seus amplos efeitos celulares. Esquema

representativo dos mecanismos moleculares decorrentes da ligação da insulina ao

receptor e suas conseqüências celulares. Adaptado de MOSTOSLAVSKY, 2008.

34

1.4.2) Resistência à insulina

A insulina é um dos principais hormônios anabólicos em mamíferos e por

isso se torna essencial para a homeostase metabólica. A ligação da insulina ao

seu receptor dispara a fosforilação em resíduos de tirosina nos seus substratos, tal

como o IRS. Estas moléculas são cruciais para mediar os efeitos biológicos da

insulina (WELLEN e HOTAMISLIGIL, 2005; TANIGUCHI e cols., 2006). Condições

de estresse celular e inflamação podem inibir esse importante componente da via

de sinalização por insulina através de modificações pós-traducionais, tais como

fosforilação em resíduos de serina no IRS, as quais são reguladas por vias

intracelulares regulatórias (ZICK, 2005). Esta inibição é observada na maioria dos

casos em indivíduos obesos que sofrem de resistência a insulina sistêmica e

Diabetes tipo 2 (HOTAMISLIGIL, 2006).

A disfunção na sinalização por insulina resultante da fosforilação em

resíduos de serina pode ser iniciada por elevados níveis de citocinas pró-

inflamatórias, tais como TNFα e IL-1β dentre outros, indicando que mediadores

inflamatórios podem ter um papel regulatório crucial na homeostase sistêmica de

glicose (HOTAMISLIGIL e cols., 1996; HORNG e HOTAMISLIGIL, 2011). A

elucidação dos mecanismos que levam a esta inibição da sinalização por insulina

vem sendo uma das principais questões em muitos laboratórios nos últimos anos.

Estudos prévios estabelecem que o IRS-1 pode ser fosforilado em resíduos

de serina por diferentes quinases resultando na inibição desta proteína em

transmitir o sinal iniciando pela insulina em seu receptor (TANIGUCHI e cols.,

2006).

35

Dentre as principais quinases envolvidas nesta fosforilação do IRS-1, estão

a JNK (do inglês, c-Jun n-terminal kinase), a IKK (IκB kinase), a mTOR (doi inglês,

mammalian target of rapamycin) e a PKR (do inglês, dsRNA-dependent protein

kinase). A hiperativação destas porteínas é um passo crucial para levar a

resistência à insulina em resposta a diversos estímulos de estresse celular

(HIROSUMI e cols., 2002; ARKAN e cols., 2005; NAKAMURA e cols., 2010).

E embora diversas vias inflamatórias possam contribuir para uma

desregulação metabólica em diferentes níveis, a modulação da via de sinalização

por insulina talvez seja a mais crucial, por ser esta uma via metabólica altamente

dominante e conservada no controle da homeostase energética e de nutrientes.

A identificação de um link entre inflamação e a sinalização por insulina vem

apresentando uma sólida plataforma para explorar outros pontos ainda não

elucidados entre as respostas imunes e o controle metabólico (HOTAMISLIGIL e

cols., 1993; XU e cols., 2003). Adicionalmente as citocinas, outros fatores, como

por exemplo, o excesso de lipídeos circulantes, ativam sinalizações inflamatórias

que diretamente inibem a sinalização do receptor de insulina. Em paralelo, vias

pró-inflamatórias também são induzidas por estresse de organelas devido à

sobrecarga de nutrientes, resultando em um processamento anormal que culmina

com o estresse metabólico celular (HOTAMISLIGIL, 2006).

É importante destacar que esse quadro patológico é um fator de risco

chave para o desenvolvimento de diabetes tipo 2, podendo ainda ser agravado

quando associado a outros fenômenos da síndrome metabólica, como obesidade,

hipertensão e hipercolesterolemia (HOTAMISLIGIL e cols., 1996; SAVAGE e

SEMPLE, 2010; FU e cols., 2012; CALAY e HOTAMISLIGIL, 2013).

36

1.4.3) Sinalização por Insulina no Sistema Nervoso Central

Antes da década de 80, pensava-se que o Sistema Nervoso Central (SNC)

seria insensível à estimulação por insulina e que a captaçãode glicose no cérebro

ocorreria de forma independente da ação da insulina. Após as primeiras

descrições da presença da insulina e de seus receptores em diversas regiões do

cérebro, começou-se a especular qual seria a sua origem e papel neste órgão

(HAVRANKOVA e cols., 1983;HAVRANKOVA e cols., 1978;PLUM e cols., 2005).

Estudos posteriores demosntraram que a insulina periférica poderia cruzar

a barreira hematoencefálica e atingir o parênquima cerebral. Este achado pareceu

esclarecer a origem da insulina cerebral, embora trabalhos posteriores tenham

detectado a expressão local de insulina no SNC (MADADI e cols., 2008; VAN DER

HEIDE e cols., 2006). Apesar dessas evidências, a questão da produção de

insulina local no cérebro ainda permanece controversa (ZHAO e ALKON, 2001;

MCNAY e RECKNAGEL, 2011).

A existência clara dos receptores de insulina e seus efetores subjacentes

em diversas estruturas cerebrais, inclusive uma notável presença nas espinhas

dendríticas, levantou sugestões de que esse hormônio apresenta uma importante

função neuronal. Fortes evidências indicam um potencial neurotrófico e

neuroprotetor contra insultos tóxicos (BRUNING e cols., 2000; DI e cols., 2010).

Estudos recentes demonstram que a insulina pode induzir respostas

eletrofisiológicas imediatas (KOVACS e HAJNAL, 2009) e proteção contra morte

neuronal por apoptose no córtex e cerebelo (TANAKA e cols., 1995; RYU e cols.,

1999). De forma interessante, a insulina também pode estimular a plasticidade

37

sináptica, promover sinaptogênese e ainda recupera déficits de LTP induzidos por

Aβ (VAN DER HEIDE e cols., 2005; CHIU e cols., 2008; LEE e cols., 2009).

Portanto, a sinalização por insulina parace ser essencial para os processos

de formação e consolidação de memória (ZHAO e cols., 2004; MCNAY e cols.,

2010; MCNAY e RECKNAGEL, 2011). Estudos recentes demonstram a sua

capacidade de potencializar a função cognitiva em roedores e humanos saudáveis

(DOU e cols., 2005; HAJ-ALI e cols., 2009; OTT e cols., 2012).

Adicionalmente, a sinalização por insulina desenvolve um importante papel

durante o desenvolvimento do sistema nervoso, contribuindo nos processos de

expressão gênica tecidual, proliferação de progenitores, neurogênese e

organogênese (VAN DER HEIDE e cols., 2006; SOUSA-NUNES e cols., 2011;

FERNANDEZ e TORRES-ALEMAN, 2012). Neste sentido, observamos que a

insulina se apresenta como um importante neuromodulador da função cerebral.

1.5) Doença de Alzheimer e resistência à insulina

A idéia de que uma disfunção na sinalização por insulina contribui para a

patogênese da doença de Alzheimer, foi inicialmente proposta há mais de 20

anos, por Hoyer e colaboradores (HOYER e NITSCH, 1989). Poucos anos depois,

um estudo demonstrou que pacientes com diabetes tipo 2 apresentavam uma

chance duas vezes maior de desenvolver a doença de Alzheimer (OTT e cols.,

1996). Estudos clínicos e epidemiológicos posteriores reforçaram essa associação

demontrando que a longa duração do diabetes tipo 2 se correlacionam com

alterações cerebrais significativas e déficits cognitivos nos pacientes (CRAFT e

38

cols., 1998; BIESSELS e cols., 2002; CRAFT e WATSON, 2004; SCHRIJVERS e

cols., 2010).

A associação entre DA e diabetes também foi reforçada em modelos

experimentais de resistência a insulina, os quais demonstram alterações

bioquímicas e comportamentais relacionadas ao déficit cognitivo (WANG e cols.,

2010). De acordo com esta evidência, outro estudo demonstrou que animais

transgênicos para a DA, quando expostos à indução de diabetes experimental,

apresentam um agravamento do quadro cognitivo (TAKEDA e cols., 2010).

Somente recentemente, as bases moleculares para esta correlação entre o

diabetes e a doença de Alzheimer começaram a ser estabelecidas. Um achado

interessante determinou que os níveis de RNAm para insulina, IGF 1/2 e para o

receptor de insulina estão reduzidos no cérebro de pacientes da DA (STEEN e

cols., 2005). Ao mesmo tempo observou–se que a atividade da PI3K e da Akt

estão reduzidas nestes pacientes, sugerindo ums disfunção na sinalização por

insulina cerebral.

As primeiras evidências de como o cérebro de pacientes da DA apresenta

uma disfunção insulinérgica, surgem de dois estudos cruciais, demonstrando que

oligômeros de Aβ se ligam a neurônios hipocampais e disparam a remoção dos

receptores de insulina das membranas plasmáticas das células (Figura 5) (ZHAO

e cols., 2008; DE FELICE e cols., 2009). Pouco depois, foi observado este mesmo

evento no cérebro de pacientes da DA (MOLONEY e cols., 2010). Os neurônios

atacados pelos oligômeros apresentam elevados níveis de receptores de insulina

no corpo celular, sugerindo uma redistribuição subcelular dos receptores (Figura

5). O resultado é uma redução da resposta à insulina, como revelado pela

39

reduzida atividade tirosina quinase do RI em neurônios hipocampais em cultura,

expostos aos oligômeros de Aβ (ZHAO e cols., 2008). De forma interessante,

outro grupo mostrou que os oligômeros de Aβ induzem a disfunção dos RIs,

sugerindo que esta disfunção é o fator principal no mecanismo que leva a inibição

da LTP (TOWNSEND e cols., 2007).

A partir desses estudos, surgiram questionamentos sobre os mecanismos

através do qual as espécies oligoméricas de Aβ pertubariam a sinalização por

insulina. Sabe-se que no diabetes, existe uma conexão bem estabelecida entre a

participação de mecanismos inflamatórios, a ativação de quinases de estresse

celular e o quadro de resistência à insulina (HOTAMISLIGIL, 2006; VALLERIE e

HOTAMISLIGIL, 2010).

Trabalhos recentes apontam para a ativação de quinases de estresse

celular, como por exemplo, a JNK, em cérebros de pacientes da DA, assim como

em modelos experimentais expostos aos oligômeros de Aβ (MA e cols., 2009). No

diabetes tipo 2, essas quinases podem ser estimuladas em resposta à ligação de

citocinas pró-inflamatórias, como por exemplo, o TNF-α (HOTAMISLIGIL e cols.,

1996; HIROSUMI e cols., 2002).

Deste modo, vem sendo fortalecida a idéia de que a doença de Alzheimer

seria um terceiro tipo de diabetes, cujo evento fisiopatológico central é o

desenvolvimento de uma resistência à insulina especificamente no cérebro (DE LA

MONTE e WANDS, 2008). Contudo, tanto a doença de Alzheimer como o diabetes

são causas comuns de morbidade e mortalidade, ressaltando assim a urgente

necessidade de desvendar como a resistência à insulina se desenvolve no cérebro

de pacientes com Alzheimer. O desafio atual é investigar em detalhes os

40

mecanismos moleculares responsáveis por induzir a disfunção da sinalização

insulinérgica cerebral, objetivando o desenvolvimento de uma terapia eficiente

para a DA.

Figura 5: Oligômeros de Aβ removem os receptors de insulin da superfície da membrana neuronal. Uma composição criada pela sobreposição de imagens de imunofluorescencia de um neurônio controle (imagem da esquerda) e um neurônio exposto aos oligômeros de Aβ (AβO) imagem da direita. Imagem da esquerda: Um neurônio saudável na ausência de AβO (nenhuma marcação AβO- positiva em vermelho observada) apresenta uma abundante marcação dendrítica de RI (pontos verdes). Um esquema de um segmento dendrítico é representado no círculo a esquerda. Níveis fisiológicos de Aβ são produzidos e não ocorre acúmulo dos mesmos. A presence do RI na membrane permite a função da sinalização por insulina. Imagem da direita: AβO se ligam aos neurônios (pontos vermelhos) causando a remoção dos receptores de insulina da superfície da membrana (pontos verdes) (ZHAO e cols., 2008; DE FELICE e cols., 2009; FERNANDEZ e TORRES-ALEMAN, 2012). Um esquema de um segmento de um dendrito está representado na direita: AβO se acumulam pelo elevado nível de Aβ gerado pela clivagem da APP. Adaptado de DE FELICE, 2013.

41

1.6) Doença de Alzheimer e Inflamação

Inflamação é parte de um mecanismo de defesa do corpo contra multiplas

ameaças, incluindo infecções e injúrias. Trata-se de uma complexa rede de sinais

envolvendo fatores solúveis e células especializadas que são recurtados a fim de

neutralizar as ameaças e restaurar as codições fisiológias do organismo (BROWN

e cols., 2007). No sistema nervoso central, o processo inflamatório ocorre de

forma muito similar aos órgãos periféricos. No cérebro, as células da glia,

especialmente os astrócitos e as micróglias, são ativados em condições pró-

inflamatórias, elevando a produção de citocinas inflamatórias. Tanto no cérebro,

como no sistema periférico, o quadro de inflamação crônica torna-se deletério, e

pode contribuir para o desenvolvimento de diversas doenças, incluindo as

neurodegeneratvas.

A inflamação possui um papel crucial na patogênese da doença de

Alzheimer. Diversos estudos estabeleceram a presença de características

inflamatórios no cérebro de pacientes da DA, incluindo níveis elevados de

citocinas e marcadoeres de gliose nas regiões atingidas pela doença (PERRY e

cols., 2010; CZIRR e WYSS-CORAY, 2012; AGUZZI e cols., 2013). Um estudo

recente revelou que análises plasmáticas de pacientes com DA apresentam níveis

elevados de mediadores inflamatórios, incluindo fator de nercose tumoral – α

(TNF-α, do inglês tumor necrosis factor-alpha), e interleucina-1β (IL-1β do inglês,

Interleukin-1β) (SWARDFAGER e cols., 2010).

No cérebro, TNF-α é secretado principalmente pelas microglias, em

resposta a algum tipo de trauma, infecção ou estresse celular resultante de

42

acúmulo de agregados protéicos (PARK e BOWERS, 2010). Níveis elevados

desta citocina foram identificados no líquor e no cérebro de pacientes DA

(GRAMMAS e OVASE, 2001; TOBINICK, 2007), assim como no cérebro de

modelos transgênicos para a doença (JIN e cols., 2008; RUAN e cols., 2009).

Adicionalmente, pacientes com transtorno cognitivo leve apresentavam aumento

significativo de TNF-α, indicando a participação desta citocina em eventos iniciais

da patogênese da DA (TARKOWSKI e cols., 2003).

Um dos eventos neurotóxicos induzidos pelos oligômeros de Aβ é a

exacerbada ativação de microglia com elevação dos níveis de mediadores

proinflamatórios como citocinas, como o TNF-α (MCGEER e cols., 2006).

Entretanto, o preciso mecanismo pelo qual o Aβ inicia a resposta inflamatória

mediada pela microglia, ainda é desconhecido (LUE e cols., 2001).

Sabe-se que os elevados níveis de TNF-α são considerados como ponto

central na patogênese do diabetes tipo 2, induzindo um quadro de resistência a

insulina (HOTAMISLIGIL e cols., 1995; HOTAMISLIGIL e cols., 1996). Em

conjunto com essas informações, a associação entre o diabetes tipo 2 e a doença

de Alzheimer têm levantado aspectos ainda não compreendidos sobre essa

conexão. Portanto, seria razoável pensar na correlação entre o impacto tóxico e

inflamatório induzido pelos oligômeros de Aβ, mediado por citocinas como o TNF-

α, sobre a sinalização por insulina nos neurônios. Como de fato, sabe-se que

oligômeros de Aβ induzem a disfunção da sinalização por insulina em neurônios

hipocampais (ZHAO e cols., 2008; DE FELICE e cols., 2009), sinalização que é

fundamental nos processos de aprendizado e memória (DOU e cols., 2005).

43

Possivelmente esta hipótese poderia explicar o efeito deletério destas

neurotoxinas sobre a memória em pacientes com a doença de Alzheimer.

1.7) Transporte axonal

O transporte intracelular é fundamental para todas as células de mamíferos,

especialmente para os neurônios. Um típico neurônio tem uma estrutura altamente

polarizada, com corpo celular diversos dendritos curtos e afilados além de um

longo e fino axônio. A maioria das proteínas que são necessárias ao axônio e ao

terminal sináptico são sintetizadas no corpo celular e transportadas ao longo do

axônio em organelas membranosas ou complexos protéicos (GRAFSTEIN e

FORMAN, 1980). Embora as proteínas dendríticas também sejam transportadas

do corpo celular, diversos RNAm específicos, são transportados até os dendritos

para suportar a síntese protéica local (JOB e EBERWINE, 2001). Além destes,

outros componentes celulares são transportados pelos axônios, sendo estes,

endossomos, mitocôndrias, vesículas sinápticas e vesículas secretoras densas

(DCV´s, do inglês Dense Core Vesicles). Estas são formadas no corpo celular, e

percorrem grandes distancias através de axônios e dendritos, até os sítios pré e

pós sinápticos, onde finalmente liberam seu conteúdo de proteínas e

neuropeptídeos (Wu e cols., 2004). A distância percorrida pelas vesículas são

mais longas, em comparação com outros tipos celulares, pois o axônio de um

neurônio motor humano, pode medir cerca de 1 metro de comprimento (STOKIN e

44

GOLDSTEIN, 2006). O mecanismo de liberação das vesículas em seus sítios de

ação é baseado no transporte de longo alcance sobre os microtúbulos.

Em ambos, dendritos e axônios, os microtúbulos são orientados

longitudinalmente (HIROKAWA, 1982), e servem como trilhos ao longo dos quais,

organelas e macromoléculas podem ser transportadas (HIROKAWA, 1998) (Fig.

5). Os microtúbulos são estruturas protéicas que fazem parte do citoesqueleto das

células, são filamentos longos, como cilindros ocos com aproximadamente 25nm.

São constituídos de um polímero de α e β-tubilina que é dinâmico e instável

(DESAI e MITCHISON, 1997). Esta dinâmica é controlada por proteínas

associadas à microtúbulos (MAP, do inglês microtubules-associated proteins) in

vivo e in vitro (STOKIN e GOLDSTEIN, 2006). Evidências com relação aos

mecanismos que regulam a direção do transporte surgem da determinação da

polaridade dos microtúbulos, altamente organizadas nos axônios. A polaridade

dos microtúbulos direciona o transporte da região proximal (próximo ao corpo

celular) para a região distal (próximo a terminação pré-sináptica) dos axônios

(STOKIN e GOLDSTEIN, 2006). Essas regiões de diferentes polaridades são

reconhecidas por proteínas motoras, como as kinesinas e dineínas, as quais

transformam energia química em movimento mecânico (Lawrence e cols., 2004).

Enquanto o transporte axonal anterógrado regulado por diversas proteínas

motoras da família das kinesinas, estudos sugerem que as dineínas sejam as

principais proteínas motoras envolvidas no transporte axonal retrógrado (STOKIN

e GOLDSTEIN, 2006). A dineína é composta por duas cadeias pesadas, duas

cadeias leves e duas cadeias intermediárias leves. Acredita-se que o transporte

45

mediado por dineína seja regulado pela sua interação com o complexo dinactina, o

qual consiste de diversas proteínas (SCHROER, 2004).

A regulação da atividade das kinesinas e das dineínas, assim como a

regulação do transporte axonal, é pouco compreentendida até os dias de hoje. Em

principio, a regulação pode ocorrer em um dos diversos passos, incluindo o

reconhecimento do cargo e a ligação deste a proteína motora, a velocidade do

transporte, e ainda, o reconhecimento do destino correto pelo complexo proteína

motora-cargo. Dados demonstram que a kinesina-1 pode ser regulada diretamente

pela ligação do cargo (FRIEDMAN e VALE, 1999).

Dentre muitos cargos e parceiros de ligação identificados para as proteínas

motoras anterógradas e retrógradas, alguns parecem estar envolvidos com um

mecanismo de regulação. Estes parceiros incluem a proteína precurssora amilóide

(APP) (KAMAL e cols., 2000), a proteína c-Jun N-terminal kinase (JNK) e as

proteínas que com ela interagem (JIP1, JIP2, JIP3/Sunday driver) (CAVALLI e

cols., 2005). Evidências adicionais sugerem um importante papel da fosforilação

na regulação de proteínas motoras, envolvendo as quinases glicogênio sintase

cinase 3β (GSK3 β) e a quinase dependente de ciclina 5 (CDK 5).

46

Figura 6: Transporte axonal e dendrítico. a. Um típico neurônio, projetando diversos dendritos (esquerda) e um único axônio do corpo celular. b. Proteínas da família das quinesinas transportando vesículas contendo APP, apolipoproteína E, mitocôndrias e vesículas sinápticas, no sentido anterógrado c. Receptores AMPA e NMDA, grânulos de RNAm transportados por quinesinas, no sentido retrógrado, nos dendritos. Adaptado de HIROKAWA e TAKEMURA, 2005.

47

1.7.1) Transporte axonal, envelhecimento e Doença de Alzheimer

Estudos sugerem que no envelhecimento podem ocorrer problemas no

transporte axonal, já que existe uma redução dos trilhos de microtúbulos (CASH e

cols., 2003), assim como mudanças na distribuição de proteínas associadas à

microtúbulos, como a proteína tau e neurofilamentos (NIEWIADOMSKA e

BAKSALERSKA-PAZERA, 2003; UCHIDA e cols., 2004). Outro trabalho observou

um aparente acúmulo de proteínas ao longo do axônio, como a APP

(KAWARABAYASHI e cols., 1993). O entendimento dessas mudanças

relacionadas com a idade na estrutura e função dos axônios permanece não

esclarecido, pois em particular não se sabe se todas as proteínas de transporte

são afetadas, ou se somente algumas vias de sinalização são prejudicadas no

envelhecimento.

Existe uma considerável quantidade de dados consistentes com a hipótese

de que a deficiência do transporte axonal desempenha um papel central na

patogênese de algumas doenças neurodegenerativas, incluindo Huntington

(TRUSHINA e cols., 2004), ALS (LAMONTE e cols., 2002) e da DA (TERRY,

1996). Uma evidência da ampla patologia axonal é a presença de anormalidades

como o intenso acúmulo de diversas moléculas ao longo do trajeto do axônio,

como a APP (CRAS e cols., 1991) e seus metabólitos, neurotransmissores e

proteínas relacionadas, neurofilamentos, tau e glicogênio, assim como uma

redução do trilhos de microtúbulos (CASH e cols., 2003). Dados sugerem que os

defeitos axonais podem coincidir com estágios iniciais da DA (STOKIN e cols.,

2005).

48

Camundongos que possuem uma expressão reduzida de kinesina-1

desenvolveram defeitos axonais, assim como, o aumento excessivo da produção

de Aβ e da deposição de placas (Stokin e cols., 2005). Evidências vêm

demonstrando que o Aβ pode induzir anormalidades axonais (PIKE e cols., 1992),

contribuindo diretamente para a deficiência no transporte axonal (HIRUMA e cols.,

2003). A hiperfosforilação de tau, um macardor da doença de Alzheimer, pode

também afetar diretamente o transporte axonal de APP e de outras moléculas

(STAMER e cols., 2002). Em adição a estes dados, drogas que estabilizam os

microtúbulos foram capazes de prevenir os déficits de transporte observados em

camundongos transgênicos para tau (TROJANOWSKI e cols., 2005). De forma

bastante interessante, o cérebro de indivíduos com DA apresentaram uma

significante redução do transporte axonal (DAI e cols., 2002).

Dados recentes muito interessantes do nosso grupo demonstram que o

tratamento de neurônios hipocampais com oligômeros do peptídeo Aβ é capaz de

reduzir de forma significativa o transporte axonal, demonstrando um novo impacto

dos oligômeros nos neurônios (DECKER e cols., 2010).

49

1.8) Uso da exendina-4 na estimilação da sinalização por insulina cerebral

A busca por novos fármacos para a diabetes específica do cérebro tem

convergido para medicamentos que estimulem a via de sinalização da insulina,

mas cuja ação independa da atividade do receptor de insulina e de seus

substratos imediatos. Os progressos mais recentes apontam as incretinas como

abordagem mais promissora para reparação dos danos neuronais na DA.

As incretinas são hormônios sintetizados primariamente no intestino e

liberados na corrente sanguínea (BARRERA e cols., 2011). Suas funções

consistem em sinalizar o aporte de nutrientes após as refeições e estimular a

liberação de insulina pelo pâncreas. As incretinas mais conhecidas são o peptídeo

insulinotrópico dependente de glicose (GIP, do inglês glucose-dependent

insulinotropic polypeptide), primeiro a ser descrito, e o peptídeo semelhante ao

glucagon (GLP-1, do inglês glucagon-like polypeptide 1) (VILSBOLL e cols., 2003).

O GLP-1 é subproduto do glucagon derivado após clivagem proteolítica

pelas enzimas prohormônio convertases 1 e 3 (PC1/3). Sua forma biologicamente

ativa tem em torno de 30 aminoácidos e tem uma curta meia-vida sérica, sendo

rapidamente metabolizada pela enzima dipeptidil peptidase (DPP) (KAZAFEOS,

2011).

Este peptídeo apresenta um potencial terapêutico enorme para diabetes, já

que reduz rapidamente a glicemia pós-prandial e aumenta a insulinemia

(VILSBOLL e cols., 2003). Contudo, a reduzida meia-vida inviabiliza o uso clínico

da sua forma nativa. Nos últimos anos, entretanto, agonistas modificados com

50

duração ampliada foram encontrados e tem sido alvos de diversos estudos (RYAN

e HARDY, 2011).

O primeiro deles é o exendin-4, que foi isolado da saliva de um lagarto e

não possui o sitio de clivagem da DPP, permanecendo por mais tempo na

circulação (GARBER, 2011). Mais recentemente, modificações orgânicas foram

implementadas à molécula do GLP-1, o que gerou agonistas com meia vida

prolongada (Val-8-GLP1 e liraglutídeo) já aprovados para uso clinico em diabetes

(MORAN e DAILEY, 2009).

O receptor de GLP-1 e de seus agonistas é uma proteína transmembranar

associada à proteína G estimulatória que, por sua vez, ativa a adenilato ciclase

(DONNELLY, 2012). A produção aumentada de AMP cíclico é o principal

transdutor e amplificador do sinal de GLP-1 e induz a ativação da via clássica de

PKA/CREB tanto em células pancreáticas quanto em neurônios. Por outro lado, a

sinalização de GLP-1 é conhecida por ativar a enzima PI3K de uma forma

independente de insulina, o que garante a ocorrência dos efeitos insulinotrópicos

(HOLSCHER e LI, 2010). De fato, a exendina-4 apresenta propriedades protetoras

contra o estresse de reticulo endoplasmático e a morte celular em células β

expostas a insultos tóxicos (YUSTA e cols., 2006; KIM e cols., 2012).

No sistema nervoso central, sabe-se que algumas populações de neurônios

produzem GLP-1, mas o seu receptor é expresso em diversas regiões do cérebro,

inclusive em neurônios hipocampais e corticais (HAMILTON e HOLSCHER, 2009).

Dados recentes sugerem que a sinalização por GLP-1 é importante para a

plasticidade sináptica e formação de memórias (DURING e cols., 2003; MATTSON

e cols., 2003; ABBAS e cols., 2009). A exendina-4 foi capaz de prevenir contra

51

insultos oxidativos e metabólicos em modelos celulares e animais de isquemia e

de doença de Parkinson (LI e cols., 2009).

Além disso, estes agonistas de GLP-1r reverteram a inibição de LTP

induzida por oligômeros de Aβ (GAULT e HOLSCHER, 2008). Uma evidência

adicional do papel central do GLP-1r na plasticidade sináptica é suportada por

dados interessantes que demonstram o déficit cognitivo apresentado por animais

que não expressam receptor de GLP-1 (ABBAS e cols., 2009). Adicionalmente, a

estimulação por GLP-1 parece induzir neurogênese em animais adultos

(HAMILTON e cols., 2011).

Este conjunto de dados recentes que apontam para os efeitos benéficos da

estimulação por GLP-1 no cérebro e nos tecidos periféricos abre caminhos para

aplicações farmacológicas destes conceitos na terapia contra a doença de

Alzheimer e motiva a realização de mais análises para elucidar o cenário completo

de atuação desta sinalização.

52

2) OBJETIVOS

- OBJETIVO GERAL

Investigar a disfunção na sinalização por insulina cerebral induzida pelos

oligômeros de Aβ e sua possível correlação com mecanismos inflamatórios

envolvidos na resistência à insulina periférica.

-OBJETIVOS ESPECÍFICOS

- Investigar se os oligômeros de Aβ induzem aumento de marcadores de

resistência à insulina em neurônios hipocampais;

- Enteder os mecanismos moleculares que propiciam esta condição de disfunção

inulinérgica cerebral;

- Analisar a possível relação entre a inibição da sinalização por insulina no

cérebro induzida pelos oligômeros com mecanismos já estabelecidos associados

ao diabetes;

- Compreender o papel da sinalização de insulina na manutenção da uma função

fisiológica importante da célula neuronal, o transporte axonal, diante do efeito

tóxico dos oligômeros.

- Investigar o possível papel de novas terapias anti-diabetes na prevenção ou

ainda reversão destes fenômenos.

53

3 METODOLOGIA

3.1) Materiais

Peptídeo Aβ1–42, Exendin-4 e o exendin 9-39 foram adquiridos da Bachem

Inc. (Torrance, CA). Peptídeo sintético com a ordem aleatória Aβ1–42 foi

comprado da Anaspec (San Jose, CA). Insulina Bovina e humana, 1,1,1,3,3,3,-

hexafluoro-2-propanol (HFIP), DMSO, poly-L-lysina foram comprados da Sigma

(St. Louis, MO). Reagentes/Meio de Cultura Neurobasal, Anticorpos secundários

Alexa Flúor e o reagente ProLong anti-fade foram comprados da Invitrogen

(Carlsbad, CA). Tampões de Eletroforese foram comprados da BioRad (Hercules,

California). Reagente de quimioluminescencia SuperSignal e kit de dosagemde

proteína BCA foram obtidos da Pierce (Deerfield, Illinois). Anticorpos contra

Substrato do Receptor de Insulina 1 (IRS-1) total, fosforilado na Tyr465, Ser-636,

307, 616, 312 foram obtidos da Santa Cruz Biotechnology (Santa Cruz, CA). Os

peptídeos Exendin 4 and exendin 9-39 foram obtidos na Bachem (Torrance, CA).

O Inibidor da JNK, SP600125, foi obtido da Tocris Bioscience (Ellisville, MO).

Plasmid pβ-actin-BDNF-mRFP foi doado pelo Dr. G. Banker (OHSU, Oregon,

USA). Marcador mitochondrial YFP foi doado pelo Dr. G. Rintoul (SFU, BC,

Canadá).

54

3.2) Preparação e caracterização dos Oligômeros de Aβ

Aβ1-42 (Bachem Inc., Torrance, CA) foi solubilizado a 1 mM em 1,1,1,3,3,3-

hexafluoro-2-propanol (HFIP; Merck) gelado e a solução límpida e incolor

resultante foi incubada à temperatura ambiente por 60 minutos. A solução foi

então colocada em gelo por 10 minutos e aliquotada, em fluxo laminar, em

microtubos. Os microtubos foram deixados abertos na capela de fluxo laminar

durante 12 horas para evaporação do HFIP. A eliminação completa do HFIP foi

feita por centrifugação em SpeedVac® por 10 minutos. As alíquotas contendo os

filmes de Aβ foram estocadas a -20 ºC para utilização posterior.

As preparações dos oligômeros de Aβ foram feitas, a partir destes

estoques, segundo protocolo previamente descrito (LAMBERT e cols., 1998). A

cada preparação dos oligômeros, uma alíquota de Aβ foi ressuspensa em

dimetilsulfóxido anidro (DMSO; Sigma, St. Louis, MO) de forma a obter uma

solução a 5 mM. Esta solução foi diluída em PBS estéril a 100 μM e incubada a 4

ºC por 24 horas. Uma amostra de mesmo volume de DMSO 2% em PBS foi

preparada e incubada a 4ºC por 24 horas, e usada experimentalmente como

controle (veículo). Após a incubação, a preparação foi centrifugada a 14.000 g por

10 minutos a 4ºC para remoção de agregados insolúveis de Aβ. O sobrenadante

da centrifugação, contendo os oligômeros, foi mantido a 4ºC até a sua utilização

em um período máximo de até 48 horas após o preparo. Para determinar a

concentração dos oligômeros nas preparações, utilizou-se o método de BCA

(Pierce, Rockford, IL).

55

Como são metaestáveis, as preparações dos oligômeros, realizadas

semanalmente no laboratório, são rotineiramente analisadas por Western blot e

cromatografia de gelfiltração para caracterizar as espécies oligoméricas presentes.

3.2.1) Gel-filtração por HPLC

As análises foram feitas com uma coluna de sílica SynChropak® GPC 100

com as seguintes características: dimensão da coluna: 250 x 4,6 mm; tamanho do

poro: 100 Å; limite de exclusão para proteínas: 3.000-300.000 kDa. A fase móvel

usada foi PBS pH 7, filtrado através de membrana de nitrocelulose Millipore

(Billerica, MA) 0,45 μm, mantido em gelo durante toda a análise. As análises foram

feitas através de cromatografia líquida de alto desempenho (high performance

liquid chromatography, HPLC), com detecção simul-tânea de absorção a 280 nm

e fluorescência com excitação a 275 nm e emissão a 305 nm. Antes da injeção da

amostra dos oligômeros, a coluna foi lavada durante 1 hora com água Milli-Q® e

equilibrada por 1 hora com a fase móvel, ambas com fluxo de 0,5mL/min.

Inicialmente, 50 μL de veículo (DMSO 2% em PBS) foram injetados, com fluxo de

0,5mL/ min e tempo de corrida de 15 minutos. Em seguida, a coluna foi re-

equilibrada com a fase móvel durante 15 minutos e 50 μL dos oligômeros foram

injetados, e a análise feita com os mesmos parâmetros usados para o veículo. Os

dados das cromatografias foram transferidos para o programa Microsoft® Excel

2002, onde os cromatogramas foram preparados para rela- tórios semanais de

caracterização das preparações.

56

Figura 7: Cromatograma representativo de gel-filtração dos Oligômeros (ADDLs). Através da gel-filtração é possível detectar duas populações de agregados. A primeira, formada por oligômeros de alto peso molecular (50-100 kDa), elui entre 3,5 – 4 minutos. Agregados de 2-3 monômeros eluem mais tarde, entre 7,5 – 8 minutos.

3.2.2) Western blot para Oligômeros de Aβ (ADDLs)

Amostras das preparações dos oligômeros foram adicionadas a tampão de

amostra e resolvidas por eletroforese em gel de gradiente de 10–20% de

acrilamida (Invitrogen) com tampão de corrida Tris/Tricina/SDS, a 120 V por 60

minutos à temperatura ambiente. O material no gel (20 pmol de Aβ/canaleta) foi

eletrotransferido para membranas de nitrocelulose Hybond ECL (Amersham

Biosciences, Piscataway, NJ) a 100 V por 1 hora a 4 °C, usando tampão contendo

25 mM Tris, 192 mM glicina, 20% (v/v) metanol, 0,02% SDS, pH 8,3. As

57

membranas foram bloqueadas com BSA 3 % em tampão Tris-HCl 20 mM

contendo Tween-20 0,1 % (TBS-T) por 1 h à temperatura ambiente. Um anticorpo

monoclonal murino específico contra oligômeros de Aβ (NU1), gentilmente doado

pelo Dr. William Klein, da Universidade Northwestern (Evanston, USA), foi diluído

a 1 μg/mL em BSA 3 %/TBS e incubado com as membranas por 90 minutos. Após

três lavagens de 10 minutos com TBS-T, as membranas foram incubadas com

anticorpo secundário anti-IgG de camundongo conjugado com peroxidase na

diluição de 1:50.000 (em TBS-T) por 1 hora. As membranas foram lavadas três

vezes por 10 minutos com TBS-T e reveladas com o substrato SuperSignal West

Femto Maximum Sensitivity (Pierce, Rockford, IL) diluído 1:1, e expostas em filme

Kodak.

58

Figura 8: Western blot representativo das preparações de Oligômeros de Aβ. O tamanho dos agregados varia de dímeros até oligômeros de ~50-100 kDa. Imagem de microscopia eletrônica de transmissão (aumento de 85.000x) obtida pela Dra. Andréa Paula-Lima.

3.3) Culturas primárias de hipocampo rato

Embriões de rato Sprague-Dawley foram obtidos de ratas com 18 dias de

gestação para preparo de culturas de hipocampo. Após decapitação dos

embriões, os encéfalos foram removidos e transferidos para uma placa de Petri

contendo solução estéril de PBS glicose 2% a 37ºC. Com o auxílio de lupa, os

hipocampos foram dissecados, as meninges foram retiradas e o tecido foi

fracionado com tesoura cirúrgica estéril. As células foram dissociadas

59

mecanicamente com pipetas Pasteur de pontas flambadas, e centrifugadas a 1000

X g por 4 minutos à temperatura ambiente. Após remoção do sobrenadante, as

células foram ressuspensas em meio Neurobasal suplementado com 2% de

suplemento B27 (Gibco, Grand Island, NY), 0.5 mM de glutamina (Gibco), 100

U/mL de penicilina/estreptomicina (Gibco) e 10 μg/mL de fungizona (Cristália,

Itapira, SP). Após quantificação em câmara de Neubauer, as células foram

plaqueadas com densidade de 1.500 células/mm2 (placas de 6 e 96 poços, 100 e

400 μL por poço respectivamente) e 1.000 células/mm2 (placas de 6 poços, 1,5 mL

por poço) . As placas foram previamente tratadas com solução de poli-L-lisina a 10

μg/mL. As culturas foram mantidas a 37ºC, em estufa umidificada contendo 5%

CO2/95% ar atmosférico, por períodos de 18-21 dias in vitro (DIV). Um terço do

volume de meio em cada poço foi trocado por meio Neurobasal suplementado

fresco a cada 7 DIV.

3.4) Neuropatologia de cérebro humano

Pacientes com a Doença de Alzheimer e indivíduos não demenciados (Non

congnitive impairment, NCI) foram submetidos à necropsia com o consentimento

do Centro de Pesquisas de Doenças Neurodegenerativas na Universidade da

Pensilvania (Center for Neurodegenerative Disease Research at the University of

Pennsylvania). O diagnóstico clínico de Alzheimer foi concedido com base no

critério de NINCDS-ADRDA, e foi adicionalmente confirmado pelo exame pós-

mortem do córtex cerebral e do hipocampo para a presença de placas senis e

emaranhados neurofibrilares. Uma combinação dos grupos foi usada para agrupar

60

22 indivíduos de cada caso (DA e não demenciados), com o mesmo sexo, idade

similar (variando no máximo 5 anos), e um intervalo pós-mortem similar ( dentro

de 7 horas). Esse conjunto de informações encontram-se reunidos na Tabela

Suplementar 1 (artigo Bomfim e cols, 2012). Tanto os indivíduos não-demenciados

controle, quanto os pacientes com Alzheimer não apresentaram histórias ou

sintomas de condições psiquiátricas ou distúrbios neurológicos diferentes da

doença de Alzheimer.

No processo de necropsia, cada cérebro foi seccionado em cortes coronais,

dos quais um semento rostrocaudal da região do hipocampo (por exemplo, giro

denteado, subiculum e giro parahipocampal) foi dissecada e fixada em fluido de

Bouin, tampão neutro de formalina 10% ou em etanol 70% em salina por 24-48

horas. Em cada um dos grupos, as amostras de tecido do grupo controle e dos

pacientes com Alzheimer foram escolhidas do mesmo hemisfério e preservadas

nas mesmas soluções de fixação. Após ficarem embebidas na parafina, os cortes

coronais foram seccionados em fatias de 6µm, montadas nas lâminas e em

seguida, submetidas às reações de imunohistoquímica para IRS-1636/639 com o

anticorpo da Cell Signaling Technology com o código 2388 com a diluição de

1:100 usando o método avidina-biotina peroxidase (TALBOT e cols., 2004)

Os cortes de todos os grupos reagiram juntos nas mesmas soluções e

foram expostos nos mesmos tempos necessários para quantificar a

imunohistoquímica. Um segundo grupo de fatias dos dois grupos foi submetido ao

mesmo processo novamente para confirmar os resultados. Uma série de

fotomicrografias (100x) cobrindo toda a região de CA1 Do hipocampo em cada

corte foram tiradas usando um suporte motorizado, e as montagens foram criadas

61

usando um software Image-Pro Plus (Media Cybernetics Inc.). Todas as imagens

do grupo controle e do grupo Alzheimer foram capturadas sob as mesmas

condições de exposição de luz. Cincos dos casos de Alzheimer foram testados

para a especificidade do anticorpo de IRS-1pSer636/639 usado nas analises de

imunocitoquímica (Figura Suplementar 1de Bomfim e cols., 2012).

3.5) Injeção intracerebroventricular (i.c.v.) em camundongos.

Camundongos fêmeas nocaute para o receptor de IL-1β em um backgound

C56/BL6 e os camundongos selvagens foram concedidos pela Professora Maria

Bélio (Instituto de Microbiologia Paulo Góes – IMPG/CCS, na Universidade

Federal do Rio de Janeiro – UFRJ). Os animais foram alocados em grupos de 5

em cada caixa com livre acesso a comida e água sob um ciclo de 12h

claro/escuro, com a temperatura e humidade controlada. Todos os procedimentos

usados no presente estudo seguiram os “Princípios de cuidados de animais de

Laboratório” (Instituto Nacional de Saúde, EUA) e foram aprovados pelo Comitê de

cuidados e uso de animais Instituicional da Universidade Federal do Rio de

Janeiro (protocolo IBqM 075-05/16).

Para iniciar o processo de injeção i.c.v. de oligômeros, os animais foram

anestesiados por 7 minutos com 2.5% de isofluorano (Cristália, São Paulo, Brasil)

usando um sistema de vaporizador (Norwell, MA) e foi gentilmente contido neste

sistema somente durante o procedimento da ineção. Uma agulha de 2.5mm foi

inserida unilateralmente 1mm a direita do ponto médio, equidistante de cada olho.

Em seguida era injetado 10pmol de oligômeros de Aβ ou veículo, em um total de

62

volume de 1µL para os camundongos C57/BL6. O mesmo volume era injetado de

um corante azul no ventrículo lateral para verificar a capacidade de difusão

através da circulação do fluido cérebro espeinhal, e assim alcançando todo o

cérebro. Ao final do experimento, a injeção de marcador azul era realizada pata

verificar a acurácia de injeções corretas no ventrículo lateral. Os camundongos

que apresentavam qualquer sinal de erro de localização da injeção ou de

hemorragia cerebral (~5% dos animais) eram excuídos das análises posteriores.

3.6) Injeções de Aβ no cérebro de macacos e estudos neuropatológicos

Quatro macacos cinomolgos (Macaca fascicularis) com 9 anos de idade

(pesando 4.7-7.0kg) foram usados. Os animais estavam sob supervisão dos

veterinários da Queen´s Univesity. Os oligômeros de Aβ foram injetados de forma

crônica através de uma cânula i.c.v.. Três animais receberam 100µg de

oligômeros a cada três dias por 24 dias. A preparação de oligômeros de Aβ era

sempre fresca e prontamente caracterizada por cromatografia de exclusão por

tamanho (size-exclusion chromatography, SEC) antes das injeções. Um animal foi

somente operado e neste foi implantada uma cânula no ventrículo lateral no

mesmo modo que os outros animais, como controle experimental.

No final deste protocolo experimental, os animais foram sedados com

anestésico intramuscular (Ketamina 10mg/kg), com 0.01mg/kg de buprenorfina

para analgesia, seguida por uma injeção intravenosa de 25mg/kg de Pentobarbital.

A próxima etapa foi a perfusão dos animais com PBS seguido por 4% de

paraformoldeído em PSB, depois mais 4% de paraformoldeído em PSB contendo

63

2.5% de Glicerol; Seguindo PBS mais 5% de Glicerol, finalizando a perfusão com

PBS mais 10% de Glicerol. O cérebro dos quatro animais foram fatiados em cortes

de 40µm. Analise de imunohistoquímica foram realizadas usando cortes em free-

floating das regiões do hipocampo e córtex temporal. Essas fatias ficaram em PBS

contendo 1% Tritton incubado com 0.1M tampão citrato, pH6 aquecidas a 60ºC

por 5 minutos. O bloqueio dos cortes foi com BSA 5% em % de soro normal de

cabra (normal goat serum, NGS) e 1% de Triton x100 por 3 horas a temperatura

ambiente. Os anticorpos primários (IRS-1pSer636, JNKpThr183/Tyr185, GFAP)

foram diluídos em solução de bloqueio e os cortes foram incubados overnight a

4ºC, seguindo com a incubação dos anticorpos secundários conjugados com

Alexa-Fluor por 2 horas a temperatura ambiente. A autofluorescência do tecido foi

eliminada com uma incubação prévia com 0.06% de permanganato de potássio

por 10 minutos a temperatura ambiente. Os núcleos foram marcados com DAPI

por 5 minutos. As lâminas foram montadas com ProLong e as imagens foram

adquiridas usando o módulo apotome do microscópio Zeiss Axio Observer Z1.

3.7) Tarefa de Reconhecimento de Objetos

Os experimentos de reconhecimento de objetos foram realizados em uma

arena de campo aberto. Os objetos do teste são feitos de plástico ou vidro e

possuem diferentes formas, cores, texturas e tamanho. Durante as etapas do

teste, os objetos são fixados com uma fita adesiva no chão, assim os animais não

os movimentam. Nenhum dos objetos usados nos experimentos evocou uma

preferência inata nos animais. Durante as sessões de habituação, cada animai foi

64

substituído em um intervalo de 5 minutos, nas quais eles exploraram livremente a

arena vazia.

Ainda na etapa de habituação, o número de linhas que os animais cruzam

no chão (número de cruzamentos) e o número de vezes que os animais exploram

(elevação sobre as patas dianteiras, denotando comportamento de exploração)

foram controlados para verficar possível alteração locomotora ou exploratória em

função dos tratamentos. O treino consiste em sessões de 5 minutos, nas quais os

animais individualmente são colocados no centro da arena na presença de dois

objetos idênticos. O tempo de exploração de cada objeto é anotado pelo

pesquisador. A arena e os objetos são limpos para remover qualquer odor ou

estímulo para o próximo animal. Duas horas depois do treino, os animais foram

reinseridos na arena para a sessão de teste, neste momento, um dos objetos

usados no treino, foi substituído por um novo objeto. Novamente, o tempo de

exploração do objeto familiar e do novo objeto foi medido. Os resultados são

expressos como o percentual de tempo de exploração de cada objeto durante o

treino ou durante o teste. Foi usado o Teste T comparando a média do tempo de

exploração para cada objeto com um valor fixo de 50%. Por definição, animais que

reconhecem um objeto como famíliar (no aprendizado normal, por exemplo)

exploram o objeto novo por um tempo significativamente mais longo do que 50%

do tempo total.

65

3.8) Tratamento com os oligômeros de Aβ e agentes farmacológicos

Após o período de 18 a 21 DIV, as culturas dissociadas foram incubadas

com soluções de veículo (DMSO) ou de oligômeros de Aβ (AβOs), recém-

preparadas como descrito acima, a uma concentração final de 500 nM de Aβ por

tempos que variaram de acordo com a análise. Para analisar a fosforilação de

IRS-1 após 3h de exposição aos AβOs, as culturas foram pré-tratadas com 300

nM de Exendin-4 (Bachem, Torrance, CA), 1 µM de insulina recombinante humana

(Sigma, St Louis, MO), 10 µM de anticorpo neutralizante de TNF-α Infliximab ou 1

µg/mL de inibidor seletivo de JNK SP600125 (Tocris Bioscience, CA) por 30

minutos antes da adição dos AβOs.

3.9) Animais Transgênicos

Camundongos APP/PS1, contendo dois transgenes humanos (APP com

mutação sueca e mutante de deleção da presenilina-1, no qual o éxon 9 está

ausente) foram adquiridos do The Jackson Laboratories (Bar Harbor, ME) e

criados em biotério especializado até o final dos procedimentos de

experimentação. Animais selvagens (wild type) pareados por idade e linhagem

também foram criados e usados como controle em parte dos experimentos. Antes

das manipulações farmacológicas, os animais foram devidamente genotipados. A

descrição da genotipagem se encontra em material suplementar.

66

Os animais permaneceram em gaiolas com disponibilidade de alimento e

água ad libitum em salas com controle de temperatura (21,5 ± 1oC) e de período

circadiano (12h luz, 12h escuridão).

Estes animais transgênicos são gerados a partir da inserção randômica dos

dois transgenes em apenas um lócus da linhagem germinativa de camundongos

C57BL/6. Cada gene inserido é controlado pela região promotora da proteína

príon murina e contém uma seqüência de cDNA (JANKOWSKY e cols., 2001).

Um dos genes codifica para a proteína precursora amilóide quimérica

modificada contendo a mutação sueca (substituições K595N e M596L),

relacionada a casos genéticos da DA em humanos. O outro transgene

corresponde a uma versão mais ativa da presenilina-1, que sofreu a deleção do

éxon 9 humano, também relacionado a DA familiar (MULLAN e cols., 1992; ISHII e

cols., 1997; HILTUNEN e cols., 2000; RABE e cols., 2011). Estes construtos são

competentemente expressos nos animais.

A manipulação genética nestes camundongos provoca alterações

moleculares, fisiológicas e cognitivas que remetem ao fenótipo de doença de

Alzheimer em humanos e são intensificadas ao longo do envelhecimento murino

(SAVONENKO e cols., 2005; GIMBEL e cols., 2010; VOLIANSKIS e cols., 2010).

3.10) Administração de Exendina-4 em animais transgênicos.

Animais transgênicos APP/PS1 com idades entre 10 e 14 meses receberam

injeções intraperitoneais diárias do agonista farmacológico do receptor de GLP-1

67

(Exendina-4, 25 nmol/kg) ou veículo (solução salina). O exendina-4 foi

administrado por 3 semanas. O peso dos animais foi acompanhado durante os

tratamentos. Ao final, os animais foram sacrificados e tiveram seus encéfalos

removidos e dissecados.

3.11) Imunocitoquímica

Após 18-19 DIV, foi avaliado o nível de imunodetecção de IRS-1 fosforilado

em resíduos de serina 636 ou tirosina 465, além da imunodetecção de ligação dos

oligômeros em culturas de hipocampo expostas ou não aos oligômeros de Aβ (500

nM) durante 3 horas. As células foram fixadas pela adição de paraformaldeído 4%

a igual volume de meio de cultura por 5 minutos, seguido de incubação apenas

com paraformaldeído 4% durante 10 minutos. As células foram, então, bloqueadas

por 2 horas em solução contendo 10 % de NGS (Gibco, Grand Island, NY) em

PBS, em câmara úmida, temepratura ambiente. Após o bloqueio, as células foram

incubadas a 4 oC por 12 horas na presença do anticorpo primário monoclonal anti-

oligômeros de Aβ 1μg/mL (NU4) (LAMBERT e cols., 2007) gentilmente cedido

pelo Prof. William Klein (Northwestern University). Após a lavagem com abundante

volume de PBS, as células foram incubadas com os anticorpos policlonais IRS-2,

IRS-1 ou p-IRS-1(pSEr636, pSer312, pSer616, pSer317 ou pTyr 465) (Santa Cruz

Biotechnology, Santa Cruz, CA ) (1:200 por duas horas temperatura ambiente em

solução contendo 10 % de NGS, 0,1 % de Triton X-100 (Merck, Darmstadt, DE)

em PBS, em câmara úmida. Após a incubação com os anticorpos primários, as

células foram extensamente lavadas com PBS. Anticorpos secundários anti-IgG

68

murino conjugado a Alexa 555 (Molecular Probes, Carslbad, CA) e anti-IgG de

coelho conjugado a a Alexa 488 (Molecular Probes, Carslbad, CA) foram diluídos

em solução de bloqueio a 1:2.000 e adicionados às células. Após 2 horas de

incubação, as células foram lavadas três vezes com PBS. As lamínulas com as

células foram montadas em lâminas na presença de Prolong® e examinadas ao

microscópio de fluorescência invertido TE300 Nikon Eclipse. Foram obtidas

imagens de vinte a trinta campos, aleatoriamente escolhidos, em cada uma das 3

lamínulas para cada condição experimental.

3.12) Analises por Western blot das amostras de hipocampo dos animais

transgênicos e das culturas de neurônios hipocampais

Animais transgênicos APP/PS1 e controle de 13-14 meses de idade foram

usados nestes experimentos. Os grupos experimentais de animais transgênicos

tratados com exendina-4 (n=9) ou com o veículo (n=7) e animais selvagens

controle (n=8) foram eutanasiados. Para as análises de western blot o hipocampo

dos animais transgênicos e as culturas de células hipocampais foram

homogeneizadas em tampão RIPA contendo inibidores de protease e fosfatase.

As amostras foram resolvidas em um gel de poliacrilamida 4-20% com um

tampão Tris-glicina-SDS, com uma voltagem de corrida de 125V por 80 minutos a

temperatura ambiente. As amostras de 42µg de proteína total/poço foram

transferidas para uma membrana de nitrocelulose (Hybond ECL) usando um

tampão de transferência com 25mM Tris, 192mM glicina, 20% (v/v) de metanol e

0,02% SDS com o pH 8.3 a 350mA por 2 horas a 4ºC. As membranas foram

69

bloqueadas usando uma solução de bloqueio de 5% de leite em tampão Tris

contendo Tween-20 (0,1% de Tween-20 em 20mM Tris-HCL, pH 7.5 e 0,8% NaCl)

por 1 hora a temperatura ambiente.

Os anticorpos primários (anti–IRS-1pSer636; pSer312; pSer307; pSer616 or

anti–IRS-1 e anti–IRS-2 [1:200], anti–p-JNK [Thr183/Tyr185], anti-JNK [1:1,000],

anti–TNF-α, anti-TNFR1, e anti-GLP1R [1:200], ou anti–cyclophilin B [1:10,000])

foram diluídos na solução de bloqueio e incubados nas membranas por 2 horas a

temperatura ambiente. Seguimos com a incubação dos anticorpos secundários

conjugados com HRP anti-mouse ou anti-rabbit (1:10.000) por 1 hora. Após esta

etapa as membranas foram reveladas usando o substrato SuperSignal West

Femto Maximum Sensitivity em filme fotográfico.

Para a análise dos níveis de TNFα, culturas de hipocampo de rato foram

expostas a 500µM de oligômeros de Aβ ou o volume equivalente do veículo por 3

horas. O meio de cultura foi removido e concentrado por centrifugação Speedvac

(Savant Instruments Inc.). A concentração de proteína foi determinada no meio de

cultura através do ensaio por BCA protein assay kit (from Pierce). As amostras

foram resolvidas em gel de poliacrilamida 4%–20% seguindo os métodos descritos

acima para Western blotting usando um anticorpo anti-TNFα.

3.13) Análises dos dados

A intensidade de imunofluorescencia observada na marcação das proteínas

IRS-2, IRS-1, p–IRS-1(pSer636; pSer312; pSer616; pSer307; or pTyr465) e na

ligação dos oligômeros foi analisada de 3-6 experimentos diferentes usando

70

culturas neuronais independentes. Em cada experimento adquirimos de 20-30

imagens de três lamínulas diferentes em cada condição experimental. A análise

dos histogramas das intensidades de Fluorescencia foram realizadas usando o

programa NIH ImageJ (http://rsbweb.nih.gov/ij/) as described previously (DE

FELICE e cols., 2007). As análises estatísticas foram feitas pelo método ANOVA

seguido de um pos-teste Bonferroni post-hoc.

Para a neuropatologia do cérebro dos macacos, a densidade de marcação

do IRS-1pSer636 e p-JNK foram determinadas após usar um thresholding

apropriado usando o Image J. Para cada animal, a densidade de imunomarcação

foi medida em um grupo de 20-31 campos através do giro denteado do hipocampo

e através do córtex temporal. Os testes estatísticos foram realizados pelo método

ANOVA seguido do pós-teste Bonferroni post-hoc.

3.14) Expressão dos Plasmídeos

Os plasmídeos para expressar pβ-actin-BDNF-mRFP e pJPA5-YFP-JBD

foram doados pelo Dr. Gary Banker (Jungers Center for Neurosciences Research,

Oregon Health and Science University, Portland, Oregon, USA). O marcador

mitochondrial YFP foi doado pelo Dr. Gordon Rintoul (Department of Biological

Sciences, Simon Fraser University, Burnaby, British Columbia, Canada). Os

plasmídeos foram transfectados nos neurônios hipocampais em cultura (9-12 DIV)

usando lipofectamina 2000 (Invitrogen). Em seguida deixamos as células

expressarem os constructos por 24 horas, antes de desafiá-las com os oligômeros

de Aβ.

71

As céluas em seguida foram imageadas vivas usando um microscópio de

fluorescência de amplo campo (DMI 6000 B, Leica), como descrito previamente

(DECKER e cols., 2010). A distinção dos axônios e dendritos foram inicialmente

baseadas pela morfologia e em seguida confirmada pela marcação para MAP2,

uma proteína de citoesqueleto dendrítica (KWINTER e cols., 2009).

72

4) RESULTADOS

4.1) Agentes antidiabéticos protegem o cérebro de camundongos de uma

disfunção na sinalização por insulina causada pelos oligômeros de Aβ

associados a doença de Alzheimer.

(Título em inglês: “An anti-diabetes agent protects the mouse brain from defective

insulin signaling caused by Alzheimer’s disease–associated Aβ oligomers”)

Tendo em vista o crucial papel da sinalização por insulina cerebral para a

plasticidade sináptica, formação de memrória e aprendizado (ZHAO e cols., 2004;

VAN DER HEIDE e cols., 2005; CHIU e cols., 2008), funções estas, que

sabidamente estão afetadas na doença de Alzheimer (FERREIRA e KLEIN, 2011)

adicionado ao fato de que a sinalização por insulina no cérebro de pacientes com

Alzheimer está prejudicada (STEEN e cols., 2005; MOLONEY e cols., 2010)

formulamos a hipótese de que o ataque dos oligômeros de Aβ sobre esta

importante via de sinalização poderia contribuir para a perda de memória,

característica principal da doença de Alzheimer.

Os resultados apresentados no artigo estabelecem que a sinalização por

insulina está deficiente no cérebro de pacientes com Alzheimer, assim como em

roedores e nos modelos de primatas não –humanos para a doença de Alzheimer,

por mecanismos similares aos que são observados na resistência a insulina em

pacientes diabéticos. Verificamos uma ativação anormal da via de sinalização

TNFα/JNK nos neurônios expostos aos oligômeros de Aβ, tanto in vitro quanto in

vivo, resultando na fosforilação do IRS-1 em resíduos de serina, conhecida por

73

inibir a função do IRS-1 e por disparar resistência à insulina periférica em

pacientes diabéticos (HOTAMISLIGIL e cols., 1996; HIROSUMI e cols., 2002).

Nossos achados demonstraram que Exendin-4, um composto anti-diabético

que estimula a via de insulina através do receptor de GLP-1, preveniu a inibição

do IRS-1 em culturas hipocampais, e de forma significativa reverteu a disfunção da

sinalização por insulina em camundongos transgênicos para a doença de

Alzheimer. A reunião dos dados apresentados no artigo a seguir sugere que a

estimulação dos receptores de GLP1 seria uma possível e promissora abordagem

terapêutica para restaurar a disfunção da sinalização por insulina presente no

cérebro de pacientes com Alzheimer.

4.2) Resultados adicionais I

No intiuito de melhor compreender outros mecanismos que possivelmente

contribuiriam para a inibição da sinalização por insulina na doença de Alzheimer,

levantamos a hipótese de que outras vias inflamatórias, que se encontram

exacerbadamente ativadas em pacientes DA, poderiam mediar os efeitos tóxicos

induzidos pelos oligômeros de Aβ.

Nossos dados prévios demonstram a ação da via TNFα/JNK na inibição da

sinalização por insulina induzida pelos AβOs, uma vez que a utilização de um

anticorpo neutralizante de TNFα, Infliximab, ou de um inibidor farmacológico da

quinase de estresse JNK (SP600125) impediram completamente a fosforilação do

IRS-1 em resíduos inibitórios de serina (Figura 4 E-I), evidenciando a importante

participação desta via na inibição insulinérgica na doença de Alzheimer.

74

Outro mediador inflamatório que se encontra em níveis bastante elevados

no cérebro de pacientes DA é a IL-1β. A crônica deposição de Aβ estimula a

persistente ativação de células como as micróglias na doença de Alzheimer,

resultando em elevados níveis de IL-1β encontrados no cérebro dos pacientes DA

(PRINZ e cols., 2011). Considerando o envolvimento da via de IL-1β na

patogênese da DA, levantamos a hipótese de que possivelmente, essa importante

citocina poderia mediar os efeitos tóxicos dos oligômeros de Aβ sobre a via de

sinalização por insulina e a capacidade de aprendizado e memória.

Para avaliar a participação da sinalização por IL-1β na inibição do IRS-1

induzido pelos AβOs, utilizamos o modelo de cultura de neurônios hipocampais

tratados com um antagonista do receptor de IL-1β (IL-1βRa), previamente ao

desafio com os oligômeros de Aβ. Nosso resultado demonstrou um bloqueio total

do aumento dos níveis de IRS-1pSer induzido pelos AβOs nas céluas tratadas

previamente com o IL-1βRa, sugerindo fortemente que a inibição do IRS-1

induzida pelos oligômeros possivelmente é mediada pela sinalização por IL-1β.

75

Figura 1: Antagonista do receptor de IL-1β (IL-1βRa) previne a inibição do IRS-1 induzido pelos oligômeros de Aβ. Culturas de neurônios hipocampais expostas ao veículo, aos AβOs ou ao antagonista do receptor de IL-1β previamente ao desafio com os AβOs. Imagens representativas demonstram os níveis de IRS-1pSer636 nas diferentes condições. Os níveis de imunofluorescencia foram quantificados a partir de 30 imagens de 3 experimentos independentes. * P<0,05 comparado as culturas tratadas com veículo; ** P<0,05 comparado as culturas tratadas com AβO.

Nosso próximo passo foi investigar a participação da via de sinalização de

IL-1β no déficit cognitivo induzido pelo impacto tóxico dos oligômeros de Aβ no

cérebro, uma vez que nosso resultado prévio demostrou que esta via esta

envolvida com a inibição da via de insulina, que por sua vez é fundamental para

Veh AβO IL-1βRa +AβO

*

**

76

formação de memória e aprendizado. Para responder essa pergunta nós

realizamos uma única injeção de 10pmol de oligômeros de Aβ no ventrículo lateral

cerebral de camundongos que não expressam o receptor de IL-1β (IL-1βR KO),

um modelo experimental que foi recentemente introduzido em nosso grupo para

avaliar in vivo a disfunção e o déficit de memória da doença de Alzheimer (LEDO e

cols., 2013; FIGUEIREDO e cols., 2013). Os animais foram em seguida

submetidos ao teste comportamental de Reconhecimento de objetos. Nossos

resultados demonstraram que os oligômeros de Aβ foram incapazes de causar o

déficit de memória nos animais IL-1βR KO, ao contrário do que foi observado nos

animais selvagens expostos aos oligômeros. Como podemos notar na Figura 2, na

tarefa comportamental de Reconhecimento de Objetos (Figura 2), realizada 24h

depois da injeção de AβOs. Esse resultado indica que a via pró-inflamatória de IL-

1β tem um importante papel no déficit cognitivo induzido pelos AβOs, sugerindo

sua participação na patogênese da doença de Alzheimer.

77

Figura 2: Deficit cognitivo induzido pelos oligômeros de Aβ é bloqueado em animais nocaute para o receptor de IL-1β. Teste do reconhecimento objeto (RO) realizado 24h após a injeção i.c.v. dos AβOS (10 pmol/sítio) em camundongos do tipo selvagem (WT) e nocaute para o receptor de IL-1β (-/-).* P<0,05

4.3) Resultados Adicionais II

Durante o Doutorado, tive a oportunidade de participar de três trabalhos de

nosso grupo. O primeiro foi o projeto de um doutorando de nosso grupo, Mychael

Lourenço, intitulado “TNF-α mediates PKR-Dependent Memory Impairment and

Brain IRS-1 Inhibition Induced by Alzheimer’s β-Amyloid Oligomers in Mice and

Monkeys” o qual teve o obejtivo de investigar a indução do estresse de retículo

endoplasmático por oligômeros de Aβ (AβOs) e sua possível correlação com os

mecanismos da resistência à insulina em neurônios hipocampais. Os dados

reunidos neste artigo demonstram que os oligômeros de Aβ ativam a proteína

quinase R, dependente de dupla-fita de RNA (PKR) de um modo dependente da

via de sinalização de TNFα, resultando na fosforilação da subunidade alfa do fator

eucariótico de iniciação da tradução (eIF2α-p), na inibição do IRS-1, perda

0

25

50

75

100

WT WT-/- -/-Veh AEO

* *

Old Old Old Old NewNewNewNewEx

plo

rati

on

tim

e (

%)

78

sináptica e déficit de memória. Este conjunto de resultados indentificaram novos

mecanismos patogênicos comuns entre a Doença de Alzheimer e o Diabetes, e

reforçam que vias de sinalização pró-inflamatórias são mediadoras dos enfeitos

tóxicos dos oligômeros de Aβ sobre a função do IRS-1 assim como da perda

sinapses e de memória de uma forma dependente da ativação da PKR.

O artigo acima mencionado foi publicado no periódico Cell Metabolism em

2013, e encontra-se disponível nos anexos desta Tese (Anexo I).

O segundo trabalho que tive a oportunidade de participar foi o da então

Pós-doc de nosso laboratório, hoje Professora da Faculdade de Farmácia (UFRJ),

Dr. Júlia Clarke. O Título do artigo é: “Alzheimer-associated Aβ oligomers impact

the central nervous system to induce peripheral metabolic deregulation”. A

hipótese inicial deste artigo teve como base estudos que identificaram depósitos

de peptídeo beta amilóide no hipotálamo, uma estrutura cerebral que tem um

papel central na regulação metabólica, e ainda pode ser afetada em casos de

diabetes tipo 2 e obesidade. Entretanto, não era completamente compreendido se

a presença do Aβ no hipotálamo afetava a sua importante função.

Os principais resultados deste estudo demonstram que a injeção de

oligômeros de Aβ no ventrículo lateral de camundongos e macacos induziu a

ativação da vía pró-inflamatória da IKKβ/NF-κB e causou um transiente estresse

de retículo endoplasmático no hipotálamo. Este impacto dos oligômeros no

hipotálamo resultou na desregulação da homeostase periférica de glicose,

resistência à insulina nos músculos esqueléticos e inflamação do tecido adiposo

em camundongos. Os oligômeros falharam em disparar a ativação da via de

IKKβ/NF-κB e a intolerância a glicose periférica em camundongos que não

79

expressam o receptor de TNF-α. Juntos, esses resultados revelaram uma nova

ação patogênica dos oligômeros no cérebro demonstrando de forma impactante

que a doença de Alzheimer compartilha dos mesmos mecanismos observados nas

disfunções hipotalâmicas em síndromes metabólicas.

Esse conjunto de resultados encontra-se reunido no artigo que se encontra

atualmente submetido para o periódico EMBO Molecular Medicine, e que está

disponível nos anexos desta Tese (Anexo II).

Ainda no Doutorado tive a oportunidade de participar de uma Revisão cujo

título é: “Inflammation, defective insulin signaling and neuronal dysfunction in

Alzheimer's disease”. Nesta revisão nós discutimos os aspectos característicos da

inflamação no Sistema Nervoso Central e no Sistema Periférico que são comuns,

ou mesmo muito similares, entre a Doença de Alzheimer e as dinfuções

Metabólicas. Adicionalmente abordamos a contribuição da inflamação cerebral

para a disfunção da sinalização por insulina no cérebro e ainda para a disfunção

neuronal. Por último, nós reunimos as evidências que apontam para um

tratamento contra a inflamação cerebral como uma importante abordagem

terapêutica no tratamento da Doença de Alzheimer. Esta Revisão foi publicada

este ano, no periódico Alzheimer´s & Dementia e está disponível nos anexos desta

Tese (Anexo III).

The Journal of Clinical Investigation http://www.jci.org

βTheresa R. Bomfim,1 Leticia Forny-Germano,1,2 Luciana B. Sathler,1 Jordano Brito-Moreira,1

Jean-Christophe Houzel,2 Helena Decker,1,3 Michael A. Silverman,3 Hala Kazi,4 Helen M. Melo,1 Paula L. McClean,5 Christian Holscher,5 Steven E. Arnold,4 Konrad Talbot,4 William L. Klein,6

Douglas P. Munoz,7 Sergio T. Ferreira,1 and Fernanda G. De Felice1

1Institute of Medical Biochemistry and 2Institute of Biomedical Sciences, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil. 3Department of Biological Sciences, Simon Fraser University, Vancouver, British Columbia, Canada. 4Department of Psychiatry, University of Pennsylvania,

Philadelphia, Pennsylvania, USA. 5School of Biomedical Sciences, Ulster University, Coleraine, United Kingdom. 6Department of Neurobiology and Physiology, Northwestern University, Evanston, Illinois, USA. 7Centre for Neuroscience Studies, Queen’s University, Kingston, Ontario, Canada.

Defective brain insulin signaling has been suggested to contribute to the cognitive deficits in patients with Alzheimer’s disease (AD). Although a connection between AD and diabetes has been suggested, a major unknown is the mechanism(s) by which insulin resistance in the brain arises in individuals with AD. Here, we show that serine phosphorylation of IRS-1 (IRS-1pSer) is common to both diseases. Brain tissue from humans with AD had elevated levels of IRS-1pSer and activated JNK, analogous to what occurs in peripheral tissue in patients with diabetes. We found that amyloid-β peptide (Aβ) oligomers, synaptotoxins that accumulate in the brains of AD patients, activated the JNK/TNF-α pathway, induced IRS-1 phosphorylation at multiple serine residues, and inhibited physiological IRS-1pTyr in mature cultured hippocampal neurons. Impaired IRS-1 signaling was also present in the hippocampi of Tg mice with a brain condition that models AD. Impor-tantly, intracerebroventricular injection of Aβ oligomers triggered hippocampal IRS-1pSer and JNK activation in cynomolgus monkeys. The oligomer-induced neuronal pathologies observed in vitro, including impaired axonal transport, were prevented by exposure to exendin-4 (exenatide), an anti-diabetes agent. In Tg mice, exendin-4 decreased levels of hippocampal IRS-1pSer and activated JNK and improved behavioral measures of cognition. By establishing molecular links between the dysregulated insulin signaling in AD and diabetes, our results open avenues for the investigation of new therapeutics in AD.

Insulin resistance in peripheral tissue is a hallmark of type 2 diabe-tes (1). Accumulating evidence suggests that insulin resistance also develops in Alzheimer’s disease (AD) brains (2). Brain levels of insu-lin and insulin receptor (IR) are lower in AD, and insulin signaling impairments have been documented in both postmortem analysis and in animal models of AD (3–6). Brain insulin signaling is par-ticularly important for learning and memory (7, 8), suggesting that insulin resistance may contribute to cognitive deficits in AD.

We recently showed that soluble oligomers of the amyloid-β peptide (Aβ) instigate a striking loss of IRs from the membranes of neuronal processes (9). Aβ oligomers (AβOs) are small, diffus-ible aggregates that accumulate in AD brain and are recognized as potent synaptotoxins (10–12). Oligomers attach with specific-ity to synapses in particular neurons, acting as pathogenic ligands (13, 14). Recent studies have shown that oligomer binding induces

AD-like pathology, including neuronal tau hyperphosphorylation (15), oxidative stress (16, 17), synapse deterioration and loss (13, 18), and inhibition of synaptic plasticity (19). Interestingly, insu-lin signaling provides a physiological defense mechanism against oligomer-induced synapse loss (13). Insulin was found to down-regulate oligomer binding sites in neurons through a mechanism requiring IR tyrosine kinase activity (13). Stimulation of insulin signaling also protects neurons from oligomer-induced impair-ment of long-term potentiation (LTP) (20) and accumulation of hyperphosphorylated tau (21).

Understanding the molecular mechanisms accounting for impaired brain insulin signaling may illuminate new approaches to counteract neuronal damage in AD. As several pathological features, including impaired insulin signaling and inflammation, appear to be shared by patients with diabetes and those with AD, we hypothesized that, perhaps triggered by different factors, mech-anisms analogous to those that account for peripheral insulin resistance in type 2 diabetes could underlie impaired brain insu-lin signaling in AD. Results reported here establish that insulin signaling is disrupted in Alzheimer’s brains, as well as in rodent and non-human primate models of the disease, by mechanisms similar to those leading to insulin resistance in diabetes. Abnormal activation of the JNK/TNF-α pathway was verified in vivo and in

Authorship note: Theresa R. Bomfim and Leticia Forny-Germano contributed equally to this work.

Conflict of interest: William L. Klein is co-founder of Acumen Pharmaceuticals, which has been licensed by Northwestern University to develop ADDL technology for Alzheimer’s therapeutics and diagnostics.

Citation for this article: J Clin Invest doi:10.1172/JCI57256.


vitro in neurons exposed to AβOs, resulting in serine phosphoryla-tion of IRS-1, known to block downstream insulin signaling and trigger peripheral insulin resistance in diabetes (22). Significantly, exendin-4, a new antidiabetic drug that activates pathways com-mon to insulin signaling through stimulation of glucagon-like peptide 1 (GLP-1) receptors, blocked the impairment of insulin signaling in hippocampal cultures, reversed insulin pathology and improved cognition in Tg mice. Our findings suggest that stimula-tion of GLP-1 receptors (GLP1Rs) may represent a promising new approach to prevent disruption of brain insulin signaling in AD.

Because serine phosphorylation of IRS-1 is a central feature in peripheral insulin resistance (22, 23), we initially looked for IRS-1pSerine in human AD brain tissue. Results demonstrate that AD brains present abnormally high levels of IRS-1 phosphorylated at serine residues 636/639 (IRS-1pSer636/639) compared with brains from non–cognitively impaired (NCI) subjects (Figure 1, A–C, and Supplemental Table 1; supplemental material avail-able online with this article; doi:10.1172/JCI57256DS1), in line with a recent study that examined other pSer epitopes (4). In NCI controls, IRS-1pSer636/639 immunoreactivity was almost exclusively detectable in cell nuclei, appearing as puncta of vari-able sizes (Figure 1A). In some cases, extranuclear immunoreac-tivity in neuronal cell bodies was also detected, but this was rare in subjects younger than 75 years. In contrast, in AD patients a high density of neurons with IRS-1pSer636/639 labeling in cell bodies and, occasionally, in proximal dendrites was found from the earliest ages studied (i.e., 51 years). This was most conspicu-ous in the hippocampal CA1 region (Figure 1B). In 20 of 22 (91%) age- and sex-matched pairs of AD and control cases, the density of CA1 neurons with extranuclear IRS-1pSer636/639 labeling was greater in the AD case (Wilcoxon signed-ranks test; W = 239, P = 0.0001; Figure 1C and Supplemental Table 1). Control speci-ficity tests on the IRS-1pSer636/639 antibody showed that label-ing in AD brain could be fully blocked by competition with syn-thetic phosphorylated immunogen (Supplemental Figure 1), but not with the corresponding non-phosphorylated peptide. These findings are in harmony with peripheral mechanisms leading to type 2 diabetes and support the idea that AD is characterized by CNS insulin resistance.

Memory impairment in AD is now attributed, at least in part, to the synaptotoxicity of AβOs (12, 13, 19, 24), which accumulate in AD brains (14) and in animal models of AD (25). Recent studies have implicated oligomers in neuronal insulin resistance (9, 13). Thus, we next investigated whether pathological IRS-1pSer could develop from the neuronal impact of AβOs. Using highly differ-entiated hippocampal neuronal cultures, we found that AβOs induced abnormal elevation in somatodendritic IRS-1pSer636 levels (Figure 1, D–F). These results provide a salient pathogen-ic basis to account for elevated IRS-1pSer levels in AD brains. Because phosphorylation of IRS-1 at additional serine residues (other than Ser636) is also known to account for insulin resis-tance in peripheral tissue (23, 26), we searched for neuronal IRS-1 phosphorylation at other epitopes. We found that IRS-1pSer616, IRS-1pSer312, and IRS-1pSer307 levels were also increased in hippocampal neurons exposed to AβOs (Figure 1, F and M). In parallel, and consistent with the expected insulin resistance asso-ciated with serine phosphorylation of IRS-1, oligomers inhibited physiological IRS-1 phosphorylation at tyrosine residue 465 (IRS-

1pTyr465; Figure 1, G–I), an essential step in the IR-stimulated signaling pathway. Neurons targeted by AβOs exhibited increased IRS-1pSer levels, whereas non-attacked neurons showed low IRS-1pSer levels, as illustrated in Figure 1, J–L. Dysregulation of IRS-1 signaling, which we found to be prominent in AD brains (Figure 1B), is thus instigated by AβOs in central neurons.

Neuronal cultures used throughout our study were maintained in Neurobasal medium supplemented with B-27, an insulin-con-taining supplement, considered optimal conditions to preserve synapse health and function and to grow mature hippocampal cultures (27). In order to determine whether the increase in IRS-pSer described above might be related to insulin coming from B-27, we used cultures grown in insulin-free B-27. As shown in Supplemental Figure 2, in insulin-free medium AβOs triggered very similar increases in IRS-1pSer levels. Results thus establish that IRS-1pSer is specifically triggered by oligomers rather than by a possible physiological action of insulin present in B-27.

To determine whether the insulin signaling defect found in cell culture experiments also occurs in vivo, we investigated the effect of AβOs on IRS-1pSer in the brains of non-human primates. To this end, 3 adult cynomolgus monkeys (Macaca fascicularis) received intracerebroventricular (i.c.v.) injections of oligomers. A sham-operated monkey was used as a control. Remarkably, we found that the monkeys that received i.c.v. oligomer injections presented elevated levels of neuronal IRS-1pSer636 in the hippocampus compared with the control monkey (Figure 2, A–G). Interestingly, IRS-1pSer636 levels were also increased in the temporal cortex of monkeys that received injections of AβOs, indicating that the impact of oligomers on IRS-1 signaling extends to other brain regions in addition to the hippocampus (Supplemental Figure 3). These results demonstrate that AβOs instigate elevated serine phosphorylation of IRS-1 in the brains of monkeys, establishing the in vivo and in situ relevance of our findings.

We next examined IRS-1pSer levels in the brains of APPSwe,PS1ΔE9 (APP/PS1) mice, which express transgenes for human amyloid pre-cursor protein (APP) bearing the Swedish mutation and a deletion mutant form of presenilin 1. IRS-1pSer636 and IRS-1pSer312 levels, but not IRS-1pSer307 levels, were increased in hippocampi of APP/PS1 Tg mice compared with WT mice (Figure 2H).

Non-phosphopeptide antibodies were used to detect total lev-els of IRS-1 and IRS-2 in our experimental models. We noted that distinct patterns of dendritic labeling were obtained for IRS-1 and IRS-2 (Figure 3, A, B, F, and G). No differences in total IRS-1 lev-els were observed in oligomer-treated cultures or in hippocampi of APP/PS1 mice (Figure 3, A–E). In contrast, total IRS-2 levels increased in hippocampal cultures exposed to AβOs, as revealed by both immunocytochemistry and Western blot analysis (Figure 3, F–I). This impact of oligomers may be related to the fact that IRS-2 is a negative regulator of memory formation, as shown by recent studies (28, 29). However, IRS-2 levels were significantly decreased in APP/PS1 Tg mice compared with WT mice (Figure 3J). Decreased levels of IRS-2 have also been found in AD brains (4), suggesting that chronic exposure to AβOs may give rise to a compensatory mechanism aimed to decrease the negative impact of brain IRS-2 signaling on memory.

Previous studies have linked IRS-1 serine phosphorylation to JNK activation in diabetes and in obesity-related insulin resistance (22). In peripheral tissue, IRS-1 is phosphorylated at Ser636 by p-JNK (30). This prompted us to investigate the involvement of JNK in oligomer-induced IRS-1pSer in cell culture experiments. AβOs


failed to induce IRS-1pSer in hippocampal neurons transfected with GFP-fused dominant negative JNK (DN JNK; Figure 4, A–D), indi-cating a role for JNK in neuronal insulin resistance. As a control, mock transfection with a plasmid containing only GFP had no pro-tective effect (Figure 4C). Oligomer-induced accumulation of IRS-1pSer636 was also blocked by the pharmacological JNK inhibitor SP600125 (Figure 4, E–G and I). Moreover, oligomer-induced JNK activation was directly observed in hippocampal neuronal cultures

(Figure 4J). Consistent with the involvement of JNK indicated by our results, a recent study showed that AβOs induce tau hyperphos-phorylation and IRS-1 inactivation via JNK activation (31).

Subsequently, we sought to determine whether p-JNK levels were elevated in the brains of APP/PS1 mice. We found a 4-fold increase in p-JNK levels in hippocampi of Tg mice compared to WT ani-mals (Figure 4K), demonstrating that activation of JNK, first detected in cell culture experiments, occurs in vivo. No changes in

β

μ

βμ

μ

β

β

β


levels of total JNK were found in hippocampal cultures exposed to oligomers or in hippocampi of APPS/PS1 Tg mice (Supplemental Figure 4). Future studies employing mice with knockout of IRS-1, JNK1/2, or JNK3 may provide additional insight into the mecha-nistic links between insulin resistance and AD.

In peripheral insulin resistance, JNK activity is known to be stimulated by TNF-α signaling (22), and TNF-α levels are elevated in AD (32). Interestingly, we found that abnormal IRS-1pSer636 triggered by AβOs was completely blocked by infliximab, a TNF-α neutralizing antibody (Figure 4, H and I). We further detected an increase in TNF-α levels in concentrated conditioned medium from hippocampal cultures exposed to AβOs (Figure 4L). No dif-ferences were found in levels of TNF-α receptor in cultured neu-rons exposed to oligomers or in hippocampi of Tg mice (Figure 4, M and N). The results suggest that oligomer-induced elevation in proinflammatory TNF-α levels triggers aberrant activation of JNK and, ultimately, serine phosphorylation of IRS-1.

We next analyzed levels of p-JNK in AD brains and found that the density of neurons with detectable levels of activated JNK was significantly increased in AD hippocampi (Figure 5, A–D), giving strong support to our proposal that activation of the JNK pathway plays a key role in AD pathology. Finally, we examined JNK activa-tion in the brains of cynomolgus monkeys. Consistent with elevated IRS-1pSer levels, the 3 monkeys that received i.c.v. injections of AβOs presented elevated neuronal p-JNK levels in their hippocampi compared with the sham-operated monkey (Figure 5, E–I). Both cytoplasmic and nuclear p-JNK labeling were detected in NeuN-positive cells, but not in GFAP-positive cells (Figure 5, J–L), dem-onstrating neuronal specificity of JNK activation induced by oligo-mers. These results establish that abnormal activation of neuronal JNK is triggered by AβOs in the brains of non-human primates and support a key role of JNK in neuronal insulin resistance in AD.

Aberrant activation of JNK has been linked to impaired axonal transport in neurological disorders (33). Several neurodegenera-

ββ

β μ μ μβ


tive diseases, including AD, display axonal pathologies compris-ing defective transport and abnormal accumulation of proteins and organelles (34). Because AβOs were recently shown to impair axonal transport in hippocampal neurons (35), we further asked whether oligomer-induced JNK activation might be responsible for defects in axonal transport of dense core vesicles (DCVs) (see Supplemental Figure 5 for a scheme describing axonal trans-port measurements). Significantly, the JNK inhibitor SP600125 blocked axonal transport alterations induced by oligomers (Figure 6, Supplemental Table 2, and Supplemental Video 1), implicating JNK activation in impaired axonal transport in AD.

Double-stranded RNA-dependent protein kinase (PKR) and IκB kinase (IKK) are two stress-sensitive kinases that medi-ate serine phosphorylation of IRS-1 and are critical regulators of peripheral insulin resistance (36–39). In an additional set of experiments, we examined whether PKR and/or IKK were also activated by AβOs. A selective PKR inhibitor completely blocked oligomer-induced IRS-1pSer636, IRS-1pSer312, and IRS-1pSer307 in hippocampal cultures (Figure 7, A–L). IKK was also found to be involved in oligomer-induced IRS-1pSer, as ace-tylsalicylic acid completely prevented abnormal IRS-1pSer636 (Figure 7, M–O and Q). Abnormally activated mTOR signal-ing has also been implicated in peripheral insulin resistance (40). However, the mTOR inhibitor rapamycin had no effect on IRS-1pSer636 triggered by oligomers (Figure 7, P and Q), sug-gesting that mTOR is not involved in oligomer-induced serine phosphorylation of IRS-1. The involvement of PKR and IKK in AβO-induced IRS-1pSer provides additional evidence for a close parallelism between inflammation-associated brain insu-

lin resistance in AD and chronic inflammation-induced insulin resistance in peripheral tissues in type 2 diabetes.

Stimulation of brain insulin signaling has been suggested as a promising approach to prevent synapse deterioration and memory decline in AD (41, 42). We thus next tested whether bolstering insu-lin signaling might also protect neurons from aberrant activation of the JNK/IRS-1pSer pathway triggered by AβOs. We examined the effects of insulin and exendin-4 (exenatide), an incretin hormone analog that activates the insulin signaling pathway through GLP1R stimulation (43) and has been recently approved for treatment of diabetes. GLP1Rs are present and functional in cultured neurons as well as in rodent and human brains, and emerging evidence indicates that their stimulation regulates neuronal plasticity and cell sur-vival (44). Significantly, we found that both insulin and exendin-4 prevented the increase in IRS-1pSer636 (Figure 8, A–D) and the decrease in IRS-1pTyr465 levels (Figure 8, E–H) induced by oligo-mers. Exendin 9-39, a potent GLP1R antagonist and a competitive inhibitor of exendin-4, blocked the protective action of exendin-4, demonstrating that protection was specifically mediated by activa-tion of GLP1Rs (Figure 8D). Because neuronal cultures used in our study were maintained in Neurobasal B-27, an insulin-containing supplement, it is possible that these results reflect to some extent crosstalk between exendin signaling and signaling initiated by insu-lin. Control experiments showed that exendin-4 or insulin alone (i.e, in the absence of oligomers) had no significant effects on IRS-1pSer levels (Figure 8D). Interestingly, insulin and exendin-4 also protect-ed neurons from the above-described oligomer-induced impair-ment of axonal transport (Figure 8, I and J, Supplemental Tables 2 and 3, and Supplemental Videos 2 and 3). Because JNK dysregula-

ββ

μ β


βμ μ

ββ μ β μ β

μβ

β βα

β α β


tion appears to underlie axonal transport defects in a number of neurodegenerative disorders in addition to AD (33), this raises the possibility that prevention of aberrant JNK activation by bolstering insulin signaling might be beneficial in such disorders.

Protection by insulin against AβO-induced neuronal damage has been shown to involve downregulation of oligomer binding sites (13). Additional experiments thus aimed to determine whether exendin-4 also interferes with oligomer binding to neurons. Results showed that exendin-4 did not block oligomer binding (Supplemen-tal Figure 6). Along with the results presented above, this indicates that GLP1R activation by exendin-4 prevents oligomer-induced impairment in IRS-1 signaling even when oligomers are attached to neurons. We also determined GLP1R levels in AβO-treated neuronal cultures and in hippocampi of AD Tg mice. No changes in levels of GLP1R were found in hippocampal cultures exposed to oligomers or in hippocampi of APPS/PS1 Tg mice (Supplemental Figure 7). Because brain insulin signaling may decline with aging and in AD (45), exendin-4 may thus be more efficient than insulin in protecting neurons from the toxic impact of oligomers.

Finally, since exendin-4 readily crosses the blood-brain barrier and has been shown to facilitate hippocampal synaptic plasticity and cognition (44, 46, 47), we asked whether systemic adminis-

tration of a GLP1R agonist could enhance brain insulin signal-ing in APP/PS1 mice. Mice (13–14 months of age) were treated for 3 weeks with a daily intraperitoneal injection of exendin-4 (48). Exendin-4–treated mice exhibited significant reductions in brain levels of IRS-1pSer636, IRS-1pSer312, and p-JNK compared with vehicle-treated animals (Figure 8, K and L). Interestingly, spatial memory in the Morris water maze task was improved by chronic exendin-4 administration to Tg mice. Exendin-4–treated APP/PS1 mice learned the task faster, with significantly reduced escape latencies observed on days 3 and 4 of training, compared with saline-treated mice (Figure 8, M and N). Furthermore, exen-din-4–treated mice had improved memory retention, as indicated by a significantly longer time spent in the target quadrant during the probe trial conducted 24 hours after the last training session (Figure 8O). These data demonstrate the beneficial effect of exen-din-4 on cognition in AD Tg mice. Interestingly, we further found that treatment with exendin-4 entailed reductions in brain levels of amyloid plaque load and soluble Aβ in the cerebral cortices of AD Tg mice (Figure 8, P–R).

Reported effects of peripheral exendin-4 administration include reduced plasma glucose levels and decreased food intake and body weight, and these could mimic the anti-aging effects of

β

β

μ μ


caloric restriction. We note that, for the duration of the behav-ioral experiments described above, comparable and slight weight losses were observed in all groups of mice analyzed, regardless of whether they received daily intraperitoneal injections of exendin-4 or saline (Supplemental Figure 8). This may be due to the exer-cise regime to which the animals were subjected during training and trials in the Morris water maze. We also note that, in insulin-secreting cells, exendin-4 inhibits JNK activation (49), counteracts TNF-α–mediated apoptosis, and reverses inhibition of the IRS-1 pathway (50). Taken together, our results suggest that exendin-4 restored impaired brain insulin signaling, decreased plaque load and soluble Aβ levels, and improved learning and memory.

Recent studies have shown that type 2 diabetes increases the risk of AD (51), and a newly recognized form of brain insulin resistance has been connected to AD (2, 3). Here, we describe a pathogenic mechanism by which a type of “brain diabetes” is triggered by AβOs, synaptotoxins that accumulate in AD brains. Because brain insulin signaling is centrally involved in learning and memory (8), the impact of AβOs on brain insulin signaling may be a mecha-nism potentially contributing to early memory loss in AD.

Similar to mechanisms that cause peripheral insulin resistance in type 2 diabetes, our results show that AβOs induce neuronal insulin resistance by activating TNF-α and the stress kinases JNK, IKK, and PKR, leading to serine phosphorylation of IRS-1 (Figure 9A). Anal-ysis of AD brains confirmed abnormal phosphorylation of IRS-1 at Ser636/639, which is typically associated to peripheral insulin resistance in diabetes (1). IRS-1pSer redistribution was recently reported in AD brains (31). Because we have previously shown that AβOs internalize/redistribute neuronal IRs (9, 13), it is possible that IR loss may underlie, or facilitate, IRS-1pSer increases. This is con-sistent with our finding that insulin, previously shown to block IR downregulation induced by AβOs (13), further blocks IRS-1pSer.

In peripheral insulin resistance, IRS-pSer is a major target for phosphorylation by mTOR (40, 52, 53). In contrast, the mecha-

nism by which oligomers trigger neuronal IRS-1 inhibition does not appear to involve mTORC1 activation, as rapamycin had no protective effect against IRS-1pSer. The possible role of mTOR in AD is still controversial. mTOR activity was shown to be increased in brains of 3xTg-AD mice (54) and rapamycin improved cogni-tion in PDAPP Tg mice (55, 56). On the other hand, upregulat-ing mTOR signaling rescued LTP in another AD mouse model (Tg2576), suggesting that mTOR inhibition correlates with impaired synaptic plasticity in AD (57). Other studies have shown that mTOR signaling is downregulated in cellular and ani-mal models of AD (54, 58). Given these controversial literature reports, further studies aimed at unraveling the possible role of mTOR signaling in AD appear warranted.

We found significant increases in p-JNK levels in AD brains and in hippocampi of APP/PS1 Tg mice. Using a different strain of Tg mice fed a high-fat diet, a recent study reported abnormally elevated brain levels of activated JNK (31). Remarkably, elevated IRS-1pSer636 and p-JNK levels were triggered by injection of AβOs in the brains of monkeys. Considering the dearth of animal model systems that truly recapitulate the main features of AD (59), a monkey model of AD may provide insight into central aspects of pathology that could be present exclusively in primates.

Inflammation is an important mediator of insulin resistance in obesity and diabetes (23, 26, 60, 61). Supporting the notion that AβO-induced neuronal insulin resistance derives from an inflammatory response mediated by TNF-α, the TNF-α–blocking antibody infliximab protected neurons against oligomers. Used to treat inflammatory diseases (62), infliximab and etanercept, a TNF-α–blocking recombinant protein, have been proposed as novel therapeutic agents to combat insulin resistance in type 2 diabetes (63) and memory decline in AD (64). By defending neu-rons from oligomer-induced dysregulation of insulin signaling, infliximab treatment may constitute a novel approach to prevent memory impairment in AD.

We did not observe significant changes in levels of IRS-1pSer307 in hippocampi from AD Tg mice, while in cultured neurons AβOs

β μ

β


increased IRS-1pSer307. A recent study demonstrated that IRS-1pSer307 in mice is a positive regulatory site that sustains periph-eral insulin signaling and moderates the severity of insulin resis-tance (65), in contrast to results obtained in previous cell-based experiments (26, 66, 67). Based on our results, it is likely that, following oligomer attack of synapses, the stress-sensitive kinases PKR, IKK, and JNK respond and coordinately lead to IRS-1pSer at multiple residues. Therefore, even if Ser307 is not phosphorylated, the final outcome may be that multiple serine-phosphorylated res-idues may act as critical regulators of neuronal insulin resistance, similar to what occurs in peripheral tissue (36).

Total IRS-2 levels were increased in hippocampal cultures exposed to AβOs. This might be related to the fact that, as previ-ously observed, IRS-2 is a negative regulator of memory forma-tion that acts by impairing dendritic spine formation (28) and that deletion of IRS-2 reduces amyloid deposition, cognitive deficits and premature mortality in Tg2576 mice (29). Beneficial effects of IRS-2 deletion in AD pathology parallel the effects on lifespan,

with less IRS-2 signaling extending life span in mice (43). We found that IRS-2 levels were significantly decreased in APP/PS1 Tg mice compared with WT mice. Decreased levels of IRS-2 were also found in AD brains (4), suggesting a compensatory phenomenon to decrease the negative impact of brain IRS-2 signaling on memo-ry in AD (68). So far, however, it is unclear whether this is an active neuroprotective response or a secondary response to the neuro-degenerative process. Therefore, there is an apparent dichotomy between the neuroprotective effects of insulin signaling in CNS and its deleterious actions on lifespan and memory. It is possible that IRS-1 acts as a positive regulator of memory, as suggested by our current results, while IRS-2 acts as a negative modulator of memory formation.

AβOs increasingly appear to be the proximal toxins that cause synapse failure in AD (12, 13, 24, 25). Oligomers constitute key target for therapeutics, as drugs and antibodies designed to target them have provided positive results in AD clinical trials (69, 70). However, issues related to efficacy and safety of such approaches

ββ μ β

β μ β μ

β


ββ

β μβ

ββ β μ β

β

μ β


have still not been fully addressed. Stimulation of neuronal insulin signaling has been proposed as a promising approach to prevent or halt memory decline in AD (41, 42). In vitro, insulin prevents AβO-induced loss of surface IRs, neuronal oxidative stress, and synapse deterioration (13). The mechanism of insulin protection involves IR signaling–dependent downregulation of oligomer binding to neurons. Thus, the protective action of insulin in rescu-ing the impairment of IRS-1 reported here probably derives from its ability to block oligomer binding to neurons. However, since IRs are removed from the neuronal membrane in AD brains (4) and in cultured neurons exposed to oligomers (9), use of insulin itself might not be the most effective way to combat AD. Instead, alternative approaches to bypass the IR and enhance insulin-relat-ed signaling pathways might provide a safe and effective strate-gy to prevent or treat AD. Here, we show that exendin-4, a novel antidiabetic drug that stimulates the insulin signaling pathway through activation of GLP1Rs, protected neurons against oligo-mer-induced dysregulation of IRS-1 phosphorylation (Figure 9B).

Exendin-4 further decreased hippocampal IRS-1pSer and p-JNK levels, decreased amyloid pathology, and improved cognition in AD Tg mice. Enhancing brain insulin signaling through the use of exendin-4 or other GLP1R agonists may thus be a key alternative to block brain insulin resistance and memory impairment in AD.

ReagentsSynthetic Aβ1–42 peptide was from American Peptide Co. Scrambled Aβ1–42 was from Anaspec. Bovine and human insulin, 1,1,1,3,3,3,-hexafluoro-2-propanol (HFIP), DMSO, poly-L-lysine, rapamycin, and acetylsalicylic acid were from Sigma-Aldrich. Culture media/reagents, Alexa Fluor–labeled secondary antibodies, and ProLong anti-fade reagent were from Invitrogen. Electrophoresis buffers were from Bio-Rad. SuperSignal chemilumines-cence reagents and the BCA protein assay kit were from Pierce. Antibodies against total IRS-1 and IRS-2, IRS-1pTyr465, IRS-1pSer636, glucagon-like peptide–1 receptor (GLP1R), TNFR1, TNF-α, and the PKR inhibitor were from Santa Cruz Biotechnology Inc. IRS-1pSer307, IRS-1pSer312, and IRS-1pSer616 antibodies were from Invitrogen. IRS-1pSer636/639 anti-body for histology, the phosphorylated immunogen supplied as a custom order for specificity tests, glial fibrillary acidic protein (GFAP) antibody, p-JNK (Thr183/Tyr185) monoclonal antibody, and JNK polyclonal anti-body were from Cell Signaling Technology. Unphosphorylated IRS-1 peptide (aa 631–646) was from Abcam (no. 41777). Exendin-4 and exendin 9-39 were from Bachem. SP600125 was from Tocris Bioscience.

Neuropathology in human brain tissueNCI controls and patients with AD were autopsied with caregiver consent by the Center for Neurodegenerative Disease Research at the University of Pennsylvania. Clinical diagnosis of AD met NINCDS-ADRDA criteria and was confirmed by postmortem examination of the cerebral cortex and hip-pocampus for senile plaque and neurofibrillary tangle densities. A matched pairs design was used to match each of 22 AD cases with 22 NCI cases of the same sex, similar age (within 5 years), and similar postmortem interval (with-in 7 hours). Demographic and autopsy data on the subjects studied are given in Supplemental Table 1. Neither NCI controls nor AD patients had histories or symptoms of psychiatric conditions or non-AD neurological disorders.

At autopsy, each brain was cut into coronal slabs, from which an interme-diate rostrocaudal segment of the hippocampal region (i.e., hippocampus, dentate gyrus, subiculum, and parahippocampal gyrus) was dissected and fixed in Bouin’s fluid, 10% neutral buffered formalin, or 70% ethanol in saline for 24–48 hours. In each matched pair, the tissue sampled from the NCI and AD cases derived from the same hemisphere and was preserved in the same fixative. After being embedded in paraffin, 6-μm coronal sec-tions were cut, mounted on slides, and reacted immunohistochemically

β α

β


for IRS-1pSer636/639 with Cell Signaling Technology antibody 2388 at a dilution of 1:100 using an avidin-biotin-peroxidase method (71). Sections from all the case pairs were reacted together with the same solutions and exposure times to enable quantitative immunohistochemistry. A second set of sections from all pairs was run to test reliability of the results. After coverslipping and drying of the sections, the borders of hippocampal field CA1 were traced in ink on the slides under a dissecting scope aided by well-established criteria and atlas-style drawings (72). A series of ×100 photomi-crographs covering all of CA1 in each section were then taken on a micro-scope with a motorized stage, and a composite montage was created using Image-Pro Plus software (Media Cybernetics Inc.). Photomicrographs of all NCI and AD sections were taken under the same lighting conditions. Image-Pro Plus was used to determine the area covered by CA1 in each section and the number of neurons with extranuclear IRS-1 pSer636/639 in that area by identifying immunoreactive objects that were larger in area than cell nuclei seen with hematoxylin staining and still within the size and shape parameters of neurons identified by NeuN immunoreactivity. The density of neurons with extranuclear IRS-1 pSer636/639 was then cal-culated. Five AD cases were tested for specificity of the IRS-1pSer636/639 antibody used in immunohistochemistry analyses (see Results).

AβOs and scrambled Aβ1–42

Oligomers were prepared from synthetic Aβ1–42 peptide (American Peptide) as previously described (16, 19). Scrambled Aβ peptide (Anaspec) treated in exactly the same manner was used in control experiments.

Mature hippocampal culturesPrimary rat hippocampal neuronal cultures were prepared according to established procedures (15, 16) and were used after 18–21 days in vitro (DIV). Cultures were prepared and maintained in Neurobasal medium supplemented with B-27, an insulin-containing supplement (Invitrogen). Some experiments (see Supplemental Figure 2) were performed using insu-lin-free B-27 (Invitrogen). Cultures were treated at 37°C for 3 hours with 500 nM AβOs or an equivalent volume of vehicle (2% DMSO in PBS). When present, insulin (1 μM), exendin-4 (300 nM), SP600125 (10 μM), infliximab (1 μg/ml), rapamycin (0.1 μM), and PKR inhibitor (1 μM) were added to cultures 30 minutes before AβOs. Acetylsalicylic acid (5 mM) was added to cultures 120 minutes before AβOs. Exendin 9-39 (1 μM) was added 15 minutes before exendin-4.

ImmunocytochemistryCells were fixed and blocked as described previously (13, 15), incubated with AβO-selective NU4 mouse monoclonal antibody (16) and IRS-2, IRS-1, or p–IRS-1 (pSer636; pSer312; pSer616; pSer307; or pTyr465) rabbit polyclonal antibodies followed by Alexa Fluor–conjugated secondary anti-bodies. Coverslips were imaged on a Zeiss Axio Observer Z1 microscope.

Injections of AβOs into monkey brains and neuropathology studiesFour cynomolgus monkeys (Macaca fascicularis) aged 9 years (weight, 4.7–7.0 kg) were used. Animals were under the close supervision of a labo-ratory animal technician and the Queen’s University veterinarian. AβOs were infused chronically through an i.c.v. canula. Three animals were administered 100 μg AβOs, every 3 days for 24 days. Oligomers were freshly prepared and characterized by size-exclusion chromatography (SEC) before each injection. One sham-operated animal, used as a control, had the cannula implanted into the lateral ventricle in the same manner as the experimental animals. At the end of the experimental protocol, animals were sedated with intramuscular 10 mg/kg ketamine with 0.01 mg/kg buprenorphine for analgesia, followed by intravenous 25 mg/kg sodium pentobarbital, perfused with PBS followed by 4% paraformaldehyde in

PBS; 4% paraformaldehyde in PBS containing 2.5% glycerol; PBS plus 5% glycerol; and PBS plus 10% glycerol. Serial 40-μm-thick brain sections were obtained. Immunohistochemistry was performed using free-floating sec-tions from hippocampus and temporal cortex in PBS containing 1% Triton incubated with 0.1 M citrate buffer, pH 6, at 60°C for 5 minutes. Sections were blocked with 5% BSA, 5% normal goat serum (NGS), and 1% Triton X-100 for 3 hours at room temperature. Primary antibodies (IRS-1pSer636, JNKpThr183/Tyr185, GFAP) were diluted in blocking solution, and sec-tions were incubated overnight at 4°C, followed by incubation with Alexa Fluor–conjugated secondary antibodies for 2 hours at room temperature. Tissue autofluorescence was quenched by previous incubation with 0.06% potassium permanganate for 10 minutes at room temperature. Nuclei were stained with DAPI for 5 minutes. Slides were mounted with ProLong and imaged on a Zeiss Axio Observer Z1 microscope using structured illumina-tion (ApoTome module) to decrease out-of-focus light.

AD Tg mouse model and treatment with exendin-4APP/PS1 mice on a C57BL/6 background were obtained from The Jackson Laboratory. Mice not expressing the transgene were used as WT controls. Male animals were used in all studies. Animals were caged individually and maintained on a 12-hour light/12-hour dark cycle (lights on at 08:00, off at 20:00), in a temperature-controlled room (21.5 ± 1°C). Food and water were available ad libitum. Animals received daily intraperitoneal injections of exendin (25 nmol/kg, dissolved in saline) or vehicle (saline) during 3 weeks.

Western blot analysis of Tg mouse hippocampi and hippocampal neuronal culturesThirteen- to 14-month-old APP/PS1 Tg mice and WT control animals were used. Exendin-treated (n = 9) or vehicle-treated Tg (n = 7) and WT animals (n = 8) were euthanized. For Western immunoblot analysis, hippocampi of Tg mice and mature hippocampal cell cultures were homogenized in RIPA buffer containing protease and phosphatase inhibitor cocktails and resolved on a 4%–20% polyacrylamide gel with Tris/glycine/SDS buffer run at 125 V for 80 minutes at room temperature. The gel (42 μg total protein/lane) was electroblotted onto Hybond ECL nitrocellulose using 25 mM Tris, 192 mM glycine, 20% (v/v) methanol, 0.02% SDS, pH 8.3, at 350 mA for 2 hours at 4°C. Membranes were blocked with 5% nonfat milk in Tris-buffered saline containing Tween-20 (TBS-T) (0.1% Tween-20 in 20 mM Tris-HCl, pH 7.5, 0.8% NaCl) for 1 hour at room temperature. Pri-mary antibodies (anti–IRS-1pSer636; pSer312; pSer307; pSer616 or anti–IRS-1 and anti–IRS-2 polyclonal antibodies [1:200], anti–p-JNK [Thr183/Tyr185] monoclonal antibody, anti-JNK polyclonal antibody [1:1,000], anti–TNF-α, anti-TNFR1, and anti-GLP1R polyclonal antibodies [1:200], or anti–cyclophilin B polyclonal antibody [1:10,000]) were diluted in 5% milk/TBS and incubated with the membranes for 120 minutes at room temperature. After incubation with HRP-conjugated anti-mouse or anti-rabbit IgGs (1:10,000 in TBS-T) for 60 minutes, membranes were washed, developed with SuperSignal West Femto Maximum Sensitivity substrate, and imaged on photographic film.

For TNF-α analysis, cultures were exposed to 500 nM AβOs or an equiva-lent volume of vehicle for 3 hours. The medium was then removed and con-centrated by Speedvac (Savant Instruments Inc.) centrifugation. Protein concentrations were determined in the medium using the enhanced BCA protein assay kit (from Pierce). Samples containing equal protein amounts were resolved in a 4%–20% polyacrylamide gel, followed by Western blot-ting using anti–TNF-α antibody.

Data analysisIRS-2, IRS-1, p–IRS-1(pSer636; pSer312; pSer616; pSer307; or pTyr465), and AβO binding immunofluorescence intensities were each analyzed in


3–6 experiments (see figure legends) using independent neuronal cultures. In each experiment, 20–30 images were acquired from 3 coverslips in each experimental condition. Histogram analysis of fluorescence intensities at each pixel across the images was performed using NIH ImageJ (http://rsbweb.nih.gov/ij/) as described previously (16). Cell bodies were digitally removed from the images so that only immunostaining on dendritic pro-cesses was quantified. Statistical significance was assessed by ANOVA fol-lowed by Bonferroni post-hoc test.

For neuropathology in monkey brains, IRS-1pSer636 and p-JNK immu-nolabeling densities were determined after proper thresholding using NIH ImageJ by counting the particles with diameters of 500 or fewer pixels and determining the total area (in pixels) occupied by particles in each field. For each animal, immunolabeling densities were measured in a set of 20–31 microscopic fields throughout the dentate gyrus or temporal cortex (see Results). Statistical significance was assessed by ANOVA followed by Bon-ferroni post-hoc test.

Plasmids and expression of transgenespβ-actin-BDNF-mRFP and pJPA5-YFP-JBD were from Gary Banker (Jungers Center for Neurosciences Research, Oregon Health and Science University, Portland, Oregon, USA). Mitochondrially targeted YFP was from Gordon Rintoul (Department of Biological Sciences, Simon Fraser University, Burnaby, British Columbia, Canada). Constructs were trans-fected into neurons at 9–12 DIV using Lipofectamine 2000 (Invitrogen). Cells were allowed to express constructs for 24 hours prior to exposure to AβOs and then imaged.

Live imagingCells were live imaged using a wide-field fluorescence microscope (DMI 6000 B, Leica), as described previously (35). Axons and dendrites were ini-tially distinguished based on morphology and confirmed retrospectively by antibody staining against MAP-2, a dendritic cytoskeletal protein (73).

Videos and quantitative analysesVideos were processed using MetaMorph (Universal Imaging) as described previously (35). Quantitative kymograph analysis was per-formed using MetaMorph.

Behavioral testsAnimals were handled daily for 2 weeks prior to commencement of the study. Mice were 9 months of age when treatment began. They received intraperitoneal exendin-4 (25 nmol/kg BW) or saline (0.9% w/v) injections twice daily for 3 weeks before behavioral tasks were conducted (n = 12 for each group). All experiments were licensed by the UK Home Office in accordance with the Animal (scientific procedures) Act of 1986.

Morris water maze task. The maze was made of white opaque plastic with a diameter of 120 cm and 40-cm-high walls and was filled with water at 25°C to avoid hypothermia. A small escape platform (10 × 6.5 × 21.5 cm) was placed at a fixed position in the center of one quadrant, 25 cm from the perimeter, and was hidden 1 cm beneath the water surface. The room contained a number of fixed visual cues on the walls.

Acquisition phase. The acquisition trial phase consisted of 4 training days and 4 trials per day with a 15-minute inter-trial interval. Four points equally spaced along the circumference of the pool (north, south, east, west) served as the starting position, which was randomized across the 4 trials each day. If an animal did not reach the platform within 90 seconds, it was guided to the platform, where it had to remain for 30 seconds, before being returned to its home cage. Mice were kept dry, between trials, in a plastic holding cage filled with paper towels. The path length and escape latencies were recorded (n = 12 per group).

Probe trial. One day after completion of the acquisition task (day 5), a probe trial was performed in order to assess spatial memory (after a 24-hour delay). The platform was removed from the maze, and animals were allowed to swim freely for 60 seconds. Spatial acuity was expressed as the amount of time spent in the exact area where the escape platform was located.

Histology in Tg miceMice were perfused transcardially with PBS buffer followed by ice-cold 4% paraformaldehyde in PBS. Brains were removed and fixed in 4% paraformaldehyde for at least 24 hours before being transferred to 30% sucrose solution overnight. Brains were then snap-frozen using Envirofreeze, and 40-μm-thick coronal sections were cut at coordinates bregma –2 to –3 using a Leica cryostat. Sections were chosen according to stereological rules (74), with the first section taken at random, fol-lowed by every fifth section afterward. Seven to 13 sections were ana-lyzed per brain. Staining was carried out for Aβ plaques. All sections were incubated in 3% H2O2 to quench endogenous peroxidase activity. After the sections were blocked in 5% normal serum to avoid nonspecific antibody binding, they were incubated with rabbit polyclonal anti-Aβ peptide (1:250, Invitrogen 71-5800). After overnight incubation at 4°C, the sections were incubated in respective secondary antibodies. For visu-alization, Vectastain Elite and SG substrate (Vector Laboratories) were used. All staining was visualized by Axio Scope 1 (Zeiss) and analyzed using a multi-threshold plug-in with ImageJ, using stereological rules as described in ref. 75.

ELISA for total soluble Aβ levelsSoluble Aβ levels were measured using an ELISA kit (Invitrogen), used according to the manufacturer’s instructions. Briefly, right hemispheres of control and exendin-4–treated APP/PS1 mouse brains were homog-enized in Tris-buffered saline (25 mM Tris-HCl, pH 7.4, 150 mM NaCl) supplemented with protease inhibitor cocktail (Sigma-Aldrich, 250 μl per 5 ml buffer). Brain homogenates were centrifuged at 100,000 g and 4°C for 1 hour. The supernatant was then diluted 1:10 before the ELISA was carried out. Protein was quantified using the Bradford protein assay. Final soluble Aβ values were determined following normalization to total protein levels (n = 6 per group).

StatisticsStatistical analyses were performed (GraphPad Prism) using 2-tailed Stu-dent’s t test when 2 conditions were compared and 1-way ANOVA followed by Bonferroni post-hoc test for multiple comparisons. Results are repre-sented as mean ± SEM (unless stated otherwise), and the total number of independent experiments, as well as P values, are specified in each figure legend. P values less than 0.05 were considered significant.

Study approvalAll experiments involving rats and Tg mice, unless otherwise specified, were performed in certified facilities under protocols approved by the Institutional Animal Care and Use Committee of the Federal University of Rio de Janeiro (protocols IBQM 022 and IBQM 019). All animal care and experimental procedures involving non-human primates were in accordance with the Canadian Council on Animal Care policies on the use of laboratory animals and approved by the Queen’s University Animal Care Committee (Animal Care Protocol Original Munoz-2011-039-Or). For the human postmortem studies, informed consent was obtained for collection and use of clinical and postmortem data from all subjects of the present study or from their next-of-kin in accordance with the University of Pennsylvania’s Institutional Review Board and its Alzheimer Disease Center (Clinical Core Protocol 068200).


This work was supported by the Human Frontier Science Program (HFSP) and the John Simon Guggenheim Memorial Foundation (FGF). F.G. De Felice and S.T. Ferreira are also funded by Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (CNPq), Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), and Instituto Nacional de Neurociência Translacio-nal (INNT). T.R. Bomfim, L. Forny-Germano, L.B. Sathler, H.M. Melo, and J. Brito-Moreira are supported by predoctoral fellow-ships from Conselho Nacional de Desenvolvimento Cientifico e Tecnológico and Coordenação de Aperfeicoamento de Pessoal de Ensino Superior. Experiments on monkeys were funded by a grant (MOP-77734) from the Canadian Institutes of Health Research (to D.P. Munoz) and by HFSP (to F.G. De Felice). D.P. Munoz was also supported by the Canada Research Chair Program. W.L. Klein is funded by grants from the American Health Assistance Foun-dation, Alzheimer’s Association, and NIH–National Institute on

Aging grants R01-AG18877 and R01-AG22547. K. Talbot and S.E. Arnold were supported by a Temple Foundation Discovery award and by the Alzheimer’s Association. M.A. Silverman is funded by the National Science and Engineering Research Council (no. 327100-06), the Canadian Foundation for Innovation (12793), and the Canadian Institutes of Health Research (no. 90396). We thank G. Banker (Oregon Health State University) for the gift of DN JNK and L. Veloso (University of Campinas, Campinas, Brazil) for the gift of infliximab.

Received for publication January 26, 2011, and accepted in revised form January 5, 2012.

Address correspondence to: Fernanda G. De Felice, Institute of Med-ical Biochemistry, CCS, Room H2-019, Federal University of Rio de Janeiro, Rio de Janeiro, RJ 21944-590, Brazil. Phone: 5521.2562.6790; Fax: 5521.2270.8647; E-mail: [email protected].

1. White MF. IRS proteins and the common path to diabetes. Am J Physiol Endocrinol Metab. 2002; 283(3):E413–E422.

2. Craft S. Insulin resistance and Alzheimer’s disease pathogenesis: potential mechanisms and impli-cations for treatment. Curr Alzheimer Res. 2007; 4(2):147–152.

3. de la Monte SM. Insulin resistance and Alzheimer’s disease. BMB Rep. 2009;42(8):475–481.

4. Moloney AM, Griffin RJ, Timmons S, O’Connor R, Ravid R, O’Neill C. Defects in IGF-1 receptor, insulin receptor and IRS-1/2 in Alzheimer’s disease indicate possible resistance to IGF-1 and insulin signalling. Neurobiol Aging. 2010;31(2):224–243.

5. Steen E, et al. Impaired insulin and insulin-like growth factor expression and signaling mecha-nisms in Alzheimer’s disease — is this type 3 diabe-tes? J Alzheimers Dis. 2005;7(1):63–80.

6. Lester-Coll N, Rivera EJ, Soscia SJ, Doiron K, Wands JR, de la Monte SM. Intracerebral streptozotocin model of type 3 diabetes: relevance to sporadic Alzheimer’s disease. J Alzheimers Dis. 2006;9(1):13–33.

7. Chiu SL, Chen CM, Cline HT. Insulin receptor sig-naling regulates synapse number, dendritic plas-ticity, and circuit function in vivo. Neuron. 2008; 58(5):708–719.

8. Dou JT, Chen M, Dufour F, Alkon DL, Zhao WQ. Insulin receptor signaling in long-term memory consolidation following spatial learning. Learn Mem. 2005;12(6):646–655.

9. Zhao WQ, et al. Amyloid beta oligomers induce impairment of neuronal insulin receptors. FASEB J. 2008;22(1):246–260.

10. Haass C, Selkoe DJ. Soluble protein oligomers in neurodegeneration: lessons from the Alzheimer’s amyloid beta-peptide. Nat Rev Mol Cell Biol. 2007; 8(2):101–112.

11. Ferreira ST, Vieira MN, De Felice FG. Soluble protein oligomers as emerging toxins in Alzheimer’s and other amyloid diseases. IUBMB Life. 2007;59(4–5):332–345.

12. Ferreira ST, Klein WL. The Abeta oligomer hypothesis for synapse failure and memory loss in Alzheimer’s disease. Neurobiol Learn Mem. 2011; 96(4):529–543.

13. De Felice FG, et al. Protection of synapses against Alzheimer’s-linked toxins: insulin signaling pre-vents the pathogenic binding of Abeta oligomers. Proc Natl Acad Sci U S A. 2009;106(6):1971–1976.

14. Gong Y, et al. Alzheimer’s disease-affected brain: presence of oligomeric A beta ligands (ADDLs) sug-gests a molecular basis for reversible memory loss. Proc Natl Acad Sci U S A. 2003;100(18):10417–10422.

15. De Felice FG, et al. Alzheimer’s disease-type neuronal tau hyperphosphorylation induced by A beta oligo-mers. Neurobiol Aging. 2008;29(9):1334–1347.

16. De Felice FG, et al. Abeta oligomers induce neuronal oxidative stress through an N-methyl-D-aspartate receptor-dependent mechanism that is blocked by the Alzheimer drug memantine. J Biol Chem. 2007;282(15):11590–11601.

17. Decker H, et al. N-methyl-D-aspartate receptors are required for synaptic targeting of Alzheimer’s toxic amyloid-beta peptide oligomers. J Neurochem. 2010;115(6):1520–1529.

18. Lacor PN, et al. Abeta oligomer-induced aberra-tions in synapse composition, shape, and density provide a molecular basis for loss of connectivity in Alzheimer’s disease. J Neurosci. 2007;27(4):796–807.

19. Lambert MP, et al. Diffusible, nonfibrillar ligands derived from Abeta1-42 are potent central ner-vous system neurotoxins. Proc Natl Acad Sci U S A. 1998;95(11):6448–6453.

20. Townsend M, Mehta T, Selkoe DJ. Soluble Abeta inhibits specific signal transduction cascades com-mon to the insulin receptor pathway. J Biol Chem. 2007;282(46):33305–33312.

21. Escribano L, et al. Rosiglitazone rescues memory impairment in Alzheimer’s transgenic mice: mecha-nisms involving a reduced amyloid and tau patholo-gy. Neuropsychopharmacology. 2010;35(7):1593–1604.

22. Hirosumi J, et al. A central role for JNK in obesity and insulin resistance. Nature. 2002;420(6913):333–336.

23. Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, Spiegelman BM. IRS-1-mediated inhi-bition of insulin receptor tyrosine kinase activity in TNF-alpha- and obesity-induced insulin resistance. Science. 1996;271(5249):665–668.

24. Shankar GM, et al. Amyloid-beta protein dimers isolated directly from Alzheimer’s brains impair synaptic plasticity and memory. Nat Med. 2008; 14(8):837–842.

25. Lesne S, et al. A specific amyloid-beta protein assem-bly in the brain impairs memory. Nature. 2006; 440(7082):352–357.

26. Rui L, et al. Insulin/IGF-1 and TNF-alpha stimulate phosphorylation of IRS-1 at inhibitory Ser307 via distinct pathways. J Clin Invest. 2001;107(2):181–189.

27. Brewer GJ, Boehler MD, Jones TT, Wheeler BC. NbActiv4 medium improvement to Neurobasal/B27 increases neuron synapse densities and net-work spike rates on multielectrode arrays. J Neurosci Methods. 2008;170(2):181–187.

28. Irvine EE, et al. Insulin receptor substrate 2 is a negative regulator of memory formation. Learn Mem. 2011;18(6):375–383.

29. Killick R, et al. Deletion of Irs2 reduces amyloid deposition and rescues behavioural deficits in APP transgenic mice. Biochem Biophys Res Commun. 2009; 386(1):257–262.

30. Bouzakri K, et al. Reduced activation of phosphati-

dylinositol-3 kinase and increased serine 636 phos-phorylation of insulin receptor substrate-1 in prima-ry culture of skeletal muscle cells from patients with type 2 diabetes. Diabetes. 2003;52(6):1319–1325.

31. Ma QL, et al. Beta-amyloid oligomers induce phosphorylation of tau and inactivation of insu-lin receptor substrate via c-Jun N-terminal kinase signaling: suppression by omega-3 fatty acids and curcumin. J Neurosci. 2009;29(28):9078–9089.

32. Cacquevel M, Lebeurrier N, Cheenne S, Vivien D. Cytokines in neuroinflammation and Alzheimer’s disease. Curr Drug Targets. 2004;5(6):529–534.

33. Morfini GA, et al. Pathogenic huntingtin inhibits fast axonal transport by activating JNK3 and phosphory-lating kinesin. Nat Neurosci. 2009;12(7):864–871.

34. Morfini GA, et al. Axonal transport defects in neurodegenerative diseases. J Neurosci. 2009; 29(41):12776–12786.

35. Decker H, Lo KY, Unger SM, Ferreira ST, Silverman MA. Amyloid-beta peptide oligomers disrupt axonal transport through an NMDA receptor-dependent mechanism that is mediated by glycogen synthase kinase 3beta in primary cultured hippocampal neurons. J Neurosci. 2010;30(27):9166–9171.

36. Nakamura T, et al. Double-stranded RNA-dependent protein kinase links pathogen sensing with stress and metabolic homeostasis. Cell. 2010;140(3):338–348.

37. Yuan M, et al. Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disrup-tion of Ikkbeta. Science. 2001;293(5535):1673–1677.

38. Cai D, et al. Local and systemic insulin resistance resulting from hepatic activation of IKK-beta and NF-kappaB. Nat Med. 2005;11(2):183–190.

39. Hundal RS, et al. Mechanism by which high-dose aspirin improves glucose metabolism in type 2 dia-betes. J Clin Invest. 2002;109(10):1321–1326.

40. Ozes ON, et al. A phosphatidylinositol 3-kinase/Akt/mTOR pathway mediates and PTEN antago-nizes tumor necrosis factor inhibition of insulin signaling through insulin receptor substrate-1. Proc Natl Acad Sci U S A. 2001;98(8):4640–4645.

41. Dhamoon MS, Noble JM, Craft S. Intranasal insu-lin improves cognition and modulates beta-amy-loid in early AD. Neurology. 2009;72(3):292–293.

42. Benedict C, Hallschmid M, Schultes B, Born J, Kern W. Intranasal insulin to improve memory function in humans. Neuroendocrinology. 2007;86(2):136–142.

43. Taguchi A, White MF. Insulin-like signaling, nutri-ent homeostasis, and life span. Annu Rev Physiol. 2008;70:191–212.

44. During MJ, et al. Glucagon-like peptide-1 receptor is involved in learning and neuroprotection. Nat Med. 2003;9(9):1173–1179.

45. Cole GM, Frautschy SA. The role of insulin and neu-rotrophic factor signaling in brain aging and Alzheim-


er’s disease. Exp Gerontol. 2007;42(1–2):10–21. 46. Mattson MP, Perry T, Greig NH. Learning from the

gut. Nat Med. 2003;9(9):1113–1115. 47. McClean PL, Gault VA, Harriott P, Holscher C.

Glucagon–like peptide–1 analogues enhance syn-aptic plasticity in the brain: a link between diabe-tes and Alzheimer’s disease. Eur J Pharmacol. 2010; 630(1–3):158–162.

48. Toledo EM, Inestrosa NC. Activation of Wnt sig-naling by lithium and rosiglitazone reduced spa-tial memory impairment and neurodegeneration in brains of an APPswe/PSEN1DeltaE9 mouse model of Alzheimer’s disease. Mol Psychiatry. 2010;15(3):272–285.

49. Ferdaoussi M, et al. Exendin-4 protects beta-cells from interleukin-1 beta-induced apoptosis by interfering with the c-Jun NH2-terminal kinase pathway. Diabetes. 2008;57(5):1205–1215.

50. Natalicchio A, et al. Exendin-4 prevents c-Jun N-ter-minal protein kinase activation by tumor necrosis factor-alpha (TNFalpha) and inhibits TNFalpha-induced apoptosis in insulin-secreting cells. Endo-crinology. 2010;151(5):2019–2029.

51. Janson J, Laedtke T, Parisi JE, O’Brien P, Petersen RC, Butler PC. Increased risk of type 2 diabetes in Alzheimer disease. Diabetes. 2004;53(2):474–481.

52. Carlson CJ, White MF, Rondinone CM. Mamma-lian target of rapamycin regulates IRS-1 serine 307 phosphorylation. Biochem Biophys Res Commun. 2004;316(2):533–539.

53. Fisher TL, White MF. Signaling pathways: the benefits of good communication. Curr Biol. 2004; 14(23):R1005–R1007.

54. Lafay-Chebassier C, et al. mTOR/p70S6k signalling alteration by Abeta exposure as well as in APP-PS1 transgenic models and in patients with Alzheimer’s disease. J Neurochem. 2005;94(1):215–225.

55. Caccamo A, Majumder S, Richardson A, Strong R, Oddo S. Molecular interplay between mammalian

target of rapamycin (mTOR), amyloid-beta, and Tau: effects on cognitive impairments. J Biol Chem. 2010;285(17):13107–13120.

56. Spilman P, et al. Inhibition of mTOR by rapamycin abolishes cognitive deficits and reduces amyloid-beta levels in a mouse model of Alzheimer’s disease. PLoS One. 2010;5(4):e9979.

57. Ma T, et al. Dysregulation of the mTOR pathway mediates impairment of synaptic plasticity in a mouse model of Alzheimer’s disease. PLoS One. 2010; 5(9):e12845.

58. Morel M, Couturier J, Lafay-Chebassier C, Paccalin M, Page G. PKR, the double stranded RNA-depen-dent protein kinase as a critical target in Alzheimer’s disease. J Cell Mol Med. 2009;13(8A):1476–1488.

59. Selkoe DJ. Resolving controversies on the path to Alzheimer’s therapeutics. Nat Med. 2011; 17(9):1060–1065.

60. Hotamisligil GS, Murray DL, Choy LN, Spiegelman BM. Tumor necrosis factor alpha inhibits signaling from the insulin receptor. Proc Natl Acad Sci U S A. 1994;91(11):4854–4858.

61. Uysal KT, Wiesbrock SM, Marino MW, Hotamis-ligil GS. Protection from obesity-induced insulin resistance in mice lacking TNF-alpha function. Nature. 1997;389(6651):610–614.

62. Wiedmann MW, Mossner J, Baerwald C, Pierer M. TNF alpha inhibition as treatment modality for certain rheumatologic and gastrointestinal diseas-es. Endocr Metab Immune Disord Drug Targets. 2009; 9(3):295–314.

63. Araujo EP, et al. Infliximab restores glucose homeo-stasis in an animal model of diet-induced obesity and diabetes. Endocrinology. 2007;148(12):5991–5997.

64. Tobinick E. Perispinal etanercept: a new therapeu-tic paradigm in neurology. Expert Rev Neurother. 2010;10(6):985–1002.

65. Copps KD, Hancer NJ, Opare-Ado L, Qiu W, Walsh C, White MF. Irs1 serine 307 promotes insulin sen-

sitivity in mice. Cell Metab. 2010;11(1):84–92. 66. Aguirre V, Uchida T, Yenush L, Davis R, White MF.

The c-Jun NH(2)-terminal kinase promotes insulin resistance during association with insulin receptor substrate-1 and phosphorylation of Ser(307). J Biol Chem. 2000;275(12):9047–9054.

67. Aguirre V, Werner ED, Giraud J, Lee YH, Shoelson SE, White MF. Phosphorylation of Ser307 in insu-lin receptor substrate-1 blocks interactions with the insulin receptor and inhibits insulin action. J Biol Chem. 2002;277(2):1531–1537.

68. Freude S, et al. Neuronal IGF-1 resistance reduces Abeta accumulation and protects against premature death in a model of Alzheimer’s disease. FASEB J. 2009;23(10):3315–3324.

69. Klyubin I, et al. Amyloid beta protein immunother-apy neutralizes Abeta oligomers that disrupt synap-tic plasticity in vivo. Nat Med. 2005;11(5):556–561.

70. Relkin NR, et al. 18-Month study of intravenous immunoglobulin for treatment of mild Alzheimer disease. Neurobiol Aging. 2009;30(11):1728–1736.

71. Talbot K, et al. Dysbindin-1 is reduced in intrin-sic, glutamatergic terminals of the hippocampal formation in schizophrenia. J Clin Invest. 2004; 113(9):1353–1363.

72. Amaral DG, Insausti R, Campbell MJ. Distribution of somatostatin immunoreactivity in the human dentate gyrus. J Neurosci. 1988;8(9):3306–3316.

73. Kwinter DM, Lo K, Mafi P, Silverman MA. Dynactin regulates bidirectional transport of dense-core vesicles in the axon and dendrites of cultured hippocampal neurons. NeuroScience. 2009;162(4):1001–1010.

74. Bondolfi L, et al. Amyloid-associated neuron loss and gliogenesis in the neocortex of amyloid precursor pro-tein transgenic mice. J Neurosci. 2002;22(2):515–522.

75. Gengler S, McClean PL, McCurtin R, Gault VA, Hölscher C. Val(8)GLP–1 rescues synaptic plasticity and reduces dense core plaques in APP/PS1 mice. Neurobiol Aging. 2012;33(2):265–276.

Supplemental Figure 1 . Immunohistochemical specificity of Cell Signaling antibody CS2388 to IRS-1pS636/639 demonstrated in the hippocampus of AD cases. The antibody reveals the phosphospecific antigen in the cytoplasm of neurons (A). Immunoreactivity was blocked by preadsorption of the antibody with a 5X molar excess of the phosphorylated immunogen (B). No appreciable inhibition of immunoreactivity was seen, however, after preadsorption of the antibody with a 5X molar excess of non-phosphorylated IRS-1 amino acid sequence 631-646 (Abcam ab41777) (C). Identical results were obtained on five AD cases.

A IRS-1pSer636 B *300

200

100

AβO

veh

636

307

312

616

C

IRS

-1pS

er

(% o

f veh

icle

)

0

Supplemental Figure 2 . Aβ oligomers (AβOs) induce the increase in IRS-1pSer levels in mature hippocampal neurons grown in insulin-free medium. IRS-1pSer636 immunolabeling in hippocampal neurons exposed to vehicle (A) or 500 nM AβOs (B) for 3 h (Scale bars = 20μm). C, integrated IRS-1pSer immunofluorescence levels determined from 2 experi-ments using independent cultures (30 images analyzed/experimental condition/experiment, each experiment carried out in triplicate). * statistically significant differences (p<0.05, ANOVA followed by Bonferroni post-hoc test) relative to vehicle-treated cultures.

*

*

M1 M2 M3S

IRS-

1pSe

r636

(103 p

ixel

/fiel

d)

Supplemental Figure 3 . Elevated IRS-1pSer636 levels in the temporal corti-ces of cynomolgus monkeys that received intracerebroventricular injections of

AβOs. IRS-1pSer636 immunoreactivities in the same segments of the tempo-ral cortex from a control (sham-operated) monkey (A) and 3 different mon-

keys that received AβOs injections (B-D). E, IRS-1pSer636 immunolabeling density (pixels -ral cortices of sham (S) or oligomer-injected monkeys (M1-M3). DAPI staining is in blue. * statistically significant (p<0.001 ANOVA followed by Bonferroni

post-hoc test) differences relative to the sham-operated monkey.

Supplemental Figure 4 . Total c-Jun N terminal kinase (JNK) levels in hippo-campal cell cultures exposed to AβOs and in the hippocampi of transgenic mice. Immunoblotting for JNK levels in hippo-campal neurons exposed to vehicle or 500 nM AβOs A) and in h ip p o c a mp a l h o mo g e n a t e s f r o m APPSwe,PS1deltaE9 transgenic mice (Tg; n=7) or wild-type mice (WT; n=5) (B). Graphs show quantification of JNK levels in cultures (A) or in transgenic mice (B) normalized by cyclophilin B as a loading control.

Supplemental Figure 5 : Schematic representation of axonal transport experiments: (A) Pri-mary hippocampal neurons (prepared as described, 35) were transfected (at days in vitro) to express BDNF-mRFP, a dense-core vesicle (DCV) cargo, or mitochondrially-targeted YFP. Cells were allowed to express constructs for 24 h before imaging or immunocytochem-istry. Axons and dendrites were initially distinguished based on morphology and confirmed retrospectively by antibody staining against MAP-2, a dendritic cytoskeletal protein. (B) Axons were live imaged using a wide-field fluorescent microscope (DMI 6000 B, Leica) as previously described (35). (C) Bidirectional axonal transport of BDNF-DCVs or mitochondria. Anterograde and retrograde transport are indicated by green and red arrows, respectively. Pannels A and B adapted from 73.

A B

C

Supplemental Figure 6 . Exendin-4 does not interfere with AβO binding to neurons. Hippocampal neurons were exposed for 3 h to vehicle (A), 500 nM AβOs (B nM AβOs (C) or 300 nM exendin-4 + 500 nM AβOs (D) and AβO binding was detected using oligomer-specific NU4 E) Integrated oligomer immunofluorescence determined from 5 experiments using indepen-dent cultures (25 images analyzed per experimental condition per experiment). * and #, statistically significant differences relative to vehi-cle-treated neurons (p < 0.01) or AβO-treated neurons (p < 0.001), respectively.

E

#

*

AβO

imm

unor

eact

. (%

of v

ehic

le)

GL

P1

-R/

cycl

o B

(%

of W

T)

\\\ WT Tg

GLP1R 56

cyclo B 19

BWT Tg KDa

GL

P1

-R/

cycl

o B

(%

of

veh

icle

)

Veh AβO

cyclo B 19

GLP1R 56

Vehicle A βOs KDa

A

Supplemental Figure 7 . Glucagon-like peptide 1 receptor (GLP1R) levels in hippocampal cell cultures exposed to AβOs and in the hippocampi of transgenic mice. Immunoblotting for GLP1R levels in hippocampal neurons exposed to vehicle or 500 nM AβOs A) and in hippocampal h o mo g e n a t e s f r o m A P P S we , P S 1 d e l t a E 9 transgenic mice (Tg; n=7) or wild-type mice (WT; n=5) (B). Graphs show densitomteric quantification of GLP1R levels in cultures (A) or in transgenic mice (B) normalized by cyclophilin B as a loading control.

WT PBS WT Ex-4 Tg PBSU Tg Ex-4

Before treatment

After treatment

Bod

y w

eigh

t (g)

Supplemental Figure 8 . Lack of effect of 3-week intraperitoneal administration of exendin-4 on body weight of APPSwe,PS1deltaE9 mice. Transgenic mice (n=7) or wild-type mice (n=5) received intra-peritoneal exendin-4 (25 nmol/kg bw) or saline (0.9% w/v) injections once-daily for 3 weeks, for the duration of the behavioral experiments.

Supplemental Videos Legends Supplemental Video 1. Role of JNK in disruption of dense-core vesicle (DCV) transport

induced by Aβ oligomers. Hippocampal cultures expressing the DCV cargo BDNF-mRFP were

used. Video shows live imaging of DCV transport in representative axons from hippocampal

neurons exposed to vehicle, 500 nM oligomers or 10μM SP600125 + 500 nM oligomers for 18

hours. Quantification of DCV transport parameters is shown in Supplemental Table 2.

Supplemental Video 2. Exendin-4 and insulin prevent Aβ oligomer-induced disruption of

dense-core vesicle (DCV) transport. Hippocampal cultures expressing the DCV cargo BDNF-

mRFP were used. Video shows live imaging of DCV transport in representative axons from

cultures exposed to 300 nM exendin-4 + 500 nM oligomers or 1 μM insulin + 500 nM oligomers

for 18 hours. Quantification of DCV transport parameters is shown in Supplemental Table 2.

Supplemental Video 3. Exendin-4 blocks Aβ-oligomer-induced disruption of mitochondria

transport. Hippocampal cultures expressing mitochondrially-targeted YFP were used. Video

shows time-lapse imaging of mitochondria transport in representative axons from cultures

exposed to 300 nM exendin-4 + 500 nM oligomers for 18 hours. Quantification of mitochondria

transport parameters is shown in Supplemental Table 3.

1

Supplemental Table 1: Demographic and Autopsy Data on Human Cases

Variable NCI AD Number of Cases 22 22 Age (y, mean ± SD)* 71.45 ± 13.44 72.77 ± 11.13 Males/Females 10/12 10/12 PMI (h, mean ± SD)*

12.43 ± 5.18 10.43 ± 4.24

AD = Alzheimer’s Disease Cases, NCI = Non-Cognitively Impaired Cases *No significant difference between NCI and AD groups

2

Supplemental Table 2: Quantitative analysis of DCV transport

DCVs Traffic values % All events Anterograde Retrograde All events

Flux (min-1)

Vehicle 18h 10.28 ± 1.49 6.60 ± 1.18 3.67 ± 0.55 # 100.00 ±

14.09 AβOs 18h 2.55 ± 0.49 1.40 ± 0.29 ** 1.15 ± 0.23 ** 24.78 ± 4.75 **

AβOs scrambled 8.05 ± 0.55 4.42 ± 0.41 ++ 3.63 ± 0.44 ++ 78.29 ± 5.35 ++

Insulin + AβOs 10.35 ± 1.39 6.13 ± 0.85 ++ 4.26 ± 0.81 ++ 100.64 ±

13.55 ++

Exendin + AβOs 8.68 ± 1.29 4.35 ± 0.59 ++ 4.34 ± 0.85 ++ 84.43 ± 12.63 ++

JNK inhibitor + AβOs 7.23 ± 1.23 3.40 ± 0.80 */+ 3.83 ± 0.54 ++ 70.35 ± 12.01 ++

Velocity (µm/s) Vehicle 18h 1.66 ± 0.11 1.66 ± 0.10 1.63 ± 0.12 100.00 ± 6.68AβOs 18h 1.32 ± 0.07 1.32 ± 0.08 * 1.34 ± 0.07 79.94 ± 4.15 *

AβOs scrambled 1.68 ± 0.09 1.74 ± 0.07 + 1.59 ± 0.14 101.51 ± 5.72 +

Insulin + AβOs 1.75 ± 0.09 1.82 ± 0.09 ++ 1.63 ± 0.09 + 105.44 ± 5.19 +Exendin + AβOs 1.39 ± 0.10 1.37 ± 0.10 1.40 ± 0.11 83.81 ± 6.19

JNK inhibitor + AβOs 1.96 ± 0.08 1.94 ± 0.09 1.95 ± 0.08 * 118.61 ± 4.73 *

Run length (µm) Vehicle 18h 8.45 ± 0.69 9.53 ± 0.83 6.78 ± 0.51 100.00 ± 8.19AβOs 18h 5.20 ± 0.33 5.30 ± 0.42 ** 4.90 ± 0.39 * 61.56 ± 3.85 **

AβOs scrambled 6.03 ± 0.45 6.52 ± 0.56 * 5.39 ± 0.46 71.32 ± 5.31 *

Insulin + AβOs 7.72 ± 0.61 8.53 ± 0.76 ++ 6.70 ± 0.56 + 91.37 ± 7.23 ++Exendin + AβOs 7.71 ± 0.76 8.02 ± 0.73 + 7.26 ± 0.94 + 91.19 ± 8.98 +

JNK inhibitor + AβOs 5.13 ± 0.31 5.00 ± 0.42 ** 5.06 ± 0.30 * 60.74 ± 3.71 ** Vehicle n=21 kymographs (21 cells, 3669 vesicles) / AβOs n=25 kymographs (25 cells, 1332 vesicles) AβOscr. n=10 kymographs (10 cells, 1917 vesicles) / Insulin + AβOs n=14 kymographs (14 cells, 3273 vesicles)

Exendin + AβOs n=15 kymographs (15 cells, 2280 vesicles) / JNK inhibitor + AβOs n=15 kymographs (15 cells, 2336 vesicles)* p<0.05 when compared with vehicle (from each column) ** p<0.0001 when compared with vehicle (from each column) + p<0.05 when compared with AβOs (from eachcolumn) ++ p<0.0001 when compared with AβOs (from each column) # p<0.0001 when compared vehicle anterograde with vehicle retrograde

3

4

Supplemental Table 3: Quantitative analysis of mitochondria transport

Mitochondria

Traffic values % All events Anterograde Retrograde All events

Flux (min-1) Vehicle 18h 0.24 ± 0.05 0.10 ± 0.03 0.14 ± 0.03 100.00 ± 21.70 AβOs 18h 0.06 ± 0.03 0.02 ± 0.01 * 0.04 ± 0.02 * 26.56 ± 11.84 *

Insulin + AβOs 0.24 ± 0.03 0.12 ± 0.02 + 0.12 ± 0.03 100.83 ± 14.51 ++

Exendin + AβOs 0.22 ± 0.03 0.11 ± 0.03 + 0.11 ± 0.02

+ 93.79 ± 13.94 +

Velocity (µm/s) Vehicle 18h 0.42 ± 0.04 0.30 ± 0.04 0.46 ± 0.07 100.00 ± 8.73 AβOs 18h 0.27 ± 0.07 0.10 ± 0.05 * 0.35 ± 0.10 63.32 ± 18.02

Insulin + AβOs 0.46 ± 0.05 0.39 ±0.07 + 0.46 ± 0.07 108.89 ± 12.91 Exendin + AβOs 0.42 ± 0.04 0.34 ± 0.04 + 0.47 ± 0.07 100.32 ± 9.31

Run length (µm)

Vehicle 18h 8.26 ± 1.07 5.36 ± 0.89 8.89 ± 1.50 100.00 ± 12.93

AβOs 18h 3.27 ± 0.98 1.89 ± 0.94

** 2.80 ± 0.95 * 39.60 ± 11.87 *

Insulin + AβOs 6.88 ± 0.79 6.49 ± 1.12

++ 5.83 ± 0.98

+ 83.31 ± 9.51 +

Exendin + AβOs 9.06 ± 1.06 9.35 ± 2.11 + 8.99 ± 1.90

+ 109.70 ± 12.86 +

Vehicle n=15 kymographs (15 cells, 229 vesicles) / AβOs n=15 kymographs (15 cells, 56 vesicles) Insulin + AβOs n=13 kymographs (13 cells, 275 vesicles) / Exendin + AβOs n=14 kymographs (14 cells, 257 vesicles) *p<0.05, when compared with vehicle (from each column)

**p<0.0001, when compared with vehicle (from each column)

+ p<0.05 when compared with AβOs (from each column) ++ p<0.0001 when compared with AβOs (from each column)

80

5) DISCUSSÃO

O entendimento das bases moleculares da disfunção neuronal e na perda

de memória na doença de Alzheimer se tornou um grande desafio da

neurosciencia e de saúde pública. E embora a espectativa do número de casos

estimados para as próximas décadas seja assustadora, a idéia de um tratamento

efetivo capaz de desacelerar a progressão da doença ainda está longe de ser

concreta (SELKOE, 2011; SELKOE, 2012). Apenas um pequeno número dos

casos de Alzheimer são atribuidos a fatores genéticos (VAN ES e VAN DEN

BERG, 2009), por outro lado, os mecanismos da patogênese e a etiologia da

forma esporádica da doença de Alzheimer, que compreende a grande maioria dos

casos, não foram totalmente elucidados. Desta forma, a identificação de

componestes moleculares que contribuem para esta doença neurológica complexa

se tornou o foco de intensos esforços nas pesquisas científicas nos últimos anos.

Uma das questões misteriosas da DA é porque especificamente a memória

é afetada nesta doença, e isto tem sido alvo de intensa investigação. É crescente

o número de evidências que mostram que a deterioração morfológica e funcional

de sinapses, centrais para a formação de memória, é induzida pelos oligômeros

do peptídeo β amilóide (Aβ). Sabe-se que estes oligômeros de Aβ, conhecidos

como AβOs, são bastante tóxicos aos neurônios e interferem em diversas vias de

sinalização fisiológicas, desencadeando mecanismos que culminam em disfunção

neuronal (KRAFFT e KLEIN, 2010). Dentre estes, são de destaque o estresse

oxidativo (DE FELICE e cols., 2007; SARAIVA e cols., 2010), que promove

alterações celulares globais, e a desregulação no tráfego e inserção de receptores

81

sinápticos (LACOR e cols., 2007; DE FELICE e cols., 2009; JURGENSEN e cols.,

2011), que certamente prejudica os mecanismos de plasticidade e resposta

sinápticas no cérebro. Somado a estes efeitos deletérios, surge o evidente

prejuízo à sinalização insulinérgica induzida por essas toxinas no cérebro (ZHAO e

cols., 2008; DE FELICE e cols., 2009; MA e cols., 2009).

Estudos clínicos e epdemioógicos demonstram que pacientes de diabetes

tipo II apresentam um risco de desenlvover a doença de Alzheimer (DE LA

MONTE e WANDS, 2008; MURTHY e cols., 2008). No presente trabalho, nós

descrevemos um mecanismo patogênico pelo qual um tipo de diabetes cerebral é

induzida pelos oligômeros de Aβ, sinaptoxinas que acumulam no cérebro de

pacientes com Alzheimer. Sabendo que a sinalização por insulina é central para o

aprendizado e memória (DOU e cols., 2005), podemos propor que o impacto dos

AβO sobre a sinalização insulinérgica cerebral poderia contribuir potencialmente

para a perda de memória, principal sintoma da doença de Alzeimer.

Inicialmente, o trabalho focalizou os efeitos locais dos oligômeros, isto é,

alterações observadas nos prolongamentos neuronais, já descritos como

principais regiões de contato sináptico (KAECH e BANKER, 2006) e também sítios

preferenciais de ligação dos AβOs (LACOR e cols., 2004).

De forma interessante, nossos achados indicam que mecanismos bastante

similares aos que causam resistência à insulina periférica no diabetes tipo 2 foram

observados em nossos modelos para a doença de Alzheimer. Os resultados

demonstram que os AβOs induzem resistência à insulina nos neurônios

hipocampais através da sinalização de TNF-α ativando quinases de estresse

celular como JNK, Ikk e PKR, levando a fosforilação do IRS-1 em resíduos de

82

serina. De fato, análises histopatológicas do cérebro de pacientes com Alzheimer

confirmaram a presença da marcação aumentada para o IRS-1 fosforilado em

resíduos de serina636/639, o qual é tipicamente associado com a resistência à

insulina periférica (HOTAMISLIGIL e cols., 1996; WHITE, 2002) em comparação

com indivíduos não demenciados. Recentemente foi demonstrada a redistribuição

do IRS-1pSer no cérebro de pacientes da DA (MA e cols., 2009). Esta questão foi

descrita previamente por nosso grupo, onde demonstramos que os AβOs

promovem a internalização dos receptores de insulina (RI) (ZHAO e cols., 2008;

DE FELICE e cols., 2009), de forma que é possível que a remoção do RI pode

contribuir, ou mesmo, facilitar o aumento dos níveis de IRSpSer. O que se dispões

de acordo com o nosso achado que revela a insulina, que previamente bloqueou a

internalização dos RI induzida pelos AβO (DE FELICE e cols., 2009), ainda

previne a fosforilação do IRS-1 em resíduos de serina.

Nos casos de resistência à insulina periférica, a fosforilação do IRS-1 em

resíduos de serina tem como uma de suas principais quinases a mTOR (OZES e

cols., 2001; CARLSON e cols., 2004). No entanto, em nosso modelo de culturas

maduras primárias de hipocampo de embrião de rato expostas aos oligômeros de

Aβ, a inibição do IRS-1 parece não envolver a ativação de mTORC1, uma vez que

o uso prévio da rapamicina não protegeu contra o aumento dos níveis de IRS-

1pSer. Embora o papel da mTOR na patogênese da DA seja bastante

controverso, a analise de sua atividade se revelou aumentada no cérebro de

camundongos triplo transgênico para a DA (LAFAY-CHEBASSIER e cols., 2005).

Por outro lado, a inibição de sua atividade pela rapamicina promoveu melhora

cognitiva em outro modelo transgênico da doença de Alhzeimer (CACCAMO e

83

cols., 2010; SPILMAN e cols., 2010). Outro estudo demonstrou que a

desregulação de sua atividade está correlacionada com a disfunção da

paslticidade sináptica, uma vez que a superexpressão de mTOR foi capaz de

recuperar a LTP no modelo transgênico para Alzheimer Tg2576 (MA e cols.,

2010). Portanto, diante dos controversos achados, é plausível pensar que estudos

adicionais seria necessários para desvendar o possível papel da sinalização da

mTOR na DA.

De acordo com a hipótese inicial, nós encontramos elevados níveis de

pJNK, em sua forma ativa, no cérebro de pacientes com Alzheimer, assim como

no hipocampo de animais transgênicos (APP/PS1) para a DA. Esta ativação da

JNK também foi identificada em um modelo de triplo transgênico para DA,

alimentados com uma dieta hipercalórica (MA e cols., 2009). Supreendentemente,

a injeção de AβOs no cérebro de macacos, revelou elevados níveis de IRS-

1pSer636 e pJNK, indicando que a inibição da via de sinalização por insulina

cerebral induzida pelos AβO, através da ativação de JNK, ocorre in vivo e

possivelmente é uma característica patológica da doença de Alzheimer.

Sabe-se, que a inflamação é central na indução de resistência à insulina na

obesidade e no diabetes tipo 2 (HOTAMISLIGIL, 2006; HOTAMISLIGIL e ERBAY,

2008). Nossos dados confirmam a idéia proposta, de que a resistência à insulina

neuronal induzida pelos oligômeros de Aβ é mediada por uma resposta

inflamatória através da via de TNF-α, uma vez que o anticorpo capaz de neutraliza

o TNF-α solúvel, protegeu os neurônios contra a inibição do IRS-1 induzida pelos

oligômeros. Também usado no tratamento de doenças inflamatórias (WIEDMANN

e cols., 2009), o infliximab e seu similar, etanercept, foram recentemente

84

sugeridos como novos agentes terapêuticos no tratamento da resistência à

insulina no diabetes tipo 2 (ARAUJO e cols., 2007). Outro estudo sugere sua

eficácia na melhora do declínio cognitivo em pacientes com Alzheimer (TOBINICK,

2009). Esse achado é reforçado por nossos dados, uma vez que o infliximab

protegeu os neurônios da desregulação insulinérgica induzida pelos oligômeros, e

desse modo se torna um potente candidato para o desenvolvimento de novas

abordagens terapêuticas contra a perda de memória na DA.

Diferentes do que esperávamos, nós não observamos mudanças nos níveis

de IRS-1p307 no hipocampo de animais transgênicos para DA, por outro lado, nas

culturas de hipocampo expostas aos oligômeros de Aβ apresentaram um aumento

siginificativo dos níveis do IRS-1 fosforilado nesta serina 307. Um estudo recente

demonstrou que em camundongos, este sítio de fosforilação o qual favorece

regulação positiva da sinalização por insulina (COPPS e cols., 2010), contraiando

resultados prévios obtidos em culturas de células (AGUIRRE e cols., 2000; RUI e

cols., 2001; AGUIRRE e cols., 2002). Com base em nossos resultados, seria

coerente afirmar que após a ligação dos oligômeros nas sinapses, a fosforialção

do IRS-1 em múltiplos resíduos de serina poderia acontecer através da ativação

das quinases de estresse celular, PKR, Ikk e JNK. Além disso, mesmo que a

serina 307 não seja fosforilada, o resultado final com a fosforilação do IRS-1 em

outros diferentes resíduos já seria suficientemente crítico para regular a

resistência à insulina neuronal, da mesma forma como ocorre nos tecidos

periféricos (WHITE, 2002).

Os níveis de IRS-2 estavam aumentados nas culturas hipocampis espoxtas

aos oligômeros de Aβ. Este fenômenopode estar relacionado ao fato de que em

85

estudos prévios, o IRS-2 foi considerado um regulador negativo na formação de

memória (IRVINE e cols., 2011). Outro estudo demonstrou que a deleção de IRS-2

reduz a deposição de Aβ e o déficit cognitivo em camundongos transgênicos para

a DA (KILLICK e cols., 2009). Em nossas análises do hipocampo de camundongos

transgênicos (APP/PS1) para DA, observamos que os níveis de IRS-2 estavam

reduzidos significativamente. Resultado semelhante foi encontrado em cérebros

de pacientes com Alzheimer (MOLONEY e cols., 2010), o que sugere um

fenômeno compensatório para reduzir o impacto negativo do IRS-2 sobre a

memória na DA (FREUDE e cols., 2009). Portanto, é possível que o IRS-1 tenha

um papel de regular positivamente a memória, como sugerido pelos nossos

resultados e por estudos prévios, enquanto o IRS-2 contribua como um regulador

negativo da formação da memória.

Diante do crescente número de evidências, os oligômeros de Aβ vem sendo

considerados como os principais responsáveis pela disfunção sináptica (LESNE e

cols., 2006; SHANKAR e cols., 2008; DE FELICE e cols., 2009; FERREIRA e

KLEIN, 2011). Desta forma, os AβO constituem um alvo central para o

desenvolvimento de fármacos que forneçam resultados positivos nos teste clínicos

da DA (KLYUBIN e cols., 2005; RELKIN e cols., 2009). Recentemente a

estimulação da sinalização por insulina tem sido sugerida como uma abordagem

promissora para prevenir ou impedir o declínio cognitivo dos pacientes com DA

(DHAMOON e cols., 2009). Neste sentido, a ação protetora da insulina em

recuperar a inibição do IRS-1, como demonstrado por nossos resultados,

provavelmente deriva da sua habilidade de bloquear a ligação dos oligômeros aos

neurônios (DE FELICE e cols., 2009).

86

Entretanto, uma vez que os receptores de insulina são removidos das

membrandas neuronais no cérebro de pacientes com DA (MOLONEY e cols.,

2010) e ainda nas culturas expostas aos oligômeros de Aβ (ZHAO e cols., 2008),

sugerindo que o uso da insulin somente pode não ser a forma de combate mais

efetiva.

Neste sentido, abordagens alternativas que estimulem a via de sinalização

por insulina por um caminho diferente do receptor de insulina podem representar

uma estratégia segura para tratar a DA. Nossos resultados demonstram que a

exendina-4, um composto antidiabético que estimula a sinalização por insulina

através da ativação do receptor de GLP-1, foi capaz de proteger os neurônios

contra a inibição do IRS-1 induzida pelos AβO. Adicionalmente, aexendina-4

reduziu a ativação da JNK e consequentemente, a inibição do IRS-1 em animais

transgênicos para a DA. Sugreindo essa via de sinalização através do receptor de

GLP1 como uma importante alternativa para bloquear a resistência à insulina e a

perda de memória na DA.

Buscando entender melhor o papel dos componentes inflamatórios na

patogênese da DA, investigamos outra importante via pró-inflamatória, a

sinalização por IL-1β. Recentes estudos fornecem fortes evidencias sugerindo um

papel central da inflamação crônica na patogênese de doenças metabólicas,

assim como na DA. Elevados níveis de mediadores inflamatórios circulantes,

como TNF-α e IL-1β são marcadores importantes da inflamação crônica no

diabetes tipo 2 (HOTAMISLIGIL, 2010; DONATH e SHOELSON, 2011). Nossos

dasos adicionais demonstraram um importante papel da sinzalição de IL-1β na

toxicidade induzida pelos oligômeros de Aβ em neurônios hipocampais. A inibição

87

do IRS-1 pelos oligômeros foi prevenida na presença de um antagonista do

receptor de IL-1β, sugerindo sua participação neste fenômeno. Da mesma forma,

o déficit de memória induzido pela injeção i.c.v. de oligômeros não ocorreu nos

animais IIL-1β KO, o que nos indica um importante papel desta citocina em mediar

esse quadro característico da doença de Alzheimer.

Sabe-se que a IL-1β é uma potente citocina pró-inflamatória do sistema

nervoso central e que se apresenta cronicamente superexpressa em pacientes

com Alzheimer, estimulando um ciclo inflamatório que contrinui para o quadro

patoglógico na DA. De forma interessante, a sinalização por IL-1β na patogênese

de diabetes tipo 2 apresenta mecanismos de ação distintos muito similares aos

observados na doença de Alzheimer. O primeiro deles é a ativação a JNK, o

segundo é a eficiente evoacação da expressão de outras citocinas pró-

inflamatórias favorecendo a amplificação do sinal inflamatório (AREND e cols.,

2008). Por último, IL-1β é capaz de induzir estesse celular, tal como estresse de

retículo e estresse oxidativo, os quais estão relacionados com a patogênese do

diabetes tipo 2 e também da DA (CARDOZO e cols., 2005; VERMA e DATTA,

2010). Os resultados adicionais obtidos por nosso grupo, em conjunto com essas

informações, destacam a sinalização de IL-1β, como um importante alvo

terapêutico para tratamento da DA.

Os resultados apresentados, portanto, descrevem um mecanismo

patogênico comum entre a doença de Alzheimer e o diabetes tipo 2, além de

apontar alvos moleculares para o desenvolvimento de estratégias antidiabéticas

na terapia desta demência. Esta possibilidade se torna especialmente interessante

por se tratarem de medicamentos já em uso na prática clínica, com risco e

88

farmacologia conhecidos, o que reduz as etapas para sua possível implementação

na terapia da doença de Alzheimer.

89

6) CONCLUSÕES

Os resultados apresentados nesta tese permitem concluir que:

9 Os oligômeros de Aβ induzem a inibição do IRS-1 in vitro e in vivo.

9 Os elevados níveis de IRS-1pSer é um indicativo da presença de disfunção

da sinalização por insulina no cérebro de um modelo transgênico para a

doença de Alzheimer, no cérebro de um novo modelo de primatas não-

humanos para Alzheimer e ainda no cérebro de pacientes acometidos pela

doença.

9 A indução da fosforilação inibitória do IRS-1 em neurônios parece ocorrer

através de vias de pró-inflamatórias, incluindo a sinalização por TNF-α e

JNK, bem como a via de IL-1β. Estes mecanimos são similares aos

envolvidos na patogênese do diabetes.

9 A estimulação da sinzalização por GLP-1, estratégia já utilizada como

terapia anti-diabetes, bloqueia a inibição do IRS-1 induzida pelos

oligômeros de Aβ in vitro e in vivo.

90

7) REFERÊNCIAS

ABBAS T., FAIVRE E. e HOLSCHER C. (2009) Impairment of synaptic plasticity and memory formation in GLP-1 receptor KO mice: Interaction between type 2 diabetes and Alzheimer's disease. Behav.Brain Res. 205:1, 265-271.

AGUIRRE V., UCHIDA T., YENUSH L., DAVIS R. e WHITE M. F. (2000) The c-Jun NH(2)-terminal kinase promotes insulin resistance during association with insulin receptor substrate-1 and phosphorylation of Ser(307). J.Biol.Chem. 275:12, 9047-9054.

AGUIRRE V., WERNER E. D., GIRAUD J., LEE Y. H., SHOELSON S. E. e WHITE M. F. (2002) Phosphorylation of Ser307 in insulin receptor substrate-1 blocks interactions with the insulin receptor and inhibits insulin action. J.Biol.Chem. 277:2, 1531-1537.

AGUZZI A., BARRES B. A. e BENNETT M. L. (2013) Microglia: scapegoat, saboteur, or something else? Science 339:6116, 156-161.

ALZHEIMER A., STELZMANN R. A., SCHNITZLEIN H. N. e MURTAGH F. R. (1995) An English translation of Alzheimer's 1907 paper, "Uber eine eigenartige Erkankung der Hirnrinde". Clin.Anat. 8:6, 429-431.

ARAUJO E. P., DE SOUZA C. T., UENO M., CINTRA D. E., BERTOLO M. B., CARVALHEIRA J. B., SAAD M. J. e VELLOSO L. A. (2007) Infliximab restores glucose homeostasis in an animal model of diet-induced obesity and diabetes. Endocrinology 148:12, 5991-5997.

AREND W. P., PALMER G. e GABAY C. (2008) IL-1, IL-18, and IL-33 families of cytokines. Immunol.Rev. 223, 20-38.

ARKAN M. C., HEVENER A. L., GRETEN F. R., MAEDA S., LI Z. W., LONG J. M., WYNSHAW-BORIS A., POLI G., OLEFSKY J. e KARIN M. (2005) IKK-beta links inflammation to obesity-induced insulin resistance. Nat.Med. 11:2, 191-198.

BALLARD C., GAUTHIER S., CORBETT A., BRAYNE C., AARSLAND D. e JONES E. (2011) Alzheimer's disease. Lancet 377:9770, 1019-1031.

91

BANTING F. G., BEST C. H., COLLIP J. B., CAMPBELL W. R. e FLETCHER A. A. (1922) Pancreatic Extracts in the Treatment of Diabetes Mellitus. Can.Med.Assoc.J. 12:3, 141-146.

BARRERA J. G., SANDOVAL D. A., D'ALESSIO D. A. e SEELEY R. J. (2011) GLP-1 and energy balance: an integrated model of short-term and long-term control. Nat.Rev.Endocrinol 7:9, 507-516.

BAUMANN C. A., RIBON V., KANZAKI M., THURMOND D. C., MORA S., SHIGEMATSU S., BICKEL P. E., PESSIN J. E. e SALTIEL A. R. (2000) CAP defines a second signalling pathway required for insulin-stimulated glucose transport. Nature 407:6801, 202-207.

BENNETT M. R. (2000) The concept of long term potentiation of transmission at synapses. Prog.Neurobiol. 60:2, 109-137.

BERTRAM L., LILL C. M. e TANZI R. E. (2010) The genetics of Alzheimer disease: back to the future. Neuron 68:2, 270-281.

BERTRAM L. e TANZI R. E. (2005) The genetic epidemiology of neurodegenerative disease. J.Clin.Invest 115:6, 1449-1457.

BIESSELS G. J., VAN DER HEIDE L. P., KAMAL A., BLEYS R. L. e GISPEN W. H. (2002) Ageing and diabetes: implications for brain function. Eur.J.Pharmacol. 441:1-2, 1-14.

BROOKMEYER R., JOHNSON E., ZIEGLER-GRAHAM K. e ARRIGHI H. M. (2007) Forecasting the global burden of Alzheimer's disease. Alzheimers.Dement. 3:3, 186-191.

BROWN K. L., COSSEAU C., GARDY J. L. e HANCOCK R. E. (2007) Complexities of targeting innate immunity to treat infection. Trends Immunol. 28:6, 260-266.

BRUNING J. C., GAUTAM D., BURKS D. J., GILLETTE J., SCHUBERT M., ORBAN P. C., KLEIN R., KRONE W., MULLER-WIELAND D. e KAHN C. R. (2000) Role of brain insulin receptor in control of body weight and reproduction. Science 289:5487, 2122-2125.

92

BURDICK D., SOREGHAN B., KWON M., KOSMOSKI J., KNAUER M., HENSCHEN A., YATES J., COTMAN C. e GLABE C. (1992) Assembly and aggregation properties of synthetic Alzheimer's A4/beta amyloid peptide analogs. J.Biol.Chem. 267:1, 546-554.

CACCAMO A., MAJUMDER S., RICHARDSON A., STRONG R. e ODDO S. (2010) Molecular interplay between mammalian target of rapamycin (mTOR), amyloid-beta, and Tau: effects on cognitive impairments. J.Biol.Chem. 285:17, 13107-13120.

CALAY E. S. e HOTAMISLIGIL G. S. (2013) Turning off the inflammatory, but not the metabolic, flames. Nat.Med. 19:3, 265-267.

CALON F., LIM G. P., YANG F., MORIHARA T., TETER B., UBEDA O., ROSTAING P., TRILLER A., SALEM N., Jr., ASHE K. H., FRAUTSCHY S. A. e COLE G. M. (2004) Docosahexaenoic acid protects from dendritic pathology in an Alzheimer's disease mouse model. Neuron 43:5, 633-645.

CARDOZO A. K., ORTIS F., STORLING J., FENG Y. M., RASSCHAERT J., TONNESEN M., VAN E. F., MANDRUP-POULSEN T., HERCHUELZ A. e EIZIRIK D. L. (2005) Cytokines downregulate the sarcoendoplasmic reticulum pump Ca2+ ATPase 2b and deplete endoplasmic reticulum Ca2+, leading to induction of endoplasmic reticulum stress in pancreatic beta-cells. Diabetes 54:2, 452-461.

CARLSON C. J., WHITE M. F. e RONDINONE C. M. (2004) Mammalian target of rapamycin regulates IRS-1 serine 307 phosphorylation. Biochem.Biophys.Res.Commun. 316:2, 533-539.

CARVALHEIRA J. B. C., ZECCHIN H. G. e SAAD M. J. (2002) Vias de Sinalização de Insulina. Arq Bras Endocrinol Metab 46, 419-425.

CASH A. D., ALIEV G., SIEDLAK S. L., NUNOMURA A., FUJIOKA H., ZHU X., RAINA A. K., VINTERS H. V., TABATON M., JOHNSON A. B., PAULA-BARBOSA M., AVILA J., JONES P. K., CASTELLANI R. J., SMITH M. A. e PERRY G. (2003) Microtubule reduction in Alzheimer's disease and aging is independent of tau filament formation. Am.J.Pathol. 162:5, 1623-1627.

CAVALLI V., KUJALA P., KLUMPERMAN J. e GOLDSTEIN L. S. (2005) Sunday Driver links axonal transport to damage signaling. J.Cell Biol. 168:5, 775-787.

CHARTIER-HARLIN M. C., CRAWFORD F., HOULDEN H., WARREN A., HUGHES D., FIDANI L., GOATE A., ROSSOR M., ROQUES P., HARDY J. e . (1991) Early-onset

93

Alzheimer's disease caused by mutations at codon 717 of the beta-amyloid precursor protein gene. Nature 353:6347, 844-846.

CHIU S. L., CHEN C. M. e CLINE H. T. (2008) Insulin receptor signaling regulates synapse number, dendritic plasticity, and circuit function in vivo. Neuron 58:5, 708-719.

COPPS K. D., HANCER N. J., OPARE-ADO L., QIU W., WALSH C. e WHITE M. F. (2010) Irs1 serine 307 promotes insulin sensitivity in mice. Cell Metab 11:1, 84-92.

CRAFT S., PESKIND E., SCHWARTZ M. W., SCHELLENBERG G. D., RASKIND M. e PORTE D., Jr. (1998) Cerebrospinal fluid and plasma insulin levels in Alzheimer's disease: relationship to severity of dementia and apolipoprotein E genotype. Neurology 50:1, 164-168.

CRAFT S. e WATSON G. S. (2004) Insulin and neurodegenerative disease: shared and specific mechanisms. Lancet Neurol. 3:3, 169-178.

CRAS P., KAWAI M., LOWERY D., GONZALEZ-DEWHITT P., GREENBERG B. e PERRY G. (1991) Senile plaque neurites in Alzheimer disease accumulate amyloid precursor protein. Proc.Natl.Acad.Sci.U.S.A 88:17, 7552-7556.

CZIRR E. e WYSS-CORAY T. (2012) The immunology of neurodegeneration. J.Clin.Invest 122:4, 1156-1163.

DAI J., BUIJS R. M., KAMPHORST W. e SWAAB D. F. (2002) Impaired axonal transport of cortical neurons in Alzheimer's disease is associated with neuropathological changes. Brain Res. 948:1-2, 138-144.

DAVIGLUS M. L., PLASSMAN B. L., PIRZADA A., BELL C. C., BOWEN P. E., BURKE J. R., CONNOLLY E. S., Jr., DUNBAR-JACOB J. M., GRANIERI E. C., MCGARRY K., PATEL D., TREVISAN M. e WILLIAMS J. W., Jr. (2011) Risk factors and preventive interventions for Alzheimer disease: state of the science. Arch.Neurol. 68:9, 1185-1190.

DE FELICE F. G. (2013) Alzheimer's disease and insulin resistance: translating basic science into clinical applications. J.Clin.Invest 123:2, 531-539.

DE FELICE F. G., VELASCO P. T., LAMBERT M. P., VIOLA K., FERNANDEZ S. J., FERREIRA S. T. e KLEIN W. L. (2007) Abeta oligomers induce neuronal oxidative stress

94

through an N-methyl-D-aspartate receptor-dependent mechanism that is blocked by the Alzheimer drug memantine. J.Biol.Chem. 282:15, 11590-11601.

DE FELICE F. G., VIEIRA M. N., BOMFIM T. R., DECKER H., VELASCO P. T., LAMBERT M. P., VIOLA K. L., ZHAO W. Q., FERREIRA S. T. e KLEIN W. L. (2009) Protection of synapses against Alzheimer's-linked toxins: insulin signaling prevents the pathogenic binding of Abeta oligomers. Proc.Natl.Acad.Sci.U.S.A 106:6, 1971-1976.

DE FELICE F. G., WU D., LAMBERT M. P., FERNANDEZ S. J., VELASCO P. T., LACOR P. N., BIGIO E. H., JERECIC J., ACTON P. J., SHUGHRUE P. J., CHEN-DODSON E., KINNEY G. G. e KLEIN W. L. (2008) Alzheimer's disease-type neuronal tau hyperphosphorylation induced by A beta oligomers. Neurobiol.Aging 29:9, 1334-1347.

DE LA MONTE S. M. e WANDS J. R. (2008) Alzheimer's disease is type 3 diabetes-evidence reviewed. J.Diabetes Sci.Technol. 2:6, 1101-1113.

DE S. B., VASSAR R. e GOLDE T. (2010) The secretases: enzymes with therapeutic potential in Alzheimer disease. Nat.Rev.Neurol. 6:2, 99-107.

DECKER H., LO K. Y., UNGER S. M., FERREIRA S. T. e SILVERMAN M. A. (2010) Amyloid-beta peptide oligomers disrupt axonal transport through an NMDA receptor-dependent mechanism that is mediated by glycogen synthase kinase 3beta in primary cultured hippocampal neurons. J.Neurosci. 30:27, 9166-9171.

DERKACH V. A., OH M. C., GUIRE E. S. e SODERLING T. R. (2007) Regulatory mechanisms of AMPA receptors in synaptic plasticity. Nat.Rev.Neurosci. 8:2, 101-113.

DESAI A. e MITCHISON T. J. (1997) Microtubule polymerization dynamics. Annu.Rev.Cell Dev.Biol. 13, 83-117.

DHAMOON M. S., NOBLE J. M. e CRAFT S. (2009) Intranasal insulin improves cognition and modulates beta-amyloid in early AD. Neurology 72:3, 292-293.

DI C. M., PICONE P., CARROTTA R., GIACOMAZZA D. e SAN BIAGIO P. L. (2010) Insulin promotes survival of amyloid-beta oligomers neuroblastoma damaged cells via caspase 9 inhibition and Hsp70 upregulation. J.Biomed.Biotechnol. 2010, 147835.

DONATH M. Y. e SHOELSON S. E. (2011) Type 2 diabetes as an inflammatory disease. Nat.Rev.Immunol. 11:2, 98-107.

95

DONNELLY D. (2012) The structure and function of the glucagon-like peptide-1 receptor and its ligands. Br.J.Pharmacol. 166:1, 27-41.

DOU J. T., CHEN M., DUFOUR F., ALKON D. L. e ZHAO W. Q. (2005) Insulin receptor signaling in long-term memory consolidation following spatial learning. Learn.Mem. 12:6, 646-655.

DURING M. J., CAO L., ZUZGA D. S., FRANCIS J. S., FITZSIMONS H. L., JIAO X., BLAND R. J., KLUGMANN M., BANKS W. A., DRUCKER D. J. e HAILE C. N. (2003) Glucagon-like peptide-1 receptor is involved in learning and neuroprotection. Nat.Med. 9:9, 1173-1179.

FERNANDEZ A. M. e TORRES-ALEMAN I. (2012) The many faces of insulin-like peptide signalling in the brain. Nat.Rev.Neurosci. 13:4, 225-239.

FERREIRA S. T. e KLEIN W. L. (2011) The Abeta oligomer hypothesis for synapse failure and memory loss in Alzheimer's disease. Neurobiol.Learn.Mem. 96:4, 529-543.

FERREIRA S. T., VIEIRA M. N. e DE FELICE F. G. (2007) Soluble protein oligomers as emerging toxins in Alzheimer's and other amyloid diseases. IUBMB.Life 59:4-5, 332-345.

FERRI C. P., PRINCE M., BRAYNE C., BRODATY H., FRATIGLIONI L., GANGULI M., HALL K., HASEGAWA K., HENDRIE H., HUANG Y., JORM A., MATHERS C., MENEZES P. R., RIMMER E. e SCAZUFCA M. (2005) Global prevalence of dementia: a Delphi consensus study. Lancet 366:9503, 2112-2117.

FIGUEIREDO C. P., CLARKE J. R., LEDO J. H., RIBEIRO F. C., COSTA C. V., MELO H. M., MOTA-SALES A. P., SARAIVA L. M., KLEIN W. L., SEBOLLELA A., DE FELICE F. G. e FERREIRA S. T. (2013) Memantine rescues transient cognitive impairment caused by high molecular weight AE oligomers but not the persistent impairment induced by low molecular weight oligomers. Journal of Neuroscience .

FREUDE S., HETTICH M. M., SCHUMANN C., STOHR O., KOCH L., KOHLER C., UDELHOVEN M., LEESER U., MULLER M., KUBOTA N., KADOWAKI T., KRONE W., SCHRODER H., BRUNING J. C. e SCHUBERT M. (2009) Neuronal IGF-1 resistance reduces Abeta accumulation and protects against premature death in a model of Alzheimer's disease. FASEB J. 23:10, 3315-3324.

FRIEDMAN D. S. e VALE R. D. (1999) Single-molecule analysis of kinesin motility reveals regulation by the cargo-binding tail domain. Nat.Cell Biol. 1:5, 293-297.

96

FU S., WATKINS S. M. e HOTAMISLIGIL G. S. (2012) The role of endoplasmic reticulum in hepatic lipid homeostasis and stress signaling. Cell Metab 15:5, 623-634.

GARBER A. J. (2011) Long-acting glucagon-like peptide 1 receptor agonists: a review of their efficacy and tolerability. Diabetes Care 34 Suppl 2, S279-S284.

GAULT V. A. e HOLSCHER C. (2008) GLP-1 agonists facilitate hippocampal LTP and reverse the impairment of LTP induced by beta-amyloid. Eur.J.Pharmacol. 587:1-3, 112-117.

GEULA C., WU C. K., SAROFF D., LORENZO A., YUAN M. e YANKNER B. A. (1998) Aging renders the brain vulnerable to amyloid beta-protein neurotoxicity. Nat.Med. 4:7, 827-831.

GIMBEL D. A., NYGAARD H. B., COFFEY E. E., GUNTHER E. C., LAUREN J., GIMBEL Z. A. e STRITTMATTER S. M. (2010) Memory impairment in transgenic Alzheimer mice requires cellular prion protein. J.Neurosci. 30:18, 6367-6374.

GLENNER G. G. e WONG C. W. (1984) Alzheimer's disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem.Biophys.Res.Commun. 120:3, 885-890.

GOATE A., CHARTIER-HARLIN M. C., MULLAN M., BROWN J., CRAWFORD F., FIDANI L., GIUFFRA L., HAYNES A., IRVING N., JAMES L. e . (1991) Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer's disease. Nature 349:6311, 704-706.

GONG Y., CHANG L., VIOLA K. L., LACOR P. N., LAMBERT M. P., FINCH C. E., KRAFFT G. A. e KLEIN W. L. (2003) Alzheimer's disease-affected brain: presence of oligomeric A beta ligands (ADDLs) suggests a molecular basis for reversible memory loss. Proc.Natl.Acad.Sci.U.S.A 100:18, 10417-10422.

GOTO Y., NIIDOME T., AKAIKE A., KIHARA T. e SUGIMOTO H. (2006) Amyloid beta-peptide preconditioning reduces glutamate-induced neurotoxicity by promoting endocytosis of NMDA receptor. Biochem.Biophys.Res.Commun. 351:1, 259-265.

GRAFSTEIN B. e FORMAN D. S. (1980) Intracellular transport in neurons. Physiol Rev. 60:4, 1167-1283.

97

GRALLE M. e FERREIRA S. T. (2007) Structure and functions of the human amyloid precursor protein: the whole is more than the sum of its parts. Prog.Neurobiol. 82:1, 11-32.

GRAMMAS P. e OVASE R. (2001) Inflammatory factors are elevated in brain microvessels in Alzheimer's disease. Neurobiol.Aging 22:6, 837-842.

GREENE J. D., HODGES J. R. e BADDELEY A. D. (1995) Autobiographical memory and executive function in early dementia of Alzheimer type. Neuropsychologia 33:12, 1647-1670.

GU Y., SANJO N., CHEN F., HASEGAWA H., PETIT A., RUAN X., LI W., SHIER C., KAWARAI T., SCHMITT-ULMS G., WESTAWAY D., ST GEORGE-HYSLOP P. e FRASER P. E. (2004) The presenilin proteins are components of multiple membrane-bound complexes that have different biological activities. J.Biol.Chem. 279:30, 31329-31336.

HAJ-ALI V., MOHADDES G. e BABRI S. H. (2009) Intracerebroventricular insulin improves spatial learning and memory in male Wistar rats. Behav.Neurosci. 123:6, 1309-1314.

HAMILTON A. e HOLSCHER C. (2009) Receptors for the incretin glucagon-like peptide-1 are expressed on neurons in the central nervous system. Neuroreport 20:13, 1161-1166.

HAMILTON A., PATTERSON S., PORTER D., GAULT V. A. e HOLSCHER C. (2011) Novel GLP-1 mimetics developed to treat type 2 diabetes promote progenitor cell proliferation in the brain. J.Neurosci.Res. 89:4, 481-489.

HARDY J. e SELKOE D. J. (2002) The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeutics. Science 297:5580, 353-356.

HARDY J. A. e HIGGINS G. A. (1992) Alzheimer's disease: the amyloid cascade hypothesis. Science 256:5054, 184-185.

HAVRANKOVA J., ROTH J. e BROWNSTEIN M. J. (1983) Insulin receptors in brain. Adv.Metab Disord. 10, 259-268.

HAVRANKOVA J., SCHMECHEL D., ROTH J. e BROWNSTEIN M. (1978) Identification of insulin in rat brain. Proc.Natl.Acad.Sci.U.S.A 75:11, 5737-5741.

98

HILTUNEN M., HELISALMI S., MANNERMAA A., ALAFUZOFF I., KOIVISTO A. M., LEHTOVIRTA M., PIRSKANEN M., SULKAVA R., VERKKONIEMI A. e SOININEN H. (2000) Identification of a novel 4.6-kb genomic deletion in presenilin-1 gene which results in exclusion of exon 9 in a Finnish early onset Alzheimer's disease family: an Alu core sequence-stimulated recombination? Eur.J.Hum.Genet. 8:4, 259-266.

HIROKAWA N. (1982) Cross-linker system between neurofilaments, microtubules, and membranous organelles in frog axons revealed by the quick-freeze, deep-etching method. J.Cell Biol. 94:1, 129-142.

HIROKAWA N. (1998) Kinesin and dynein superfamily proteins and the mechanism of organelle transport. Science 279:5350, 519-526.

HIROKAWA N. e TAKEMURA R. (2005) Molecular motors and mechanisms of directional transport in neurons. Nat.Rev.Neurosci. 6:3, 201-214.

HIROSUMI J., TUNCMAN G., CHANG L., GORGUN C. Z., UYSAL K. T., MAEDA K., KARIN M. e HOTAMISLIGIL G. S. (2002) A central role for JNK in obesity and insulin resistance. Nature 420:6913, 333-336.

HIRUMA H., KATAKURA T., TAKAHASHI S., ICHIKAWA T. e KAWAKAMI T. (2003) Glutamate and amyloid beta-protein rapidly inhibit fast axonal transport in cultured rat hippocampal neurons by different mechanisms. J.Neurosci. 23:26, 8967-8977.

HOLSCHER C. e LI L. (2010) New roles for insulin-like hormones in neuronal signalling and protection: new hopes for novel treatments of Alzheimer's disease? Neurobiol.Aging 31:9, 1495-1502.

HORNG T. e HOTAMISLIGIL G. S. (2011) Linking the inflammasome to obesity-related disease. Nat.Med. 17:2, 164-165.

HOTAMISLIGIL G. S. (2006) Inflammation and metabolic disorders. Nature 444:7121, 860-867.

HOTAMISLIGIL G. S. (2010) Endoplasmic reticulum stress and the inflammatory basis of metabolic disease. Cell 140:6, 900-917.

99

HOTAMISLIGIL G. S., ARNER P., CARO J. F., ATKINSON R. L. e SPIEGELMAN B. M. (1995) Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J.Clin.Invest 95:5, 2409-2415.

HOTAMISLIGIL G. S. e ERBAY E. (2008) Nutrient sensing and inflammation in metabolic diseases. Nat.Rev.Immunol. 8:12, 923-934.

HOTAMISLIGIL G. S., PERALDI P., BUDAVARI A., ELLIS R., WHITE M. F. e SPIEGELMAN B. M. (1996) IRS-1-mediated inhibition of insulin receptor tyrosine kinase activity in TNF-alpha- and obesity-induced insulin resistance. Science 271:5249, 665-668.

HOTAMISLIGIL G. S., SHARGILL N. S. e SPIEGELMAN B. M. (1993) Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science 259:5091, 87-91.

HOYER S. e NITSCH R. (1989) Cerebral excess release of neurotransmitter amino acids subsequent to reduced cerebral glucose metabolism in early-onset dementia of Alzheimer type. J.Neural Transm. 75:3, 227-232.

HSIEH H., BOEHM J., SATO C., IWATSUBO T., TOMITA T., SISODIA S. e MALINOW R. (2006) AMPAR removal underlies Abeta-induced synaptic depression and dendritic spine loss. Neuron 52:5, 831-843.

IRVINE E. E., DRINKWATER L., RADWANSKA K., AL-QASSAB H., SMITH M. A., O'BRIEN M., KIELAR C., CHOUDHURY A. I., KRAUSS S., COOPER J. D., WITHERS D. J. e GIESE K. P. (2011) Insulin receptor substrate 2 is a negative regulator of memory formation. Learn.Mem. 18:6, 375-383.

ISHII K., II K., HASEGAWA T., SHOJI S., DOI A. e MORI H. (1997) Increased A beta 42(43)-plaque deposition in early-onset familial Alzheimer's disease brains with the deletion of exon 9 and the missense point mutation (H163R) in the PS-1 gene. Neurosci.Lett. 228:1, 17-20.

JACOBSEN J. S., WU C. C., REDWINE J. M., COMERY T. A., ARIAS R., BOWLBY M., MARTONE R., MORRISON J. H., PANGALOS M. N., REINHART P. H. e BLOOM F. E. (2006) Early-onset behavioral and synaptic deficits in a mouse model of Alzheimer's disease. Proc.Natl.Acad.Sci.U.S.A 103:13, 5161-5166.

100

JANKOWSKY J. L., SLUNT H. H., RATOVITSKI T., JENKINS N. A., COPELAND N. G. e BORCHELT D. R. (2001) Co-expression of multiple transgenes in mouse CNS: a comparison of strategies. Biomol.Eng 17:6, 157-165.

JARRETT J. T., BERGER E. P. e LANSBURY P. T., Jr. (1993) The carboxy terminus of the beta amyloid protein is critical for the seeding of amyloid formation: implications for the pathogenesis of Alzheimer's disease. Biochemistry 32:18, 4693-4697.

JIN J. J., KIM H. D., MAXWELL J. A., LI L. e FUKUCHI K. (2008) Toll-like receptor 4-dependent upregulation of cytokines in a transgenic mouse model of Alzheimer's disease. J.Neuroinflammation. 5, 23.

JOB C. e EBERWINE J. (2001) Localization and translation of mRNA in dendrites and axons. Nat.Rev.Neurosci. 2:12, 889-898.

JURGENSEN S., ANTONIO L. L., MUSSI G. E., BRITO-MOREIRA J., BOMFIM T. R., DE FELICE F. G., GARRIDO-SANABRIA E. R., CAVALHEIRO E. A. e FERREIRA S. T. (2011) Activation of D1/D5 Dopamine Receptors Protects Neurons from Synapse Dysfunction Induced by Amyloid-{beta} Oligomers. J.Biol.Chem. 286:5, 3270-3276.

KAECH S. e BANKER G. (2006) Culturing hippocampal neurons. Nat.Protoc. 1:5, 2406-2415.

KALARIA R. N., MAESTRE G. E., ARIZAGA R., FRIEDLAND R. P., GALASKO D., HALL K., LUCHSINGER J. A., OGUNNIYI A., PERRY E. K., POTOCNIK F., PRINCE M., STEWART R., WIMO A., ZHANG Z. X. e ANTUONO P. (2008) Alzheimer's disease and vascular dementia in developing countries: prevalence, management, and risk factors. Lancet Neurol. 7:9, 812-826.

KAMAL A., STOKIN G. B., YANG Z., XIA C. H. e GOLDSTEIN L. S. (2000) Axonal transport of amyloid precursor protein is mediated by direct binding to the kinesin light chain subunit of kinesin-I. Neuron 28:2, 449-459.

KAWARABAYASHI T., SHOJI M., YAMAGUCHI H., TANAKA M., HARIGAYA Y., ISHIGURO K. e HIRAI S. (1993) Amyloid beta protein precursor accumulates in swollen neurites throughout rat brain with aging. Neurosci.Lett. 153:1, 73-76.

KAZAFEOS K. (2011) Incretin effect: GLP-1, GIP, DPP4. Diabetes Res.Clin.Pract. 93 Suppl 1, S32-S36.

101

KILLICK R., SCALES G., LEROY K., CAUSEVIC M., HOOPER C., IRVINE E. E., CHOUDHURY A. I., DRINKWATER L., KERR F., AL-QASSAB H., STEPHENSON J., YILMAZ Z., GIESE K. P., BRION J. P., WITHERS D. J. e LOVESTONE S. (2009) Deletion of Irs2 reduces amyloid deposition and rescues behavioural deficits in APP transgenic mice. Biochem.Biophys.Res.Commun. 386:1, 257-262.

KIM J. Y., LIM D. M., PARK H. S., MOON C. I., CHOI K. J., LEE S. K., BAIK H. W., PARK K. Y. e KIM B. J. (2012) Exendin-4 protects against sulfonylurea-induced beta-cell apoptosis. J.Pharmacol.Sci. 118:1, 65-74.

KLEIN W. L. (2006) Synaptic targeting by A beta oligomers (ADDLS) as a basis for memory loss in early Alzheimer's disease. Alzheimers.Dement. 2:1, 43-55.

KLYUBIN I., WALSH D. M., LEMERE C. A., CULLEN W. K., SHANKAR G. M., BETTS V., SPOONER E. T., JIANG L., ANWYL R., SELKOE D. J. e ROWAN M. J. (2005) Amyloid beta protein immunotherapy neutralizes Abeta oligomers that disrupt synaptic plasticity in vivo. Nat.Med. 11:5, 556-561.

KOVACS P. e HAJNAL A. (2009) In vivo electrophysiological effects of insulin in the rat brain. Neuropeptides 43:4, 283-293.

KRAFFT G. A. e KLEIN W. L. (2010) ADDLs and the signaling web that leads to Alzheimer's disease. Neuropharmacology 59:4-5, 230-242.

KWINTER D. M., LO K., MAFI P. e SILVERMAN M. A. (2009) Dynactin regulates bidirectional transport of dense-core vesicles in the axon and dendrites of cultured hippocampal neurons. Neuroscience 162:4, 1001-1010.

LACOR P. N., BUNIEL M. C., CHANG L., FERNANDEZ S. J., GONG Y., VIOLA K. L., LAMBERT M. P., VELASCO P. T., BIGIO E. H., FINCH C. E., KRAFFT G. A. e KLEIN W. L. (2004) Synaptic targeting by Alzheimer's-related amyloid beta oligomers. J.Neurosci. 24:45, 10191-10200.

LACOR P. N., BUNIEL M. C., FURLOW P. W., CLEMENTE A. S., VELASCO P. T., WOOD M., VIOLA K. L. e KLEIN W. L. (2007) Abeta oligomer-induced aberrations in synapse composition, shape, and density provide a molecular basis for loss of connectivity in Alzheimer's disease. J.Neurosci. 27:4, 796-807.

LAFAY-CHEBASSIER C., PACCALIN M., PAGE G., BARC-PAIN S., PERAULT-POCHAT M. C., GIL R., PRADIER L. e HUGON J. (2005) mTOR/p70S6k signalling

102

alteration by Abeta exposure as well as in APP-PS1 transgenic models and in patients with Alzheimer's disease. J.Neurochem. 94:1, 215-225.

LAMBERT M. P., BARLOW A. K., CHROMY B. A., EDWARDS C., FREED R., LIOSATOS M., MORGAN T. E., ROZOVSKY I., TROMMER B., VIOLA K. L., WALS P., ZHANG C., FINCH C. E., KRAFFT G. A. e KLEIN W. L. (1998) Diffusible, nonfibrillar ligands derived from Abeta1-42 are potent central nervous system neurotoxins. Proc.Natl.Acad.Sci.U.S.A 95:11, 6448-6453.

LAMBERT M. P., VELASCO P. T., CHANG L., VIOLA K. L., FERNANDEZ S., LACOR P. N., KHUON D., GONG Y., BIGIO E. H., SHAW P., DE FELICE F. G., KRAFFT G. A. e KLEIN W. L. (2007) Monoclonal antibodies that target pathological assemblies of Abeta. J.Neurochem. 100:1, 23-35.

LAMONTE B. H., WALLACE K. E., HOLLOWAY B. A., SHELLY S. S., ASCANO J., TOKITO M., VAN W. T., HOWLAND D. S. e HOLZBAUR E. L. (2002) Disruption of dynein/dynactin inhibits axonal transport in motor neurons causing late-onset progressive degeneration. Neuron 34:5, 715-727.

LANZ T. A., CARTER D. B. e MERCHANT K. M. (2003) Dendritic spine loss in the hippocampus of young PDAPP and Tg2576 mice and its prevention by the ApoE2 genotype. Neurobiol.Dis. 13:3, 246-253.

LAU C. G. e ZUKIN R. S. (2007) NMDA receptor trafficking in synaptic plasticity and neuropsychiatric disorders. Nat.Rev.Neurosci. 8:6, 413-426.

LEDO J. H., AZEVEDO E. P., CLARKE J. R., RIBEIRO F. C., FIGUEIREDO C. P., FOGUEL D., DE FELICE F. G. e FERREIRA S. T. (2013) Amyloid-beta oligomers link depressive-like behavior and cognitive deficits in mice. Mol.Psychiatry 18:10, 1053-1054.

LEE H. K., KUMAR P., FU Q., ROSEN K. M. e QUERFURTH H. W. (2009) The insulin/Akt signaling pathway is targeted by intracellular beta-amyloid. Mol.Biol.Cell 20:5, 1533-1544.

LESNE S., KOH M. T., KOTILINEK L., KAYED R., GLABE C. G., YANG A., GALLAGHER M. e ASHE K. H. (2006) A specific amyloid-beta protein assembly in the brain impairs memory. Nature 440:7082, 352-357.

LEVY E., CARMAN M. D., FERNANDEZ-MADRID I. J., POWER M. D., LIEBERBURG I., VAN DUINEN S. G., BOTS G. T., LUYENDIJK W. e FRANGIONE

103

B. (1990) Mutation of the Alzheimer's disease amyloid gene in hereditary cerebral hemorrhage, Dutch type. Science 248:4959, 1124-1126.

LI Y., PERRY T., KINDY M. S., HARVEY B. K., TWEEDIE D., HOLLOWAY H. W., POWERS K., SHEN H., EGAN J. M., SAMBAMURTI K., BROSSI A., LAHIRI D. K., MATTSON M. P., HOFFER B. J., WANG Y. e GREIG N. H. (2009) GLP-1 receptor stimulation preserves primary cortical and dopaminergic neurons in cellular and rodent models of stroke and Parkinsonism. Proc.Natl.Acad.Sci.U.S.A 106:4, 1285-1290.

LIZCANO J. M. e ALESSI D. R. (2002) The insulin signalling pathway. Curr.Biol. 12:7, R236-R238.

LUE L. F., WALKER D. G. e ROGERS J. (2001) Modeling microglial activation in Alzheimer's disease with human postmortem microglial cultures. Neurobiol.Aging 22:6, 945-956.

MA Q. L., YANG F., ROSARIO E. R., UBEDA O. J., BEECH W., GANT D. J., CHEN P. P., HUDSPETH B., CHEN C., ZHAO Y., VINTERS H. V., FRAUTSCHY S. A. e COLE G. M. (2009) Beta-amyloid oligomers induce phosphorylation of tau and inactivation of insulin receptor substrate via c-Jun N-terminal kinase signaling: suppression by omega-3 fatty acids and curcumin. J.Neurosci. 29:28, 9078-9089.

MA T., HOEFFER C. A., CAPETILLO-ZARATE E., YU F., WONG H., LIN M. T., TAMPELLINI D., KLANN E., BLITZER R. D. e GOURAS G. K. (2010) Dysregulation of the mTOR pathway mediates impairment of synaptic plasticity in a mouse model of Alzheimer's disease. PLoS.One. 5:9.

MADADI G., DALVI P. S. e BELSHAM D. D. (2008) Regulation of brain insulin mRNA by glucose and glucagon-like peptide 1. Biochem.Biophys.Res.Commun. 376:4, 694-699.

MASTERS C. L., MULTHAUP G., SIMMS G., POTTGIESSER J., MARTINS R. N. e BEYREUTHER K. (1985) Neuronal origin of a cerebral amyloid: neurofibrillary tangles of Alzheimer's disease contain the same protein as the amyloid of plaque cores and blood vessels. EMBO J. 4:11, 2757-2763.

MATTSON M. P. (2012) Energy intake and exercise as determinants of brain health and vulnerability to injury and disease. Cell Metab 16:6, 706-722.

MATTSON M. P., PERRY T. e GREIG N. H. (2003) Learning from the gut. Nat.Med. 9:9, 1113-1115.

104

MCGEER P. L., ROGERS J. e MCGEER E. G. (2006) Inflammation, anti-inflammatory agents and Alzheimer disease: the last 12 years. J.Alzheimers.Dis. 9:3 Suppl, 271-276.

MCNAY E. C., ONG C. T., MCCRIMMON R. J., CRESSWELL J., BOGAN J. S. e SHERWIN R. S. (2010) Hippocampal memory processes are modulated by insulin and high-fat-induced insulin resistance. Neurobiol.Learn.Mem. 93:4, 546-553.

MCNAY E. C. e RECKNAGEL A. K. (2011) Brain insulin signaling: a key component of cognitive processes and a potential basis for cognitive impairment in type 2 diabetes. Neurobiol.Learn.Mem. 96:3, 432-442.

MINERS J. S., BARUA N., KEHOE P. G., GILL S. e LOVE S. (2011) Abeta-degrading enzymes: potential for treatment of Alzheimer disease. J.Neuropathol.Exp.Neurol. 70:11, 944-959.

MOCHIZUKI A., PETERSON J. W., MUFSON E. J. e TRAPP B. D. (1996) Amyloid load and neural elements in Alzheimer's disease and nondemented individuals with high amyloid plaque density. Exp.Neurol. 142:1, 89-102.

MOLONEY A. M., GRIFFIN R. J., TIMMONS S., O'CONNOR R., RAVID R. e O'NEILL C. (2010) Defects in IGF-1 receptor, insulin receptor and IRS-1/2 in Alzheimer's disease indicate possible resistance to IGF-1 and insulin signalling. Neurobiol.Aging 31:2, 224-243.

MOOLMAN D. L., VITOLO O. V., VONSATTEL J. P. e SHELANSKI M. L. (2004) Dendrite and dendritic spine alterations in Alzheimer models. J.Neurocytol. 33:3, 377-387.

MORAN T. H. e DAILEY M. J. (2009) Minireview: Gut peptides: targets for antiobesity drug development? Endocrinology 150:6, 2526-2530.

MOSTOSLAVSKY R. (2008) DNA repair, insulin signaling and sirtuins: at the crossroads between cancer and aging. Front Biosci. 13, 6966-6990.

MUCKE L., MASLIAH E., YU G. Q., MALLORY M., ROCKENSTEIN E. M., TATSUNO G., HU K., KHOLODENKO D., JOHNSON-WOOD K. e MCCONLOGUE L. (2000) High-level neuronal expression of abeta 1-42 in wild-type human amyloid protein precursor transgenic mice: synaptotoxicity without plaque formation. J.Neurosci. 20:11, 4050-4058.

105

MULLAN M., CRAWFORD F., AXELMAN K., HOULDEN H., LILIUS L., WINBLAD B. e LANNFELT L. (1992) A pathogenic mutation for probable Alzheimer's disease in the APP gene at the N-terminus of beta-amyloid. Nat.Genet. 1:5, 345-347.

MURTHY S. B., JAWAID A. e SCHULZ P. E. (2008) Diabetes mellitus and dementia: advocating an annual cognitive screening in patients with diabetes mellitus. J.Am.Geriatr.Soc. 56:10, 1976-1977.

NAKAMURA T., FURUHASHI M., LI P., CAO H., TUNCMAN G., SONENBERG N., GORGUN C. Z. e HOTAMISLIGIL G. S. (2010) Double-stranded RNA-dependent protein kinase links pathogen sensing with stress and metabolic homeostasis. Cell 140:3, 338-348.

NIEWIADOMSKA G. e BAKSALERSKA-PAZERA M. (2003) Age-dependent changes in axonal transport and cellular distribution of Tau 1 in the rat basal forebrain neurons. Neuroreport 14:13, 1701-1706.

NOWRANGI M. A., RAO V. e LYKETSOS C. G. (2011) Epidemiology, assessment, and treatment of dementia. Psychiatr.Clin.North Am. 34:2, 275-94, vii.

OTT A., STOLK R. P., HOFMAN A., VAN H. F., GROBBEE D. E. e BRETELER M. M. (1996) Association of diabetes mellitus and dementia: the Rotterdam Study. Diabetologia 39:11, 1392-1397.

OTT V., BENEDICT C., SCHULTES B., BORN J. e HALLSCHMID M. (2012) Intranasal administration of insulin to the brain impacts cognitive function and peripheral metabolism. Diabetes Obes.Metab 14:3, 214-221.

OZES O. N., AKCA H., MAYO L. D., GUSTIN J. A., MAEHAMA T., DIXON J. E. e DONNER D. B. (2001) A phosphatidylinositol 3-kinase/Akt/mTOR pathway mediates and PTEN antagonizes tumor necrosis factor inhibition of insulin signaling through insulin receptor substrate-1. Proc.Natl.Acad.Sci.U.S.A 98:8, 4640-4645.

PARK K. M. e BOWERS W. J. (2010) Tumor necrosis factor-alpha mediated signaling in neuronal homeostasis and dysfunction. Cell Signal. 22:7, 977-983.

PATTI M. E. e KAHN C. R. (1998) The insulin receptor--a critical link in glucose homeostasis and insulin action. J.Basic Clin.Physiol Pharmacol. 9:2-4, 89-109.

PAWSON T. (1995) Protein modules and signalling networks. Nature 373:6515, 573-580.

106

PAWSON T. e SCOTT J. D. (1997) Signaling through scaffold, anchoring, and adaptor proteins. Science 278:5346, 2075-2080.

PERRY V. H., NICOLL J. A. e HOLMES C. (2010) Microglia in neurodegenerative disease. Nat.Rev.Neurol. 6:4, 193-201.

PETERSEN R. C., ROBERTS R. O., KNOPMAN D. S., BOEVE B. F., GEDA Y. E., IVNIK R. J., SMITH G. E. e JACK C. R., Jr. (2009) Mild cognitive impairment: ten years later. Arch.Neurol. 66:12, 1447-1455.

PIKE C. J., CUMMINGS B. J. e COTMAN C. W. (1992) beta-Amyloid induces neuritic dystrophy in vitro: similarities with Alzheimer pathology. Neuroreport 3:9, 769-772.

PIROLA L., JOHNSTON A. M. e VAN O. E. (2004) Modulation of insulin action. Diabetologia 47:2, 170-184.

PLUM L., SCHUBERT M. e BRUNING J. C. (2005) The role of insulin receptor signaling in the brain. Trends Endocrinol.Metab 16:2, 59-65.

PRINZ M., PRILLER J., SISODIA S. S. e RANSOHOFF R. M. (2011) Heterogeneity of CNS myeloid cells and their roles in neurodegeneration. Nat.Neurosci. 14:10, 1227-1235.

RABE S., REICHWALD J., AMMATURO D., DE S. B., SAFTIG P., NEUMANN U. e STAUFENBIEL M. (2011) The Swedish APP mutation alters the effect of genetically reduced BACE1 expression on the APP processing. J.Neurochem. 119:1, 231-239.

RELKIN N. R., SZABO P., ADAMIAK B., BURGUT T., MONTHE C., LENT R. W., YOUNKIN S., YOUNKIN L., SCHIFF R. e WEKSLER M. E. (2009) 18-Month study of intravenous immunoglobulin for treatment of mild Alzheimer disease. Neurobiol.Aging 30:11, 1728-1736.

RICHARDSON J. C., KENDAL C. E., ANDERSON R., PRIEST F., GOWER E., SODEN P., GRAY R., TOPPS S., HOWLETT D. R., LAVENDER D., CLARKE N. J., BARNES J. C., HAWORTH R., STEWART M. G. e RUPNIAK H. T. (2003) Ultrastructural and behavioural changes precede amyloid deposition in a transgenic model of Alzheimer's disease. Neuroscience 122:1, 213-228.

ROBERSON E. D. e MUCKE L. (2006) 100 years and counting: prospects for defeating Alzheimer's disease. Science 314:5800, 781-784.

107

ROSELLI F., TIRARD M., LU J., HUTZLER P., LAMBERTI P., LIVREA P., MORABITO M. e ALMEIDA O. F. (2005) Soluble beta-amyloid1-40 induces NMDA-dependent degradation of postsynaptic density-95 at glutamatergic synapses. J.Neurosci. 25:48, 11061-11070.

RUAN L., KANG Z., PEI G. e LE Y. (2009) Amyloid deposition and inflammation in APPswe/PS1dE9 mouse model of Alzheimer's disease. Curr.Alzheimer Res. 6:6, 531-540.

RUI L., AGUIRRE V., KIM J. K., SHULMAN G. I., LEE A., CORBOULD A., DUNAIF A. e WHITE M. F. (2001) Insulin/IGF-1 and TNF-alpha stimulate phosphorylation of IRS-1 at inhibitory Ser307 via distinct pathways. J.Clin.Invest 107:2, 181-189.

RYAN G. J. e HARDY Y. (2011) Liraglutide: once-daily GLP-1 agonist for the treatment of type 2 diabetes. J.Clin.Pharm.Ther. 36:3, 260-274.

RYU B. R., KO H. W., JOU I., NOH J. S. e GWAG B. J. (1999) Phosphatidylinositol 3-kinase-mediated regulation of neuronal apoptosis and necrosis by insulin and IGF-I. J.Neurobiol. 39:4, 536-546.

SALTIEL A. R. e KAHN C. R. (2001) Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414:6865, 799-806.

SARAIVA L. M., SEIXAS DA SILVA G. S., GALINA A., DA-SILVA W. S., KLEIN W. L., FERREIRA S. T. e DE FELICE F. G. (2010) Amyloid-beta triggers the release of neuronal hexokinase 1 from mitochondria. PLoS.One. 5:12, e15230.

SAVAGE D. B. e SEMPLE R. K. (2010) Recent insights into fatty liver, metabolic dyslipidaemia and their links to insulin resistance. Curr.Opin.Lipidol. 21:4, 329-336.

SAVONENKO A., XU G. M., MELNIKOVA T., MORTON J. L., GONZALES V., WONG M. P., PRICE D. L., TANG F., MARKOWSKA A. L. e BORCHELT D. R. (2005) Episodic-like memory deficits in the APPswe/PS1dE9 mouse model of Alzheimer's disease: relationships to beta-amyloid deposition and neurotransmitter abnormalities. Neurobiol.Dis. 18:3, 602-617.

SCHEFF S. W. e PRICE D. A. (2003) Synaptic pathology in Alzheimer's disease: a review of ultrastructural studies. Neurobiol.Aging 24:8, 1029-1046.

108

SCHEIBEL M. E., LINDSAY R. D., TOMIYASU U. e SCHEIBEL A. B. (1975) Progressive dendritic changes in aging human cortex. Exp.Neurol. 47:3, 392-403.

SCHRIJVERS E. M., WITTEMAN J. C., SIJBRANDS E. J., HOFMAN A., KOUDSTAAL P. J. e BRETELER M. M. (2010) Insulin metabolism and the risk of Alzheimer disease: the Rotterdam Study. Neurology 75:22, 1982-1987.

SCHROER T. A. (2004) Dynactin. Annu.Rev.Cell Dev.Biol. 20, 759-779.

SELKOE D. J. (2000) The genetics and molecular pathology of Alzheimer's disease: roles of amyloid and the presenilins. Neurol.Clin. 18:4, 903-922.

SELKOE D. J. (2004) Alzheimer disease: mechanistic understanding predicts novel therapies. Ann.Intern.Med. 140:8, 627-638.

SELKOE D. J. (2011) Resolving controversies on the path to Alzheimer's therapeutics. Nat.Med. 17:9, 1060-1065.

SELKOE D. J. (2012) Preventing Alzheimer's disease. Science 337:6101, 1488-1492.

SESHADRI S., BEISER A., KELLY-HAYES M., KASE C. S., AU R., KANNEL W. B. e WOLF P. A. (2006) The lifetime risk of stroke: estimates from the Framingham Study. Stroke 37:2, 345-350.

SHANKAR G. M., LI S., MEHTA T. H., GARCIA-MUNOZ A., SHEPARDSON N. E., SMITH I., BRETT F. M., FARRELL M. A., ROWAN M. J., LEMERE C. A., REGAN C. M., WALSH D. M., SABATINI B. L. e SELKOE D. J. (2008) Amyloid-beta protein dimers isolated directly from Alzheimer's brains impair synaptic plasticity and memory. Nat.Med. 14:8, 837-842.

SHIM K. S. e LUBEC G. (2002) Drebrin, a dendritic spine protein, is manifold decreased in brains of patients with Alzheimer's disease and Down syndrome. Neurosci.Lett. 324:3, 209-212.

SLOANE J. A., PIETROPAOLO M. F., ROSENE D. L., MOSS M. B., PETERS A., KEMPER T. e ABRAHAM C. R. (1997) Lack of correlation between plaque burden and cognition in the aged monkey. Acta Neuropathol. 94:5, 471-478.

109

SNYDER S. W., LADROR U. S., WADE W. S., WANG G. T., BARRETT L. W., MATAYOSHI E. D., HUFFAKER H. J., KRAFFT G. A. e HOLZMAN T. F. (1994) Amyloid-beta aggregation: selective inhibition of aggregation in mixtures of amyloid with different chain lengths. Biophys.J. 67:3, 1216-1228.

SOUSA-NUNES R., YEE L. L. e GOULD A. P. (2011) Fat cells reactivate quiescent neuroblasts via TOR and glial insulin relays in Drosophila. Nature 471:7339, 508-512.

SPILMAN P., PODLUTSKAYA N., HART M. J., DEBNATH J., GOROSTIZA O., BREDESEN D., RICHARDSON A., STRONG R. e GALVAN V. (2010) Inhibition of mTOR by rapamycin abolishes cognitive deficits and reduces amyloid-beta levels in a mouse model of Alzheimer's disease. PLoS.One. 5:4, e9979.

SPIRES T. L., MEYER-LUEHMANN M., STERN E. A., MCLEAN P. J., SKOCH J., NGUYEN P. T., BACSKAI B. J. e HYMAN B. T. (2005) Dendritic spine abnormalities in amyloid precursor protein transgenic mice demonstrated by gene transfer and intravital multiphoton microscopy. J.Neurosci. 25:31, 7278-7287.

STAMER K., VOGEL R., THIES E., MANDELKOW E. e MANDELKOW E. M. (2002) Tau blocks traffic of organelles, neurofilaments, and APP vesicles in neurons and enhances oxidative stress. J.Cell Biol. 156:6, 1051-1063.

STEEN E., TERRY B. M., RIVERA E. J., CANNON J. L., NEELY T. R., TAVARES R., XU X. J., WANDS J. R. e DE LA MONTE S. M. (2005) Impaired insulin and insulin-like growth factor expression and signaling mechanisms in Alzheimer's disease--is this type 3 diabetes? J.Alzheimers.Dis. 7:1, 63-80.

STOKIN G. B. e GOLDSTEIN L. S. (2006) Axonal transport and Alzheimer's disease. Annu.Rev.Biochem. 75, 607-627.

STOKIN G. B., LILLO C., FALZONE T. L., BRUSCH R. G., ROCKENSTEIN E., MOUNT S. L., RAMAN R., DAVIES P., MASLIAH E., WILLIAMS D. S. e GOLDSTEIN L. S. (2005) Axonopathy and transport deficits early in the pathogenesis of Alzheimer's disease. Science 307:5713, 1282-1288.

SWARDFAGER W., LANCTOT K., ROTHENBURG L., WONG A., CAPPELL J. e HERRMANN N. (2010) A meta-analysis of cytokines in Alzheimer's disease. Biol.Psychiatry 68:10, 930-941.

110

TAGUCHI A., WARTSCHOW L. M. e WHITE M. F. (2007) Brain IRS2 signaling coordinates life span and nutrient homeostasis. Science 317:5836, 369-372.

TAGUCHI A. e WHITE M. F. (2008) Insulin-like signaling, nutrient homeostasis, and life span. Annu.Rev.Physiol 70, 191-212.

TAKEDA S., SATO N., UCHIO-YAMADA K., SAWADA K., KUNIEDA T., TAKEUCHI D., KURINAMI H., SHINOHARA M., RAKUGI H. e MORISHITA R. (2010) Diabetes-accelerated memory dysfunction via cerebrovascular inflammation and Abeta deposition in an Alzheimer mouse model with diabetes. Proc.Natl.Acad.Sci.U.S.A 107:15, 7036-7041.

TALBOT K., EIDEM W. L., TINSLEY C. L., BENSON M. A., THOMPSON E. W., SMITH R. J., HAHN C. G., SIEGEL S. J., TROJANOWSKI J. Q., GUR R. E., BLAKE D. J. e ARNOLD S. E. (2004) Dysbindin-1 is reduced in intrinsic, glutamatergic terminals of the hippocampal formation in schizophrenia. J.Clin.Invest 113:9, 1353-1363.

TANAKA M., SAWADA M., YOSHIDA S., HANAOKA F. e MARUNOUCHI T. (1995) Insulin prevents apoptosis of external granular layer neurons in rat cerebellar slice cultures. Neurosci.Lett. 199:1, 37-40.

TANIGUCHI C. M., EMANUELLI B. e KAHN C. R. (2006) Critical nodes in signalling pathways: insights into insulin action. Nat.Rev.Mol.Cell Biol. 7:2, 85-96.

TANTI J. F. e JAGER J. (2009) Cellular mechanisms of insulin resistance: role of stress-regulated serine kinases and insulin receptor substrates (IRS) serine phosphorylation. Curr.Opin.Pharmacol. 9:6, 753-762.

TARKOWSKI E., ANDREASEN N., TARKOWSKI A. e BLENNOW K. (2003) Intrathecal inflammation precedes development of Alzheimer's disease. J.Neurol.Neurosurg.Psychiatry 74:9, 1200-1205.

TERRY R. D. (1996) The pathogenesis of Alzheimer disease: an alternative to the amyloid hypothesis. J.Neuropathol.Exp.Neurol. 55:10, 1023-1025.

THAL D. R., CAPETILLO-ZARATE E., DEL T. K. e BRAAK H. (2006) The development of amyloid beta protein deposits in the aged brain. Sci.Aging Knowledge.Environ. 2006:6, re1.

111

TOBINICK E. (2007) Perispinal etanercept for treatment of Alzheimer's disease. Curr.Alzheimer Res. 4:5, 550-552.

TOBINICK E. (2009) Perispinal etanercept for neuroinflammatory disorders. Drug Discov.Today 14:3-4, 168-177.

TOWNSEND M., MEHTA T. e SELKOE D. J. (2007) Soluble Abeta inhibits specific signal transduction cascades common to the insulin receptor pathway. J.Biol.Chem. 282:46, 33305-33312.

TROJANOWSKI J. Q., SMITH A. B., HURYN D. e LEE V. M. (2005) Microtubule-stabilising drugs for therapy of Alzheimer's disease and other neurodegenerative disorders with axonal transport impairments. Expert.Opin.Pharmacother. 6:5, 683-686.

TRUSHINA E., DYER R. B., BADGER J. D., URE D., EIDE L., TRAN D. D., VRIEZE B. T., LEGENDRE-GUILLEMIN V., MCPHERSON P. S., MANDAVILLI B. S., VAN H. B., ZEITLIN S., MCNIVEN M., AEBERSOLD R., HAYDEN M., PARISI J. E., SEEBERG E., DRAGATSIS I., DOYLE K., BENDER A., CHACKO C. e MCMURRAY C. T. (2004) Mutant huntingtin impairs axonal trafficking in mammalian neurons in vivo and in vitro. Mol.Cell Biol. 24:18, 8195-8209.

UCHIDA A., TASHIRO T., KOMIYA Y., YORIFUJI H., KISHIMOTO T. e HISANAGA S. (2004) Morphological and biochemical changes of neurofilaments in aged rat sciatic nerve axons. J.Neurochem. 88:3, 735-745.

UCHIDA T., MYERS M. G., Jr. e WHITE M. F. (2000) IRS-4 mediates protein kinase B signaling during insulin stimulation without promoting antiapoptosis. Mol.Cell Biol. 20:1, 126-138.

ULLRICH A., BELL J. R., CHEN E. Y., HERRERA R., PETRUZZELLI L. M., DULL T. J., GRAY A., COUSSENS L., LIAO Y. C., TSUBOKAWA M. e . (1985) Human insulin receptor and its relationship to the tyrosine kinase family of oncogenes. Nature 313:6005, 756-761.

VALLERIE S. N. e HOTAMISLIGIL G. S. (2010) The role of JNK proteins in metabolism. Sci.Transl.Med. 2:60, 60rv5.

VAN DER HEIDE L. P., KAMAL A., ARTOLA A., GISPEN W. H. e RAMAKERS G. M. (2005) Insulin modulates hippocampal activity-dependent synaptic plasticity in a N-methyl-

112

d-aspartate receptor and phosphatidyl-inositol-3-kinase-dependent manner. J.Neurochem. 94:4, 1158-1166.

VAN DER HEIDE L. P., RAMAKERS G. M. e SMIDT M. P. (2006) Insulin signaling in the central nervous system: learning to survive. Prog.Neurobiol. 79:4, 205-221.

VAN ES M. A. e VAN DEN BERG L. H. (2009) Alzheimer's disease beyond APOE. Nat.Genet. 41:10, 1047-1048.

VERMA G. e DATTA M. (2010) IL-1beta induces ER stress in a JNK dependent manner that determines cell death in human pancreatic epithelial MIA PaCa-2 cells. Apoptosis. 15:7, 864-876.

VILSBOLL T., KRARUP T., MADSBAD S. e HOLST J. J. (2003) Both GLP-1 and GIP are insulinotropic at basal and postprandial glucose levels and contribute nearly equally to the incretin effect of a meal in healthy subjects. Regul.Pept. 114:2-3, 115-121.

VOLIANSKIS A., KOSTNER R., MOLGAARD M., HASS S. e JENSEN M. S. (2010) Episodic memory deficits are not related to altered glutamatergic synaptic transmission and plasticity in the CA1 hippocampus of the APPswe/PS1deltaE9-deleted transgenic mice model of ss-amyloidosis. Neurobiol.Aging 31:7, 1173-1187.

WALSH D. M., KLYUBIN I., FADEEVA J. V., CULLEN W. K., ANWYL R., WOLFE M. S., ROWAN M. J. e SELKOE D. J. (2002) Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal long-term potentiation in vivo. Nature 416:6880, 535-539.

WALSH D. M. e SELKOE D. J. (2004) Deciphering the molecular basis of memory failure in Alzheimer's disease. Neuron 44:1, 181-193.

WANG Q., ROWAN M. J. e ANWYL R. (2004) Beta-amyloid-mediated inhibition of NMDA receptor-dependent long-term potentiation induction involves activation of microglia and stimulation of inducible nitric oxide synthase and superoxide. J.Neurosci. 24:27, 6049-6056.

WANG X., ZHENG W., XIE J. W., WANG T., WANG S. L., TENG W. P. e WANG Z. Y. (2010) Insulin deficiency exacerbates cerebral amyloidosis and behavioral deficits in an Alzheimer transgenic mouse model. Mol.Neurodegener. 5, 46.

113

WELANDER H., FRANBERG J., GRAFF C., SUNDSTROM E., WINBLAD B. e TJERNBERG L. O. (2009) Abeta43 is more frequent than Abeta40 in amyloid plaque cores from Alzheimer disease brains. J.Neurochem. 110:2, 697-706.

WELLEN K. E. e HOTAMISLIGIL G. S. (2005) Inflammation, stress, and diabetes. J.Clin.Invest 115:5, 1111-1119.

WHITE M. F. (2002) IRS proteins and the common path to diabetes. Am.J.Physiol Endocrinol.Metab 283:3, E413-E422.

WHITE M. F. (2003) Insulin signaling in health and disease. Science 302:5651, 1710-1711.

WHITE M. F. (2006) Regulating insulin signaling and beta-cell function through IRS proteins. Can.J.Physiol Pharmacol. 84:7, 725-737.

WHITE M. F. e YENUSH L. (1998) The IRS-signaling system: a network of docking proteins that mediate insulin and cytokine action. Curr.Top.Microbiol.Immunol. 228, 179-208.

WIEDMANN M. W., MOSSNER J., BAERWALD C. e PIERER M. (2009) TNF alpha inhibition as treatment modality for certain rheumatologic and gastrointestinal diseases. Endocr.Metab Immune.Disord.Drug Targets. 9:3, 295-314.

WOODS S. C., LOTTER E. C., MCKAY L. D. e PORTE D., Jr. (1979) Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature 282:5738, 503-505.

XU H., BARNES G. T., YANG Q., TAN G., YANG D., CHOU C. J., SOLE J., NICHOLS A., ROSS J. S., TARTAGLIA L. A. e CHEN H. (2003) Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J.Clin.Invest 112:12, 1821-1830.

YUSTA B., BAGGIO L. L., ESTALL J. L., KOEHLER J. A., HOLLAND D. P., LI H., PIPELEERS D., LING Z. e DRUCKER D. J. (2006) GLP-1 receptor activation improves beta cell function and survival following induction of endoplasmic reticulum stress. Cell Metab 4:5, 391-406.

114

ZHANG H., MA Q., ZHANG Y. W. e XU H. (2012) Proteolytic processing of Alzheimer's beta-amyloid precursor protein. J.Neurochem. 120 Suppl 1, 9-21.

ZHAO W. Q. e ALKON D. L. (2001) Role of insulin and insulin receptor in learning and memory. Mol.Cell Endocrinol. 177:1-2, 125-134.

ZHAO W. Q., CHEN H., QUON M. J. e ALKON D. L. (2004) Insulin and the insulin receptor in experimental models of learning and memory. Eur.J.Pharmacol. 490:1-3, 71-81.

ZHAO W. Q., DE FELICE F. G., FERNANDEZ S., CHEN H., LAMBERT M. P., QUON M. J., KRAFFT G. A. e KLEIN W. L. (2008) Amyloid beta oligomers induce impairment of neuronal insulin receptors. FASEB J. 22:1, 246-260.

ZICK Y. (2005) Ser/Thr phosphorylation of IRS proteins: a molecular basis for insulin resistance. Sci.STKE. 2005:268, e4.

115

ANEXOS

Inflammation, defective insulin signaling, and neuronal dysfunction inAlzheimer’s disease

Sergio T. Ferreira*, Julia R. Clarke, Theresa R. Bomfim, Fernanda G. De FeliceInstitute of Medical Biochemistry Leopoldo de Meis, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil

Abstract A link between Alzheimer’s disease (AD) and metabolic disorders has been established, withpatients with type 2 diabetes at increased risk of developing AD and vice versa. The incidence ofmetabolic disorders, including insulin resistance and type 2 diabetes is increasing at alarming ratesworldwide, primarily as a result of poor lifestyle habits. In parallel, as the world population ages,the prevalence of AD, the most common form of dementia in the elderly, also increases. In additionto their epidemiologic and clinical association, mounting recent evidence indicates shared mecha-nisms of pathogenesis between metabolic disorders and AD. We discuss the concept that peripheraland central nervous system inflammation link the pathogenesis of AD and metabolic diseases. Wealso explore the contribution of brain inflammation to defective insulin signaling and neuronaldysfunction. Last, we review recent evidence indicating that targeting neuroinflammation mayprovide novel therapeutic avenues for AD.! 2014 The Alzheimer’s Association. All rights reserved.

Keywords: Inflammation; Alzheimer’s disease; Diabetes; Obesity; Insulin signaling; Aging

1. Introduction

Inflammation is part of the body’s defense mechanismsagainst multiple threats, including infections and injury.Inflammation is complex and involves both soluble factorsand specialized cells that are mobilized to neutralize andfight threats to restore normal body physiology [1]. Similarinflammatory processes are thought to occur in thebrain and in peripheral tissues. In the brain, glial cells,especially astrocytes and microglia, undergo activationunder pro-inflammatory conditions. In a process similarto that described for peripheral immune cells, activatedmicroglia in the central nervous system (CNS) increaseproduction of inflammatory cytokines. Both in the brainand in peripheral tissues, unchecked or chronic inflamma-tion becomes deleterious, leading to progressive tissuedamage in degenerative diseases.

Inflammation plays critical roles in the pathogenesis ofAlzheimer’s disease (AD) and metabolic diseases, including

type 2 diabetes. These disorders are chronic, debilitating,and extremely costly for health programs in developed anddeveloping countries. Since the Rotterdam study waspublished, suggesting that diabetes almost doubles the riskof AD [2], a number of clinical and epidemiologic studieshave strengthened the link between these diseases [3–6].

Several studies have established further the presence ofinflammatory markers in the AD brain, including elevatedlevels of cytokines/chemokines and gliosis (notably micro-gliosis) in damaged regions [7–10]. A recent meta-analysisshowed that blood concentrations of several inflammatorymediators, including tumor necrosis factor-alpha (TNF-a),interleukin (IL) 6, and IL-1b are increased in AD patients[11].

Overproduction of pro-inflammatory cytokines, includingTNF-a, is a key feature of the pathophysiology of metabolicdisorders. TNF-a is overexpressed in adipose tissue of obeseindividuals [12], and landmark studies by Hotamisligil andcolleagues [12,13] demonstrated that elevated TNF-alevels cause peripheral insulin resistance. Interestingly,brain inflammation has recently been proposed to underliedefective neuronal insulin signaling in AD [14]. Severalpathological features, including impaired insulin signaling

The authors have no conflicts of interest to report.*Corresponding author. Tel.: 155 21 2562 6790; Fax: +55 21 2270-

8647.E-mail address: [email protected]

1552-5260/$ - see front matter ! 2014 The Alzheimer’s Association. All rights reserved.http://dx.doi.org/10.1016/j.jalz.2013.12.010

Alzheimer’s & Dementia 10 (2014) S76–S83

Delta:1_surname

Delta:1_surname

mailto:[email protected]

http://dx.doi.org/10.1016/j.jalz.2013.12.010

http://dx.doi.org/10.1016/j.jalz.2013.12.010

and inflammation, appear to be shared by patients withdiabetes and patients with AD. Therefore, it is likely thatmechanisms analogous to those that account for peripheralinsulin resistance in type 2 diabetes underlie impairedbrain insulin signaling and neuronal dysfunction in AD.In the following sections, we discuss molecular/cellularmechanisms underlying defective brain insulin signalingand neuronal dysfunction in AD, with emphasis on evidencethat AD and diabetes share common inflammatory signalingpathways.

2. Elevated TNF-a and activation of stress kinasesunderlie defective neuronal insulin signaling in AD

The molecular events and pathways leading to disruptedbrain insulin signaling in AD have only recently begun to beunraveled. Insulin and insulin-like growth factor receptorsbelong to the tyrosine kinase receptor family and signalvia insulin receptor substrate (IRS) proteins. These areclosely related, high-molecular weight proteins namedIRS-1 through IRS-4, of which IRS-1 and IRS-2 are themost important and best studied [15,16]. Physiologically,activation of insulin receptors (IRs) in peripheral tissuesstimulates tyrosine phosphorylation of IRS to initiateintracellular signaling pathways. In type 2 diabetes,elevated TNF-a levels trigger serine phosphorylation ofIRS-1 by stress kinases [13,17,18], which interferes withits ability to engage in IR signaling and blocks theintracellular actions of insulin [19–22]. Underlining itsrole in disrupted insulin signaling, blockade of TNF-a inobese mouse models results in improved insulin sensitivityand glucose homeostasis [23,24].

In the brain, TNF-a is secreted mainly by microglial cellsin response to trauma, infection, or abnormal accumulationof protein aggregates [25]. TNF-a levels are elevated inAD cerebrospinal fluid and AD brain microvessels [26,27],as well as in the brain of transgenic mouse models of AD[28,29]. Initial evidence that impaired neuronal insulinsignaling in AD is linked to pro-inflammatory signalingcame from the finding that soluble oligomers of theamyloid-b (Ab) peptide—synaptotoxins that accumulatein AD [30,31]—cause IRS-1 inhibition through TNF-aactivation [14]. In fact, IRS-1 phosphoserine triggeredby Ab oligomers in hippocampal neurons is blocked byinfliximab, a TNF-a neutralizing antibody [14].

Ab oligomers have been shown to instigate removal ofIRs from the membranes of neuronal processes [32,33]and to cause defective insulin signaling (revealed byincreased serine phosphorylation and reduced tyrosinephosphorylation of IRS-1) in postmortem AD brains andin several experimental models of AD [14,34,35]. Insulinsignaling in the CNS promotes neuronal survivaland regulates key processes underlying learning andmemory, including synapse density, dendritic plasticity,and circuit function [36–38]. Thus, the induction of pro-inflammatory pathways and ensuing defective insulin

signaling instigated by Ab oligomers are thought to belinked to neuronal dysfunction in AD.

c-Jun N-terminal kinase (JNK) is the major intracellularstress kinase linking TNF-a to inhibitory serine phosphory-lation of IRS-1 in type 2 diabetes [13,20], and activated JNK(pJNK) is also a feature of human obesity [39–41]. Aboligomer-induced activation of JNK was recently proposedto participate in AD pathology. AD brains exhibit elevatedlevels of pJNK [14,35], and increased pJNK has beendemonstrated in hippocampi of a transgenic mouse modelof AD and in cynomolgus monkeys that receivedintracerebral infusions of Ab oligomers [14].

IkBa kinase (IKK), another stress kinase activated byTNF-a in peripheral insulin resistance [42], also mediatesAb oligomer-induced neuronal IRS-1 inhibition [14]. Ithas now been established that overnutrition induces aninflammatory response in peripheral metabolic tissues.This form of metabolic inflammation, or “metaflammation,”as proposed by Calay and Hotamisligil [43], causesmetabolic defects that underlie type 2 diabetes and obesity[44,45]. In this context, IKK has been identified as a targetfor anti-inflammatory therapy in obesity-associated type 2diabetes [46]. Positive results were obtained in obese micetreated with pharmacological inhibitors of IKK [47,48],providing preclinical support to clinical trials aiming toassess the potential benefit of salsalate, an inhibitor ofIKK, to type 2 diabetes patients [49]. The recently estab-lished involvement of IKK in IRS-1 inhibition in AD pro-vides additional evidence for a close parallelism betweeninflammation-associated defective brain insulin signalingin AD and chronic inflammation-induced insulin resistancein peripheral tissues. Additional studies aiming to explorethe role of IKK in neuronal dysfunction are warranted andmay bring novel clues on mechanisms underlying AD pa-thology.

The double-stranded RNA-dependent protein kinase(PKR), originally identified as a pathogen sensor and a regu-lator of the innate immune response against viral infectionsin higher eukaryotes [50], can regulate or act in conjunctionwith major inflammatory kinases/signaling pathwaysimplicated in metabolic homeostasis, including JNKand IKK [51–53]. Interestingly, PKR is involved in Aboligomer-induced neuronal IRS-1 inhibition [14], rein-forcing the hypothesis that common mechanisms underlieperipheral insulin resistance in type 2 diabetes and impairedbrain insulin signaling in AD.

Current evidence thus indicates that pro-inflammatoryTNF-a signaling and activation of stress-sensitive kinasesplay a key role in inducing IRS-1 inhibition and neuronaldysfunction in AD (Fig. 1).

3. Possible links between peripheral and CNSinflammation

Inflammation in AD has been associated primarily withactivation of CNS-resident microglia induced by Ab

S.T. Ferreira et al. / Alzheimer’s & Dementia 10 (2014) S76–S83 S77

aggregates [55]. Increased brain levels of pro-inflammatorycytokines can lead to several pathological features of AD.Cytokines have been associated with increased tauphosphorylation and decreased synaptophysin levels,establishing their roles in cytoskeletal and synapticalteration in AD [56,57]. Targeting the increasedcirculating levels of IL-1b with a neutralizing antibody hasbeen shown to reduce the activity of several tau kinasesand levels of phosphorylated tau (p-tau), and also toreduce the load of oligomeric and fibrillar Ab in brains oftriple-transgenic AD mice (3xTg) [58]. Interestingly, treat-ment with a drug targeting the p38 mitogen-activated proteinkinase pathway [59] was shown recently to normalize levelsof pro-inflammatory cytokines and to attenuate synaptic pro-tein loss and impaired synaptic plasticity in AD mousemodels [60].

In addition, the clinical association between type 2diabetes and AD has led to the hypothesis that periphery-derived pro-inflammatory molecules could also influencepathogenesis in the Central Nervous System (CNS)(Fig. 1). Adipose tissue inflammation is one of the majortraits of diabetes and obesity [61,62], and both adipocytesand adipose-resident macrophages may participate in a

crosstalk between periphery and CNS. In obese patients, ad-ipocytes react by producing pro-inflammatory cytokines,adipokines, and chemokines, whereas resident macrophagesundergo a phenotypic change to a so-called M1, classically-activated pro-inflammatory state [63]. This leads toincreased TNF-a, IL-1b, and IL-6 production, all of whichcan cross the blood–brain barrier (BBB) [64]. Therefore,adipose-derived inflammatory mediators could be an impor-tant addition to cytokines produced by CNS-resident micro-glia in triggering brain inflammation.

Fat-derived hormones, such as adiponectin and leptin,have been associated with AD and could play a role inconnecting peripheral and central pathogenic mechanisms.Adiponectin, derived from visceral fat, helps sensitize thebody to insulin by acting on receptors that are distributedubiquitously, including the brain [65–68]. Plasmaadiponectin levels are decreased in animal models ofobesity and in obese patients [69–71]. Plasma andcerebrospinal fluid levels of adiponectin show a positivecorrelation [72] and, intriguingly, recent studies foundincreased levels of this adipokine in patients with mildcognitive impairment and with AD [72,73]. It will beinteresting to determine the precise role, if any, of

Fig. 1. Peripheral inflammatory mediators contribute to brain inflammation, neuronal insulin resistance, and neuronal dysfunction in Alzheimer’s disease.Inflamm-aging [54], metaflammation [43] (as seen in obesity-related metabolic disorders, including type 2 diabetes and insulin resistance), and peripheralinfection/inflammation (caused by pathogens or systemic inflammatory disorders) give rise to states of chronic, low-grade systemic inflammation, leadingto overproduction of pro-inflammatory cytokines, such as tumor necrosis factor-a (TNF-a), interleukin (IL) 1b, and IL-6. Elevated levels of adipokinesmay also link peripheral and central nervous system (CNS) inflammation in obesity. Peripheral inflammatory mediators cross the blood–brain barrier and,in conjunction withmediators produced by activatedmicroglia, may lead to CNS inflammation. Activation of neuronal cytokine receptors (e.g., TNF-a receptor)induces aberrant activation of stress kinases (c-Jun N-terminal kinase [JNK], IkBa kinase [IKK], and double-stranded RNA-dependent protein kinase [PKR]),which phosphorylate insulin receptor substrate 1 (IRS-1) at serine residues and inhibit insulin-induced physiological tyrosine phosphorylation of IRS-1.This interferes with the ability of IRS-1 to engage in insulin signaling and blocks the intracellular actions of insulin.

S.T. Ferreira et al. / Alzheimer’s & Dementia 10 (2014) S76–S83S78

adiponectin in impaired neuronal insulin signaling andneuronal dysfunction in AD.

Leptin was found to reduce Ab generation and tauphosphorylation in vitro [74,75], and leptin replacementtherapy induces hippocampal neurogenesis [76] and im-proves cognitive performance [74] in transgenic models ofAD. Initially described for its role in satiety and long-termbody weight maintenance, leptin has recently been proposedto regulate cognition, axonal growth, and synaptogenesis inextrahypothalamic regions [77]. Lower plasma levels ofleptin have been associated with a fourfold increased risk ofdevelopment of AD in a 12-year follow-up period comparedwith patients in whom leptin levels were greater [78].

Dyslipidemia is another important trait of metabolicdisorders. Cholesterol- and sphingolipid-enriched cellmembrane domains, called lipid rafts, appear to be preferen-tial sites for Ab generation. In these microdomains, amyloidprecursor protein is cleaved preferentially through theamyloidogenic pathway [79–81]. Particular types ofsphingolipids, namely ceramide and its metabolites, causeinflammation [82,83], and have been increasinglyassociated with type 2 diabetes [84]. Peripherally generatedceramides can cross the BBB [85] and thus could contributeto AD pathogenesis in two ways: (i) by changing themicroenvironment of lipid rafts, thereby favoring Abgeneration, and/or (ii) by inducing central inflammationand disruption in neuronal insulin signaling [84].

Preclinical and clinical observations are accumulatingthat support the notion that peripheral inflammatorymediators link peripheral and central events in metabolicdyshomeostasis and AD. Novel findings in this area couldpotentially bring about advances for patients with type 2diabetes and patients with AD, allowing for improvedprevention strategies for the former and raising hopes ofnovel therapeutic approaches for the latter. As pointed outrecently, available evidence indicates that a healthy lifestyleand long-term metabolic control are to be encouraged,especially so for diabetic and obese patients, as a preventivemeasure to reduce the risk of AD [86,87].

4. Aging and other general causes of inflammation

Aging is the single most important risk factor for AD. Theconcept of “inflamm-aging” was introduced by ClaudioFranceschi to define changes in the immune system thataccompany physiological aging, leading to chronic systemicinflammation [54,88]. Inflammatory dyshomeostasis inaging can result from loss of control over finely tunedlevels of pro- and anti-inflammatory cytokines, or as aconsequence of the incapacity to restore equilibrium afterit is perturbed by external factors. Whatever the underlyingcauses, they can lead to a chronic state of low-gradeinflammation [89]. Interestingly, it has been proposed thatinflamm-aging is a result of lifetime exposure to acute andchronic infections [90], with human longevity associatedat least in part with an increased capacity to maintain a

inflammatory response at low levels [91]. Nevertheless, apro-inflammatory profile has also been described incentenarians [90]. Increased levels of pro-inflammatorycytokines and markers, such as IL-1b, IL-6, TNF-a, andC-reactive protein [89], have been reported in “normal”aging, but it is not clear whether this is a cause or aconsequence of aging. Inflammation, or rather the defectiveability to maintain low inflammation levels as one ages, maycontribute to the onset of AD (Fig. 1).

Additional conditions in which low-grade inflammationcan be maintained throughout a period of life includerecurrent or persistent infections (Fig. 1), and it is temptingto speculate that chronically elevated peripheral inflamma-tory mediators could be associated with acceleratedneuronal dysfunction and cognitive decline, or predisposeto earlier onset of AD [86]. Significantly, type 2 diabetesinduces changes in BBB permeability [92], and postmortemanalysis of diabetic AD brains showed increased levels ofIL-6 compared with nondiabetic AD brains [93]. Moreover,the BBB of a transgenic mouse model of AD has beenreported to be more permeable to peripheral inflammatorycytokines [94]. These findings raise the possibility that ADbrains could be more susceptible to changes in peripheralinflammatory dyshomeostasis.

5. Anti-inflammatory therapies for AD?

Nonsteroidal anti-inflammatory drugs (NSAIDs) are aheterogeneous family of cyclooxygenase (COX) inhibitors.Their therapeutic outcomes vary according to theirspecificity toward the two known isoforms of the enzyme,COX-1 and COX-2 [95]. Early reports on patients treatedchronically with anti-inflammatory drugs for rheumatoidarthritis or leprosy [96,97] found that a 2-year treatmentwith NSAIDs reduced significantly the risk of AD later inlife [98], and that clinical benefit increased with longerperiods of treatment [99]. Similar results were reported forpatients using aspirin [99] or combined NSAID/steroidtherapy [100]. On the other hand, subsequent studies failedto find a correlation between NSAID use and AD risk[101,102]. Intriguingly, considerable variability was foundeven among studies that showed beneficial actions ofNSAIDs, possibly reflecting differences in dosage, periodof treatment, and patient age. Benefits were observedwhen anti-inflammatory drugs were used well beforethe onset of dementia [103] and might be limited to apolipo-protein E4-carrying patients [104].

Results from the observational studies described hereencouraged the search for molecular mechanisms that couldexplain the positive effects of NSAIDs in delaying/prevent-ing AD. Interestingly, certain NSAIDs, such as ibuprofen,indomethacin, and sulindac sulfide, were shown to decreaseAb42 production by up to 80% in cultured cells, apparentlyvia a COX-independent mechanism [105]. Mice overex-pressing amyloid precursor protein and treated withibuprofen also showed a significant reduction in cortical


amyloid plaque load along with reduced microglial activa-tion [106]. Neurons treated with COX-1 preferential inhibi-tors, such as ibuprofen and aspirin, were more resistant to Abtoxicity than neurons treated with COX-2-specific inhibitors[107]. In line with this finding, Kotilinek and colleagues[108] reported improved synaptic plasticity and memoryformation in AD transgenic mice treated with COX-2- butnot COX-1-selective inhibitors.

Despite the initial optimism generated by these observa-tions, randomized clinical trials provided disappointingresults that did not support disease-modifying or beneficialactions of NSAIDs in AD [109–114]. As suggestedpreviously [103], such trials might have failed in findingpositive effects of NSAIDs for one or more of the followingreasons: (i) most trials tested NSAIDs for treatment ofclinically established AD, whereas epidemiologic datasupported a positive effect of anti-inflammatory agentsbefore clinical onset of symptoms; and (ii) beneficial effectsmay be specific to anti-inflammatory drugs capable ofreducing Ab levels (the so-called selective Ab-loweringagents). Thus, general testing of several members of thelarge NSAID family may have generated conflicting results.

Renewed hopes for an effective AD treatment based onanti-inflammatory therapy came from a report that perispinaladministration of etanercept, an anti-TNF-a fusion protein,improved cognitive performance of patients with AD in a6-month pilot study [115], followed by another report onrapid cognitive improvement (within minutes) on etanerceptadministration to one patient with late-onset AD [116].Despite their possible significance, additional clinicalstudies using larger cohorts and patients at different stagesof the disease are needed to validate these initial findings.In this regard, two clinical trials are currently under way[49] to investigate the possible benefit of etanercept to treatmild/moderate AD.

Anti-TNF-a strategies for AD have received furthersupport from experimental findings using the neutralizingantibody infliximab. Intracerebroventricular infusion ofinfliximab in AD transgenic mice reduced the number ofamyloid plaques and the levels of p-tau [117]. The samegroupsubsequently reported cognitive improvement in one patientwith AD after intrathecal administration of infliximab[118]. Clinical trials are currently investigating the efficacyof infliximab in a wide range of pathologies, including majordepression, obesity-associated insulin resistance, anddiabetic complications, among many others [49]. There areno trials, however, investigating specifically the safety andefficacy of infliximab to treat patients with AD. Given theevidence indicating a role of TNF-a in the pathogenesis ofAD [119], clinical trials using infliximab appear warranted.

Neither etanercept nor infliximab cross the BBB,thus requiring invasive forms of central administration.Thalidomide, on the other hand, is a BBB-permeable drugthat has attracted increasing attention because of itsanti-inflammatory actions related to the inhibition of TNF-aproduction [120]. When given intraperitoneally, both

thalidomide and a synthetic analog (3,60-dithiothalidomide[3,60-DTT] [121]) inhibited lipopolysaccharide-inducedincreases in TNF-a messenger RNA and protein levelsin the cortex [122] and hippocampus [123] of rodents.3,60-DTT, but not thalidomide, reversed cognitive impairmentin 3xTg mice in the eight-arm radial maze [122] and in theMorris Water Maze [123]. Whether 3,60-DTT also rescuesbrain levels of synaptic proteins, p-tau, and Ab42 remainscontroversial [122,123]. Thalidomide also protected ratsfrom neuronal loss induced by intrahippocampal Ab42injection [124]. Thalidomide is now in a phase 2 clinical trialto treat mild to moderate AD, and results will establish itspotential value in preventing or halting the progression ofAD.

6. Conclusion

CNS inflammation, impaired neuronal insulin signaling,and neuronal dysfunction in AD may be a consequence ofsystemic inflammatory processes that occur throughout life[86,87]. Inflammatory mediators—notably pro-inflammatorycytokines such as TNF-a, IL-6, and IL-1b—may play a rolein the crosstalk between peripheral tissues and the brain.Inflamm-aging [54], metaflammation [43], and peripheralinfection/inflammation caused by pathogens or systemicinflammatory disorders may give rise to chronic, low-gradeinflammation. Ultimately, this could contribute to oraccelerate the onset of clinical manifestations of AD.

Acknowledgments

Work in our laboratory was supported by grants from theNational Institute for Translational Neuroscience (INNT/Brazil) (STF), the Human Frontiers Science Program(FGF), and the Brazilian funding agencies ConselhoNacional de Desenvolvimento Cient!ıfico e Tecnol!ogico(CNPq) and Fundac~ao de Amparo "a Pesquisa do Estado doRio de Janeiro (FAPERJ) (FGF, STF, and JRC). TRB issupported by a CNPq predoctoral fellowship and JRC isrecipient of a postdoctoral fellowship from Coordenacaode Aperfeicoamento de Pessoal de Ensino Superior.

References

[1] Brown KL, Cosseau C, Gardy JL, Hancock RE. Complexities oftargeting innate immunity to treat infection. Trends Immunol 2007;28:260–6.

[2] Ott A, Stolk RP, Hofman A, van HF, Grobbee DE, Breteler MM.Association of diabetes mellitus and dementia: the Rotterdam study.Diabetologia 1996;39:1392–7.

[3] Ahtiluoto S, Polvikoski T, Peltonen M, Solomon A, Tuomilehto J,Winblad B, et al. Diabetes, Alzheimer disease, and vasculardementia: a population-based neuropathologic study. Neurology2010;75:1195–202.

[4] Sims-Robinson C, Kim B, Rosko A, Feldman EL. How does diabetesaccelerate Alzheimer disease pathology? Nat Rev Neurol 2010;6:551–9.

[5] Wang KC, Woung LC, Tsai MT, Liu CC, Su YH, Li CY. Risk ofAlzheimer’s disease in relation to diabetes: a population-based cohortstudy. Neuroepidemiology 2012;38:237–44.


http://refhub.elsevier.com/S1552-5260(13)02924-5/sref1
















[6] Crane PK,Walker R, Hubbard RA, Li G, Nathan DM, Zheng H, et al.Glucose levels and risk of dementia. N Engl J Med 2013;369:540–8.

[7] Perry VH, Nicoll JA, Holmes C. Microglia in neurodegenerativedisease. Nat Rev Neurol 2010;6:193–201.

[8] Czirr E, Wyss-Coray T. The immunology of neurodegeneration.J Clin Invest 2012;122:1156–63.

[9] Aguzzi A, Barres BA, Bennett ML. Microglia: scapegoat, saboteur,or something else? Science 2013;339:156–61.

[10] Lee CY, Landreth GE. The role of microglia in amyloid clearancefrom the AD brain. J Neural Transm 2010;117:949–60.

[11] Swardfager W, Lanctot K, Rothenburg L, Wong A, Cappell J,Herrmann N. A meta-analysis of cytokines in Alzheimer’s disease.Biol Psychiatry 2010;68:930–41.

[12] Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM.Increased adipose tissue expression of tumor necrosis factor-alphain human obesity and insulin resistance. J Clin Invest 1995;95:2409–15.

[13] Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF,Spiegelman BM. IRS-1-mediated inhibition of insulin receptortyrosine kinase activity in TNF-alpha- and obesity-induced insulinresistance. Science 1996;271:665–8.

[14] Bomfim TR, Forny-Germano L, Sathler LB, Brito-Moreira J,Houzel JC, Decker H, et al. An anti-diabetes agent protects the mousebrain from defective insulin signaling caused byAlzheimer’s disease-associated Abeta oligomers. J Clin Invest 2012;122:1339–53.

[15] White MF, Yenush L. The IRS-signaling system: a network ofdocking proteins that mediate insulin and cytokine action. Curr TopMicrobiol Immunol 1998;228:179–208.

[16] White MF. IRS proteins and the common path to diabetes. AmJ Physiol Endocrinol Metab 2002;283:E413–22.

[17] Kanety H, Feinstein R, PapaMZ, Hemi R, Karasik A. Tumor necrosisfactor alpha-induced phosphorylation of insulin receptor substrate-1(IRS-1): possible mechanism for suppression of insulin-stimulatedtyrosine phosphorylation of IRS-1. J Biol Chem 1995;270:23780–4.

[18] Rui L, Aguirre V, Kim JK, Shulman GI, Lee A, Corbould A, et al.Insulin/IGF-1 and TNF-alpha stimulate phosphorylation of IRS-1at inhibitory Ser307 via distinct pathways. J Clin Invest 2001;107:181–9.

[19] Nakamura T, Furuhashi M, Li P, Cao H, Tuncman G, Sonenberg N,et al. Double-stranded RNA-dependent protein kinase links pathogensensing with stress and metabolic homeostasis. Cell 2010;140:338–48.

[20] Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K,et al. A central role for JNK in obesity and insulin resistance. Nature2002;420:333–6.

[21] Gao Z, Hwang D, Bataille F, Lefevre M, York D, Quon MJ, et al.Serine phosphorylation of insulin receptor substrate 1 by inhibitorkappa B kinase complex. J Biol Chem 2002;277:48115–21.

[22] Yang J, Park Y, Zhang H, Xu X, Laine GA, Dellsperger KC, et al.Feed-forward signaling of TNF-alpha and NF-kappaB via IKK-beta pathway contributes to insulin resistance and coronary arteriolardysfunction in type 2 diabetic mice. Am J Physiol Heart Circ Physiol2009;296:H1850–8.

[23] Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expressionof tumor necrosis factor-alpha: direct role in obesity-linked insulinresistance. Science 1993;259:87–91.

[24] Samad F, Uysal KT, Wiesbrock SM, Pandey M, Hotamisligil GS,Loskutoff DJ. Tumor necrosis factor alpha is a key component inthe obesity-linked elevation of plasminogen activator inhibitor 1.Proc Natl Acad Sci U S A 1999;96:6902–7.

[25] Park KM, Bowers WJ. Tumor necrosis factor-alpha mediatedsignaling in neuronal homeostasis and dysfunction. Cell Signal2010;22:977–83.

[26] Tobinick E. Perispinal etanercept for treatment of Alzheimer’sdisease. Curr Alzheimer Res 2007;4:550–2.

[27] Grammas P, Ovase R. Inflammatory factors are elevated in brain mi-crovessels in Alzheimer’s disease. Neurobiol Aging 2001;22:837–42.

[28] Ruan L, Kang Z, Pei G, Le Y. Amyloid deposition and inflammationin APPswe/PS1dE9 mouse model of Alzheimer’s disease. CurrAlzheimer Res 2009;6:531–40.

[29] Jin JJ, Kim HD, Maxwell JA, Li L, Fukuchi K. Toll-like receptor4-dependent upregulation of cytokines in a transgenic mouse modelof Alzheimer’s disease. J Neuroinflammation 2008;5:23.

[30] Ferreira ST, Klein WL. The Abeta oligomer hypothesis for synapsefailure and memory loss in Alzheimer’s disease. Neurobiol LearnMem 2011;96:529–43.

[31] Ferreira ST, Vieira MN, De Felice FG. Soluble protein oligomers asemerging toxins in Alzheimer’s and other amyloid diseases. IUBMBLife 2007;59:332–45.

[32] Zhao WQ, De Felice FG, Fernandez S, Chen H, Lambert MP,Quon MJ, et al. Amyloid beta oligomers induce impairment ofneuronal insulin receptors. FASEB J 2008;22:246–60.

[33] De Felice FG, Vieira MN, Bomfim TR, Decker H, Velasco PT,Lambert MP, et al. Protection of synapses against Alzheimer’s-linkedtoxins: insulin signaling prevents the pathogenic binding of Abetaoligomers. Proc Natl Acad Sci U S A 2009;106:1971–6.

[34] Craft S. Alzheimer disease: Insulin resistance and AD: extending thetranslational path. Nat Rev Neurol 2012;8:360–2.

[35] Ma QL, Yang F, Rosario ER, Ubeda OJ, Beech W, Gant DJ,, et al.Beta-amyloid oligomers induce phosphorylation of tau andinactivation of insulin receptor substrate via c-Jun N-terminalkinase signaling: suppression by omega-3 fatty acids and curcumin.J Neurosci 2009;29:9078–89.

[36] Chiu SL, Chen CM, Cline HT. Insulin receptor signaling regulatessynapse number, dendritic plasticity, and circuit function in vivo.Neuron 2008;58:708–19.

[37] Costello DA, Claret M, Al-Qassab H, Plattner F, Irvine EE,Choudhury AI, et al. Brain deletion of insulin receptor substrate2 disrupts hippocampal synaptic plasticity and metaplasticity.PLoS One 2012;7:e31124.

[38] Zhao WQ, Chen H, Quon MJ, Alkon DL. Insulin and the insulinreceptor in experimental models of learning and memory. EurJ Pharmacol 2004;490:71–81.

[39] Boden G, Duan X, Homko C, Molina EJ, Song W, Perez O, et al.Increase in endoplasmic reticulum stress-related proteins and genesin adipose tissue of obese, insulin-resistant individuals. Diabetes2008;57:2438–44.

[40] Gregor MF, Yang L, Fabbrini E, Mohammed BS, Eagon JC,Hotamisligil GS, et al. Endoplasmic reticulum stress is reducedin tissues of obese subjects after weight loss. Diabetes 2009;58:693–700.

[41] Sourris KC, Lyons JG, De Court MP, Dougherty SL, Henstridge DC,CooperME, et al. c-JunNH2-terminal kinase activity in subcutaneousadipose tissue but not nuclear factor-kappaB activity in peripheralblood mononuclear cells is an independent determinant of insulinresistance in healthy individuals. Diabetes 2009;58:1259–65.

[42] Cai D, Yuan M, Frantz DF, Melendez PA, Hansen L, Lee J, et al.Local and systemic insulin resistance resulting from hepaticactivation of IKK-beta and NF-kappaB. Nat Med 2005;11:183–90.

[43] Calay ES, Hotamisligil GS. Turning off the inflammatory, but not themetabolic, flames. Nat Med 2013;19:265–7.

[44] Hotamisligil GS. Inflammation and metabolic disorders. Nature2006;444:860–7.

[45] Lehrke M, Reilly MP, Millington SC, Iqbal N, Rader DJ, Lazar MA.An inflammatory cascade leading to hyperresistinemia in humans.PLoS Med 2004;1:e45.

[46] YuanM, Konstantopoulos N, Lee J, Hansen L, Li ZW, Karin M, et al.Reversal of obesity- and diet-induced insulin resistance withsalicylates or targeted disruption of IKKbeta. Science 2001;293:1673–7.

[47] Reilly SM, Chiang SH, Decker SJ, Chang L, Uhm M, Larsen MJ,et al. An inhibitor of the protein kinases TBK1 and IKK-varepsilonimproves obesity-related metabolic dysfunctions in mice. Nat Med2013;19:313–21.








































































































































[48] Tobar N, Oliveira AG, Guadagnini D, Bagarolli RA, Rocha GZ,Araujo TG, et al. Diacerhein improves glucose tolerance and insulinsensitivity in mice on a high-fat diet. Endocrinology 2011;152:4080–93.

[49] US National Institute of Health. Available at: clinicaltrials.gov. Ac-cessed: November 1, 2013.

[50] Thomis DC, Samuel CE. Mechanism of interferon action: evidencefor intermolecular autophosphorylation and autoactivation of theinterferon-induced, RNA-dependent protein kinase PKR. J Virol1993;67:7695–700.

[51] Bonnet MC, Weil R, Dam E, Hovanessian AG, Meurs EF. PKRstimulates NF-kappaB irrespective of its kinase function byinteracting with the IkappaB kinase complex. Mol Cell Biol 2000;20:4532–42.

[52] Goh KC, deVeer MJ, Williams BR. The protein kinase PKR isrequired for p38 MAPK activation and the innate immune responseto bacterial endotoxin. EMBO J 2000;19:4292–7.

[53] Takada Y, Ichikawa H, Pataer A, Swisher S, Aggarwal BB. Geneticdeletion of PKR abrogates TNF-induced activation of IkappaBalphakinase, JNK, Akt and cell proliferation but potentiates p44/p42MAPK and p38 MAPK activation. Oncogene 2007;26:1201–12.

[54] Vitale G, Salvioli S, Franceschi C. Oxidative stress and the ageingendocrine system. Nat Rev Endocrinol 2013;9:228–40.

[55] Combs CK. Inflammation and microglia actions in Alzheimer’sdisease. J Neuroimmune Pharmacol 2009;4:380–8.

[56] Quintanilla RA, Orellana DI, Gonzalez-Billault C, Maccioni RB.Interleukin-6 induces Alzheimer-type phosphorylation of tauprotein by deregulating the cdk5/p35 pathway. Exp Cell Res 2004;295:245–57.

[57] Li Y, Liu L, Barger SW, Griffin WS. Interleukin-1 mediatespathological effects of microglia on tau phosphorylation and onsynaptophysin synthesis in cortical neurons through a p38-MAPKpathway. J Neurosci 2003;23:1605–11.

[58] Kitazawa M, Cheng D, Tsukamoto MR, Koike MA, Wes PD,Vasilevko V, et al. Blocking IL-1 signaling rescues cognition,attenuates tau pathology, and restores neuronal beta-catenin pathwayfunction in an Alzheimer’s disease model. J Immunol 2011;187:6539–49.

[59] Munoz L, Ralay RH, Roy SM, HuW, Craft JM, McNamara LK, et al.A novel p38 alpha MAPK inhibitor suppresses brain proinflamma-tory cytokine up-regulation and attenuates synaptic dysfunctionand behavioral deficits in an Alzheimer’s disease mouse model.J Neuroinflammation 2007;4:21.

[60] Bachstetter AD, Norris CM, Sompol P, Wilcock DM, Goulding D,Neltner JH, et al. Early stage drug treatment that normalizesproinflammatory cytokine production attenuates synaptic dys-function in a mouse model that exhibits age-dependent progressionof Alzheimer’s disease-related pathology. J Neurosci 2012;32:10201–10.

[61] Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, et al.Chronic inflammation in fat plays a crucial role in the develop-ment of obesity-related insulin resistance. J Clin Invest 2003;112:1821–30.

[62] Ouchi N, Parker JL, Lugus JJ, Walsh K. Adipokines in inflammationand metabolic disease. Nat Rev Immunol 2011;11:85–97.

[63] Lumeng CN, Bodzin JL, Saltiel AR. Obesity induces a phenotypicswitch in adipose tissue macrophage polarization. J Clin Invest2007;117:175–84.

[64] Banks WA. Blood–brain barrier transport of cytokines: a mechanismfor neuropathology. Curr Pharm Des 2005;11:973–84.

[65] Yamauchi T, Kamon J, Waki H, Terauchi Y, Kubota N, Hara K, et al.The fat-derived hormone adiponectin reverses insulin resistanceassociated with both lipoatrophy and obesity. NatMed 2001;7:941–6.

[66] Berg AH, Combs TP, Du X, Brownlee M, Scherer PE. Theadipocyte-secreted protein Acrp30 enhances hepatic insulin action.Nat Med 2001;7:947–53.

[67] Fruebis J, Tsao TS, Javorschi S, Ebbets-Reed D, Erickson MR,Yen FT, et al. Proteolytic cleavage product of 30-kDa adipocytecomplement-related protein increases fatty acid oxidation in muscleand causes weight loss in mice. Proc Natl Acad Sci U S A 2001;98:2005–10.

[68] Yamauchi T, Kamon J, Ito Y, Tsuchida A, Yokomizo T, Kita S, et al.Cloning of adiponectin receptors that mediate antidiabetic metaboliceffects. Nature 2003;423:762–9.

[69] Hotta K, Funahashi T, Bodkin NL, Ortmeyer HK, Arita Y,Hansen BC, et al. Circulating concentrations of the adipocyte proteinadiponectin are decreased in parallel with reduced insulin sensitivityduring the progression to type 2 diabetes in rhesus monkeys. Diabetes2001;50:1126–33.

[70] Ryo M, Nakamura T, Kihara S, Kumada M, Shibazaki S,Takahashi M, et al. Adiponectin as a biomarker of the metabolicsyndrome. Circ J 2004;68:975–81.

[71] Yatagai T, Nagasaka S, Taniguchi A, Fukushima M, Nakamura T,Kuroe A, et al. Hypoadiponectinemia is associated with visceralfat accumulation and insulin resistance in Japanese men with type2 diabetes mellitus. Metabolism 2003;52:1274–8.

[72] Une K, Takei YA, Tomita N, Asamura T, Ohrui T, Furukawa K, et al.Adiponectin in plasma and cerebrospinal fluid in MCI andAlzheimer’s disease. Eur J Neurol 2011;18:1006–9.

[73] van Himbergen TM, Beiser AS, Ai M, Seshadri S, Otokozawa S,Au R, et al. Biomarkers for insulin resistance and inflammationand the risk for all-cause dementia and Alzheimer disease:results from the Framingham Heart Study. Arch Neurol 2012;69:594–600.

[74] Greco SJ, Sarkar S, Johnston JM, Zhu X, Su B, Casadesus G, et al.Leptin reduces Alzheimer’s disease-related tau phosphorylation inneuronal cells. Biochem Biophys Res Commun 2008;376:536–41.

[75] Fewlass DC, Noboa K, Pi-Sunyer FX, Johnston JM, Yan SD,Tezapsidis N. Obesity-related leptin regulates Alzheimer’s Abeta.FASEB J 2004;18:1870–8.

[76] Perez-Gonzalez R, Antequera D, Vargas T, Spuch C, Bolos M,Carro E. Leptin induces proliferation of neuronal progenitorsand neuroprotection in a mouse model of Alzheimer’s disease.J Alzheimers Dis 2011;24:17–25.

[77] Paz-Filho G, Wong ML, Licinio J. The procognitive effects of leptinin the brain and their clinical implications. Int J Clin Pract 2010;64:1808–12.

[78] Lieb W, Beiser AS, Vasan RS, Tan ZS, Au R, Harris TB, et al.Association of plasma leptin levels with incident Alzheimer diseaseand MRI measures of brain aging. JAMA 2009;302:2565–72.

[79] Simons M, Keller P, De SB, Beyreuther K, Dotti CG, Simons K.Cholesterol depletion inhibits the generation of beta-amyloid inhippocampal neurons. Proc Natl Acad Sci U S A 1998;95:6460–4.

[80] Kalvodova L, Kahya N, Schwille P, Ehehalt R, Verkade P,Drechsel D, et al. Lipids as modulators of proteolytic activity ofBACE: involvement of cholesterol, glycosphingolipids, and anionicphospholipids in vitro. J Biol Chem 2005;280:36815–23.

[81] Osenkowski P, Ye W, Wang R, Wolfe MS, Selkoe DJ. Direct andpotent regulation of gamma-secretase by its lipid microenvironment.J Biol Chem 2008;283:22529–40.

[82] Bikman BT. A role for sphingolipids in the pathophysiology ofobesity-induced inflammation. Cell Mol Life Sci 2012;69:2135–46.

[83] Holland WL, Bikman BT, Wang LP, Yuguang G, Sargent KM,Bulchand S, et al. Lipid-induced insulin resistance mediated bythe proinflammatory receptor TLR4 requires saturated fattyacid-induced ceramide biosynthesis in mice. J Clin Invest 2011;121:1858–70.

[84] de la Monte SM. Triangulated mal-signaling in Alzheimer’s disease:roles of neurotoxic ceramides, ER stress, and insulin resistancereviewed. J Alzheimers Dis 2012;30:S231–49.

[85] Smith QR, Nagura H. Fatty acid uptake and incorporation in brain:studies with the perfusion model. J Mol Neurosci 2001;16:167–72.






http://clinicaltrials.gov































































































































[86] De Felice FG. Alzheimer’s disease and insulin resistance: translatingbasic science into clinical applications. J Clin Invest 2013;123:531–9.

[87] Mattson MP. Energy intake and exercise as determinants of brainhealth and vulnerability to injury and disease. Cell Metab 2012;16:706–22.

[88] Salvioli S, Monti D, Lanzarini C, ConteM, Pirazzini C, Bacalini MG,et al. Immune system, cell senescence, aging and longevity:inflamm-aging reappraised. Curr Pharm Des 2013;19:1675–9.

[89] Salvioli S, Capri M, Valensin S, Tieri P, Monti D, Ottaviani E, et al.Inflamm-aging, cytokines and aging: state of the art, new hypotheseson the role of mitochondria and new perspectives from systemsbiology. Curr Pharm Des 2006;12:3161–71.

[90] Franceschi C, Bonafe M, Valensin S, Olivieri F, De LM, Ottaviani E,et al. Inflamm-aging: an evolutionary perspective on immuno-senescence. Ann N YAcad Sci 2000;908:244–54.

[91] Baggio G, Donazzan S, Monti D, Mari D, Martini S, Gabelli C, et al.Lipoprotein(a) and lipoprotein profile in healthy centenarians:a reappraisal of vascular risk factors. FASEB J 1998;12:433–7.

[92] Acharya NK, Levin EC, Clifford PM, Han M, Tourtellotte R,Chamberlain D, et al. Diabetes and hypercholesterolemia increaseblood–brain barrier permeability and brain amyloid deposition:beneficial effects of the LpPLA2 inhibitor darapladib. J AlzheimersDis 2013;35:179–98.

[93] Sonnen JA, Larson EB, Brickell K, Crane PK,Woltjer R, Montine TJ,et al. Different patterns of cerebral injury in dementia with or withoutdiabetes. Arch Neurol 2009;66:315–22.

[94] Takeda S, Sato N, Ikimura K, Nishino H, Rakugi H, Morishita R.Increased blood–brain barrier vulnerability to systemic inflammationin an Alzheimer disease mouse model. Neurobiol Aging 2013;34:2064–70.

[95] Kaufmann WE, Andreasson KI, Isakson PC, Worley PF. Cyclo-oxygenases and the central nervous system. Prostaglandins 1997;54:601–24.

[96] McGeer PL, Rogers J. Anti-inflammatory agents as a therapeuticapproach to Alzheimer’s disease. Neurology 1992;42:447–9.

[97] Wyss-Coray T, Rogers J. Inflammation in Alzheimer disease: a briefreview of the basic science and clinical literature. Cold Spring HarbPerspect Med 2012;2:a006346.

[98] in t’ Veld BA, Ruitenberg A, Hofman A, Launer LJ, van Duijn CM,Stijnen T, et al. Nonsteroidal antiinflammatory drugs and the risk ofAlzheimer’s disease. N Engl J Med 2001;345:1515–21.

[99] Zandi PP, Anthony JC, Hayden KM, Mehta K, Mayer L, Breitner JC.Reduced incidence of AD with NSAID but not H2 receptorantagonists: the Cache County study. Neurology 2002;59:880–6.

[100] McGeer PL, Schulzer M, McGeer EG. Arthritis and anti-inflammatory agents as possible protective factors for Alzheimer’sdisease: a review of 17 epidemiologic studies. Neurology 1996;47:425–32.

[101] Breitner JC, Haneuse SJ, Walker R, Dublin S, Crane PK, Gray SL,et al. Risk of dementia and AD with prior exposure to NSAIDs inan elderly community-based cohort. Neurology 2009;72:1899–905.

[102] Arvanitakis Z, Grodstein F, Bienias JL, Schneider JA, Wilson RS,Kelly JF, et al. Relation of NSAIDs to incident AD, change incognitive function, and AD pathology. Neurology 2008;70:2219–25.

[103] Szekely CA, Zandi PP. Non-steroidal anti-inflammatory drugsand Alzheimer’s disease: the epidemiological evidence. CNS NeurolDisord Drug Targets 2010;9:132–9.

[104] Szekely CA, Breitner JC, Fitzpatrick AL, Rea TD, Psaty BM,Kuller LH, et al. NSAID use and dementia risk in the CardiovascularHealth Study: role of APOE and NSAID type. Neurology 2008;70:17–24.

[105] Weggen S, Eriksen JL, Das P, Sagi SA, Wang R, Pietrzik CU, et al. Asubset of NSAIDs lower amyloidogenic Abeta42 independently ofcyclooxygenase activity. Nature 2001;414:212–6.

[106] Yan Q, Zhang J, Liu H, Babu-Khan S, Vassar R, Biere AL, et al.Anti-inflammatory drug therapy alters beta-amyloid processing and

deposition in an animal model of Alzheimer’s disease. J Neurosci2003;23:7504–9.

[107] Bate C, Veerhuis R, Eikelenboom P, Williams A. Neurones treatedwith cyclo-oxygenase-1 inhibitors are resistant to amyloid-beta1-42. Neuroreport 2003;14:2099–103.

[108] Kotilinek LA, Westerman MA, Wang Q, Panizzon K, Lim GP,Simonyi A, et al. Cyclooxygenase-2 inhibition improves amyloid-beta-mediated suppression of memory and synaptic plasticity. Brain2008;131:651–64.

[109] Scharf S, Mander A, Ugoni A, Vajda F, Christophidis N.A double-blind, placebo-controlled trial of diclofenac/misoprostolin Alzheimer’s disease. Neurology 1999;53:197–201.

[110] Aisen PS, Schmeidler J, Pasinetti GM. Randomized pilot studyof nimesulide treatment in Alzheimer’s disease. Neurology 2002;58:1050–4.

[111] Aisen PS, Schafer KA, Grundman M, Pfeiffer E, Sano M, Davis KL,et al. Effects of rofecoxib or naproxen vs placebo on Alzheimerdisease progression: a randomized controlled trial. JAMA 2003;289:2819–26.

[112] Reines SA, Block GA, Morris JC, Liu G, Nessly ML, Lines CR, et al.Rofecoxib: no effect on Alzheimer’s disease in a 1-year, randomized,blinded, controlled study. Neurology 2004;62:66–71.

[113] Pasqualetti P, Bonomini C, Dal FG, Paulon L, Sinforiani E, Marra C,et al. A randomized controlled study on effects of ibuprofen oncognitive progression of Alzheimer’s disease. Aging Clin Exp Res2009;21:102–10.

[114] Green RC, Schneider LS, Amato DA, Beelen AP, Wilcock G,Swabb EA, et al. Effect of tarenflurbil on cognitive decline andactivities of daily living in patients with mild Alzheimer disease:a randomized controlled trial. JAMA 2009;302:2557–64.

[115] Tobinick E, Gross H, Weinberger A, Cohen H. TNF-alphamodulation for treatment of Alzheimer’s disease: a 6-month pilotstudy. Med Gen Med 2006;8:25.

[116] Tobinick EL, Gross H. Rapid cognitive improvement inAlzheimer’s disease following perispinal etanercept administration.J Neuroinflammation 2008;5:2.

[117] Shi JQ, Shen W, Chen J, Wang BR, Zhong LL, Zhu YW, et al.Anti-TNF-alpha reduces amyloid plaques and tau phosphorylationand induces CD11c-positive dendritic-like cell in the APP/PS1transgenic mouse brains. Brain Res 2011;1368:239–47.

[118] Shi JQ,WangBR, JiangWW,Chen J,ZhuYW,ZhongLL, et al. Cogni-tive improvement with intrathecal administration of infliximab in awoman with Alzheimer’s disease. J Am Geriatr Soc 2011;59:1142–4.

[119] Clark I, Atwood C, Bowen R, Paz-Filho G, Vissel B. Tumornecrosis factor-induced cerebral insulin resistance in Alzheimer’sdisease links numerous treatment rationales. Pharmacol Rev 2012;64:1004–26.

[120] Tweedie D, Sambamurti K, Greig NH. TNF-alpha inhibition as atreatment strategy for neurodegenerative disorders: new drugcandidates and targets. Curr Alzheimer Res 2007;4:378–85.

[121] Zhu X, Giordano T, Yu QS, Holloway HW, Perry TA, Lahiri DK, et al.Thiothalidomides: novel isosteric analogues of thalidomide withenhancedTNF-alpha inhibitory activity. JMedChem2003;46:5222–9.

[122] Gabbita SP, Srivastava MK, Eslami P, Johnson MF, Kobritz NK,Tweedie D, et al. Early intervention with a small molecule inhibitorfor tumor necrosis factor-alpha prevents cognitive deficits in a tripletransgenic mouse model of Alzheimer’s disease. J Neuroinflamma-tion 2012;9:99.

[123] Tweedie D, Ferguson RA, Fishman K, Frankola KA, Van PH,Holloway HW, et al. Tumor necrosis factor-alpha synthesis inhibitor3,60-dithiothalidomide attenuatesmarkers of inflammation, Alzheimerpathology and behavioral deficits in animalmodels of neuroinflamma-tion and Alzheimer’s disease. J Neuroinflammation 2012;9:106.

[124] Ryu JK, McLarnon JG. Thalidomide inhibition of perturbedvasculature and glial-derived tumor necrosis factor-alpha in ananimal model of inflamed Alzheimer’s disease brain. Neurobiol Dis2008;29:254–66.









































































































































Cell Metabolism

Article

TNF-a Mediates PKR-Dependent Memory Impairmentand Brain IRS-1 Inhibition Induced by Alzheimer’sb-Amyloid Oligomers in Mice and MonkeysMychael V. Lourenco,1 Julia R. Clarke,1 Rudimar L. Frozza,1 Theresa R. Bomfim,1 Letıcia Forny-Germano,1

Andre F. Batista,1 Luciana B. Sathler,1 Jordano Brito-Moreira,1 Olavo B. Amaral,1 Cesar A. Silva,1 Leo Freitas-Correa,1

Sheila Espırito-Santo,3 Paula Campello-Costa,3 Jean-Christophe Houzel,2 William L. Klein,4 Christian Holscher,5

Jose B. Carvalheira,6 Aristobolo M. Silva,7,8 Lıcio A. Velloso,9 Douglas P. Munoz,10 Sergio T. Ferreira,1,*and Fernanda G. De Felice1,*1Institute of Medical Biochemistry Leopoldo de Meis2Institute of Biomedical SciencesFederal University of Rio de Janeiro, Rio de Janeiro, RJ 21941-902, Brazil3Institute of Biology, Fluminense Federal University, Niteroi, RJ 24001-970, Brazil4Department of Neurobiology, Northwestern University, Evanston, IL 60208, USA5Faculty of Health and Medicine, Biomedical and Life Sciences, Lancaster University, Lancaster LA1 4YQ, UK6Department of Internal Medicine, Faculty of Medical Sciences, UNICAMP, Campinas, SP 13084-761, Brazil7Institute of Biological Sciences, Federal University of Minas Gerais, Belo Horizonte, MG 31270-901, Brazil8Rene Rachou Research Center, Fiocruz, Minas Gerais, Belo Horizonte, MG 30190-002, Brazil9Laboratory of Cell Signalling, Obesity and Comorbidities Research Centre, University of Campinas, DCM-FCM UNICAMP, Campinas,SP 13084-761, Brazil10Centre for Neuroscience Studies, Queen’s University, Kingston, ON K7L3N6, Canada*Correspondence: [email protected] (S.T.F.), [email protected] (F.G.D.F.)http://dx.doi.org/10.1016/j.cmet.2013.11.002

SUMMARY

Alzheimer’s disease (AD) and type 2 diabetesappear to share similar pathogenic mechanisms.dsRNA-dependent protein kinase (PKR) underliesperipheral insulin resistance in metabolic disor-ders. PKR phosphorylates eukaryotic translationinitiation factor 2a (eIF2a-P), and AD brainsexhibit elevated phospho-PKR and eIF2a-P levels.Whether and how PKR and eIF2a-P participatein defective brain insulin signaling and cognitiveimpairment in AD are unknown. We report thatb-amyloid oligomers, AD-associated toxins, acti-vate PKR in a tumor necrosis factor a (TNF-a)-dependent manner, resulting in eIF2a-P, neuronalinsulin receptor substrate (IRS-1) inhibition, syn-apse loss, and memory impairment. Brain phos-pho-PKR and eIF2a-P were elevated in AD animalmodels, including monkeys given intracerebro-ventricular oligomer infusions. Oligomers failedto trigger eIF2a-P and cognitive impairment inPKR!/! and TNFR1!/! mice. Bolstering insulinsignaling rescued phospho-PKR and eIF2a-P.Results reveal pathogenic mechanisms shared byAD and diabetes and establish that proinflamma-tory signaling mediates oligomer-induced IRS-1 in-hibition and PKR-dependent synapse and memoryloss.

INTRODUCTION

Recent evidence suggests that Alzheimer’s disease (AD) is anovel, brain-specific form of diabetes (de la Monte and Wands,2008; De Felice, 2013). AD brains exhibit defective insulinsignaling with altered levels and cellular distribution of insulin re-ceptors (Moloney et al., 2010). Insulin signaling is central toneuronal survival, regulation of synapse number, and dendriticplasticity (van der Heide et al., 2005; Chiu et al., 2008; McNayand Recknagel, 2011), raising the possibility that deficient insulinsignaling may be linked to neuronal dysfunction in AD.b-amyloid oligomers (AbOs), toxins that accumulate in AD

brains and instigate synapse damage (Ferreira and Klein,2011), remove insulin receptors from the neuronal surface(Zhao et al., 2008; De Felice et al., 2009) and activate c-JunN-terminal kinase (JNK) to trigger insulin receptor substrate(IRS-1) inhibition (Bomfim et al., 2012). These findings providedinitial clues on how impaired neuronal insulin signaling de-velops in AD (De Felice, 2013). Landmark studies from the dia-betes field have established that activation of proinflammatorytumor necrosis factor alpha and JNK signaling (TNF-a/JNKsignaling) is a key mechanism leading to peripheral insulinresistance (Hotamisligil et al., 1994, 1996; Gregor and Hotami-sligil, 2011). Therefore, it is likely that a molecular parallel existsbetween defective brain insulin signaling in AD and peripheralinsulin resistance in diabetes (Bomfim et al., 2012; De Felice,2013).The double-stranded RNA-dependent protein kinase (PKR) is

a critical player in the integration of an inflammatory responsethat leads to peripheral insulin resistance in metabolic disorders,

Cell Metabolism 18, 831–843, December 3, 2013 ª2013 Elsevier Inc. 831

including diabetes (Ozcan et al., 2004; Hotamisligil, 2010; Naka-mura et al., 2010). Elevated levels of phosphorylated PKR and ofits key target, eukaryotic translation initiation factor 2a (eIF2a),have been reported in AD brains and in animal models of AD(Hoozemans et al., 2009; Yoon et al., 2012). eIF2a phosphoryla-tion is critical for memory regulation (Costa-Mattioli et al., 2007)and has been recently shown to mediate prion-related neurode-generation in the hippocampus (Moreno et al., 2012), a memorycenter that is affected early in AD.

Given the pathophysiological roles of TNF-a and PKR inperipheral insulin resistance (Nakamura et al., 2010; Gregorand Hotamisligil, 2011), we investigated here whether similarproinflammatory mechanisms might underlie neuronal dysfunc-tion in AD. We hypothesized that the TNF-a pathway mightcause phosphorylation of PKR and eIF2a-P, IRS-1 inhibition,and impact synapses and memory in AD. We show that phos-pho-PKR and eIF2a-P are elevated in the brains of a transgenicmouse model of AD, of mice and cynomolgus monkeys givenintracerebroventricular (i.c.v.) infusions of AbOs, and in culturedhippocampal neurons exposed to oligomers. AbOs failed totrigger eIF2a-P and cognitive impairment in both PKR!/! andTNFR1!/! mice, as well as in mice treated with infliximab, aTNF-a neutralizing antibody. Salubrinal and thapsigargin, agentsthat increase eIF2a-P levels and endoplasmic reticulum (ER)stress, respectively, inducedmemory impairment inmice. Insulintreatment prevented oligomer-induced phosphorylation of PKRand eIF2a-P in hippocampal cultures. Glucagon-like peptide 1(GLP-1) receptor agonists blocked eIF2a-P in hippocampal cul-tures and in the brains of transgenic mice and oligomer-injectedmonkeys. Collectively, results provide the grounds for targetingTNF-a/PKR/eIF2a-P signaling as a potential disease-modifyingtherapy for AD.

RESULTS

Ab Oligomers Instigate Neuronal eIF2a-P and Other ERStress Markers In Vitro and In VivoWe first explored whether AbOs abnormally activate theunfolded protein response (UPR), which has been described tointersect with inflammatory and stress signaling pathways thatlead to peripheral insulin resistance in chronic metabolic dis-eases (Ozcan et al., 2004, 2006; Hotamisligil, 2010). To thisend, we searched for ER stress markers in cell cultures and indifferent animal models of AD. Usingmature cultured hippocam-pal neurons, we found increased IRE1a-pSer724 in dendritesand cell bodies after exposure of neurons to AbOs for 3 hr (Fig-ures 1A and 1B and S1A available online). Consistent withincreased endonuclease activity of IRE1a upon phosphorylation,levels of spliced X box binding protein 1 (XBP1), a downstreameffector of IRE1a recently proposed as a connection betweenUPR and insulin signaling (Park et al., 2010; Winnay et al.,2010), were increased in oligomer-exposed neurons (Fig-ure S1B). Further, AbOs increased dendritic and cell body levelsof eIF2a-pSer51 (eIF2a-P) in hippocampal neurons (Figures 1C,1D, and S1C).

We also detected increased neuronal levels of 78 kDaglucose-regulated protein (GRP78) (also known as bindingimmunoglobulin protein, Bip), an ER chaperone that is part ofthe stress response program and is upregulated in AD brains

(Hoozemans et al., 2005), in oligomer-exposed neurons (Fig-ure S1D). Prolonged ER stress is known to trigger apoptosismediated by C/EBP homologous protein (CHOP; also knownas growth arrest- and DNA damage-inducible gene 153,GADD153) (Lai et al., 2007). In line with the results describedabove, CHOP mRNA levels were increased in neurons exposedto AbOs for 24 hr (Figure S1E).Next, we looked for eIF2a-P in the brains of APPSwe,PS1DE9

(APP/PS1) mice. These mice harbor transgenes for human amy-loid precursor protein (APP) bearing the Swedish mutationand a deletion mutant form of presenilin 1 (PS1) and presentincreased Ab production and cognitive deficits (Jankowskyet al., 2001). APP/PS1 mice displayed increased hippocampallevels of eIF2a-P compared to wild-type animals (Figure 1E).

AbO-Induced PKR Activation Leads to eIF2aPhosphorylationFour kinases (PKR-like endoplasmic reticulum kinase [PERK],double-stranded RNA-dependent protein kinase [PKR], generalcontrol nonrepressed kinase 2 [GCN2], and heme-regulated in-hibitor [HRI]) have been reported to phosphorylate eIF2a understress conditions (Gkogkas et al., 2010). PERK and PKR arethe main eIF2a kinases in response to ER stress and inflamma-tion, respectively (Raven and Koromilas, 2008). We thereforeexamined the roles of both kinases in oligomer-induced eIF2a-Pin neurons. In hippocampal neuronal cultures, the distributionof activated PERK (PERK-pThr981) was mainly restricted tocell bodies, and its levels were not altered by exposure toAbOs for 3 hr (Figures S2A and S2B). In contrast, PKR was acti-vated by exposure to AbOs in neuronal cultures (Figures 1F and1G). AbOs are known to selectively target a subset of neurons inhippocampal cultures (Lacor et al., 2007). Importantly, elevatedphospho-PKR levels were found independent of whether or notneurons exhibited oligomers bound to their dendrites (Figure 1H).This indicates that PKR phosphorylation is not triggered bydirect binding of oligomers to individual neurons, but rather isinstigated by soluble factors released to the medium uponexposure of cultures to AbOs.To establish the in vivo relevance of these findings, we

analyzed levels of active PKR and PERK in the brains of APP/PS1 transgenic mice. Increased phosphorylation of PKR, butnot of PERK, was detected in hippocampi of APP/PS1 micecompared to wild-type controls (Figures 1I and S2C). Consistentwith the role of PKR in AbO-dependent eIF2a-P, pharmacolog-ical inhibition of PKR completely blocked oligomer-inducedeIF2a-P aswell as IRE1a-pSer724 (Figures 1J and 1K) in culturedhippocampal neurons. We further extended our investigationto the brains of monkeys that received i.c.v. infusions of AbOs.In monkeys, AbOs promoted eIF2a-P (Figures 1L, 1M, S1F,and S1G) and PKR activation in the hippocampus and entorhinalcortex (Figures 1N, 1O, S1H, and S1I).

TNF-a Receptor Activation Lies Upstream of PKRDysregulationIn metabolic disorders, ER stress has been linked to insulin resis-tance and proinflammatory TNF-a signaling (Ozcan et al., 2006;Steinberg et al., 2006). To determine whether TNF-a activationwas involved in AbO-induced phospho-PKR and eIF2a-P in hip-pocampal neurons, we first treated cultures with infliximab, a

Cell Metabolism

TNF-a, PKR, Insulin Signaling, and Memory

832 Cell Metabolism 18, 831–843, December 3, 2013 ª2013 Elsevier Inc.

Figure 1. b-Amyloid Oligomers Trigger Neuronal IRE1a-pSer724, eIF2a-P, and Phospho-PKR In Vitro and In Vivo(A and B) Shown is IRE1a-pSer724 immunolabeling (A) and levels (B) in cultured hippocampal neurons exposed to vehicle or 500 nM AbOs for 3 hr (scale bars =

10 mm). Boxes under each panel show optical zoom images of selected dendrite segments (white dashed rectangles).

(C and D) Shown is eIF2a-P immunolabeling (C) and levels (D), determined from 4–6 experiments using independent cultures (30 images analyzed per experi-

mental condition per experiment).

(E) Immunoblot analysis of eIF2a-P in hippocampal homogenates from 13- to 16-month-old WT (n = 5) or APP/PS1 (n = 7) mice.

(F) Immunolabeling for PKR-pThr451 in cultured hippocampal neurons exposed to vehicle or 500 nM AbOs for 3 hr (scale bar = 10 mm).

(G) PKR-pThr451 immunofluorescence levels, determined from four experiments using independent cultures (30 images analyzed per experimental condition per

experiment).

(H) Immunolabeling of DAPI (blue), AbOs (red), and phospho-PKR (green) in cultured hippocampal neurons. White arrow indicates a neuron with elevated

phospho-PKR level in the absence of oligomer binding.

(I) Immunoblot analysis of PKR-pThr451 (normalized by total PKR) in hippocampal homogenates of 13- to 16-month-old WT (n = 5) and APP/PS1 (n = 7) mice.

*p < 0.05, **p < 0.01, Student’s t test.

(J) eIF2a-P immunolabeling in hippocampal neurons exposed to vehicle, 500 nM AbOs, or 1 mM PKR inhibitor + 500 nM AbOs for 3 hr (scale bar = 10 mm).

(K) eIF2a-P and IRE1a-pSer724 immunofluorescence levels, determined from three experiments using independent cultures. *p < 0.05, ANOVA followed by

Bonferroni post hoc test.

(L and M) Shown is eIF2a-P immunolabeling (L) and integrated optical densities (M) in hippocampus (CA1 and CA3 regions) and entorhinal cortex of cynomolgus

monkeys that received i.c.v. injections of AbOs (n = 4) (right panels) compared to sham-operated control monkeys (n = 3) (left panels) (scale bars = 200 mm).

(N and O) Shown is phosphor-PKR labeling (N) and integrated optical densities (O) (see Experimental Procedures) in hippocampus (CA1 and CA3 regions) and

entorhinal cortex of sham-operated (Sham) (n = 3) or AbO-injected monkeys (AbOs) (n = 4). *p < 0.05, ANOVA followed by Bonferroni post hoc test. Graphs show

means ± SEM. See also Figure S1.

Cell Metabolism



TNF-a neutralizing monoclonal antibody. Infliximab suppressedboth PKR activation and eIF2a-P triggered by oligomers (Figures2A–2E). SP600125, a specific JNK inhibitor, also blocked eI-F2a-P in neuronal cultures (Figures 2D and 2E), in line with recentstudies that have implicated the JNK pathway in AbO-inducedIRS-1 inhibition (Ma et al., 2009; Bomfim et al., 2012; De Felice,2013). It is notable that infliximab did not block oligomer bindingto neurons (Figures 2F and 2G), substantiating the notion thatactivation of TNF-a/PKR signaling is independent of direct bind-ing of AbOs to individual neurons and is likely mediated by TNF-asecreted to the medium. Indeed, TNF-a levels were increased inthe culture medium after exposure to oligomers (Figure 2H),consistent with our recent finding that TNF-a levels are increasedin the brains of mice that received i.c.v infusions of oligomers(Ledo et al., 2013).

Activation of the TNF-a/PKR/eIF2a-P Pathway Is Linkedto Synapse Loss and Memory ImpairmentWe next hypothesized that activation of proinflammatory TNF-asignaling might be connected to AbO-induced memory impair-ment. Supporting this hypothesis, infliximab prevented memoryimpairment triggered by AbOs in mice (Figure 3A). We nextinjected oligomers i.c.v. in TNFR1!/! mice. Significantly, AbOscaused memory deficits in wild-type (WT) mice, but not inTNFR1!/!mice (Figures 3B, 3C, and S3). These results implicateTNF-a signaling in the mechanism underlying memory impair-ment induced by AbOs in mice. Interestingly, AbOs triggered

phosphorylation of PKR and eIF2a in the hippocampus of WTmice, but not in TNFR1!/!mice (Figures 3D and 3E), establishingthat activation of TNF-a receptors lies upstream of PKR andeIF2a-P in vivo.Synapse loss has been proposed to be the best pathological

correlate of the extent of dementia in AD (Terry et al., 1991), andAD brains present reduced levels of synaptophysin and PSD-95, pre- and postsynaptic markers, respectively (Terry et al.,1991).AbOshavebeenshown to reduce the levelsof synaptophy-sin or PSD-95 in neuronal cultures, in the brains of mice receivingi.c.v. infusions of oligomers, and in ex vivo human cortical slices(Sebollela et al., 2012; Figueiredo et al., 2013). Supporting thenotion that synaptic deterioration underlies TNF-a-dependentmemory impairment induced by AbOs, levels of synaptophysinand PSD-95 were decreased in hippocampi of WT, but not ofTNFR1!/!, mice i.c.v. injected with AbOs (Figure 3F).To determine the role of PKR in AbO-induced cognitive impair-

ment, we investigated the effect of AbOs in PKR!/!mice. Impor-tantly, we found that AbOs induced hippocampal eIF2a-P andcognitive deficits in WT mice, but not in PKR!/! mice (Figures3G–3I). These results demonstrate that PKR, recently implicatedin metabolic stress and impaired insulin signaling in diabetes(Nakamura et al., 2010), is a key mediator of neuronal eIF2a-Pand memory impairment induced by AbOs.To further examine the impact of PKR activation on synapses,

we exposed hippocampal cultures to AbOs in the absence orpresence of a PKR inhibitor. After 24 hr, we evaluated synapse

Figure 2. TNF-a Mediates Oligomer-Induced Phospho-PKR and eIF2a-P(A) PKR-pThr451 immunolabeling in cultured hip-

pocampal neurons exposed for 3 hr to vehicle or

500 nM AbOs in the absence or presence of in-

fliximab.

(B) PKR-pThr451 immunofluorescence levels,

determined from four experiments using inde-

pendent cultures.

(C) eIF2a-P immunolabeling in cultured hippo-

campal neurons exposed for 3 hr to vehicle or

500 nM AbOs in the absence or presence of in-

fliximab.

(D) eIF2a-P immunofluorescence levels, deter-

mined from four experiments using independent

cultures exposed to vehicle or 500 nM AbOs in the

absence or presence of infliximab or JNK inhibitor

(JNKi). **p < 0.01, ANOVA followed by Bonferroni

post hoc test.

(E) Immunoblot analysis for eIF2a-P in hippo-

campal neuronal cultures exposed for 3 hr to

500 nM AbOs in the absence or in the presence of

infliximab or JNK inhibitor (n = 3 independent ex-

periments; *p < 0.05, Student’s t test).

(F) Imunolabeling of AbOs (with anti-AbO, NU4) in

neurons exposed to 500 nM AbOs in the absence

or presence of infliximab.

(G) Quantification of Ab oligomer binding in neu-

rons exposed to AbOs in the absence or presence

of infliximab.

(H) Soluble TNF-a levels in conditionedmedia from

cultured hippocampal neurons exposed to vehicle

(veh) or 500 nM AbOs for 3 hr (n = 3 independent

experiments; *p < 0.05, Student’s t test). Graphs

show means ± SEM. See also Figure S2.

Cell Metabolism



density by determining immunoreactivities of synaptophysin andPSD-95 and their colocalization at synapses. Results showedthat inhibition of PKR attenuated synapse loss induced byAbOs (Figures 3J–3M) and suggest that loss of synaptic proteinsunderlies the deleterious effects of PKR on memory.Based on the findings described above, we hypothesized that

elevated eIF2a-P in the brain might be linked to memory impair-ment triggered by AbOs. To explore the connection betweeneIF2a-P and memory, we performed i.p. injections of salubrinal(an inhibitor of GADD34, a phosphatase that preferentiallydephosphorylates eIF2a) to increase eIF2a-P levels in the hippo-campi of adult Swiss mice. Animals that received salubrinalpresented memory impairment in the novel object recognitiontask (Figure 4A) and significantly lower hippocampal levels of

PSD-95 and synaptophysin (Figures 4B–4D). We also founddecreased levels of synaptophysin and PSD-95, as well asdecreased synapse density, in salubrinal-exposed hippocampalcultures (Figures 4E–4H). Control experiments indicated that, asexpected, salubrinal promoted eIF2a-P in hippocampal neuronsin culture (Figures S4A and S4B) and in the brains of injectedmice (Figure S4C). These results show that eIF2a-P is sufficientto cause memory impairment and synapse loss in mice.Additional support to the notion that eIF2a-P and, more gener-

ally, ER stress cause memory impairment came from the obser-vation that thapsigargin, a classical inducer of ER stress thattriggers eIF2a-P in neurons (Figures S5A and S5B), caused braineIF2a-P and memory impairment when injected i.c.v. in mice(Figures S5C and S5D). 4-phenylbutyrate (4-PBA), a chemical

Figure 3. TNF-a/PKR Signaling MediatesCognitive Impairment and Synapse LossCaused by Ab Oligomers(A) Exploration times of mice i.c.v. injected with

vehicle or 10 pmol AbOs in the absence or pres-

ence of infliximab in the novel object recognition

task (n = 7–10 per experimental group).

(B) Exploration times of WT or TNFR1!/! mice

i.c.v. injected with vehicle or 10 pmol AbOs in

the novel object recognition task (n = 9–10 per

experimental group). Asterisks denote a statisti-

cally significant (p < 0.05) difference from 50%

(reference value).

(C) Freezing times of WT or TNFR1!/! mice in-

jected with vehicle or 10 pmol AbOs in the

contextual fear conditioning task. n = 9–10 animals

per experimental group. *p < 0.05, ANOVA fol-

lowed by Bonferroni post hoc test.

(D–F) Immunoblot analysis of PKR-pThr451

(normalized by total PKR) (D), eIF2a-P (normalized

by total eIF2a) (E), and synaptophysin (white bars)

and PSD95 (gray bars) (normalized by b-actin) (F)

in hippocampal homogenates of 2- to 3-month-old

WT or TNFR1!/! mice injected with vehicle or 10

pmol AbOs (n = 7 for each experimental condition).

(G) Exploration times of WT or PKR!/! mice

exposed to vehicle or 10 pmol AbOs in the novel

object recognition task (n = 8–10 per experimental

group). Asterisks denote a statistically significant

difference (p < 0.05) from 50% (reference value).

(H) Freezing times of WT or PKR!/! mice injected

with vehicle or 10 pmol AbOs in the contextual fear

conditioning task. n = 9–10 animals per experi-

mental group. *p < 0.05, ANOVA followed by

Bonferroni post hoc test.

(I) Immunoblot analysis of eIF2a-P (normalized

by total eIF2a) in hippocampal homogenates of 2-

to 3-month-old WT or PKR!/! mice injected with

vehicle or 10 pmol AbOs (n = 7 for each experi-

mental condition).

(J) Synaptophysin (green) and PSD95 (red) im-

munolabeling of cultured hippocampal neurons

exposed for 3 hr to vehicle or 500 nM AbOs in

the absence or presence of PKR inhibitor. Syn-

apses, evidenced by colocalized puncta, appear

in yellow.

(K–M) Quantification of synaptophysin (K), PSD95

(L), and colocalized (M) puncta, determined from

four experiments using independent cultures. *p <

0.05, ANOVA followed by Bonferroni post hoc test.

Graphs show means ± SEM. See also Figure S3.

Cell Metabolism



chaperone known to alleviate ER stress, prevented memory loss(Figure S5D). 4-PBA further blockedmemory impairment causedby i.c.v. injection of AbOs in mice (Figure S5E), implicating ERstress in the impact of AbOs on memory.

TNF-a, but Not eIF2a-P, Causes Neuronal IRS-1InhibitionWe recently reported that TNF-a and PKR mediate IRS-1 inhibi-tion in cultured hippocampal neurons (Bomfim et al., 2012).Establishing that proinflammatory TNF-a mediates oligomer-induced IRS-1 inhibition in vivo, AbOs triggered IRS-1pSer636in the hippocampus of WT mice, but not in TNFR1!/! mice(Figure 5A). We also examined IRS-1pSer levels in the brains ofPKR!/! mice. Possibly reflecting different patterns of activationof IRS-1 in mice of different genetic backgrounds (Xu et al.,2013), we did not detect increased IRS-1pSer levels inducedby AbOs in WT 129/SvEv mice. Interestingly, however, lowerIRS-1pSer levels were found in PKR!/! mice than in WT mice,demonstrating that suppression of PKR per se attenuatesIRS-1 inhibition (Figure 5B).

Because ER stress and eIF2a-P are linked to insulin resistancein peripheral tissues (Ozcan et al., 2004; Birkenfeld et al., 2011),we next aimed to determine whether they might be related toIRS-1 inhibition in neurons. We initially found that thapsgargintriggered dendritic IRS-1pSer636 in hippocampal neurons, andthis was blocked by 4-PBA (Figures 5C and 5D). However, levelsof IRS-1pSer were unaffected in hippocampal neuronal culturesexposed to salubrinal (Figures 5E and 5F) or in hippocampi ofsalubrinal-injected mice (Figures 5G and 5H). These results sug-gest that, while eIF2a-P alone is not sufficient to cause IRS-1pSer in neurons, aberrant activation of TNF-a signaling andinduction of ER stress play important roles in brain IRS-1 inhibi-tion. In this regard, it is interesting to note that AbOs inducedupregulation of XBP1s, GRP78, and CHOP (Figures S1B, S1D,and S1E), implicated in mechanisms by which ER dysfunctionis linked to inflammatory signaling and insulin resistance(Hotamisligil, 2010). Based on our results, it is conceivable that

different branches of the UPR respond to AbO-induced TNFRactivation and lead to IRS-1 inhibition.

Stimulation of Insulin Signaling Prevents Phospho-PKRand eIF2a-PLastly, we asked whether antidiabetic agents, recently shown torestore insulin signaling and exert neuroprotective actions in ADmodels (De Felice et al., 2009; McClean et al., 2011; Bomfimet al., 2012), could alleviate phospho-PKR and eIF2a-P triggeredby AbOs. Insulin treatment blocked the phosphorylation of PKRand eIF2a in hippocampal cultures exposed to AbOs (Figures6A–6D). Exendin-4, a GLP-1 receptor agonist approved fortreatment of diabetes (Ryan and Hardy, 2011), blocked AbO-trig-gered eIF2a-P in hippocampal neurons (Figures 6C and 6D), andintraperitoneal treatment with exendin-4 rescued eIF2a-P levelsin hippocampi of APP/PS1 mice (Figure 6E).Liraglutide, a long-lasting GLP-1 receptor agonist, has also

been shown to exhibit neuroprotective actions in animal models.Liraglutide treatment significantly reduced the levels of ER stressmarker GRP78 in the brains of APP/PS1 mice (Figure S6A). Thiswas accompanied by a significant increase in levels of synapticmarker drebrin (Figure S6B), suggesting increased synapticdensity. We further asked whether liraglutide would affect hippo-campal Ab oligomer burden in APP/PS1 mice. Interestingly,levels of 28 kDa and 108 kDa Ab oligomers, recently implicatedin AD pathogenesis (Tomiyama et al., 2010; Bao et al., 2012),were markedly reduced in liraglutide-treated animals (FiguresS6C and S6D).The neuroprotective actions of exendin-4 and liraglutide trans-

lated into memory benefit, as indicated by improved perfor-mance in fear conditioning learning in APP/PS1 mice (FiguresS6E and S6F). These findings are in line with recent studiesthat demonstrated beneficial effects of GLP-1R activationon memory in AD mice (McClean et al., 2011; Bomfim et al.,2012). Control measurements showed that neither exendin-4nor liraglutide altered animal body weight during experiments(Figures S6G and S6H). Collectively, these data suggest that

Figure 4. eIF2a-P Triggers CognitiveImpairment and Synapse Loss(A) Exploration times ofmice i.p. injected for 7 days

with vehicle or 1 mg/kg salubrinal in the novel

object recognition task (n = 10 per experimental

group). Asterisks denote a statistically significant

difference (p < 0.05) from 50% (reference value).

(B) Representative immunoblot analysis of syn-

aptophysin and PSD-95 levels in hippocampal

homogenates of 2- to 3-month-old mice receiving

vehicle or salubrinal intraperitoneally for 7 days

(n = 7 per experimental group).

(C and D) Levels of synaptophysin (C) and PSD-95

(D) were normalized by b-actin.

(E) Synaptophysin (green) and PSD95 (red) im-

munolabeling in cultured hippocampal neurons

exposed for 3 hr to vehicle or 10 mM salubrinal.

Synapses, evidenced by colocalized puncta,

appear in yellow.

(F–H) Quantification of synaptophysin (F), PSD-95

(G), and colocalized (H) puncta, determined from

four experiments using independent cultures.

*p < 0.05, Student’s t test. Graphs show means ±

SEM. See also Figure S4.

Cell Metabolism



the beneficial actions of GLP-1R agonists in AD transgenicmice involve decreased AbO levels and attenuation of braineIF2a-P and ER stress. Finally, systemic treatment with liraglu-tide reduced eIF2a-P induced by AbOs in the entorhinal cortexand hippocampus of two cynomolgus monkeys (Figures 6F–6H).

DISCUSSION

An intriguingmolecular connectionhasbeenestablishedbetweentype 2 diabetes and AD, following the discovery that impairedinsulin signaling, a hallmark of diabetes, is present in AD brains(Bomfimetal., 2012;Talbotet al., 2012).Clinical andepidemiolog-ical studies have further linked AD and diabetes, with each dis-ease increasing the risk of developing the other (Ott et al., 1996;Janson et al., 2004). AbOs, synaptotoxins that accumulate in ADbrains (Gong et al., 2003), were recently found to disrupt neuronalinsulin signaling by causing cellular redistribution of insulin recep-tors and inhibitory serine phosphorylation of IRS-1 (Zhao et al.,2008; De Felice et al., 2009; Bomfim et al., 2012). These studies

Figure 5. TNF-a and ER Stress, But NoteIF2a-P Alone, Cause Neuronal IRS-1Inhibition(A) Immunoblot analysis of IRS-1pSer636

(normalized by total IRS-1) in hippocampal ho-

mogenates of 2- to 3-month-old WT or TNFR1!/!

mice injected with vehicle or 10 pmol AbOs (n = 7

for each experimental condition).

(B) Immunoblot analysis of IRS-1pSer636 in hip-

pocampal homogenates of 2- to 3-month-old WT

or PKR!/! mice injected with vehicle or 10 pmol

AbOs (n = 8 for each experimental condition).

(C) MAP2 and IRS-1pSer636 immunolabeling

in dendrite segments from hippocampal neurons

exposed to vehicle, 1 mM thapsigargin, or 1 mM

4-PBA + 1 mM thapsigargin for 3 hr.

(D) Graph shows IRS-1pSer636 immunofluo-

rescence levels (3 independent experiments; 80

dendrite segments analyzed per experimental

condition per experiment). Scale bar: 5 mm.

(E) IRS-1pSer636 immunolabeling in cultured

hippocampal neurons exposed to vehicle or 10 mM

salubrinal for 3 hr (scale bars = 10 mm).

(F) Graph shows IRS-1pSer636 immunofluo-

rescence levels (3 independent experiments; 30

images/experimental condition/experiment).

(G) Representative immunoblot analysis of IRS-

1pSer636 (normalized by total IRS-1) in hippo-

campal homogenates of 2- to 3-month-old mice

injected intraperitoneally with vehicle or 1 mg/kg

salubrinal (n = 7 per experimental condition).

(H) IRS-1pSer636 levels were normalized by total

IRS-1. *p < 0.05, ANOVA followed by Bonferroni

post hoc test. Graphs show means ± SEM. See

also Figure S5.

have provided initial evidence thatmechanisms similar to those underlyingperipheral insulin resistance in metabolicdiseases lead to impaired brain insulinsignaling in AD (Bomfim et al., 2012;Talbot et al., 2012; De Felice, 2013).

AbOs exert multiple neurotoxic actions, including disruptionof neuronal calcium homeostasis (De Felice et al., 2007;Mattson, 2010), abnormal ER calcium release (Paula-Limaet al., 2011), and activation of JNK (De Felice et al., 2009;De Felice, 2013), conditions known to favor the developmentof ER stress in peripheral tissues (Hotamisligil, 2010). ER stressplays a key role in metabolic disorders, including type 2 dia-betes and obesity, and is linked to peripheral insulin resistanceand inflammation (Ozcan et al., 2004). Elevated ER stressmarkers, including eIF2a-P, have been reported in AD brains(Hoozemans et al., 2009; Yoon et al., 2012). Using differentexperimental models, including monkeys that received i.c.v.injections of AbOs, we demonstrate here that eIF2a-P andother ER stress responses are induced by AbOs in neurons.Our findings are in accord with studies that reported elevatedlevels of ER stress markers in other cellular and animalmodels of AD (Yoon et al., 2012) and indicate that pathologicalfindings in those studies can be attributed to the toxic impactof AbOs.

Cell Metabolism



We have identified PKR as an eIF2a kinase activated by AbOsin neurons and shown that AbOs act via the TNF-a pathway toactivate PKR. Active PKR, present in AD brains (Bullido et al.,2008), is an important mediator of inflammation and IRS-1dysfunction in metabolic diseases (Nakamura et al., 2010; Gre-gor and Hotamisligil, 2011). Of direct relevance to the currentstudy, PKR was recently shown to be involved in neuronal IRS-1 inhibition triggered by AbOs (Bomfim et al., 2012) and to nega-

tively regulate memory in mice (Zhu et al., 2011). Current findingsindicate that PKR activation does not depend on direct bindingof oligomers to individual neurons, as elevated phospho-PKRand eIF2a-P levels were detected in neurons regardless ofwhether or not they had oligomers bound to their dendrites.Therefore, AbOs do not seem to act directly on neurons to phos-phorylate PKR. Rather, it is likely that a crosstalk between neu-rons and microglia leads to elevated levels of TNF-a, causing

Figure 6. Antidiabetic Agents Block AbO-Induced eIF2a-P In Vitro and In Vivo(A) PKR-pThr451 immunolabeling in cultured hippocampal neurons exposed for 3 hr to vehicle or 500 nM AbOs in the absence or presence of 1 mM insulin (scale

bar = 10 mm).

(B) PKR-pThr451 immunofluorescence levels, determined from four experiments using independent cultures.

(C) eIF2a-P immunolabeling in hippocampal neurons exposed for 3 hr to 500 nM AbOs, 1 mM insulin + 500 nM AbOs, or 300 nM exendin-4 + 500 nM AbOs (scale

bar = 10 mm).

(D) eIF2a-P immunofluorescence levels (n = 3 experiments using independent cultures; 30 images analyzed per experimental condition per experiment).

**p < 0.01, ANOVA followed by Bonferroni post hoc test.

(E) Immunoblot analysis for eIF2a-P in hippocampal homogenates from 13- to 16-month-old APP/PS1 (n = 7) or exendin-4-treated APP/PS1 (n = 5) mice. Graphs

show means ± SEM.

(F) eIF2a-P immunolabeling in the hippocampi (representative images from CA3) and entorhinal cortices of sham, AbO-injected, or liraglutide-treated and

AbO-injected monkeys (scale bars = 200 mm).

(G and H) eIF2a-P immunolabeling densities in the hippocampi (G) and entorhinal cortices (H) of sham (white bar), AbO-injected (black bar), or liraglutide-treated

AbO-injectedmonkeys (gray bars). Graphs showmeans ± SEM obtained for different animals in each experimental group (n = 3, 4, or 2 for sham, AbO-injected, or

liraglutide-treated AbO-injected monkeys, respectively). Asterisk indicates a statistically significant (p < 0.05; Student’s t test) difference between sham and

AbO-injected animals. See also Figure S6.

Cell Metabolism



activation of neuronal TNF-a/PKR/eIF2a signaling. TNF-a hasbeen suggested recently to play a role in brain dysfunction indifferent diseases, including AD, trauma, and Parkinson’s dis-ease (Clark et al., 2012). Elevated TNF-a levels are furtherthought to cause cognitive deficits (He et al., 2007). Our resultsdemonstrate that, in Alzheimer’s disease, elevated TNF-a is aconsequence of the brain accumulation and impact of AbOs.AD is a devastating disease affecting memory. AbOs are

recognized as potent synaptotoxins that inhibit synaptic plas-ticity (Ferreira and Klein, 2011). We found here that oligomer-induced aberrant TNF-a/PKR/eIF2a signaling and induction ofER stress are linked to synapse loss and memory impairment.It is notable that, in both PKR!/! and TNFR1!/! mice, oligomersfailed to trigger eIF2a-P and cognitive impairment. Inhibition ofPKRwas found to attenuate the loss of PSD-95 and synaptophy-sin induced by AbOs, implicating synapse deterioration in thedeleterious impact of PKR on memory. PKR and eIF2a haveboth been shown to be critical for memory regulation (Costa-Mattioli et al., 2007; Zhu et al., 2011). More recently, eIF2a-Pwas found to mediate prion-related neurodegeneration in thehippocampus (Moreno et al., 2012). We report that upregulationof eIF2a-P triggers synapse loss, suggesting that synergisticneurotoxic events that culminate in eIF2a-P may respond, atleast in part, for synapse damage in AD. Our results thus estab-lish that AbO-induced TNFa, PKR, and eIF2a-P dysregulation isdirectly linked to synapse failure and cognitive impairment,revealing a mechanism by which AbOs disrupt memory in AD.The parallel we now describe between AD and diabetes sheds

light on how insulin signaling is impaired in AD. In peripheral tis-sues, inflammatory and metabolic stress signaling cascadestrigger disruption of insulin signaling (Ozcan et al., 2004, 2006;Hotamisligil, 2010; Nakamura et al., 2010; Gregor and Hotamisli-gil, 2011). Our findings show that ER stress is linked to neuronalIRS-1 inhibition andmay act synergistically with proinflammatorysignals to disrupt brain insulin signaling in AD.It is important to note that our conclusion that TNF-a signaling

is linked to both defective brain insulin signaling and memoryimpairment caused by AbOs does not necessarily imply thatmemory impairment is a direct consequence of disruption ofinsulin signaling. In fact, the role of insulin signaling in memoryformation, and how defective signaling might result in neurode-generative disorders, is still under investigation. While insulinhas been shown to positively modulate hippocampal synapticplasticity (van der Heide et al., 2005), an early study showedthat spatial memory was preserved in neural-specific insulin re-ceptor b subunit knockout (NIRKO) mice (Schubert et al., 2004).A recent study, however, reported that IRb haploinsufficiencycauses reduced brain insulin signaling and defects in late-phaselong-term potentiation (LTP) and long-term recognition memorystorage, thereby implicating insulin receptor (IR)-dependentmechanisms in memory formation (Nistico et al., 2012). Signifi-cantly, intranasal insulin administration has been found toenhance memory in healthy volunteers (Benedict et al., 2004),in memory-impaired older subjects (Reger et al., 2006), andin early AD patients (Craft et al., 2012), with beneficial effectsreported for both acute and long-term insulin treatment.In sporadic AD (which corresponds to >90% of AD cases), the

exact mechanism that leads to accumulation of Ab oligomersand amyloid in the brain remains to be fully elucidated. Interest-

ingly, eIF2a-P promotes BACE1 expression, Ab production, anddeposition (O’Connor et al., 2008). Thus, AbO-induced TNF-a,PKR, and eIF2a-P may constitute a hub in a feedforward delete-rious cycle involving increased AbO generation and perpetuationand amplification of neuronal dysfunction.In line with our current findings demonstrating that eIF2a-P is

sufficient to cause synapse loss and memory impairment inmice, a very recent study showed that the eIF2a kinase PERKmediates Ab-induced LTP impairment (Ma et al., 2013). Impor-tantly, Ma and colleagues further reported that genetically sup-pressing the eIF2a kinases PERK or GCN2 prevented spatialmemory impairment in AD mice (Ma et al., 2013). Althoughthe complete set of mechanisms remains to be elucidated,it is conceivable that TNF-a represents an initial trigger thatorchestrates activation of multiple stress response pathways(of which the PKR/eIF2a pathway here investigated appearscentral) that culminate with synapse dysfunction and memoryloss in AD.In recent years, there has been a strong effort to develop new

therapeutic strategies for diabetes and insulin resistance disor-ders. GLP-1 receptor stimulation has insulinotropic action andrestores glucose homeostasis in peripheral tissue (Yusta et al.,2006), and several GLP-1 analogs are currently used for diabetestreatment. We demonstrate here that both insulin and GLP-1Ragonists prevent abnormal neuronal phospho-PKR and eIF2a-P.The observation that insulin protects neurons from AbO-inducedPKR activation suggests that blockade of eIF2a-P by insulin ismediated by suppression of PKR. Recent studies have shownbeneficial effects of exendin-4 and liraglutide against neurode-generation and cognitive impairment in AD models (Li et al.,2010; McClean et al., 2011; Bomfim et al., 2012), but thesignaling mechanisms involved in neuroprotection are onlypartially known. The present study suggests a mechanisminvolving suppression of the PKR/eIF2a-P pathway. Importantly,we provide evidence that a GLP-1R agonist exerts neuroprotec-tive actions in the primate brain, as systemic liraglutide treatmentreduced oligomer-induced brain eIF2a-P in monkeys.In conclusion, our findings establish that activation of TNF-a

signaling mediates AbO-induced brain IRS-1 inhibition andPKR-dependent eIF2a-P, synapse loss, and memory impair-ment (Figure 7), revealing mechanisms that lead to synapseloss and memory impairment. Identifying a pathogenic mecha-nism that is shared between AD and diabetes and contributesto memory loss in AD may open avenues for rapid implemen-tation of clinically approved antidiabetic drugs as therapeuticsin AD.

EXPERIMENTAL PROCEDURES

Ab OligomersOligomers were prepared weekly from synthetic Ab1–42 and routinely charac-

terized by size-exclusion chromatography and, occasionally, by western

immunoblots and transmission electron microscopy, as previously described

(De Felice et al., 2007, 2008; Sebollela et al., 2012). Oligomers were kept at

4"C and used within 48 hr of preparation.

Mature Hippocampal CulturesPrimary rat hippocampal neuronal cultures, prepared and developed in

Neurobasal Medium supplemented with B27 (Invitrogen) and antibiotics

according to established procedures (De Felice et al., 2007), were used after

Cell Metabolism



18–21 days in vitro. All procedures were approved by the Institutional Animal

Care and Use Committee of the Federal University of Rio de Janeiro (protocol

#IBqM 022). Cultures were exposed at 37!C for 3 hr to 500 nM Ab oligomers

or an equivalent volume of vehicle (2% DMSO in phosphate-buffered

saline [PBS]). When present, PKR inhibitor (1 mM), insulin (1 mM), exendin-4

(300 nM), SP600125 (10 mM), or infliximab (1 mg/mL) were added to cultures

30 min before Ab oligomers. For ER stress induction, thapsigargin (1 mM)

was added to neuronal cultures for 3 hr, and in some experiments 4-phenylbu-

tyrate (4-PBA) (1 mM) was also present. For eIF2a-P induction, salubrinal

(10 mM) was added to neuronal cultures for 3 hr.

Experimental SubjectsMale Swiss mice were obtained from our animal facility (Federal University of

Rio de Janeiro) and were 2.5–3 months old at the beginning of experiments.

PKR"/" male mice on a 129/SvEv background (Yang et al., 1995) were

obtained from Federal University of Minas Gerais Animal Centre. 129/SvEv

wild-type mice were used as controls for experiments with PKR"/" mice.

TNFR1"/" (TNFRp55"/") female mice on a C57BL/6 background (Pfeffer

et al., 1993) were obtained from the University of Campinas Breeding Centre

and were 3 months old at the beginning of experiments. C57BL/6 wild-type

mice were used as controls for experiments with TNFR1"/" mice. APPSwe/

PS1DE9 mice on a C57BL/6 background (Jankowsky et al., 2001) were ob-

tained from The Jackson Laboratory. Wild-type littermates were used as con-

trols. Animals were genotyped prior to the studies using specific primers (see

Supplemental Experimental Procedures). All procedures involving transgenic

mice were approved by the Institutional Animal Care and Use Committee of

the Federal University of Rio de Janeiro (protocol #IBqM 055). Animals were

housed in groups of five in each cage with free access to food and water,

with controlled room temperature and humidity, and under a 12 hr light/12 hr

dark cycle. All procedures used in the present study followed the ‘‘Principles

of Laboratory Animal Care’’ from the US National Institutes of Health.

Intracerebroventricular Injections in MiceFor i.c.v. injection of AbOs, animals were anesthetized for 7 min with 2.5% iso-

flurane (Cristalia) using a vaporizer system and gently restrained only during

the injection procedure itself, as described (Figueiredo et al., 2013; Ledo

et al., 2013). A 2.5 mm long needle was unilaterally inserted 1 mm to the right

of the midline point equidistant from each eye and 1 mm posterior to a line

drawn through the anterior base of the eye (Ledo et al., 2013). AbOs

(10 pmol), 1 mg of thapsigargin, or vehicle was injected in a final volume of

3 ml, and the needle was kept in place for 30 s to avoid backflow. Before eutha-

nasia, blue staining was injected using the same hole used previously, which

allowed us to determine the accurate placement of the injection. Behavioral

results from mice that showed signs of misplaced injections or any sign of

hemorrhage were excluded from the final statistical analysis (this happened

in 5% of cases, on average).

In Vivo Drug Treatments in MiceExendin-4 and liraglutide are two long-lasting GLP-1 receptor agonists.

APPSwe/PS1DE9 mice (13–14 months old) received daily intraperitoneal

injections of exendin-4 (25 nmol/kg, dissolved in saline) or vehicle (saline) for

3 weeks. Another group of transgenic animals received daily intraperitoneal

injections of liraglutide (25 nmol/kg, dissolved in saline) or vehicle (saline) for

6 weeks. Male Swiss mice (3 months old) received daily intraperitoneal

injections of infliximab (20 mg/day) for 7 days starting immediately after i.c.v.

injection of AbOs. In experiments with 4-phenylbutyrate (4-PBA), one intraper-

itoneal injection (200mg/kg) was given immediately after i.c.v injection of either

Ab oligomers or thapsigargin in male Swiss mice. 4-PBA solutions were pre-

pared as described. Salubrinal (1 mg/kg) or vehicle was administered intraper-

itoneally in 3-month-old male Swiss mice for 7 days before cognitive analysis

(Moreno et al., 2012). All procedures were approved by the Institutional Animal

Care and Use Committee of the Federal University of Rio de Janeiro (protocol

#IBqM 041).

Intracerebroventricular Injection of Ab Oligomers in Monkeys andTreatment with LiraglutideAdult cynomolgus monkeys (Macaca fascicularis) (n = 9) were used (body

weights 4.7–7.0 kg). All procedures were approved by the Queen’s University

Animal Care Committee and were in full compliance with the Canada Council

on Animal Care (Animal Care Protocol Original Munoz-2011-039-Or). Animals

were under the close supervision of an animal technician and the institute

veterinarian. AbOs were infused chronically through an intracerebroventricular

cannula. A total of 4 animals received 100 mg Ab oligomers i.c.v. every 3 days

for 24 days, while 3 animals served as sham-operated controls. Control

animals had the cannula implanted into the lateral ventricle in the samemanner

as the experimental animals but did not receive oligomer injections. Two addi-

tional animals received daily subcutaneous injections of liraglutide (25 nmol/kg)

beginning 1 week prior to Ab oligomer infusion and continuing until the end

of AbO injections. Oligomers were freshly prepared and characterized by

size-exclusion chromatography before each injection. Upon completion of

the experimental protocol, animals were sedated with ketamine (10 mg/kg,

intramuscular) plus buprenorphine (0.01 mg/kg) for analgesia, followed by

intravenous sodium pentobarbital (25 mg/kg). Next, animals were perfused

with PBS and then, sequentially, by 4% paraformaldehyde in PBS, 4% para-

formaldehyde in PBS containing 2.5% glycerol, PBS + 5% glycerol, and

PBS + 10% glycerol. Serial brain sections (40 mm thick) were obtained, and

neuropathological analyses were performed.

Figure 7. Ab Oligomers Trigger Synapse Loss, Memory Impairment,and IRS-1 Inhibition via TNF-a/PKR SignalingAb oligomers lead to increased brain levels of TNF-a, leading to TNFR1-

mediated activation of PKR and other stress kinases. Activated PKR phos-

phorylates neuronal IRS-1 (see also Bomfim et al., 2012) and eIF2a. Increased

eIF2a-P levels trigger synapse loss and memory impairment.

Cell Metabolism



Statistical AnalysisAll analyses were performed with GraphPad Prism, and data sets were

assessed for normality parameters prior to significance determination. Values

are expressed as means ± SEM, unless otherwise stated. Statistical confi-

dence levels are indicated in each figure (*p < 0.05; **p < 0.01).

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures

and six figures and can be found with this article online at http://dx.doi.org/

10.1016/j.cmet.2013.11.002.

AUTHOR CONTRIBUTIONS

M.V.L., D.P.M., S.T.F., and F.G.D.F. designed the study. M.V.L., J.R.C., R.L.F.,

T.R.B., L.F.-G., A.F.B., J.B.-M., L.F.-C., C.A.S., L.B.S., and S.E.-S. performed

the research. M.V.L., T.R.B., C.A.S., L.F.-C., P.C.-C., O.B.A., S.T.F., and

F.G.D.F. analyzed data. J.B.C., L.A.V., A.M.S., W.L.K., and C.H. contributed

animals, reagents, materials, and analysis tools. M.V.L., J.-C.H., D.P.M.,

S.T.F., and F.G.D.F. analyzed and discussed results. M.V.L., S.T.F., and

F.G.D.F. wrote the manuscript.

ACKNOWLEDGMENTS

This work was supported by grants from Human Frontiers Science Program

(HFSP) (to F.G.D.F.), National Institute for Translational Neuroscience (INNT/

Brazil), the Brazilian funding agencies Conselho Nacional de Desenvolvimento

Cientıfico e Tecnologico (CNPq) and Fundacao de Amparo a Pesquisa do

Estado do Rio de Janeiro (FAPERJ) (to S.T.F. and F.G.D.F.), Canadian Insti-

tutes for Health Research (CIHR), and Canada Research Chair Program

(to D.P.M.). M.V.L., L.B.S., L.F.-G., A.F.B., T.R.B., J.B.-M., and L.F.-C. were

supported by CNPq predoctoral fellowships. We thank Drs. Matthias Gralle,

Wagner Seixas, and Claudio A. Masuda (Federal University of Rio de Janeiro,

Brazil) for insightful discussions and Dr. Claudia P. Figueiredo (Federal Univer-

sity of Rio de Janeiro, Brazil) for advice on immunohistochemical analysis of

monkey brains. W.L.K. is a cofounder of Acumen Pharmaceuticals, which

has been licensed by Northwestern University to develop ADDL technology

for Alzheimer’s therapeutics and diagnostics.

Received: August 2, 2012

Revised: September 17, 2013

Accepted: October 18, 2013

Published: December 3, 2013

REFERENCES

Bao, F., Wicklund, L., Lacor, P.N., Klein, W.L., Nordberg, A., and Marutle, A.

(2012). Different b-amyloid oligomer assemblies in Alzheimer brains correlate

with age of disease onset and impaired cholinergic activity. Neurobiol. Aging

33, e1–e13.

Benedict, C., Hallschmid, M., Hatke, A., Schultes, B., Fehm, H.L., Born, J.,

and Kern, W. (2004). Intranasal insulin improves memory in humans.

Psychoneuroendocrinology 29, 1326–1334.

Birkenfeld, A.L., Lee, H.Y., Majumdar, S., Jurczak, M.J., Camporez, J.P.,

Jornayvaz, F.R., Frederick, D.W., Guigni, B., Kahn, M., Zhang, D., et al.

(2011). Influence of the hepatic eukaryotic initiation factor 2alpha (eIF2alpha)

endoplasmic reticulum (ER) stress response pathway on insulin-mediated

ER stress and hepatic and peripheral glucose metabolism. J. Biol. Chem.

286, 36163–36170.

Bomfim, T.R., Forny-Germano, L., Sathler, L.B., Brito-Moreira, J., Houzel, J.C.,

Decker, H., Silverman, M.A., Kazi, H., Melo, H.M., McClean, P.L., et al. (2012).

An anti-diabetes agent protects the mouse brain from defective insulin

signaling caused by Alzheimer’s disease- associated Ab oligomers. J. Clin.

Invest. 122, 1339–1353.

Bullido, M.J., Martınez-Garcıa, A., Tenorio, R., Sastre, I., Munoz, D.G., Frank,

A., and Valdivieso, F. (2008). Double stranded RNA activated EIF2 a kinase

(EIF2AK2; PKR) is associated with Alzheimer’s disease. Neurobiol. Aging 29,

1160–1166.

Chiu, S.-L., Chen, C.-M., and Cline, H.T. (2008). Insulin receptor signaling

regulates synapse number, dendritic plasticity, and circuit function in vivo.

Neuron 58, 708–719.

Clark, I., Atwood, C., Bowen, R., Paz-Filho, G., and Vissel, B. (2012). Tumor

necrosis factor-induced cerebral insulin resistance in Alzheimer’s disease links

numerous treatment rationales. Pharmacol. Rev. 64, 1004–1026.

Costa-Mattioli, M., Gobert, D., Stern, E., Gamache, K., Colina, R., Cuello, C.,

Sossin, W., Kaufman, R., Pelletier, J., Rosenblum, K., et al. (2007). eIF2a

phosphorylation bidirectionally regulates the switch from short- to long-term

synaptic plasticity and memory. Cell 129, 195–206.

Craft, S., Baker, L.D., Montine, T.J., Minoshima, S., Watson, G.S., Claxton, A.,

Arbuckle, M., Callaghan, M., Tsai, E., Plymate, S.R., et al. (2012). Intranasal

insulin therapy for Alzheimer disease and amnestic mild cognitive impairment:

a pilot clinical trial. Arch. Neurol. 69, 29–38.

De Felice, F.G. (2013). Alzheimer’s disease and insulin resistance: translating

basic science into clinical applications. J. Clin. Invest. 123, 531–539.

De Felice, F.G., Velasco, P.T., Lambert, M.P., Viola, K., Fernandez, S.J.,

Ferreira, S.T., and Klein, W.L. (2007). Abeta oligomers induce neuronal

oxidative stress through an N-methyl-D-aspartate receptor-dependent

mechanism that is blocked by the Alzheimer drug memantine. J. Biol. Chem.

282, 11590–11601.

De Felice, F.G., Wu, D., Lambert, M.P., Fernandez, S.J., Velasco, P.T., Lacor,

P.N., Bigio, E.H., Jerecic, J., Acton, P.J., Shughrue, P.J., et al. (2008).

Alzheimer’s disease-type neuronal tau hyperphosphorylation induced by A

beta oligomers. Neurobiol. Aging 29, 1334–1347.

De Felice, F.G., Vieira, M.N.N., Bomfim, T.R., Decker, H., Velasco, P.T.,

Lambert, M.P., Viola, K.L., Zhao, W.Q., Ferreira, S.T., and Klein, W.L. (2009).

Protection of synapses against Alzheimer’s-linked toxins: insulin signaling

prevents the pathogenic binding of Abeta oligomers. Proc. Natl. Acad. Sci.

USA 106, 1971–1976.

de la Monte, S.M., and Wands, J.R. (2008). Alzheimer’s disease is type 3

diabetes-evidence reviewed. J. Diabetes Sci. Tech. 2, 1101–1113.

Ferreira, S.T., and Klein, W.L. (2011). The Ab oligomer hypothesis for synapse

failure and memory loss in Alzheimer’s disease. Neurobiol. Learn. Mem. 96,

529–543.

Figueiredo, C.P., Clarke, J.R., Ledo, J.H., Ribeiro, F.C., Costa, C.V., Melo,

H.M., Mota-Sales, A.P., Saraiva, L.M., Klein, W.L., Sebollela, A., et al. (2013).

Memantine rescues transient cognitive impairment caused by high-molecu-

lar-weight ab oligomers but not the persistent impairment induced by low-

molecular-weight oligomers. J. Neurosci. 33, 9626–9634.

Gkogkas, C., Sonenberg, N., and Costa-Mattioli, M. (2010). Translational con-

trol mechanisms in long-lasting synaptic plasticity and memory. J. Biol. Chem.

285, 31913–31917.

Gong, Y., Chang, L., Viola, K.L., Lacor, P.N., Lambert, M.P., Finch, C.E., Krafft,

G.A., and Klein, W.L. (2003). Alzheimer’s disease-affected brain: presence of

oligomeric A b ligands (ADDLs) suggests a molecular basis for reversible

memory loss. Proc. Natl. Acad. Sci. USA 100, 10417–10422.

Gregor, M.F., and Hotamisligil, G.S. (2011). Inflammatory mechanisms in

obesity. Annu. Rev. Immunol. 29, 415–445.

He, P., Zhong, Z., Lindholm, K., Berning, L., Lee, W., Lemere, C., Staufenbiel,

M., Li, R., and Shen, Y. (2007). Deletion of tumor necrosis factor death receptor

inhibits amyloid b generation and prevents learning and memory deficits in

Alzheimer’s mice. J. Cell Biol. 178, 829–841.

Hoozemans, J.J.M., Veerhuis, R., Van Haastert, E.S., Rozemuller, J.M., Baas,

F., Eikelenboom, P., and Scheper, W. (2005). The unfolded protein response

is activated in Alzheimer’s disease. Acta Neuropathol. 110, 165–172.

Hoozemans, J.J.M., van Haastert, E.S., Nijholt, D.A.T., Rozemuller, A.J.M.,

Eikelenboom, P., and Scheper, W. (2009). The unfolded protein response is

activated in pretangle neurons in Alzheimer’s disease hippocampus. Am. J.

Pathol. 174, 1241–1251.

Hotamisligil, G.S. (2010). Endoplasmic reticulum stress and the inflammatory

basis of metabolic disease. Cell 140, 900–917.

Cell Metabolism



Hotamisligil, G.S., Budavari, A., Murray, D., and Spiegelman, B.M. (1994).

Reduced tyrosine kinase activity of the insulin receptor in obesity-diabetes.

Central role of tumor necrosis factor-a. J. Clin. Invest. 94, 1543–1549.

Hotamisligil, G.S., Peraldi, P., Budavari, A., Ellis, R., White, M.F., and

Spiegelman, B.M. (1996). IRS-1-mediated inhibition of insulin receptor

tyrosine kinase activity in TNF-a- and obesity-induced insulin resistance.

Science 271, 665–668.

Jankowsky, J.L., Slunt, H.H., Ratovitski, T., Jenkins, N.A., Copeland, N.G., and

Borchelt, D.R. (2001). Co-expression of multiple transgenes in mouse CNS:

a comparison of strategies. Biomol. Eng. 17, 157–165.

Janson, J., Laedtke, T., Parisi, J.E., O’Brien, P.O., Petersen, R.C., and Butler,

P.C. (2004). Increased risk of type 2 diabetes in Alzheimer disease. Diabetes

53, 474–481.

Lacor, P.N., Buniel, M.C., Furlow, P.W., Clemente, A.S., Velasco, P.T., Wood,

M., Viola, K.L., and Klein, W.L. (2007). Abeta oligomer-induced aberrations in

synapse composition, shape, and density provide a molecular basis for loss

of connectivity in Alzheimer’s disease. J. Neurosci. 27, 796–807.

Lai, E., Teodoro, T., and Volchuk, A. (2007). Endoplasmic reticulum stress:

signaling the unfolded protein response. Physiology (Bethesda) 22, 193–201.

Ledo, J.H., Azevedo, E.P., Clarke, J.R., Ribeiro, F.C., Figueiredo, C.P., Foguel,

D., De Felice, F.G., and Ferreira, S.T. (2013). Amyloid-b oligomers link

depressive-like behavior and cognitive deficits in mice. Mol. Psychiatry 18,

1053–1054.

Li, Y., Duffy, K.B., Ottinger, M.A., Ray, B., Bailey, J.A., Holloway, H.W.,

Tweedie, D., Perry, T., Mattson, M.P., Kapogiannis, D., et al. (2010). GLP-1

receptor stimulation reduces amyloid-b peptide accumulation and cytotoxicity

in cellular and animal models of Alzheimer’s disease. J. Alzheimers Dis. 19,

1205–1219.

Ma, Q.L., Yang, F., Rosario, E.R., Ubeda, O.J., Beech, W., Gant, D.J., Chen,

P.P., Hudspeth, B., Chen, C., Zhao, Y., et al. (2009). b-amyloid oligomers

induce phosphorylation of tau and inactivation of insulin receptor substrate

via c-Jun N-terminal kinase signaling: suppression by omega-3 fatty acids

and curcumin. J. Neurosci. 29, 9078–9089.

Ma, T., Trinh, M.A., Wexler, A.J., Bourbon, C., Gatti, E., Pierre, P., Cavener,

D.R., and Klann, E. (2013). Suppression of eIF2a kinases alleviates

Alzheimer’s disease-related plasticity and memory deficits. Nat. Neurosci.

16, 1299–1305.

Mattson, M.P. (2010). ER calcium and Alzheimer’s disease: in a state of flux.

Sci. Signal. 3, pe10.

McClean, P.L., Parthsarathy, V., Faivre, E., and Holscher, C. (2011). The

diabetes drug liraglutide prevents degenerative processes in a mouse model

of Alzheimer’s disease. J. Neurosci. 31, 6587–6594.

McNay, E.C., and Recknagel, A.K. (2011). Brain insulin signaling: a key compo-

nent of cognitive processes and a potential basis for cognitive impairment

in type 2 diabetes. Neurobiol. Learn. Mem. 96, 432–442.

Moloney, A.M., Griffin, R.J., Timmons, S., O’Connor, R., Ravid, R., and O’Neill,

C. (2010). Defects in IGF-1 receptor, insulin receptor and IRS-1/2 in

Alzheimer’s disease indicate possible resistance to IGF-1 and insulin signal-

ling. Neurobiol. Aging 31, 224–243.

Moreno, J.A., Radford, H., Peretti, D., Steinert, J.R., Verity, N., Martin, M.G.,

Halliday, M., Morgan, J., Dinsdale, D., Ortori, C.A., et al. (2012). Sustained

translational repression by eIF2a-P mediates prion neurodegeneration.

Nature 485, 507–511.

Nakamura, T., Furuhashi, M., Li, P., Cao, H., Tuncman, G., Sonenberg, N.,

Gorgun, C.Z., and Hotamisligil, G.S. (2010). Double-stranded RNA-dependent

protein kinase links pathogen sensing with stress and metabolic homeostasis.

Cell 140, 338–348.

Nistico, R., Cavallucci, V., Piccinin, S., Macrı, S., Pignatelli, M., Mehdawy, B.,

Blandini, F., Laviola, G., Lauro, D., Mercuri, N.B., and D’Amelio, M. (2012).

Insulin receptor b-subunit haploinsufficiency impairs hippocampal late-phase

LTP and recognition memory. Neuromolecular Med. 14, 262–269.

O’Connor, T., Sadleir, K.R., Maus, E., Velliquette, R.A., Zhao, J., Cole, S.L.,

Eimer, W.A., Hitt, B., Bembinster, L.A., Lammich, S., et al. (2008).

Phosphorylation of the translation initiation factor eIF2a increases BACE1

levels and promotes amyloidogenesis. Neuron 60, 988–1009.

Ott, A., Stolk, R.P., Hofman, A., van Harskamp, F., Grobbee, D.E., and

Breteler, M.M.B. (1996). Association of diabetes mellitus and dementia: the

Rotterdam Study. Diabetologia 39, 1392–1397.

Ozcan, U., Cao, Q., Yilmaz, E., Lee, A.H., Iwakoshi, N.N., Ozdelen, E.,

Tuncman, G., Gorgun, C.Z., Glimcher, L.H., and Hotamisligil, G.S. (2004).

Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes.

Science 306, 457–461.

Ozcan, U., Yilmaz, E., Ozcan, L., Furuhashi, M., Vaillancourt, E., Smith, R.O.,

Gorgun, C.Z., and Hotamisligil, G.S. (2006). Chemical chaperones reduce

ER stress and restore glucose homeostasis in a mouse model of type 2 dia-

betes. Science 313, 1137–1140.

Park, S.W., Zhou, Y., Lee, J., Lu, A., Sun, C., Chung, J., Ueki, K., and Ozcan, U.

(2010). The regulatory subunits of PI3K, p85alpha and p85beta, interact with

XBP-1 and increase its nuclear translocation. Nat. Med. 16, 429–437.

Paula-Lima, A.C., Adasme, T., SanMartın, C., Sebollela, A., Hetz, C., Carrasco,

M.A., Ferreira, S.T., and Hidalgo, C. (2011). Amyloid b-peptide oligomers stim-

ulate RyR-mediated Ca2+ release inducing mitochondrial fragmentation in

hippocampal neurons and prevent RyR-mediated dendritic spine remodeling

produced by BDNF. Antioxid. Redox Signal. 14, 1209–1223.

Pfeffer, K., Matsuyama, T., Kundig, T.M., Wakeham, A., Kishihara, K.,

Shahinian, A., Wiegmann, K., Ohashi, P.S., Kronke, M., and Mak, T.W.

(1993). Mice deficient for the 55 kd tumor necrosis factor receptor are resistant

to endotoxic shock, yet succumb to L. monocytogenes infection. Cell 73,

457–467.

Raven, J.F., and Koromilas, A.E. (2008). PERK and PKR: old kinases learn new

tricks. Cell Cycle 7, 1146–1150.

Reger, M.A., Watson, G.S., Frey, W.H., 2nd, Baker, L.D., Cholerton, B.,

Keeling, M.L., Belongia, D.A., Fishel, M.A., Plymate, S.R., Schellenberg,

G.D., et al. (2006). Effects of intranasal insulin on cognition in memory-

impaired older adults: modulation by APOE genotype. Neurobiol. Aging 27,

451–458.

Ryan, G.J., and Hardy, Y. (2011). Liraglutide: once-daily GLP-1 agonist for the

treatment of type 2 diabetes. J. Clin. Pharm. Ther. 36, 260–274.

Schubert, M., Gautam, D., Surjo, D., Ueki, K., Baudler, S., Schubert, D.,

Kondo, T., Alber, J., Galldiks, N., Kustermann, E., et al. (2004). Role for

neuronal insulin resistance in neurodegenerative diseases. Proc. Natl. Acad.

Sci. USA 101, 3100–3105.

Sebollela, A., Freitas-Correa, L., Oliveira, F.F., Paula-Lima, A.C., Saraiva, L.M.,

Martins, S.M., Mota, L.D., Torres, C., Alves-Leon, S., de Souza, J.M., et al.

(2012). Amyloid-b oligomers induce differential gene expression in adult

human brain slices. J. Biol. Chem. 287, 7436–7445.

Steinberg, G.R., Michell, B.J., van Denderen, B.J.W., Watt, M.J., Carey, A.L.,

Fam, B.C., Andrikopoulos, S., Proietto, J., Gorgun, C.Z., Carling, D., et al.

(2006). Tumor necrosis factor a-induced skeletal muscle insulin resistance

involves suppression of AMP-kinase signaling. Cell Metab. 4, 465–474.

Talbot, K.,Wang, H.Y., Kazi, H., Han, L.Y., Bakshi, K.P., Stucky, A., Fuino, R.L.,

Kawaguchi, K.R., Samoyedny, A.J., Wilson, R.S., et al. (2012). Demonstrated

brain insulin resistance in Alzheimer’s disease patients is associated with

IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. J. Clin. Invest.

122, 1316–1338.

Terry, R.D., Masliah, E., Salmon, D.P., Butters, N., DeTeresa, R., Hill, R.,

Hansen, L.A., and Katzman, R. (1991). Physical basis of cognitive alterations

in Alzheimer’s disease: synapse loss is the major correlate of cognitive impair-

ment. Ann. Neurol. 30, 572–580.

Tomiyama, T., Matsuyama, S., Iso, H., Umeda, T., Takuma, H., Ohnishi, K.,

Ishibashi, K., Teraoka, R., Sakama, N., Yamashita, T., et al. (2010). A mouse

model of amyloid b oligomers: their contribution to synaptic alteration,

abnormal tau phosphorylation, glial activation, and neuronal loss in vivo.

J. Neurosci. 30, 4845–4856.

van der Heide, L.P., Kamal, A., Artola, A., Gispen, W.H., and Ramakers, G.M.J.

(2005). Insulin modulates hippocampal activity-dependent synaptic plasticity

Cell Metabolism



in a N-methyl-d-aspartate receptor and phosphatidyl-inositol-3-kinase-

dependent manner. J. Neurochem. 94, 1158–1166.

Winnay, J.N., Boucher, J., Mori, M.A., Ueki, K., and Kahn, C.R. (2010). A reg-

ulatory subunit of phosphoinositide 3-kinase increases the nuclear accumula-

tion of X-box-binding protein-1 to modulate the unfolded protein response.

Nat. Med. 16, 438–445.

Xu, J., Gontier, G., Chaker, Z., Lacube, P., Dupont, J., and Holzenberger, M.

(2013). Longevity effect of IGF-1R(+/-) mutation depends on genetic back-

ground-specific receptor activation. Aging Cell. Published online July 30,

2013. http://dx.doi.org/10.1111/acel.12145.

Yang, Y.L., Reis, L.F., Pavlovic, J., Aguzzi, A., Schafer, R., Kumar, A.,

Williams, B.R., Aguet, M., and Weissmann, C. (1995). Deficient signaling in

mice devoid of double-stranded RNA-dependent protein kinase. EMBO J.

14, 6095–6106.

Yoon, S.O., Park, D.J., Ryu, J.C., Ozer, H.G., Tep, C., Shin, Y.J., Lim, T.H.,

Pastorino, L., Kunwar, A.J., Walton, J.C., et al. (2012). JNK3 perpetuates

metabolic stress induced by Ab peptides. Neuron 75, 824–837.

Yusta, B., Baggio, L.L., Estall, J.L., Koehler, J.A., Holland, D.P., Li, H.,

Pipeleers, D., Ling, Z., and Drucker, D.J. (2006). GLP-1 receptor activation

improves b cell function and survival following induction of endoplasmic

reticulum stress. Cell Metab. 4, 391–406.

Zhao, W.Q., De Felice, F.G., Fernandez, S., Chen, H., Lambert, M.P., Quon,

M.J., Krafft, G.A., and Klein, W.L. (2008). Amyloid beta oligomers induce

impairment of neuronal insulin receptors. FASEB J. 22, 246–260.

Zhu, P.J., Huang, W., Kalikulov, D., Yoo, J.W., Placzek, A.N., Stoica, L., Zhou,

H., Bell, J.C., Friedlander, M.J., Krnjevi!c, K., et al. (2011). Suppression of

PKR promotes network excitability and enhanced cognition by interferon-g-

mediated disinhibition. Cell 147, 1384–1396.

Cell Metabolism



Cell Metabolism, Volume 18

Supplemental Information

TNF-� Mediates Brain IRS-1 Inhibition and PKR-Dependent Memory Impairment Induced by

��-Linked �-Amyloid Oligomers in Mice and Monkeys

Mychael V. Lourenco, Julia R. Clarke, Rudimar L. Frozza, Theresa R. Bomfim, Letícia Forny-Germano, André F. Batista, Luciana B. Sathler, Jordano Brito-Moreira, Olavo B. Amaral, Cesar A. Silva, Léo Freitas-Correa, Sheila Espírito-Santo, Paula Campello-Costa, Jean-Christophe Houzel, William L. Klein, Christian Holscher, José B. Carvalheira, Aristobolo M. Silva, Lício A. Velloso, Douglas P. Munoz, Sergio T. Ferreira, and Fernanda G. De Felice!

Lourenco et al, Figure S1






Supplemental Figure Legends

Figure S1, related to Figure 1 – A"O-induced ER stress, eIF2!-P and phospho-

PKR in hippocampal neurons. A, IRE1!-pSer724 immunolabeling (yellow),

showing intense perinuclear immunoreactivity. B-D, Immunoblot analyses for spliced

XBP1 (XBP1s) (B), eIF2!-pSer51 (C) and GRP78 (D) in hippocampal cultures

exposed to vehicle or 500 nM A!Os for 3 hours (n = 3 independent experiments). E,

CHOP/gadd153 mRNA levels in hippocampal cultures exposed to vehicle or 500 nM

A!Os for 24 hours (n = 4 independent experiments). No change in CHOP mRNA

level was detected after 12-hour exposure to A"Os. * p < 0.05, Student’s t-test.

Graphs show means ± standard error (SEM). F,H, eIF2!-P (F) and phospho-PKR (H)

immunoreactivities in hippocampal CA2 and subiculum of cynomolgus monkeys that

received i.c.v. injections of A!Os compared to sham-operated control monkeys (scale

bar = 200 µm). G, I, eIF2!-P (G) and phospho-PKR (I) immunolabeling densities in

CA2 (upper graph) and subiculum (lower graph) in sham or A!O-injected monkeys.

Graphs show means ± standard error (SEM) of animals. N = 3 and 4 for sham and

A"O-injected monkeys, respectively. *p < 0.05, Student’s t-test.

Figure S2, related to Figure 2 – PERK is not activated by A"Os in hippocampal

neurons. A, PERK-pThr981 immunolabeling (green) in hippocampal neurons exposed

to vehicle or 500 nM A!Os for 3 h (scale bar = 10 µm). A!O-specific NU4 labeling is

shown in red. B, PERK-pThr981 immunofluorescence levels (n = 4 experiments using

independent cultures; 30 images analyzed/experimental condition/experiment). C,

Immunoblot analysis for PERK-pThr981 (normalized by total PERK) in hippocampal

homogenates from 13-16 month old WT (n = 5) or APP/PS1 (n = 7) mice. Graphs

show means ± standard error (SEM).

Figure S3, related to Figure 3 – TNFR1-mediated memory impairment caused

by A" oligomers. Step-down latency of WT or TNFR1-/- mice injected with vehicle

or 10 pmol A!Os in the step-down inhibitory avoidance task (* p < 0.05, ANOVA

followed by Bonferroni post-hoc test). Graphs show means ± standard error (SEM).

Figure S4, related to Figure 4 – Salubrinal causes neuronal eIF2!-P in vitro and

in vivo. A, eIF2!-P immunolabeling in cultured hippocampal neurons exposed for 3

hours to vehicle or 10 µM salubrinal. B, Quantification of eIF2!-P levels, determined

from 3 experiments using independent cultures. C, Immunoblot analysis of eIF2!-P

levels (normalized by total eIF2!) in hippocampal homogenates of 2-3 month old

mice injected i.p. with vehicle or 1 mg/kg salubrinal for 7 days (n=7 per experimental

group). * p < 0.05, Student’s t-test. Graphs show means ± standard error (SEM).

Figure S5, related to Figure 5 – ER stress promotes neuronal eIF2!-P and

cognitive impairment in mice. A, eIF2!-P immunolabeling in cultured hippocampal

neurons exposed for 3 hours to vehicle or 1 µM thapsigargin. B, Quantification of

eIF2!-P levels, determined from 3 experiments using independent cultures. C,

Immunoblot analysis of eIF2!-P levels (normalized by total eIF2!) in hippocampal

homogenates of 2-3 month old mice injected i.c.v. with vehicle or a single dose of 1

µg thapsigargin (n=7 per experimental group). D, Exploration times of mice i.c.v.

injected with vehicle (white bars), 1 µg thapsigargin in the absence (black bars) or

presence (dashed bars) of 200 mg/kg 4-phenylbutyrate (4-PBA) in the novel object

recognition task (n=10 per experimental group). E, Exploration times of mice injected

with vehicle (white bars), 10 pmol A!Os (i.c.v.) in the absence (black bars) or

presence (hatched bars) of 200 mg/kg 4-PBA (i.p.) in the novel object recognition

task (n=10 per experimental group). Asterisks denote a statistically significant (p <

0.05) difference from 50% (reference value). Graphs show means ± standard error

(SEM).

Figure S6, related to Figure 6 – Anti-diabetic agents rescue AD pathology and

cognitive impairment in APP/PS1 mice. A-D, Immunoblot analyses for GRP78 (A),

drebrin (B), 28 kDa A"Os (C) and 108 kDa A!Os (D) in hippocampal homogenates

from 13-16 month old APP/PS1 (n = 7) or liraglutide-treated APP/PS1 (n = 5) mice. *

p < 0.05, Student’s t-test. E-F, Freezing times of exendin-4- (E) or liraglutide-treated

(F) APP/PS1 mice compared to untreated APP/PS1 mice in a contextual fear

conditioning task. N = 10 animals per experimental group. *p < 0.05, Student t-test.

G-H, Body weights (means !" SEM; n = 12 animals per experimental group) in

exendin-4- (G) or liraglutide-treated mice (H), before (B) and after (A) treatments.

Graphs show means ± standard error (SEM).

Supplemental Experimental Procedures

Reagents

Synthetic A!1–42 peptide was from American Peptide (Sunnyvale, CA). Human

insulin, 3,3’-diaminobenzidine, 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), DMSO,

DAPI, 4-phenylbutyrate, thapsigargin and poly-L-lysine were from Sigma (St. Louis,

MO). Culture media/reagents, RNA extraction and qPCR kits, Alexa-labeled

secondary antibodies and ProLong anti-fade reagent were from Invitrogen (Carlsbad,

CA). Electrophoresis buffers were from BioRad (Hercules, CA). SuperSignal

chemiluminescence reagents, BCA protein assay kits and the antibody against

synaptophysin were from Pierce (Deerfield, IL). The antibody against the ! subunit of

eukaryotic initiation factor 2 phosphorylated at serine 51 (eIF2!#P) was from Enzo

Life Sciences (Farmingdale, NY). Antibodies against inositol-requiring enzyme 1!

phosphorylated at serine 724 (IRE1!#pSer724), glucose-regulated protein 78

(GRP78/Bip), drebrin, beta-actin and cyclophilin B were from Abcam (Cambridge,

MA). Antibodies against total or phosphorylated double-stranded RNA-dependent

protein kinase (PKR or PKR-pThr451, respectively), total or phosphorylated PKR-like

endoplasmic reticulum kinase (PERK or PERK-pThr981, respectively), PSD95, insulin

receptor substrate 1 phosphorylated at serine 636 (IRS-1pSer636) and total IRS-1 were

from Santa Cruz Biotechnology (Santa Cruz, CA). Exendin-4 was from Bachem

(Torrance, CA). Liraglutide was from GL Biochem (Shanghai). SP600125 was from

Tocris Bioscience (Ellisville, MO). Infliximab (Remicade) was from Schering-Plough

(Kenilworth, NJ). PKR inhibitor was from Santa Cruz Biotechnology (Santa Cruz,

CA). Salubrinal was from Merck Millipore (Darmstadt, Germany). Rat TNF-!

ELISA kits were from Peprotech (Rocky Hill, NJ). A! oligomer-specific antibody

(NU4) was generated in William L. Klein’s laboratory (Northwestern University,

Evanston, IL) and has been previously characterized (Lambert et al., 2007).

Immunofluorescence

Cells were fixed and blocked as described (De Felice et al., 2009) and were double-

labeled with oligomer-specific NU4 mouse monoclonal antibody plus eIF2!#P,

IRE1!#pSer724, IRS-1pSer636, PKR-pThr451 or PERK-pThr981 rabbit polyclonal

antibodies followed by Alexa-conjugated secondary antibodies (Invitrogen, CA).

Nuclei were counterstained with DAPI. Coverslips were imaged on a Zeiss

AxioObserver Z1 microscope. eIF2!#P, IRE1!#pSer724, PKR-pThr451 and PERK-

pThr981 immunofluorescence intensities were each analyzed in 4–6 experiments (see

figure legends) using independent neuronal cultures. In each experiment, 20-30

images were acquired from each of 3 coverslips in each experimental condition.

Histogram analysis of fluorescence intensities at each pixel across the images was

performed using NIH Image J (Abràmoff et al., 2004) as described (De Felice et al.,

2007). When indicated, cell bodies were digitally removed from the images so that

only immunostaining on dendritic processes was quantified. For synapse

quantification, cells were double-labeled for synaptophysin (pre-synaptic marker) and

PSD95 (post-synaptic marker) and imaged on a Nikon TE Eclipse microscope.

Synapses were defined as a juxtaposition between synaptophysin and PSD95 puncta

(Diniz et al., 2012) and were quantified using Puncta Analyzer ImageJ plugin,

developed by Ben Barres lab (Stanford University) and previously described

(Christopherson et al., 2005). Results are from at least 3 independent experiments,

carried out in duplicates. Statistical significances of differences between experimental

groups were assessed by ANOVA followed by post-hoc Bonferroni test, and p-values

are indicated in figure legends.

RNA extraction

Total RNA was extracted from hippocampal cultures using Trizol (Invitrogen, CA),

following manufacturer’s instructions. Briefly, 1 mL of Trizol was used to extract

RNA from 1.5 x 106 cells. Purity and integrity of RNA preparations were checked by

the 260/280 nm absorbance ratio and by agarose gel electrophoresis. Only

preparations with 260/280 nm OD ratios higher than 1.8 and no signs of rRNA

degradation were used. RNA concentrations were determined by absorption at 260

nm.

Quantitative RT-PCR

One microgram of total RNA was used for cDNA synthesis using 50 pmol of oligo

dT20 and the Superscript III First Strand cDNA kit. Quantitative expression analysis

of CHOP was performed by qRT-PCR on a 7500 Applied Biosystems Real-Time

PCR system with the Power SYBR kit (Applied Biosystems, Foster City, USA). "-

actin (actb) was used as an endogenous reference gene for data normalization. qRT-

PCR was performed in 20 "L reaction volumes according to manufacturer’s

protocols. The following primer sequences were used: CHOP/gadd153 forward

(5’GAAAGCAGAAACCGGTCCAAT3’), CHOP/gadd153 reverse (5’GGATGAGA

TATAGGTGCCCCC3’), actb forward (5’GTCTTCCCCTCCATCGTG3’) and actb

reverse (5’AGGATGCCTCTCTTGCTCTG3’) for the detection of CHOP/gadd153

and "-actin mRNAs, respectively. Cycle threshold (Ct) values were used to calculate

fold changes in gene expression using the 2-##Ct method (Livak and Schmittgen,

2001). Statistical significance of changes in expression was assessed using Student’s t

test.

Neuropathology in monkey brains

Immunohistochemistry was performed as described (Medeiros et al., 2007). Briefly,

hippocampal or cortical sections from monkey brains were incubated in pre-heated

0.1 M citrate buffer, pH 6.0, for 40 min. Sections were blocked with 5% bovine serum

albumin (BSA), 5% normal goat serum (NGS) and 1% Triton X-100 for 3 h at room

temperature. Primary antibodies (eIF2!#P, PKR-pThr451, 1:200 dilution) were diluted

in blocking solution and sections were incubated overnight at 4 °C followed by

incubation with biotinylated secondary antibodies for 1 h at room temperature.

Sections were then exposed to HRP-conjugated biotin-streptavidin (Vector Labs, CA)

for 30 min and developed with 3,3’-diaminobenzidine. Light counterstaining was

performed with hematoxylin. Slides were imaged on a Zeiss AxioPlan microscope

using brightfield illumination under the same acquisition settings for all conditions.

eIF2!#P and PKR-pThr451 immunolabeling densities were determined using a

multithreshold plugin within NIH ImageJ software. For each animal, densities were

measured in a set of 15-30 microscopic fields (20X objective) throughout defined

hippocampal and/or cortical regions (see “Results”). Brain regions were defined

according to a cynomolgus stereotaxical atlas (Martin and Bowden, 1996). Statistical

significances were assessed by one-way ANOVA followed by post-hoc Bonferroni

test and p-values are indicated in figure legends.

Western blotting

Thirteen to fourteen month-old APPSwe/PS1#E9 transgenic mice and three month-

old PKR-/- and TNFR1-/- mice and wild-type control animals were used. Animals (at

least 5 per experimental group) were euthanized and their hippocampi were removed.

For Western blot analysis, hippocampi or mature hippocampal cell cultures were

homogenized in RIPA buffer containing protease and phosphatase inhibitor cocktails

and resolved on 4-20% polyacrylamide pre-cast gels with Tris/glycine/SDS buffer run

at 125 V for 80 min at room temperature. The gel (30 µg total protein/lane) was

electroblotted onto Hybond ECL nitrocellulose using 25 mM Tris, 192 mM glycine,

20% (v/v) methanol, 0.02% SDS, pH 8.3, at 350 mA for 2 h at 4 °C. Membranes were

blocked with 5% non-fat milk in Tris-buffered saline containing Tween-20 (TBS-T)

(0.1% Tween-20 in 20 mM Tris-HCl, pH 7.5, 0.9% NaCl) for 1 h at room

temperature. Primary antibodies (anti-eIF2!-P, anti-p-PERK, anti-PERK, anti-

synaptophysin, anti-PSD95, anti-drebrin antibodies; anti-A"Os NU4 (1:1000), anti-

phospho-PKR, anti-PKR, anti-IRS-1pSer636, anti-IRS-1 (1:400), anti-beta actin

monoclonal antibodies or anti-cyclophilin B polyclonal antibody (1:10,000)) were

diluted in 5% milk/TBS and incubated with the membranes overnight at 4 oC. After

incubation with HRP-conjugated anti-mouse or anti-rabbit IgGs (1:10,000 in TBS-T)

for 60 min, membranes were washed, developed with SuperSignal West Femto

Maximum Sensitivity substrate and imaged on photographic film.

Soluble TNF-! levels

Soluble TNF-! levels were measured using an ELISA kit (Peprotech) according to

manufacturer instructions. Prior to analysis, hippocampal cultures were exposed to

500 nM A!Os or an equivalent volume of vehicle for 3 hours. The medium was then

removed and concentrated by Speedvac (Savant Instruments, Inc., Farmingdale, NY)

centrifugation. Statistical significance of differences between experimental groups

was determined by Student’s t-test.

Novel object recognition paradigm

Object recognition experiments were carried out in an open field arena measuring 0.3

(w) x 0.3 (d) x 0.45 (h) m. Test objects were made of glass or plastic and had different

shapes, colors, sizes and textures. During behavioral sessions, objects were fixed with

tape to the floor so that the animals could not move them. None of the objects used in

our experiments evoked innate preference. Before training, each animal was

submitted to a 5 minute-long habituation session, in which they were allowed to

freely explore the empty arena. During habituation sessions, the number of lines that

each animal crossed on the floor (number of crossings) and the number of rearings

(elevation on rear paws, denoting exploratory behavior) were recorded to verify

possible effects of treatments on locomotor exploratory activities. Training consisted

in a 5 minute-long session during which animals were placed at the centre of the arena

in the presence of two identical objects. The time spent exploring each object was

recorded by a trained researcher. Sniffing and touching the object were considered as

exploratory behavior. The arena and stimulus objects were cleaned thoroughly

between trials with 20% alcohol (vol/vol) to ensure minimal olfactory cues. Two

hours after training, animals were reinserted into the arena for the test session, when

one of the two objects used in the training session was replaced by a new one. Again,

the time exploring familiar and novel object was measured. Results are expressed as

percentage of time exploring each object during the training or test session and were

analyzed using a one-sample Student’s t-test comparing the mean exploration time for

each object with the fixed value of 50%. By definition, animals that recognize the

familiar object as such (i.e., normal learning) explore the novel object for a time

significantly longer than 50% of the total time.

Contextual Fear Conditioning

For the experiments using APPSwe/PS1#E9 mice, vehicle-, exendin-4- or liraglutide-

treated transgenic mice were allowed to freely explore the conditioning chamber (0.4

x 0.25 x 0.3 m) for 2 min, after which a 2-sec shock stimulus (0.8 mA) was applied to

the floor. Animals remained in the cage for 1 additional minute. After 24 h, mice were

placed in the same cage and allowed to explore it for 3 min in the absence of electric

shock. Freezing time was measured during this period. Statistical significances of

differences between groups were evaluated by Student’s t-test. In experiments with

TNFR1-/- and PKR-/- mice and corresponding wild-type controls, animals were

allowed to freely explore the training chamber (0.25 x 0.25 x 0.25 m; Harvard

Apparatus) for 3 minutes, after which they received two 2-sec long 0.35 mA foot-

shocks with a 30 sec interval. Animals were removed from the cage after 30 seconds.

Twenty-four hours thereafter, they were reinserted into the box for 5 minutes and total

freezing time during test session was determined. Statistical significances of

differences between groups were evaluated by two-way ANOVA followed by

Bonferroni post-hoc test.

Inhibitory Avoidance

The inhibitory avoidance apparatus consists on a box measuring 0.5 x 0.25 x 0.25 m

(Insight) with a 0.09 x 0,05 x 0.02 m platform placed in the centre, surrounded by a

floor made of parallel bronze bars and connected to a power source. During the

training sessions, TNFR1-/- mice and wild-type controls were gently placed on the

platform and when they stepped down with four paws onto the grip they received a 2

sec-long, 0.7 mA scrambled foot shock. Latency to step down during training session

was registered. For evaluation of memory retention, animals were again placed on top

of the platform twenty-four hours after training, and latency to step down was

registered again. Prior to test, all animals were subjected to an open field session to

eliminate any locomotor or exploratory deficit that could interfere with the latency to

step down.

Mouse genotyping

Genotyping was carried out according to Jackson’s Laboratories protocols

(http://jaxmice.jax.org/strain/004462.html). Briefly, genomic DNA was extracted

from thin sections of mice tails and subjected to amplification using specific forward

and reverse primers for the transgenic construct. The thermal cycling consisted of 35

cycles at 94 °C for 1 min, 52 °C for 1 min, and 72 °C for 1 min. PCR products were

resolved on a 1.5% agarose gel, evidencing a band of 377 bp for transgenic animals

after ethidium bromide staining.

!

!

!

!

!

!

!

"#$$%&'&()*%!+&,&-&(.&/!

Abràmoff, M.D., Magalhães, P.J., and Ram, S.J. (2004). Image processing with ImageJ. Biophotonics International 2004, 1-7.

Christopherson, K.S., Ullian, E.M., Stokes, C.C.A., Mullowney, C.E., Hell, J.W., Agah, A., Lawler, J., Mosher, D.F., Bornstein, P., and Barres, B.A. (2005). Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell 120, 421-433.

Diniz, L.P., Almeida, J.C., Tortelli, V., Vargas Lopes, C., Setti-Perdigao, P., Stipursky, J., Kahn, S.A., Romao, L.F., de Miranda, J., Alves-Leon, S.V., et al. (2012). Astrocyte-induced synaptogenesis is mediated by transforming growth factor ! signaling through modulation of D-serine levels in cerebral cortex neurons. Journal of Biological Chemistry 287, 41432-41445.

Lambert, M.P., Velasco, P.T., Chang, L., Viola, K.L., Fernandez, S., Lacor, P.N., Khuon, D., Gong, Y., Bigio, E.H., Shaw, P., et al. (2007). Monoclonal antibodies that target pathological assemblies of A!. Journal of Neurochemistry 100, 23-35.

Livak, K.J., and Schmittgen, T.D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402-408.

Martin, R.F., and Bowden, D.M. (1996). A stereotaxic template atlas of the macaque brain for digital imaging and quantitative neuroanatomy. Neuroimage 4, 119-150.

Medeiros, R., Prediger, R.D.S., Passos, G.F., Pandolfo, P., Duarte, F.S., Franco, J.L., Dafre, A.L., Di Giunta, G., Figueiredo, C.P., Takahashi, R.N., et al. (2007). Connecting TNF-$ signaling pathways to iNOS expression in a mouse model of Alzheimer's disease: relevance for the behavioral and synaptic deficits induced by amyloid-! protein. Journal of Neuroscience 27, 5394-5404.

"""

"

Research Article

Alzheimer-associated Ab oligomers impact thecentral nervous system to induce peripheralmetabolic deregulationJulia R Clarke1,2,†, Natalia M Lyra e Silva1,†, Claudia P Figueiredo2, Rudimar L Frozza1, Jose H Ledo1,

Danielle Beckman1, Carlos K Katashima3, Daniela Razolli3, Bruno M Carvalho3, Renata Frazão4,

Marina A Silveira4, Felipe C Ribeiro1, Theresa R Bomfim1, Fernanda S Neves2, William L Klein5,

Rodrigo Medeiros6, Frank M LaFerla6, Jose B Carvalheira3, Mario J Saad3, Douglas P Munoz7,

Licio A Velloso3, Sergio T Ferreira1,8 & Fernanda G De Felice1,*

Abstract

Alzheimer’s disease (AD) is associated with peripheral metabolicdisorders. Clinical/epidemiological data indicate increased risk ofdiabetes in AD patients. Here, we show that intracerebroventricu-lar infusion of AD-associated Ab oligomers (AbOs) in mice triggeredperipheral glucose intolerance, a phenomenon further verified intwo transgenic mouse models of AD. Systemically injected AbOsfailed to induce glucose intolerance, suggesting AbOs target brainregions involved in peripheral metabolic control. Accordingly, weshow that AbOs affected hypothalamic neurons in culture, induc-ing eukaryotic translation initiation factor 2a phosphorylation(eIF2a-P). AbOs further induced eIF2a-P and activated pro-inflammatory IKKb/NF-jB signaling in the hypothalamus of miceand macaques. AbOs failed to trigger peripheral glucose intoler-ance in tumor necrosis factor-a (TNF-a) receptor 1 knockout mice.Pharmacological inhibition of brain inflammation and endoplasmicreticulum stress prevented glucose intolerance in mice, indicatingthat AbOs act via a central route to affect peripheral glucosehomeostasis. While the hypothalamus has been largely ignored inthe AD field, our findings indicate that AbOs affect this brain regionand reveal novel shared molecular mechanisms between hypotha-lamic dysfunction in metabolic disorders and AD.

Keywords Alzheimer’s disease; ER stress; hypothalamus; inflammation;

insulin resistance

Subject Categories Metabolism; Neuroscience

DOI 10.15252/emmm.201404183 | Received 30 April 2014 | Revised 12

December 2014 | Accepted 17 December 2014

EMBO Mol Med (2015) 7: 190–210

Introduction

Increasing evidence suggests an association between metabolic

disorders, notably type 2 diabetes (T2D), and Alzheimer’s disease

(AD) (Craft, 2007; De Felice, 2013). Clinical and epidemiological

studies indicate that diabetic patients have increased risk of devel-

oping AD (Ott et al, 1999; Sims-Robinson et al, 2010; Wang et al,

2012) and AD brains exhibit defective insulin signaling (Moloney

et al, 2010; Bomfim et al, 2012; Craft, 2012; Talbot et al, 2012).

Recent studies have shown that soluble amyloid-b peptide oligomers

(AbOs), toxins that build up in AD brains and have been proposed

to be major players in synapse failure in AD (reviewed in Ferreira

& Klein, 2011; Selkoe, 2011; Mucke & Selkoe, 2012), are linked to

impaired hippocampal insulin signaling. AbOs were found to cause

internalization and cellular redistribution of insulin receptors, to

block downstream hippocampal insulin signaling (De Felice et al,

2009; Ma et al, 2009; Bomfim et al, 2012) and to cause hippocampal

endoplasmic reticulum (ER) stress (Lourenco et al, 2013), establish-

ing molecular parallels between AD and T2D. Hyperinsulinemic/

hyperglycemic individuals and mice show increased plasma and

brain levels of Ab (Ho et al, 2004; Takeda et al, 2010; Zhang et al,

2012), suggesting that altered peripheral metabolic homeostasis

1 Institute of Medical Biochemistry Leopoldo de Meis, Federal University of Rio de Janeiro, Rio de Janeiro, RJ, Brazil2 School of Pharmacy, Federal University of Rio de Janeiro, Rio de Janeiro, RJ, Brazil3 Department of Internal Medicine, Faculty of Medical Sciences, State University of Campinas, Campinas, SP, Brazil4 Department of Anatomy, Institute of Biomedical Sciences, University of São Paulo, SP, Brazil5 Department of Neurobiology, Northwestern University, Evanston, IL, USA6 Institute for Memory Impairments and Neurological Disorders, University of California, Irvine, CA, USA7 Center for Neuroscience Studies, Queen’s University, Kingston, ON, Canada8 Institute of Biophysics Carlos Chagas Filho, Federal University of Rio de Janeiro, Rio de Janeiro, RJ, Brazil

*Corresponding author. Tel: +55 21 38888308; E-mail: [email protected]†These authors contributed equally to this work

EMBO Molecular Medicine Vol 7 | No 2 | 2015 ª 2015 The Authors. Published under the terms of the CC BY 4.0 license190

may increase Ab levels and influence AD development (De Felice,

2013; De Felice & Ferreira, 2014).

Intriguingly, AD has been associated with increased risk of T2D

development (Janson et al, 2004), suggesting that the connection

between AD and T2D may be a two-way road. Early studies demon-

strated peripheral glucose intolerance in AD patients (Craft et al,

1992). Recently, hyperglycemia and hyperinsulinemia, cardinal

features of T2D and other metabolic disorders, were found to posi-

tively correlate with the development of AD-like brain pathology in

humans (Matsuzaki et al, 2010). Obesity-induced insulin resistance

is exacerbated in transgenic mouse models of AD (Takeda et al,

2010; Jimenez-Palomares et al, 2012). However, the molecular

mechanisms underlying these observations are still largely

unknown.

We hypothesized that AbOs could impact brain regions responsi-

ble for metabolic control and therefore represent a key pathogenic

link between AD and deregulated peripheral glucose homeostasis.

The hypothalamus plays a central role in neuroendocrine interaction

between the central nervous system and the periphery (Schwartz &

Porte, 2005; Koch et al, 2008). Emerging evidence further indicates

that hypothalamic inflammation and ER stress are critical patho-

genic events in the establishment of peripheral insulin resistance in

metabolic disorders (Zhang et al, 2008; Milanski et al, 2009; Denis

et al, 2010; Arruda et al, 2011; Thaler et al, 2012). An interesting

recent study showed that hypothalamic inflammation accelerates

aging and shortens lifespan in mice (Zhang et al, 2013). In post-

mortem AD brains, early studies identified Ab deposits in the hypo-

thalamus (Ogomori et al, 1989; Standaert et al, 1991). More

recently, voxel-based morphometry revealed reduced hypothalamic

volume in early AD compared to healthy controls (Loskutova et al,

2010), and a decrease in the number of hypothalamic orexin

neurons has been reported in AD brains (Fronczek et al, 2012). In

rats that received an intracerebroventricular injection of amyloid-

b25–35 fibrils, Zussy et al, (2011) detected accumulation of fibrillar

aggregates in the hypothalamus for as long as 3 weeks after the

injection, as well as hypothalamic astrocytosis. In addition, oligo-

meric species of the amyloid-b peptide were recently shown to

induce oxidative stress in a hypothalamic cell line (Gomes et al,

2014). While the hypothalamus has been largely ignored in the AD

field, these studies indicate that this brain region could indeed be

affected in AD. If so, hypothalamic dysfunction may have important

consequences, predisposing AD patients to develop diabetes.

Several studies have established that AbOs target hippocampal

neurons and induce synapse loss and neuronal dysfunction, eventu-

ally leading to memory impairment in AD (Ferreira & Klein, 2011;

Mucke & Selkoe, 2012; Selkoe, 2012). Intracerebroventricular

(i.c.v.) administration of AbOs has been shown to cause synapse

loss and behavioral alterations linked to AD in mice (Figueiredo

et al, 2013; Ledo et al, 2013) and AD-like pathology in non-human

primates (Forny-Germano et al, 2014), providing a suitable model

to investigate mechanisms germane to AD. Here, we show that

i.c.v.-injected AbOs induce peripheral glucose intolerance and hall-

marks of insulin resistance, including adipose tissue inflammation

and impaired insulin-induced surface translocation of GLUT-4 in

skeletal muscle. Peripheral glucose intolerance appeared to be medi-

ated by a direct effect of AbOs in the central nervous system, and

not by leakage of oligomers to peripheral tissues, as peripherally

administered AbOs failed to induce glucose intolerance in mice.

Glucose intolerance was further verified in two transgenic mouse

models of AD, namely 3xTg-AD (Oddo et al, 2003) and APP/PS1

(Jankowsky et al, 2001) mice. We show that AbOs target primary

hypothalamic neurons in vitro and accumulate in the hypothalamus

of cynomolgus macaques given i.c.v. infusions of AbOs. AbOsfurther triggered aberrant generation of reactive oxygen species

(ROS) and phosphorylation of eIF2a in cultured hypothalamic

neurons, as well as activation of IKKb/NF-jB inflammatory signal-

ing in the hypothalamus of mice and macaques. The impact of AbOsin the hypothalamus of mice preceded alterations in peripheral

glucose homeostasis. In TNF-a receptor 1 knockout mice (Romanatto

et al, 2009), AbOs failed to trigger hypothalamic IKK activation

and IRS-1 inhibition. AbO-associated glucose intolerance was

prevented in TNFR1!/! mice as well as in wild-type mice given

i.c.v. infusions of tauroursodeoxycholic acid (TUDCA), an ER stress

inhibitor. i.c.v treatment with infliximab, a TNF-a neutralizing anti-

body, further prevented glucose intolerance in AbO-injected mice

and in APP/PS1 mice. Collectively, results establish a novel patho-

genic mechanism by which AbOs impact the hypothalamus, causing

peripheral metabolic deregulation.

Results

Mouse models of AD exhibit impaired glucose tolerance

Alzheimer’s disease has been associated with increased risk of

T2D development. We hypothesized that brain accumulation of

AbOs could represent a key pathogenic link between AD and

deregulated peripheral glucose homeostasis. To test this hypothe-

sis, we initially performed a single injection of 10 pmol AbOs into

the right lateral cerebral ventricle of adult Swiss mice (Supplemen-

tary Fig S1; Figueiredo et al, 2013; Ledo et al, 2013). AbOs were

freshly prepared before each experiment and were routinely char-

acterized by size-exclusion chromatography, Western blots using

anti-oligomer monoclonal antibody NU4 (Lambert et al, 2007) and,

occasionally, by transmission electron microscopy, as previously

described (Jurgensen et al, 2011; Sebollela et al, 2012; Figueiredo

et al, 2013). Interestingly, mice that received an i.c.v. injection of

AbOs exhibited impaired peripheral glucose tolerance and insulin

resistance 7 days after injection (Fig 1A and B). Control experi-

ments showed that peripheral glucose tolerance was unaffected by

i.c.v. injection of a preparation of scrambled Ab peptide submitted

to the same oligomerization protocol used for regular AbO prepa-

rations (Supplementary Fig S2A). The impairment in glucose toler-

ance induced by i.c.v. AbOs was comparable to that verified in

mice submitted to a high-fat diet for 7 days (Supplementary Fig

S2B). Impaired glucose tolerance could be detected as early as

36 h, but not 12 h after i.c.v. injection of AbOs (Supplementary

Fig S2D and C), and persisted for at least 14 days post-injection

(Supplementary Fig S2E). We further examined the possibility that

leakage of AbOs from the brain might explain the observed effects

on peripheral glucose metabolism. To this end, we injected 10

pmol AbOs (the same amount used in i.c.v. injections) directly

into the caudal vein or into the peritoneum of mice. In either case,

systemic administration of AbOs failed to impair glucose tolerance

(Fig 1C and D), ruling out a direct action of AbOs on peripheral

tissues in our conditions.

ª 2015 The Authors EMBO Molecular Medicine Vol 7 | No 2 | 2015

Julia R Clarke et al AbOs trigger peripheral metabolic deregulation EMBO Molecular Medicine

191

A B

C D

E F

Figure 1. AD mouse models show peripheral glucose intolerance.

A Adult Swiss mice (n = 11 Veh; 15 AbOs) received a single i.c.v. injection of vehicle or 10 pmol AbOs and were assessed in a glucose tolerance test (2 g glucose/kgbody weight, i.p.) 7 days after injection. Blood levels of glucose were measured at several time points following glucose administration. Bar graph represents areasunder the curves in the time course plot. Data are representative of three independent experiments with similar results. Left panel: ***P = 0.0006, two-way ANOVAfollowed by Bonferroni post hoc test; right panel: *P = 0.0207, Student’s t-test.

B Insulin tolerance test (1 IU insulin/kg body weight, i.p.) (n = 7 Veh; 8 AbOs). Blood levels of glucose were measured at several time points following insulinadministration. Bar graph represents the kinetic constants for glucose disappearance (Kitt) calculated from the time course plot. Data are representative of twoindependent experiments with similar results. Left panel: *P = 0.0456 and ***P = 0.0007, two-way ANOVA followed by Bonferroni post hoc test; right panel:**P = 0.0033, Student’s t-test.

C, D Glucose tolerance test (2 g glucose/kg body weight, i.p.) in mice 7 days after a single intracaudal (C; n = 8 animals/group) or intraperitoneal (D; n = 13 animals/group) injection of AbOs (10 pmol) or vehicle.

E, F Glucose tolerance test (2 g glucose/kg body weight, i.p.) in 8- to 13-month-old APP/PS1 mice (E; n = 9 animals/group) or 6-month-old 3xTg-AD male mice(F; n = 10 WT; 9 3xTg), or their corresponding wild-type littermates. Bar graph represents areas under the curves in the time course plots. In (E), left panel:*P = 0.0466, two-way ANOVA followed by Bonferroni post hoc test; right panel: &P = 0.072, Student’s t-test. In (F), left panel: *P = 0.0171 and #P = 0.0781,two-way ANOVA followed by Bonferroni post hoc test; right panel: *P = 0.0101, Student’s t-test.

Data information: Data are expressed as means ! SEM.

EMBO Molecular Medicine Vol 7 | No 2 | 2015 ª 2015 The Authors

EMBO Molecular Medicine AbOs trigger peripheral metabolic deregulation Julia R Clarke et al

192

Significantly, altered peripheral glucose homeostasis was also

verified in 9- to 13-month-old APPSwePS1∆E9 (APP/PS1) mice

compared to wild-type animals (Fig 1E). Those mice harbor trans-

genes for human amyloid precursor protein (APP) bearing the Swed-

ish mutation and a deletion mutant form of presenilin 1 (Shi et al,

2011a), and present increased Ab production and cognitive deficits

(Jankowsky et al, 2001). Similar results were obtained using the

triple-transgenic mouse model of AD (3xTg-AD), which presents

increased Ab levels and develops tau and synaptic pathology, hall-

mark features of AD (Oddo et al, 2003). We found that 6-month-old

3xTg-AD mice show glucose intolerance compared to wild-type

littermates (Fig 1F). The fact that altered peripheral glucose homeo-

stasis was detected in both mouse models exhibiting progressive Abaccumulation in the brain underscores the notion that our observa-

tions in the acute model consisting of brain infusion of AbOs are

relevant when compared to clinical observations in early AD

patients (Craft et al, 1992).

i.c.v. injection of AbOs induces metabolic changes in muscle andadipose tissue and increases plasma noradrenaline levels

We next sought to analyze metabolic changes and insulin respon-

siveness in metabolically active tissues. We found increased CD68

immunoreactivity in adipose tissue of mice that received an i.c.v.

injection of AbOs (Fig 2A), indicating macrophage/myeloid cell

infiltration. Further, AbO-injected mice had higher amounts of

epididymal fat (Fig 2B) and increased expression of leptin and pro-

inflammatory cytokines, TNF-a and IL-6, in white adipose tissue

(Fig 2C–E). In obese mice, adipose-derived TNF-a is involved in

insulin resistance through the activation of JNK, leading to

increased inhibitory serine phosphorylation of insulin receptor

substrate-1 (IRS-1pSer) in muscle (Hotamisligil et al, 1996; Ozcan

et al, 2004). Therefore, we investigated whether this pathway was

affected in AbO-injected mice. Indeed, skeletal muscle from mice

i.c.v. injected with AbOs showed increased levels of activated JNK

(Fig 2F) and IRS-1pSer312 (Fig 2G). Physiologically, insulin signaling

in muscle induces translocation of glucose transporter-4 (GLUT-4)

from intracellular compartments to the plasma membrane (Huang &

Czech, 2007). In line with our finding of IRS-1 inhibition, insulin-

stimulated translocation of GLUT-4 to the plasma membrane was

severely impaired in skeletal muscle of mice that received an i.c.v.

injection of AbOs (Fig 2H), while GLUT-4 expression and total

protein levels in muscle remained unaltered (Fig 2I and J).

In order to provide a more comprehensive view of metabolic

deregulation in AbO-injected mice, we next measured serum levels

of leptin and insulin in mice 7 days after i.c.v. injection of AbOs. We

found no changes in serum levels of insulin or leptin under these

conditions (Fig 2K and L). As noted above, the impairment in

glucose tolerance induced by i.c.v. administration of AbOs is compa-

rable to that verified in mice submitted to a high-fat diet (HFD) for

7 days (Supplementary Fig S2B). In harmony with our results, previ-

ous studies have shown that plasma leptin and insulin levels are not

affected in mice (wild-type or ob/ob) submitted to a short-term

(4–7 days) high-fat diet (HFD), whereas glucose tolerance and insu-

lin sensitivity are clearly impaired under the same conditions (e.g.,

El-Haschimi et al, 2000; Ji et al, 2012; Le et al, 2014). Further, short-

term HFD induces increases in epididymal white adipose tissue

weight, adipocyte hypertrophy and increased transcript levels of

TNF-a and IL-6 (e.g., Lee et al, 2011; Ji et al, 2012), similar to our

observations in mice i.c.v. injected with AbOs. Moreover, plasma

levels of cholesterol and triglycerides were comparable between

vehicle- and AbO-injected animals (Fig 2M and N). We further found

elevated plasma noradrenaline (NA) levels (Fig 2O), indicating that

AbOs cause deregulation of peripheral sympathetic control.

AbOs bind to hypothalamic neurons in culture and induceaberrant ROS generation and TNF-a-dependent increasein eIF2a-P

Since i.p. or i.v. administration of AbOs had no effect on peripheral

glucose homeostasis, we hypothesized that AbOs could target brain

regions involved in control of peripheral glucose homeostasis.

Because interference in the hypothalamus of mice has been

shown to be sufficient to induce peripheral metabolic deregulation

(Purkayastha et al, 2011), and early studies showed that Abaccumulates in the hypothalamus of AD patients, we next aimed to

determine whether this brain region was particularly affected in our

experimental models. Initially, highly differentiated primary hypo-

thalamic neuronal cultures were exposed to AbOs (500 nM) for 3 h

and AbO binding to neurons was investigated by double immunoflu-

orescence labeling using oligomer-sensitive antibody NU4 (Lambert

et al, 2007) and microtubule-associated protein 2 (MAP-2). Results

showed that AbOs bind to the soma and, especially, to dendrites of

selected hypothalamic neurons (Fig 3A), similar to previous results

demonstrating that oligomers bind to a specific subset of neurons in

hippocampal cultures, rather than to all neurons (Lacor et al, 2004;

Zhao et al, 2008; Bomfim et al, 2012; Lourenco et al, 2013). To

examine the possibility that AbOs could bind to astrocytes, we

further double-labeled cultures with anti-GFAP and NU4. Results

indicate that oligomers do not bind to astrocytes in culture (Fig 3B).

We further asked whether AbOs would instigate oxidative stress in

primary hypothalamic neurons in culture, as previously shown in

hippocampal neurons (De Felice et al, 2007) and in a hypothalamic

cell line (Gomes et al, 2014). We found that AbOs induce a robust

increase in reactive oxygen species (ROS) levels in cultured hypo-

thalamic neurons (Fig 3C). Under the same conditions, the lactate

dehydrogenase cytotoxicity assay provided no evidence of cell death

induced by exposure to AbOs in culture (Fig 3D).

Because phosphorylation of eIF2a-P, one of the branches of the

unfolded protein response (UPR) activated upon ER stress, was

recently shown to underlie AbO toxicity in the hippocampus (Costa

et al, 2012; Lourenco et al, 2013; Ma et al, 2013), and hypothalamic

ER stress has been proposed to play an important role in the patho-

genesis of metabolic disorders (Ozcan et al, 2004, 2006; Hotamisli-

gil, 2010), we asked whether AbOs might trigger eIF2a-P in mature

cultured hypothalamic neurons. We found increased eIF2a-pSer51(eIF2a-P) in neuronal dendrites and cell bodies after exposure of

neurons to AbOs for 3 h (Fig 3E). Importantly, elevated eIF2a-Plevels were found independent of whether or not neurons exhibited

oligomers bound to their dendrites (Fig 3F). This indicates that

eIF2a phosphorylation is not triggered by direct binding of oligo-

mers to individual neurons, but rather is instigated by soluble

factors released to the medium upon exposure of cultures to AbOs.In a recent study, we found that pro-inflammatory TNF-a signaling

induced eIF2a-P in hippocampal neurons (Lourenco et al, 2013). To

determine whether TNF-a activation was involved in AbO-induced



193

eIF2a-P in hypothalamic neurons, we treated cultures with inflix-

imab, a TNF-a neutralizing monoclonal antibody. Infliximab attenu-

ated eIF2a-P triggered by AbOs (Fig 3E). It is noteworthy that

infliximab did not block oligomer binding to neurons (Fig 3G),

substantiating the notion that activation of TNF-a/eIF2a-P signaling

is independent of direct binding of AbOs to individual neurons and

is likely mediated by TNF-a secreted to the medium.

i.c.v. injection of AbOs induces increased hypothalamicinflammation and eIF2a-P in mice and macaques

We next asked whether i.c.v.-infused AbOs might trigger eIF2a-Pin the mouse hypothalamus. We found a significant increase in

hypothalamic levels of eIF2a-P 4 h after i.c.v. injection of AbOs

(Fig 4A), but not 7 days after oligomer injection (Fig 4B). We next

investigated levels of other components of the UPR 4 h after i.c.v.

injection of AbOs. Consistent with increased eIF2a-P, levels of

ATF4, a downstream effector of eIF2a, were increased in AbO-injected mice (Fig 4C). Other ER stress markers analyzed remained

unaltered, including PERKpThr980, ATF6, IRE1a-pSer724, spliced

Xbp1 and Grp78 (Supplementary Fig S3A–G). We note that we have

examined ER stress markers at a single time point (4 h post-AbOinjection) and future studies aimed to analyze in more detail the

time course of changes in levels of ER stress markers may provide

additional insight into the mechanisms by which AbOs instigate

hypothalamic deregulation.

In animal models of T2D and obesity, an inflammatory response

in the hypothalamus, notably via the activation of the IKKb/NF-jB

A

C

H I J

D E F G

K L M N O

B

Figure 2.



194

pathway, is an important part of the mechanism underlying patho-

genesis (Zhang et al, 2008; Thaler et al, 2012). Compared to vehicle-

injected mice, AbO-injected mice exhibited early activation of IKKbin the hypothalamus (Fig 4D, 4 h after i.c.v. injection), which

persisted for 7 days after i.c.v. injection of AbOs (Fig 4E). Once acti-

vated, IKKb phosphorylates IjBa, which undergoes ubiquitination

and proteasomal degradation, allowing NF-jB phosphorylation and

nuclear translocation. Accordingly, we found decreased levels of

IjBa (Fig 4F), a trend of increase in cytoplasmic NF-jB phosphory-

lation (Fig 4G), and significantly increased levels of NF-jB in the

nucleus (Fig 4H) in the hypothalamus of AbO-injected mice. On the

other hand, no differences in activated JNK or PKR levels were

detected in the hypothalamus of AbO-injected mice compared with

vehicle-injected mice 4 h or 7 days after i.c.v. injection of oligomers

(Supplementary Fig S4G–J).We further found that IRS-1pSer636 levels were increased and

IRS-1pTyr465 levels were decreased in the hypothalamus of mice

7 days after oligomer injection (Fig 4I and J), indicating that AbOsimpaired hypothalamic insulin signaling. To determine whether

AbO-induced insulin resistance in neuroendocrine brain regions

impaired the ability of the brain to respond to insulin signaling by

reducing food intake, mice were kept in metabolic cages for 7 days

following i.c.v. injection of AbOs (or vehicle) and food intake was

measured following an acute i.c.v. infusion of insulin (Schwartz

et al, 2000; Sanchez-Lasheras et al, 2010). Significantly, AbO-injected mice failed to exhibit the expected suppression in acute

food intake upon i.c.v. administration of insulin, indicating central

insulin resistance (Fig 4K).

To determine the impact of AbOs in an animal model with

greater proximity to humans, we have recently developed a non-

human primate model of AD by delivering i.c.v. infusions of oligo-

mers in adult cynomolgus macaques (Forny-Germano et al, 2014).

Our previous studies showed that this macaque model of AD

presents hippocampal IRS-1 pathology and elevated hipocampal

eIF2a-P levels (Bomfim et al, 2012; Lourenco et al, 2013). Three

macaques received i.c.v. injections of AbOs, while three sham-

operated animals were used as controls, and their hypothalami were

analyzed (Supplementary Fig S4). Strong AbO immunoreactivity

was found in the hypothalamus of oligomer-injected macaques, but

not in sham animals (Fig 5A). We next investigated whether similar

effects to those found in mice could be observed in AbO-injectedmacaques. We found significantly elevated hypothalamic levels of

eIF2a-P (Fig 5B) and pIKKb (Fig 5C), as well as a trend of decrease

in hypothalamic IjBa levels in AbO-injected macaques (Fig 5D).

Results indicate that abnormal inflammatory signaling and ER stress

are triggered by AbOs in the primate hypothalamus.

AbOs induce increased expression of orexigenic peptides andchow intake in mice

Intriguingly, AbO-injected mice presented increased chow intake

(Fig 6A), even though no significant differences in body weight

(Fig 6B) were found between experimental groups. Consistent with

increased chow ingestion, elevated hypothalamic expression of orexi-

genic neuropeptides AgRP and NPY (but no alterations in anorexigenic

POMC mRNA levels) was detected in AbO-injected mice (Fig 6C–E).To gain insight into how AbOs cause the observed peripheral meta-

bolic alterations, we asked whether AbO injection might lead to death

of hypothalamic cells. We carried out Fluorojade staining in brain

tissue from vehicle- or AbO-injected mice (7 days post-injection).

Results showed no evidence of cell degeneration in AbO-injected mice

compared to vehicle-injected animals (Fig 6F). We next performed

whole-cell patch-clamp recordings in brain slices from AbO-injectedmice to determine whether AbOs affected hypothalamic neuron elec-

trophysiology. We targeted cells from the arcuate nucleus, a region

enriched in NPY neurons (Allen Brain Atlas [http://mouse.brain-

map.org]; Hahn et al, 1998). No changes were detected in frequency

or amplitude of either excitatory or inhibitory post-synaptic currents,

or in resting membrane potential of the recorded neurons (Supplemen-

tary Fig S5A–E), suggesting that the mechanism by which AbOs inducefunctional deregulation of hypothalamic neurons does not include

major alterations in their electrophysiological properties.

Blockade of brain ER stress or inflammation attenuates glucoseintolerance and normalizes plasma noradrenaline levels in mice

Recent observations indicate that transient hypothalamic ER stress

is sufficient to deregulate peripheral insulin signaling and upregulate

◀ Figure 2. i.c.v-injected AbOs induce adipose tissue inflammation and insulin resistance in muscle.

A CD68 immunoreactivity in white adipose tissue (scale bar = 25 lm, images representative of one animal each from a total of four animals per experimental group).Arrow points to a region stained with CD68 antibody. *P = 0.0109, Student’s t-test.

B Epididymal fat mass was analyzed in mice (n = 6 animals/group) 7 days after i.c.v. injection of vehicle or AbOs. Data are representative of three independentexperiments with similar results. *P = 0.0255.

C–E Relative expression of leptin (C), TNF-a (D) and IL-6 (E), respectively, in white adipose tissue of mice (n = 7 Veh; 9 AbOs) 7 days after i.c.v. injection of vehicle orAbOs. In (C), *P = 0.0394; in (D), **P = 0.0038; in (E), *P = 0.0305; Student’s t-test.

F, G p-JNK (F; n = 5 animals/group) and IRS-1pSer312 (G; n = 6 animals/group) levels (normalized by total JNK and total IRS-1, respectively) in skeletal muscle of mice7 days after i.c.v. injection of vehicle or AbOs. In (F), *P = 0.0464; in (G), *P = 0.0081; Student’s t-test.

H Representative images of GLUT-4 immunofluorescence in insulin-stimulated skeletal muscle from mice that were i.c.v.-injected with vehicle (Veh) or 10 pmol AbOs.Bar graphs show quantification of GLUT-4 surface immunoreactivity in skeletal muscle of mice that received intraperitoneal injections of PBS or insulin (1 IU/kgbody weight) 7 days after i.c.v. injection of vehicle or AbOs, as indicated (n = 5 animals/group). Scale bar = 25 lm. *P = 0.0144, one-way ANOVA followed byBonferroni post hoc test.

I, J GLUT-4 mRNA (n = 4 animals/group) and total protein levels (normalized to actin levels; n = 5 Veh; 6 AbOs) were unchanged in skeletal muscle of Swiss miceinjected with vehicle (Veh) or 10 pmol AbOs.

K–O Plasma levels of insulin (K; n = 12 animals/group), leptin (L; n = 11 Veh; 12 AbOs), cholesterol (M; n = 8 Veh; 6 AbOs), triglycerides (N; n = 8 Veh; 6 AbOs) ornoradrenaline (O; n = 7 Veh; 8 AbOs) measured 7 days after i.c.v. injection of vehicle (Veh) or 10 pmol AbOs. In (O), *P = 0.0361, Student’s t-test.

Data information: Data are expressed as means ! SEM, and data are representative of two independent experiments with similar results. To assess statisticalsignificance, AbO-injected mice were compared to vehicle-injected mice.Source data are available online for this figure.



195

http://mouse.brain-map.org

http://mouse.brain-map.org

A

C D

E

F

G

B

Figure 3. AbOs bind to and impact hypothalamic neurons.

A Representative immunocytochemistry images of mature hypothalamic neurons in culture exposed to vehicle (Veh) or AbOs (500 nM) for 3 h. Binding of AbOs wasdetected using anti-oligomer monoclonal antibody NU4 (red). Neurons were double-labeled using MAP-2 antibody (green). Images represent typical results fromexperiments with three independent hypothalamic cultures (three coverslips/experimental condition per independent experiment). Scale bar = 30 and 10 lm formain panels and insets, respectively.

B Representative immunocytochemistry image of mature hypothalamic culture exposed to AbOs (500 nM) and immunolabeled with anti-GFAP (green) and NU4 (red)antibodies. Insets show AbOs binding to neuronal dendrites, whereas no binding was detected to GFAP-positive cells.

C Representative DCF fluorescence images from hypothalamic neuronal cultures exposed to vehicle or AbOs (500 nM, 4 h). Insets show optical zoom images of theindicated areas. Scale bars = 100 and 50 lm for main panels and insets, respectively. Graph shows integrated DCF fluorescence intensities (relative units; seeMaterials and Methods) (n = 3 independent hypothalamic cultures; three wells/experimental condition per experiment; three images acquired per well). Barsrepresent means ! SEM. #P = 0.0604, one-sample t-test compared with a fixed value of 100 RUs.

D LDH activity (IU/l) in culture media of hypothalamic cultures exposed to vehicle or AbOs (500 nM, 3 h).E Representative immunofluorescence images of eIF2a-P in hypothalamic cultures exposed to vehicle or AbOs (500 nM, 3 h) in the absence or presence of infliximab

(1 lg/ml). Scale bar = 30 lm. Graph represents integrated immunofluorescence intensities of eIF2a-P levels from three independent hypothalamic cultures (threecoverslips/experimental condition per experiment, 20 images per coverslip). Bars represent means ! SEM. *P = 0.0489, one-way ANOVA followed by Bonferroni posthoc test comparing AbO-treated versus vehicle-treated cultures.

F Representative images of hypothalamic cultures exposed to AbOs (500 nM, 3 h) and double-labeled with NU4 (oligomer-sensitive) and eIF2a-P antibodies. Arrowpoints to a neuron presenting high levels of eIF2a-P in the absence of AbO binding. Nuclear staining (DAPI) is shown in blue. Scale bar = 30 lm.

G Representative images of hypothalamic neurons labeled with NU4 antibody exposed to AbOs (500 nM, 3 h) in the absence or presence of infliximab (1 lg/ml). Similarpatterns of AbO binding were observed in both conditions. Scale bar = 20 lm.



196

peripheral sympathetic tonus (Purkayastha et al, 2011). Since we

found that AbOs induce transient hypothalamic eIF2a-P (Fig 4A and

B) and increased plasma noradrenaline levels in mice (Fig 2O), we

next investigated whether prevention of brain ER stress could atten-

uate AbO-induced defects in peripheral glucose homeostasis and in

plasma noradrenaline levels. We found that i.c.v. injections of

tauroursodeoxycholic acid (TUDCA), a chemical chaperone that

alleviates ER stress, prevented both the impairment in glucose toler-

ance and the increase in plasma noradrenaline levels induced by

i.c.v.-injected AbOs (Fig 7A and B). These results indicate that AbOsuse a central route to cause deregulation of peripheral glucose

homeostasis.

We recently reported that aberrant TNF-a signaling in the hippo-

campus mediates impaired neuronal insulin signaling, synapse dete-

rioration and memory loss in mice receiving i.c.v. infusions of AbOs(Bomfim et al, 2012; Lourenco et al, 2013). In addition, pioneering

studies have established that activation of pro-inflammatory TNF-ais a key mechanism leading to peripheral insulin resistance in diabe-

tes (Hotamisligil & Spiegelman, 1994; Hotamisligil et al, 1996;

Gregor & Hotamisligil, 2011) and that inhibition of hypothalamic

inflammation prevents peripheral insulin resistance (Milanski et al,

2012). Furthermore, our in vitro results indicated that TNF-a medi-

ates AbO-induced eIF2a-P (Fig 3E). Thus, we hypothesized that the

TNF-a pathway might be involved in AbO-induced deregulation of

glucose homeostasis in mice. To this end, we investigated the effects

of i.c.v.-injected AbOs in TNF-a receptor 1 knockout mice (Romanatto

et al, 2009). AbOs failed to induce glucose intolerance in TNFR1!/!

mice (Fig 7C). In metabolic disorders, ER stress has been linked to

insulin resistance and pro-inflammatory TNF-a signaling (Ozcan

et al, 2006; Steinberg et al, 2006). TNF-a signaling has further been

shown to activate intracellular stress kinases, including IKKb(Hotamisligil et al, 1996; Cai et al, 2005). i.c.v.-injected AbOstriggered IKKb activation and IRS-1pSer636 in the hypothalamus of

wild-type mice, but failed to do so in TNFR1!/! mice (Fig 7D and

E). Because whole-body TNF-a signaling would be expected to be

affected in TNFR1!/! mice, and to further investigate the specific

role of brain TNF-a signaling in deregulation of glucose metabolism,

we performed additional experiments in mice that were treated

i.c.v. with infliximab. We found that AbOs failed to trigger glucose

intolerance in mice that were previously treated with infliximab

(Supplementary Fig S6).

Our recent studies on the effects of oligomers in the hippocam-

pus indicate that, in addition to a direct effect on neurons, oligomers

also seem to impact microglial cells, the cellular components of the

innate immune system in the brain, to induce increased TNF-alevels and to deregulate hippocampal function (Lourenco et al,

2013). Therefore, we decided to test whether a similar indirect effect

of oligomers might lead to alterations in AgRP and NPY expressions

in the hypothalamus. To this end, we carried out experiments in

mice that had been treated intraperitoneally with minocycline, an

antibiotic known to prevent microglial activation and polarization

to an M1 proinflammatory profile. For reasons that are unclear to

us, minocycline treated-mice injected with vehicle showed increased

hypothalamic expression of AgRP and NPY (albeit not statistically

significant when compared to vehicle-injected mice) (Fig 7F and G).

Importantly, AbOs failed to induce increases in AgRP and NPY

levels in mice that had been treated with minocycline (Fig 7F and

G). This indicates that oligomers act on microglial cells, which likely

secret soluble factors (including TNF-a) to increase neuronal AgRP

and NPY expressions. Results thus indicate that a crosstalk between

neuronal and microglial cells is key to the effects of AbOs in the

hypothalamus. Finally, we tested whether infliximab treatment

would alleviate glucose intolerance in APP/PS1 mice. To this end,

we performed i.c.v. injections of infliximab in APP/PS1 mice and

found that infliximab rescued glucose intolerance in transgenic mice

(Fig 7H).

Discussion

Diabetes and AD are chronic degenerative diseases increasing in

prevalence in aging populations worldwide. Although clinical and

epidemiological studies have linked AD to diabetes, with each

disease increasing the risk of developing the other, the mechanisms

of pathogenesis connecting them at the molecular and cellular levels

remain to be elucidated. In particular, why AD patients present

increased probability of developing diabetes is unknown. Here, we

show that AbOs, toxins that accumulate in the AD brain and have

been linked to neuronal dysfunction in brain areas related to learn-

ing and memory, impact the hypothalamus of mice and macaques.

Intriguingly, infusion of AbOs in the brain triggers peripheral insulin

resistance in mice. Alterations in peripheral glucose homeostasis

were further detected in two transgenic mouse models of AD. These

results provide initial evidence implicating Ab oligomers in the

biological mechanisms underlying the clinical observations linking

AD to diabetes.

Numerous studies have investigated the impact of AbOs in

memory centers, specially the hippocampus (Ferreira & Klein,

2011), known to be fundamentally involved in the acquisition,

consolidation and recollection of new memories. This is because AD

is classically recognized as a disease of memory, and indeed

memory-related brain regions have long been known to be affected

in the course of disease (Walsh et al, 2002; Chhatwal & Sperling,

2012). However, early studies indicated that other brain regions, not

necessarily involved in learning and memory, might also be affected

in AD. For example, postmortem analysis of AD brains identified Abdeposits in the hypothalamus (Ogomori et al, 1989; Standaert et al,

1991), and evidence of peripheral glucose intolerance in AD patients

has been reported (Craft et al, 1992). More recently, voxel-based

morphometry analysis showed reduced hypothalamic volume and a

decreased number of orexin neurons in AD patients compared to

healthy controls (Loskutova et al, 2010; Fronczek et al, 2012).

Furthermore, hyperglycemia and hyperinsulinemia were shown to

positively correlate with the development of AD pathology (Matsuzaki

et al, 2010). In transgenic mouse models of AD, obesity-induced

insulin resistance is exacerbated (Takeda et al, 2010; Jimenez-

Palomares et al, 2012). Collectively, these observations raise the

intriguing possibility that the neuroendocrine axis, including the

hypothalamus, may be affected in AD. However, studies investigat-

ing the mechanisms underlying such clinical and postmortem obser-

vations are lacking. Using different experimental models, including

cell-based assays, mice and macaques that received i.c.v. injections

of AbOs, we now report that the hypothalamus is affected by AbOs.In both mice and macaques, i.c.v. infusion of AbOs induced

hypothalamic inflammation and eIF2a-P, recently implicated as

important pathogenic events in the onset of peripheral insulin



197

A B C D E

F

K

G H I J

Figure 4. AbOs induce hypothalamic inflammation, eIF2a phosphorylation and impaired insulin signaling.

A, B Western blot analysis of eIF2a-P levels in the hypothalamus of mice 4 h (A; n = 4 animals/group) or 7 days (B; n = 6 Veh; 5 AbOs) after a single i.c.v. injection ofvehicle (Veh) or 10 pmol AbOs. Graphs show densitometric data normalized by total eIF2a levels. *P = 0.0213.

C Western blot analysis of ATF4 levels in the hypothalamus of mice 4 h after i.c.v. injection of vehicle (Veh) or 10 pmol AbOs; graph shows densitometric datanormalized by b-actin (n = 7 Veh; 8 AbOs). #P = 0.0731; Student’s t-test.

D, E Western blot analysis of hypothalamic phospho-IKKb levels in the hypothalamus of mice 4 h (C; n = 6 animals/group) or 7 days (D; n = 4 Veh; 5 AbOs) after i.c.v.injection of vehicle or 10 pmol AbOs. Graphs show densitometric data normalized by total IKKb levels. In (D), *P = 0.0437; in (E), *P = 0.0444; Student’s t-test.

F, G Western blot analysis of IjBa (F; n = 6 animals/group) and cytoplasmic phospho-p65-NF-jB (G; n = 4 Veh; 5 AbOs) in the hypothalamus of mice 4 h after i.c.v.injection of vehicle or 10 pmol AbOs. Graphs show densitometric data normalized by actin (F) or total NF-jB levels (G). *P = 0.0207.

H Nuclear NF-jB levels in the hypothalamus 6 h after i.c.v. injection of vehicle or 10 pmol AbOs in mice. Graphs show NF-jB levels normalized by nuclear marker,lamin B (n = 6 animals/group). **P = 0.0024; Student’s t-test.

I, J IRS-1pSer636 (I; n = 4 animals/group) and pTyr465 (J; n = 6 Veh; 4 AbOs) levels in the hypothalamus 7 days after i.c.v. injection of vehicle or AbOs in mice. Graphsshow IRS-1pSer or IRS-1pTyr levels normalized by total IRS-1. In (I), *P = 0.0043; in (J), *P = 0.0275; Student’s t-test.

K Twelve-hour food intake after a single i.c.v. infusion of insulin (200 mU) in mice. Experiment was performed 7 days after i.c.v. injection of vehicle or AbOs (n = 5PBS; 5 Veh + Insulin; 9 AbOs + Insulin), data are representative of two independent experiments with similar results. ***P < 0.0001, one-way ANOVA followed byBonferroni post hoc test comparing Veh-Insulin versus PBS groups.

Data information: Data are expressed as means ! SEM. In (A–J), to assess statistical significance, AbO-injected mice were compared to vehicle-injected mice.Source data are available online for this figure.



198

resistance in metabolic disorders (Zhang et al, 2008; Denis et al,

2010; Arruda et al, 2011; Thaler et al, 2012). Interestingly, while in

mice we found a transient increase in hypothalamic eIF2a-P levels

following a single i.c.v. injection of oligomers, persistently elevated

eIF2a-P levels were found in macaques after a series of AbO injec-

tions. This suggests that persistently elevated oligomer levels in the

brain may induce prolonged effects in eIF2a-P. AbOs further

induced IRS-1 inhibition (IRS-1pSer) in the hypothalamus of mice. It

is noteworthy that oligomers failed to trigger both hypothalamic

IKKb activation and IRS-1pSer in TNFR1!/! mice. Results thus indi-

cate that AbO-induced TNF-a/pIKK deregulation is directly linked to

disrupted insulin signaling in the hypothalamus.

Activated JNK and PKR were recently implicated in AbO-induceddefective hippocampal insulin signaling (Bomfim et al, 2012; Lourenco

et al, 2013). However, at the time points investigated (4 h or 7 days

post-AbO injection), no differences in pJNK and pPKR levels were

detected in the hypothalamus of AbO-injected mice compared to

vehicle-injected mice. Further, no changes were detected in other

markers of ER stress (4 h post-AbO injection), including phospho-

PERK, IRE1a-pSer724, ATF6 and Grp78. We note that future studies

aimed to analyze in more detail the time course of changes in levels

of ER stress markers may provide additional insight into the mecha-

nisms by which AbOs instigate hypothalamic deregulation.

Transient hypothalamic ER stress has been shown to induce

increased plasma levels of noradrenaline in mice, and this was

reported to be sufficient to induce peripheral glucose intolerance in

mice (Purkayastha et al, 2011). Consistent with that interesting

study, we found that prevention of brain ER stress by i.c.v. adminis-

tration of TUDCA normalized plasma noradrenaline levels and

blocked AbO-induced peripheral glucose intolerance. Moreover,

AbOs failed to induce glucose intolerance, hypothalamic IKKbactivation and IRS-1 inhibition in TNFR1!/! mice, or glucose

intolerance in mice treated i.c.v. with infliximab. These results

suggest that brain ER stress and inflammation underlie alterations in

peripheral glucose homeostasis induced by AbOs, and indicate that

AbOs hijack key signaling pathways in the CNS to deregulate

peripheral glucose handling.

We recently demonstrated that i.c.v.-injected AbOs disrupt insu-

lin signaling and induce inflammation in the hippocampus of mice

and macaques (Bomfim et al, 2012; Ledo et al, 2013; Lourenco

et al, 2013). AbO-induced abnormal hippocampal TNF-a signaling

was found to be directly linked to synapse deterioration and cogni-

tive impairment (Lourenco et al, 2013). It is thus possible that, in

the AD brain, progressive accumulation of Ab oligomers (due to

elevated Ab production or reduced clearance) brings about different

functional outcomes in different brain regions. While the impact of

A

C

B

D

Figure 5. AbOs accumulate in the hypothalamus of macaques and induce inflammation and eIF2a phosphorylation.

A Representative images of AbO immunoreactivity (detected using anti-oligomer monoclonal antibody NU4) in the hypothalamus of control, sham-operated adultcynomolgus macaques (Sh; n = 3) or macaques that received i.c.v. injections of AbOs (n = 3; see Materials and Methods). Nuclear staining (DAPI) is shown in blue.Insets show optical zoom images of the areas indicated by white dashed rectangles in the main panels. Scale bars = 100 and 20 lm for main panels and insets,respectively.

B–D Representative images showing eIF2a-P (B), phospho-IKKb (C) and IjBa (D) immunoreactivities in the hypothalamus of cynomolgus macaques that received i.c.v.injections of AbOs or control (sham-operated; Sh) macaques (n = 3 animals/group). Graphs show immunolabeling optical density analysis from three imagesacquired in the hypothalamus of each macaque (three control versus three AbO-injected animals). In (B), #P = 0.0523; in (C) #P = 0.1123; unpaired Student’s t-testwith Welch’s correction for unequal variances; AbO-injected monkeys compared to sham-operated monkeys. Scale bars = 50 lm



199

AbOs in the hippocampus involves inflammation, ER stress and

synapse deterioration, leading to memory deficits, AbO-inducedinflammation and eIF2a-P in the hypothalamus may be especially

relevant in terms of disrupting hypothalamic insulin signaling. The

hypothalamus is well known for its ability to respond to changes in

circulating insulin levels by regulating food ingestion (Sanchez-

Lasheras et al, 2010). We found that an acute i.c.v. injection of insu-

lin failed to suppress short-term food ingestion in AbO-injectedmice, suggesting that AbOs rendered the hypothalamus resistant to

insulin. Remarkably, activation of a hypothalamic inflammatory

pathway similar to the pathway we report in our model has been

implicated as a central mechanism regulating energy imbalance in

obese mice, and its suppression has been proposed to represent a

potential strategy to combat obesity-related diseases (Zhang et al,

2008). These findings further indicate that AbO- and obesity-

induced hypothalamic inflammation share common pathogenic

pathways.

Current findings indicate that AgRP and NPY levels remained

unaltered in AbO-injected mice that had been treated with minocy-

cline, suggesting that oligomers impact microglial cells, the cellular

components of the innate immune system in the brain, likely induc-

ing secretion of soluble factors (including TNF-a) to increase neuro-

nal AgRP and NPY expressions. Furthermore, it is noteworthy that

eIF2a-P does not depend on direct binding of oligomers to individ-

ual neurons, as elevated eIF2a-P levels were detected in neurons

regardless of whether or not they had oligomers bound to their

A

F

B C D E

Figure 6. i.c.v.-injected AbOs induce increased food intake, hypothalamic expression of orexigenic neuropeptides but no hypothalamic cell degeneration.

A Accumulated chow intake (normalized by body weight) measured during 7 days following a single i.c.v. injection of vehicle or 10 pmol AbOs in mice (n = 13 Veh;10 AbOs; data are representative of two independent experiments with similar results). ***P < 0.0001; Student’s t-test.

B Daily body weight measured during 7 days after i.c.v. injection of vehicle or AbOs (n = 7 animals/group; data are representative of two independent experimentswith similar results).

C–E Adult Swiss mice received a single i.c.v. injection of vehicle or 10 pmol AbOs, and hypothalamic levels of mRNA for AgRP (C; n = 6 Veh; 5 AbOs), NPY (D; n = 6 Veh;5 AbOs) and POMC (E; n = 7 animals/group) were analyzed 7 days after injection. In (C), *P = 0.0191; in (D), *P = 0.0115; Student’s t-test.

F Swiss mice received a single i.c.v. injection of vehicle (Veh) or 10 pmol AbOs, and their brains were analyzed by Fluorojade staining of degenerating cells 7 daysafter the injection. Representative images of Fluorojade staining in the hypothalamus of vehicle- or AbO-injected mice (n = 4/group). Scale bar = 100 lm in leftpanels (top and bottom) and 20 lm in right panels (top and bottom). Positive control (bottom left panel) was the hippocampus of a mouse that received one i.c.v.injection of quinolinic acid (36.8 nmol) and was analyzed 24 h after.

Data information: Data are expressed as means ! SEM. To assess statistical significance, AbO-injected mice were compared to vehicle-injected mice.



200

A

C

F G H

D E

B

Figure 7. AbO-induced peripheral glucose intolerance and hypothalamic insulin resistance are mediated by TNF-a signaling and hypothalamic ER stress.

A Glucose tolerance test (2 g glucose/kg body weight, i.p.) in mice that received i.c.v. injections of vehicle, vehicle + TUDCA, AbOs or AbOs + TUDCA (when used,TUDCA was administered in 5 i.c.v. injections of 5 lg TUDCA each, before and after oligomer injection; see Materials and Methods. Control groups received injectionsof saline). Glucose tolerance test (GTT) was performed 7 days after i.c.v. injection of vehicle or AbOs. Bar graph represents areas under the curves (AUC) in the timecourse plots (n = 15 Veh; 15 AbOs; 10 Veh + TUDCA; 16 AbOs + TUDCA). Data are representative of two independent experiments with similar results. Left panel:**P = 0.0048, **P = 0.003, two-way ANOVA followed by Bonferroni post hoc test; right panel: *P = 0.0384, one-way ANOVA followed by Bonferroni post hoc test.

B Plasma noradrenaline (NA) levels measured 7 days after i.c.v. injection of vehicle, vehicle + TUDCA, AbOs or AbOs + TUDCA in mice (n = 7 animals/group). Data arerepresentative of two independent experiments with similar results. *P = 0.0071, one-way ANOVA followed by Bonferroni post hoc test.

C Glucose tolerance test (2 g glucose/kg body weight, i.p.) in TNFR1!/! mice or wild-type littermates performed 7 days after i.c.v. injection of vehicle or AbOs. Bargraph represents areas under the curves (AUC) in the time course plots (n = 8 WT + Veh; 7 WT + AbOs; 7 TNFR!/! + Veh; 8 TNFR!/! + AbOs). Left panel:**P = 0.0049, ***P < 0.0001, two-way ANOVA followed by Bonferroni post hoc test; right panel: *P = 0.0001, one-way ANOVA followed by Bonferroni post hoc test.

D, E Western blot analysis of phospho-IKKb (D; n = 6 WT + Veh; 6 WT + AbOs; 4 TNFR!/! + Veh; 6 TNFR!/! + AbOs) and IRS-1pSer636 levels (E; n = 6 WT + Veh; 5WT + AbOs; 6 TNFR!/! + Veh; 5 TNFR!/! + AbOs) in the hypothalamus of wild-type (WT) or TNFR1!/! mice 10 days after i.c.v. injection of vehicle or AbOs.Representative images from Western blot experiments were always run on the same gels but represent noncontiguous lanes. In (D), *P = 0.0088, Student’s t-test;in (E), *P = 0.0428, one-way ANOVA followed by Bonferroni post hoc test.

F, G Adult Swiss mice pre-treated with minocycline or PBS received a single i.c.v. injection of vehicle or 10 pmol AbOs, and hypothalamic levels of mRNA for AgRP(F; n = 5 Veh; 6 AbOs; 5 Veh + Mino; 4 AbOs + Mino) and NPY (G; n = 14 Veh; 13 AbOs; 8 Veh + Mino; 9 AbOs + Mino) were analyzed 7 days after injection. In (F),*P = 0.0097, one-way ANOVA followed by Bonferroni post hoc test; in (G), *P = 0.0219, one-way ANOVA followed by Bonferroni post hoc test.

H Glucose tolerance test (GTT) in APP/PS1 mice before and after i.c.v. injection of infliximab (0.2 lg daily for 4 days). Bar graph represents areas under the curves(AUC) in the time course plots (n = 9 animals/group). Left panel: *P = 0.0177, two-way ANOVA followed by Bonferroni post hoc test; right panel: *P = 0.0327, pairedt-test.

Data information: Data are expressed as means " SEM. To assess statistical significance, AbO-injected mice were compared to vehicle-injected mice.Source data are available online for this figure.



201

dendrites. Therefore, AbOs do not seem to act directly on neurons

to induce phosphorylation of eIF2a. Rather, it is likely that a cros-

stalk between neurons and microglia leads to elevated levels of

TNF-a, causing activation of neuronal TNF-a/eIF2a signaling to

deregulate hypothalamic function. We note that similar observa-

tions were made in studies of the effects of AbOs on hippocampal

cells (Lourenco et al, 2013).

We showed that no alterations in peripheral glucose homeostasis

were detected 12 h after an i.c.v. injection of AbOs (Supplementary

Fig S2C), but markers of hypothalamic inflammation were found to

be elevated as soon as 4 h after AbO infusion. This supports the

notion that hypothalamic inflammation precedes and may lead to

peripheral metabolic alterations, a possibility that deserves further

investigation. In this regard, an interesting recent study reported

that, unlike inflammation in peripheral tissues, which develops as a

consequence of obesity, hypothalamic inflammatory signaling is

evident in rats within 1 to 3 days of feeding on a high-fat diet, prior

to substantial weight gain (Thaler et al, 2012) and implicating

hypothalamic inflammation in obesity pathogenesis (Thaler et al,

2013). We further note that AbO-induced deregulation of peripheral

glucose homeostasis is similar in magnitude to the deregulation

induced by a short period (7 days) of high-fat diet. Extending the

findings of a recent study using APP/PS1 mice (Zhang et al, 2012),

we found altered peripheral glucose homeostasis both in APP/PS1

mice and in 3xTg-AD mice, two different experimental models of AD.

Importantly, we further demonstrated that i.c.v. injections of

infliximab rescued glucose tolerance in APP/PS1 mice, establishing

that brain inflammation triggers alterations in peripheral glucose

homeostasis in AbO-injected mice and in the APP/PS1 mouse model

of AD. Intracerebroventricular infusion of infliximab in AD trans-

genic mice has been reported to reduce the number of amyloid

plaques and phospho-tau levels (Shi et al, 2011a). Intrathecal

administration of infliximab was further reported to improve cogni-

tion in one patient with AD (Shi et al, 2011b), and clinical trials are

currently investigating the efficacy of infliximab in a wide range of

pathologies, including major depression, obesity-associated insulin

resistance and diabetic complications, among others (US National

Institute of Health; http://clinicaltrials.gov/). However, infliximab

does not cross the blood–brain barrier, and so far, it is important to

note that anti-TNF-a strategies for AD require invasive forms of

central administration, making this a difficult strategy to treat AD.

Nevertheless, our results suggest that pharmacological or other

approaches to prevent neuroendocrine dysfunction may provide

novel therapeutics for metabolic deregulation in AD.

Our results demonstrate that brain accumulation of AbOs affects

the hypothalamus and impacts peripheral metabolism by mecha-

nisms similar to those underlying peripheral insulin resistance in

type 2 diabetes and other metabolic diseases. Similar to what has

been described in metabolic disorders (Rossmeisl et al, 2003; Thaler

et al, 2013), i.c.v.-injected AbOs induce adipose tissue inflammation

and impaired insulin-induced surface translocation of GLUT-4 in

muscle cells. A previous study reported that a very high concentra-

tion of Ab (10 lM) induced hepatic insulin resistance in vitro

through a direct effect on hepatocytes (Zhang et al, 2012). However,

in our experimental conditions, AbOs failed to cause alterations in

peripheral glucose homeostasis when delivered via the caudal vein

or by intraperitoneal injection in mice, ruling out a direct effect of

AbOs on peripheral tissues. It is important to note that, besides the

hypothalamus, other brain regions involved in neuroendocrine

control might be also affected by AbOs. Whether AbOs indeed affect

other brain regions responsible for the control of peripheral glucose

homeostasis warrants further exploration.

In conclusion, our findings establish that i.c.v.-injected AbOstrigger inflammation in the hypothalamus and cause peripheral

glucose intolerance and insulin resistance. Results support the

emerging notion that pathological hypothalamic inflammation/ER

stress leads to impaired peripheral glucose homeostasis. We propose

that the impact of AbOs on the hypothalamus comprises a key novel

pathological mechanism that disrupts metabolic homeostasis and

leads to insulin resistance, revealing an important crosstalk between

central and peripheral pathogenic mechanisms in AD. Our discovery

that AbOs instigate hypothalamic deregulation draws attention to a

brain structure that has been largely ignored to date in the study of

AD pathogenesis, and highlights the importance of recognizing AD

as a disease of both the brain and the periphery. As peripheral insu-

lin resistance has been implicated in the development of AD (Janson

et al, 2004; De Felice, 2013), current results suggest the existence of

a vicious cycle, instigated by brain accumulation of AbOs, contribut-ing to the development of both AD and metabolic disorders, includ-

ing type 2 diabetes.

Materials and Methods

Preparation of Ab oligomers

Oligomers were prepared from synthetic Ab1–42 peptide (American

Peptide, Sunnyvale, CA) as originally described Lambert et al

(1998). The peptide was solubilized in hexafluoroisopropanol

(HFIP) and the solvent was evaporated to produce dried films,

which were subsequently dissolved in sterile anhydrous dimethyl-

sulfoxide to make a 5 mM solution. This solution was diluted to

100 lM in ice-cold PBS and incubated overnight at 4°C. The prepa-

ration was centrifuged at 14,000 g for 10 min at 4°C to remove

insoluble aggregates (protofibrils and fibrils), and the supernatants

containing soluble Ab oligomers were stored at 4°C. Protein concen-

tration was determined using the BCA kit (Pierce, Deerfield, IL).

Routine characterization of preparations was performed by size-

exclusion chromatography and Western blotting using anti-Ab 6E10

(Abcam, Cambridge, MA) or anti-Ab oligomer NU1 (Lambert et al,

2007) monoclonal antibodies and, occasionally, by transmission

electron microscopy, as previously described (Jurgensen et al, 2011;

Sebollela et al, 2012; Figueiredo et al, 2013). Oligomers were used

within 48 h of preparation.

Mature hypothalamic neuronal cultures, immunocytochemistry,ROS and LDH release assays

Primary hypothalamic neuronal cultures were prepared from rat

embryos (E16) according to the procedures established for hippo-

campal neuronal cultures (De Felice et al, 2007, 2009). Cultures

were plated at a density of 70,000 cells/cm2 on poly-L-lysine-coated

coverslips and were maintained in neurobasal medium with B27

supplement and L-glutamine (0.5 mM). After 14 days in vitro,

cultures were incubated with vehicle or 500 nM AbOs for 3 h at

37°C. Infliximab was added 30 min prior to AbOs. For experiments



202

http://clinicaltrials.gov/

designed to determine reactive oxygen species (ROS) formation,

20,000 cells/cm2 were plated directly on poly-L-lysine-coated wells

of 96-well plates. After 18–21 days in vitro, cultures were incubated

for 4 h at 37°C with vehicle or 500 nM AbOs. ROS formation was

assessed using 2 lM of the fluorescent probe CM-H2DCFDA (Invitro-

gen, Carlsbad, CA), as previously described in De Felice et al (2007).

CM-H2DCFDA is sensitive to the formation of various types of ROS,

including peroxide, hydroxyl radical, peroxyl radicals and peroxyni-

trite. After 30 min of loading with the fluorescent probe, neurons

were rinsed three times with warm PBS and two times with neuro-

basal medium without phenol red. Cells were immediately imaged

on a Nikon Eclipse TE 300-U fluorescence microscope. At least three

experiments with independent neuronal cultures were performed,

each with triplicate well per experimental condition. Three images

were acquired from randomly selected fields per well. Results

obtained in independent experiments were combined to allow quan-

titative estimates of changes in neuronal ROS levels. Quantitative

analysis of immunofluorescence data was carried using ImageJ

(Windows version) using appropriate thresholding to eliminate

background signal before histogram analysis, as described by De

Felice et al (2007).

Immunocytochemistry was performed as previously described by

De Felice et al (2009). Briefly, hypothalamic cultures were treated

for 3 h at 37°C with 500 nM AbOs or equivalent volumes of vehicle

and were fixed for 10 min with 4% paraformaldehyde containing

4% sucrose in PBS. Cells were blocked for 1 h with 10% normal

goat serum in PBS and incubated at 4°C with monoclonal AbO-selective NU4 antibody (1:2,000; (Lambert et al, 2007)) overnight.

Neurons were rinsed three times with PBS, permeabilized with

0.1% Triton X-100 for 5 min and incubated overnight at 4°Cwith anti-MAP2 (Santa Cruz Biotechnology, Santa Cruz, CA;

1:200, Cat#sc20172), anti-GFAP (DAKO, Carpinteria, CA; 1:200,

Cat#Z-0334) or anti-phospho-eIF2a (Enzo Life Sciences, Farming-

dale, NY; 1:200, Cat#BML-SA405) antibodies. After rinsing, neurons

were incubated for 2 h at room temperature with Alexa Fluor-555

anti-mouse IgG and Alexa Fluor-488 anti-rabbit IgG (1:2,000). After

washing, cells were mounted on coverslips using Prolong Gold

Antifade with DAPI (Invitrogen) and were imaged on a Zeiss Axio

Observer Z1 Microscope equipped with an Apotome module.

Measurement of lactate dehydrogenase (LDH) released to the

medium was assessed as a cell death indicator. LDH was assayed by

a commercial kit (Doles, Goiania, Brazil) according to manufac-

turer’s instructions. Briefly, culture medium was collected after

exposure to AbOs (or vehicle) and LDH activity was measured.

Absorbance was measured at 510 nm.

Animals and intracerebroventricular (i.c.v.) injections

Male Swiss mice obtained from our own animal facility were 2.5–3months old at the beginning of experiments. TNFR1!/! female mice

in a C57/BL6 background and wild-type littermates were obtained

from the University of Campinas Breeding Centre (CEMIB).

Six-month-old triple-transgenic (3xTg-AD) male mice and wild-type

littermates were obtained from University of California Irvine (Xu

et al, 2003). Nine- to thirteen-month-old APP/PS1 (seven males and

two females) and littermate wild-type mice (six males and three

females) were obtained from our own breeding facilities. Animals

were housed in groups of five in each cage with free access to food

and water, under a 12-h light/dark cycle, with controlled room

temperature and humidity. Animals were randomly assigned to

different experimental groups, and researchers conducting the

experiments were blind to experimental condition. All procedures

were performed in the light phase and followed the ‘Principles of

Laboratory Animal Care’ (US National Institutes of Health) and were

approved by the Institutional Animal Care and Use Committee of

the Federal University of Rio de Janeiro (protocol IBqM 072-05/16)

and UCI Institutional Animal Care and Use Committee. For i.c.v.

injection of AbOs, animals were anesthetized for 7 min with 2.5%

isoflurane (Cristalia, Sao Paulo, Brazil) using a vaporizer system

(Norwell, MA) and were gently restrained only during the injection

procedure itself, as recently described in Figueiredo et al (2013). A

2.5-mm-long needle was unilaterally inserted 1 mm to the right of

the midline point equidistant from each eye and 1 mm posterior to a

line drawn through the anterior base of the eye (Laursen & Belknap,

1986; Figueiredo et al, 2011, 2013); see Supplementary Fig S1). Ten

pmol of AbOs (concentration expressed in terms of Ab monomers)

or vehicle was injected in 30 s, in a total volume of 3 ll for Swiss

mice. When C57/BL6 mice were used, 100 pmol of AbOs or vehiclewas injected in 30 s in a total volume of 1 ll. Injection of 3 ll of ablue dye into the lateral ventricle of Swiss mice was performed to

verify diffusion along the CSF circulation so as to reach the whole

brain (Supplementary Fig S1). At the end of experiments, injection

of blue dye in the same injection site used for AbOs or vehicle was

employed to verify the accuracy of injection into the lateral ventri-

cle. Mice showing any signs of misplaced injections or brain hemor-

rhage (~5% of animals throughout our study) were excluded from

further analysis.

In experiments using macaques, six female cynomolgus maca-

ques (Macaca fascicularis; body weights 4.7–7.0 kg) were used.

Macaques were maintained at the Centre for Neuroscience at

Queen’s University (Kingston, Canada) under the close supervision

of a laboratory animal technician and the Institute veterinarian. All

animals had a cannula implanted in the lateral ventricle by aseptic

surgery. Anesthesia was induced by ketamine (10 mg/kg, intramus-

cular). During surgery, glycopyrrolate (0.013 mg/kg) and isoflurane

(1–3%) were also used. Correct placement of the cannula was

assessed by MRI. After a recovery period, three macaques received

intracerebroventricular injections of 100 lg of AbOs (one injection

per day every 3 days for 24 days). Three sham-operated macaques

were used as controls. Upon completion of the experimental

protocol, macaques were sedated with intramuscular ketamine

(10 mg/kg) plus buprenorphine (0.01 mg/kg) for analgesia, followed

by intravenous sodium pentobarbital (25 mg/kg), perfused with

phosphate-buffered saline (PBS) followed by 4% paraformaldehyde

in PBS; 4% paraformaldehyde in PBS containing 2.5% glycerol;

PBS + 5% glycerol; and PBS + 10% glycerol. All procedures were

approved by the Queen’s University Animal Care Committee and

were in full compliance with the Canada Council on Animal Care

(Animal Care Protocol Original Munoz-2011-039-Or).

Immunohistochemistry in macaque brain sections

Immunohistochemistry was performed using free-floating serial 40-

lm-thick coronal sections in PBS containing 1% Triton X-100 incu-

bated with 0.1 M citrate buffer, pH 6, at 60°C for 5 min. Endogenous

peroxidase was inactivated by incubation of sections with 3%



203

hydrogen peroxide in methanol for 2 h. Sections were then blocked

with 5% bovine serum albumin (BSA) and 5% normal goat serum

(NGS) in 1% Triton X-100 for 3 h at room temperature. Primary anti-

bodies against phospho-eIF2a (Enzo Life Sciences; 1:200, Cat#BML-

SA405), phospho-IKKb (Abcam; 1:200, Cat#ab59195) and IjBa (Cell

Signaling; 1:200, Cat#9242) were diluted in blocking solution, and

sections were incubated at 4°C for 16 h, followed by incubation with

biotinylated secondary antibody for 2 h at room temperature, and

then processed using the Vectastain Elite ABC reagent (Vector Labo-

ratories) according to manufacturer’s instructions. The sections were

washed in PBS and developed using DAB in chromogen solution,

and counterstained with Harris’ hematoxylin. Slides were mounted

with Entellan (Merck) and imaged on a Zeiss Axio Observer Z1

microscope. Omission of primary antibody was routinely used to

certify the absence of nonspecific labeling (data not shown). For

immunofluorescence analysis, tissue autofluorescence was

quenched by incubation with 0.06% potassium permanganate for

10 min at room temperature. Sections were blocked in 5% bovine

serum albumin (BSA) and 5% normal goat serum (NGS) in 1%

Triton X-100 for 3 h at room temperature. Primary antibody against

AbOs (NU4; 1:300; (Lambert et al, 2007)) was diluted in blocking

solution, and sections were incubated at 4°C for 16 h, followed by

incubation with Alexa-555-conjugated anti-mouse secondary anti-

body (1:1,500) for 2 h at room temperature. Slides were mounted

with Prolong Gold Antifade with DAPI (Invitrogen) and imaged on a

Zeiss Axio Observer Z1 microscope equipped with an Apotome

module to minimize out-of-focus light.

Immunohistochemistry in mouse tissues

For GLUT-4 immunohistochemistry, mice received one i.c.v. injec-

tion of vehicle or 10 pmol AbOs. Seven days later, mice received

one i.p. injection of either PBS or insulin (1 IU/kg body weight)

and were killed by decapitation 15 min later. The soleus muscle

was dissected and fixed in 4% paraformaldehyde. After 48 h,

tissues were cryoprotected in sucrose (20-30%) and 20 lmsections were obtained in a cryostat (Leica CM1850). Sections

were fixed with acetone for 30 min, washed twice with PBS and

incubated for 1 h with rabbit polyclonal anti-GLUT-4 antibody

(Abcam; 1:500, ab-654). Sections were then incubated with Alexa-

555-conjugated anti-rabbit antibody (1:1,000; Invitrogen) for 1 h

and mounted in Prolong Gold Antifade with DAPI (Invitrogen).

Sections were imaged on a Zeiss Axio Observer Z1 microscope

equipped with an Apotome module. Eight images were acquired

per section, and integrated immunofluorescence intensity was

determined using ImageJ software (Windows version). For adipose

tissue immunohistochemistry, mice i.c.v. injected with vehicle or

AbOs were killed 7 days after injection and samples of epididymal

adipose tissue were removed and fixed in 4% paraformaldehyde.

After 48 h, tissues were included into paraffin blocks, and 3 lmsections were obtained using a microtome and mounted in slides.

For immunohistochemistry, slides were immersed in xylene for

10 min, sequentially rehydrated in absolute, 95 and 70% ethanol

in water, and incubated with 3% H2O2 in methanol for inactiva-

tion of endogenous peroxidase. Antigens were reactivated by the

treatment with 0.01 M citrate buffer for 40 min at 95°C. Slides

were washed in PBS and incubated with CD68 antibody (Abcam;

1:200, Cat#ab125212) for 12–16 h at 2–8°C. After washing with

PBS, slides were incubated with biotinylated secondary antibody

for 1 h, washed twice with PBS and incubated with streptavidin-biotin peroxidase for 30 min. Slides were then covered with 3,30

diaminobenzidine solution (0.06% DAB in PBS containing 2%

DMSO and 0.018% H2O2) for 1 to 5 min or until a brown precipi-

tate could be observed. Identical conditions and reaction times

were used for slides from different animals to allow comparison

between immunoreactivity densities. Reaction was stopped by

immersion of slides in distilled water. Counterstaining was

performed with Harris’ hematoxylin. Four images were randomly

acquired for each animal using a Zeiss Axio Observer Z1 micro-

scope. An optical density threshold that best discriminated staining

from background was obtained using NIH ImageJ 1.36b imaging

software (NIH, Bethesda, MD).

Fluorojade (FJ) histochemistry was used as indicative of neuro-

nal degeneration. The paraffin-embedded brain tissue sections

were immersed into 100% ethanol for 3 min, then into 70% ethanol

for 1 min and distilled water for 1 min. Slices were then immersed

into 0.06% potassium permanganate solution for 10 min to

suppress endogenous background signal, and slices were washed

with distilled water for 1 min. Fluorojade B staining solution (10 ml

of 0.01% Fluorojade B aqueous solution added to 90 ml of 0.1%

acetic acid in distilled water) was added and slices were stained for

30 min. After staining, sections were rinsed three times with

distilled water. Excess water was drained off, and the slides were

cover-slipped with dibutylphthalate in xylene (D.P.X.) mounting

media (Aldrich Chem. Co., Milwaukee, WI). Sections comprising

the arcuate nucleus (Arc) and ventromedial hypothalamus (VMH)

were examined on epifluorescence microscopes (Olympus Bx41 or

Nikon Eclipse 50i). Positive staining controls consisted of sections

from the hippocampus of a mouse i.c.v. injected with 36.8 nmol

quinolinic acid and killed 24 h thereafter.

Intraperitoneal glucose tolerance test (GTT)

Mice were fasted for 12 h and blood samples were collected from a

tail incision. After collection of a baseline sample, mice received an

i.p. injection of glucose (2 g/kg body weight). Blood glucose

measurements were repeated at 15, 30, 45, 60 and 120 min after

glucose injection, using a One-Touch Ultra Glucose Meter and strips

(Johnson & Johnson). An additional measurement of blood glucose

levels (180 min after glucose injection) was performed in experi-

ments using 3xTg-AD mice. Mice with fasting glucose levels lower

than 50 mg/dl or higher than 100 mg/dl, or whose plasma glucose

levels did not increase at any time point after glucose injection were

excluded from the study.

Intraperitoneal insulin tolerance test (ITT)

Mice were fasted for 5 h and blood samples were collected from a

tail incision. After collection of a baseline sample, mice received an

i.p. injection of insulin (1 IU/kg body weight). Blood glucose

measurements were repeated at 15, 30, 45 and 60 min after insulin

injection, using a One-Touch Ultra Glucose Meter and strips (Johnson

& Johnson). If blood glucose levels fell below 20 mg/dl, mice were

immediately given an i.p. injection of glucose and were excluded

from the experiment. Kitt was calculated as described by Ropelle

et al (2010).



204

Plasma insulin and leptin measurements

Mice were i.c.v.-injected with vehicle or AbOs and 7 days later were

fasted for 3 h before being deeply anesthetized with 100 mg/kg

ketamine and 10 mg/kg xylazine. After complete loss of reflex,

blood samples were collected in EDTA-containing tubes and kept on

ice until plasma separation (centrifugation at 3,000 g for 10 min at

4°C). Samples were kept at 4°C, and insulin detection and leptin

detection were performed using the Ultra Sensitive Mouse Insulin

ELISA kit and Mouse Leptin ELISA kit (both from Crystal Chem Inc,

Downers Grove, IL).

High-fat diet

Mice were maintained for 7 days on normal chow or a high-fat diet

containing 55% of energy derived from fat, 29% from carbohy-

drates and 16% from protein, prepared as described (Romanatto

et al, 2009; Ropelle et al, 2010).

Treatment with tauroursodeoxycholic acid (TUDCA)

Mice received 5 lg TUDCA i.c.v. per injection. Injections were

carried out 20 min prior to AbO injection, and at 2, 24 and 96 h

thereafter. An extra TUDCA injection was given 12 h before the

GTT, which was performed 7 days after AbO administration.

Twenty-four hours after the GTT, mice were deeply anesthetized

with 100 mg/kg ketamine and 10 mg/kg xylazine, and blood

samples were collected by cardiac puncture in heparinized tubes.

Plasma was separated by centrifugation at 3,000 g at 4°C for

10 min, and samples were used for noradrenaline quantification (as

described below).

Treatment with infliximab

Swiss mice were given a single i.c.v. injection of 2 ll of a 0.1 lg/llsolution of Infliximab 20 min prior to AbOs. In APP/PS1 mice, daily

i.c.v. injections of 1 ll of a 0.2 lg/ll solution of Infliximab were

administered for 4 days. The last injection was performed 12 h prior

to the glucose tolerance test.

Minocycline treatment

Swiss mice received daily i.p. injections of minocycline (25 mg/kg)

for 3 days prior to i.c.v. injection of AbOs. Control mice received

i.p. injections of PBS.

Intracaudal injections

Animals were anesthetized with halothane and aseptically injected

via the tail vein with 10 pmol AbOs or Dulbecco’s PBS, in a final

injection volume of 100 ll.

Determination of accumulated food intake andintracerebroventricular insulin injection

Swiss mice were submitted to stereotaxic surgery for implantation

of a cannula directed to the third ventricle, as described in Ropelle

et al (2010). Mice were allowed to recover from surgery in their

home cages for 4 days before being placed in individual metabolic

cages. Animals then received i.c.v. injections of vehicle or AbOs,and food intake was measured every day at the same time for

7 days. Mice then received an i.c.v. injection of PBS or insulin

(200 mU) at the beginning of the dark cycle, and food intake was

determined by the difference between chow given to mice immedi-

ately after injection and the weight of remaining chow 12 h after.

Noradrenaline extraction and quantification

Norepinephrine levels in plasma were measured by HPLC separa-

tion coupled with electrochemical detection (HPLC-ED). Perchloric

acid was added to the plasma samples to a final concentration of

0.1 M. Samples were centrifuged (10,000 g) to remove precipitated

proteins, and supernatants were used for automated injection into

the HPLC. Fast isocratic separation was obtained using a reverse-

phase LC-18 column (4.6 × 250 mm; Supelco) with the following

mobile phase: 20 mM sodium dibasic phosphate, 20 mM citric acid,

pH 2.64, containing 10% methanol, 0.12 mM Na2EDTA and

566 mg/l heptanesulfonic acid.

Western blots

Four hours, 6 h or 7 days after i.c.v. injection of AbOs (as indicated in

‘Results’), mice were euthanized by decapitation and the hypothala-

mus and gastrocnemius muscle were rapidly dissected and frozen in

liquid nitrogen. For total protein extraction, samples were thawed and

homogenized in buffer containing 25 mM Tris–HCl, pH 7.5, 150 mM

NaCl, 1% NP-40 (Invitrogen), 1% sodium deoxycholate, 0.1% SDS,

5 mM EDTA, 1% Triton X-100 and phosphatase and protease inhibi-

tor cocktail (Pierce–Thermo Scientific, Rockford, IL). Protein concen-

tration was determined using the BCA kit. Aliquots containing 30 lgprotein were resolved by SDS–PAGE in 4–20% polyacrylamide gels

(Invitrogen) and were electrotransferred to nitrocellulose or PVDF

membranes for 1 h at 300 mA. Blots were blocked for 1 h with 5%

non-fat dry milk in Tween-Tris buffer solution at room temperature or

with Odyssey blocking buffer (Licor, Lincoln, NE; 1:2 dilution in

Tween-Tris Buffer) and were incubated overnight at 2°C with primary

antibodies diluted in blocking buffer. Molecular weight markers were

run in one lane in every gel (Benchmark pre-stained protein ladder;

Life Technologies). Primary antibodies used were IRS-1pSer636 (Santa

Cruz; 1:200, Cat. #sc-33957), IRS-1pSer312 (Invitrogen; 1:200,

Cat#44814-G and Cell Signaling; 1:1,000, Cat#2381), IRS-1pTyr465

(Santa Cruz; 1:200, Cat#sc-17194), total IRS-1 (Santa Cruz; 1:200.

Cat#sc-559), pJNK (Thr183/Tyr185) monoclonal antibody (Cell Signal-

ing; 1:1,000, Cat#9255S), JNK polyclonal antibody (Cell Signaling;

1:1,000, Cat#9252S), phospho-eIF2a (Enzo Life Sciences; 1:1,000,

Cat#BML-SA405 and Cell Signaling; 1:1,000, Cat#9721), total eIF2a(Abcam; 1:1,000, Cat#ab5369 and Cell Signaling; 1:1,000, Cat#9722),

pIKKb (Abcam; 1:1,000; Cat#ab59195), total IKKb (Abcam; 1:1,000,

Cat#ab55404), IjBa (Cell Signaling; 1:1,000, Cat#9242), pNF-jB p65

(Ser536; Cell Signaling; 1:1,000, Cat#3031), total NF-jB p65 (Santa

Cruz; 1:250, Cat#sc-372), p-PKR (Santa Cruz; 1:200, Cat#sc-101784),

total PKR (Santa Cruz; 1:250, Cat#sc-366778), ATF6 (Abcam; 1:1,000,

Cat#ab11909), GLUT-4 (Abcam; 1:500, Cat#ab654), PERKpThr981

(Santa Cruz; 1:500, Cat#sc-32577), total PERK (Abcam; 1:500,

Cat#ab65142), GRP78 (Abcam; 1:500, Cat#ab53068), ATF4 (Sigma;

1:500, Cat#WH0000468M1), spliced and unspliced Xbp1 (Abcam;



205

1:500, Cat#ab37152), phospho-IRE-1 (Novus Biological; 1:1,000,

Cat#NB100-2323), total IRE-1 (Novus Biological; 1:1,000, Cat#NB100-

2324), b-tubulin III (Sigma-Aldrich, St. Louis, MO; 1:10,000,

Cat#T8660) and b-actin (Cell Signaling; 1:10,000, Cat#12262). After

overnight incubation with primary antibodies, membranes were

incubated with horseradish peroxidase-conjugated secondary anti-

body (1:30–50,000), IRDye800CW- or IRDye680RD-conjugated

secondary antibodies (Licor; 1:10,000) at room temperature for 2 h.

Chemiluminescence was developed using SuperSignal West Femto

(Thermo Fisher Scientific). Alternatively, fluorescence intensities were

quantified in an Odyssey CLx apparatus (Licor).

Nuclear-enriched fractions

For the preparation of nuclear extracts, hypothalamus of vehicle- or

AbO-injected mice was homogenized in 0.1 ml hypotonic lysis

buffer (10 mM Hepes, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM

dithiothreitol plus a phosphatase and protease inhibitor cocktail) for

15 min at 4°C. Cells were then lysed by adding 0.5% Nonidet P-40.

The homogenate was centrifuged (13,000 g for 5 min at 4°C), andsupernatants containing the cytoplasmic extracts were stored

at !80°C. The nuclear pellet was resuspended in 75 ll ice-cold

hypertonic extraction buffer (20 mM Hepes, pH 7.9, 300 mM NaCl,

1.5 mM MgCl2, 0.25 mM EDTA, 25% glycerol, 0.5 mM dithiothrei-

tol plus phosphatase and protease inhibitors). After 40 min of inter-

mittent mixing, extracts were centrifuged (13,000 g for 20 min at

4°C), and supernatants containing nuclear proteins were saved.

Total protein concentration was determined using the BCA kit.

Aliquots containing 20 lg protein were resolved by SDS–PAGE in

4–20% polyacrylamide gels (Invitrogen) and were electrotransferred

to nitrocellulose membranes for 1 h at 300 mA. Blots were

processed and incubated with antibodies as described above.

Whole-cell recording

Whole-cell patch-clamp recordings were performed in neurons of

the Arc in brain slices of male Swiss mice (2–3 months old). During

the recordings, neurons were maintained in hypothalamic slice

preparations and data analyses were performed as previously

described Frazao et al (2013). Mice were decapitated and the entire

brain was removed. After removal, brains were immediately

submerged in ice-cold, carbogen-saturated (95% O2 and 5% CO2)

artificial cerebrospinal fluid (aCSF; 126 mM NaCl, 2.8 mM KCl,

1.2 mM MgCl2, 2.5 mM CaCl2, 1.25 mM NaH2PO4, 26 mM NaHCO3

and 5 mM glucose). Coronal sections from a hypothalamic block

(250 lM thick) were cut on a Leica VT1000S vibratome and incu-

bated in oxygenated aCSF at room temperature for at least 1 h

before recording. Slices were transferred to the recording chamber

and allowed to equilibrate for 10–20 min before recording. The

slices were bathed in oxygenated aCSF (32–34°C) at a flow rate of

~2 ml/min. The pipette solution for whole-cell recording was modi-

fied to include an intracellular dye (Alexa Fluor 488): 120 mM

K-gluconate, 10 mM KCl, 10 mM HEPES, 5 mM EGTA, 1 mM CaCl2,

1 mM MgCl2, 2 mM (Mg)-ATP and 0.03 mM Alexa Fluor 488 hydra-

zide dye, pH 7.3. Infrared differential interference contrast was used

to target and obtain the whole-cell recording of neurons at the Arc

(Leica DM6000 FS equipped with a fixed stage and a Leica DFC360

FX high-speed monochrome digital camera). Electrophysiological

signals were recorded using an Axopatch 700B amplifier (Molecular

Devices), low-pass-filtered at 2–5 kHz and analyzed offline on a PC

with pCLAMP programs (Molecular Devices). Recording electrodes

had resistances of 2.5–5 MΩ when filled with the K-gluconate inter-

nal solution. Input resistance was assessed by measuring voltage

deflection at the end of the response to a hyperpolarizing rectangu-

lar current pulse (500 ms of !10 to !50 pA). Membrane potential

values were compensated to account for junction potential

(!8 mV). Solutions containing insulin (50 nM) were typically

perfused for 5 min as previously described by Hill et al (2010). The

recorded cells were randomly chosen. Alexa Fluor 488 hydrazide

dye was used to verify the position of the recorded cells related to

the third ventricle. Only cells located laterally to the third ventricle

at a maximal distance of up to 100 micrometers were recorded.

RNA extraction and quantitative real-time PCR analysis

Hypothalamus and adipose tissue from vehicle- or AbO-injectedmice were homogenized in 500 or 1,000 ll Trizol (Invitrogen),

respectively, and RNA extraction was performed according to manu-

facturer’s instructions. Purity and integrity of RNA were determined

by the 260/280 nm absorbance ratio and by agarose gel electropho-

resis. Only preparations with ratios >1.8 and no signs of RNA degra-

dation were used. In adipose tissue samples, a 30-min-long

incubation at 30°C was performed, the lipid layer was removed and

discarded, and RNA extraction was performed in the water soluble

phase. One lg RNA was used for cDNA synthesis using the Super-

Strand III Reverse Transcriptase kit (Invitrogen). Expression of

genes of interest was analyzed by qPCR on an Applied Biosystems

7500 RT–PCR system using the Power SYBR kit (Applied Biosys-

tems). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or

actin was used as endogenous control. Primer pairs used are shown

in Supplementary Table S1. Cycle threshold (Ct) values were used to

calculate fold changes in gene expression using the 2!DCt method. In

all cases, reactions were performed in 15 ll reaction volumes.

Statistical analysis

No previous statistical calculation was employed to determine sample

size. Instead, sample size in our experiments was chosen based on

usual procedures and best practices in the field. Gaussian distribution

of data was assessed using the D’Agostino-Pearson normality test.

Sample variances were assessed using the F test, when comparing two

independent groups, and using Bartletts’ test and Brown-Forsythe

test, when comparing three or more groups. Variances were equal

between groups, except when stated otherwise. Glucose tolerance test

curves were analyzed by two-way ANOVA followed by Bonferroni

post hoc test. Two-tailed Student’s t-test was performed when

comparing two groups with comparable variances. All data in

macaques shown unequal variances, and therefore, unpaired t-test

with Welch’s correction was applied. For experiments using APP/PS1

mice, a paired t-test was performed to compare groups before and

after treatment with infliximab, since the same animals were assessed

before and after drug administration. In Western blot experiments, a

few lanes (indicated by a red ‘X’ symbol in the source data) were

excluded from final analysis due to (i) excessive background, (ii) faint

or undetectable bands in either phospho- or total proteins or (iii)

fitting the mathematical definition of outliers.



206

Supplementary information for this article is available online:

http://embomolmed.embopress.org

AcknowledgementsThis study is dedicated to the memory of Prof. Leopoldo de Meis (1938–2014),

founder of the Institute of Medical Biochemistry of the Federal University of

Rio de Janeiro. This work was supported by grants from Human Frontiers

Science Program (HFSP) (to FGF), National Institute for Translational Neurosci-

ence (INNT/Brazil) (to STF), the Brazilian funding agencies Conselho Nacional

de Desenvolvimento Científico e Tecnológico (CNPq) (to STF, FGF, CPF and JRC),

Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) (to STF,

FGF, CPF and JRC), Fundação de Amparo à Pesquisa do Estado de São Paulo

(FAPESP2012/12202-4 to RF), Canadian Institutes for Health Research (CIHR)

and Canada Research Chair Program (to DPM). MAS, JRC, NMLS, RLF, JHL, DB,

FCR, TRB and FSN received fellowships from Brazilian agencies CAPES or CNPq.

We thank Drs. Matthias Gralle and Jordano Brito-Moreira for performing AbO

injections in macaques, Drs. Brito-Moreira and Adriano Sebollela for chromato-

graphic characterization of oligomer preparations, Dr. Leonardo M. Saraiva for

help with hypothalamic neuronal cultures and ROS assays, Maíra S. Oliveira,

Mariângela M. Viana, Sandra Bambrilla and Dioze Guadagnini for technical

support, Prof. Carla Tasca (Federal University of Santa Catarina) for the hippo-

campal sections used as positive controls in Fluorojade staining experiments

and Ana Claudia Rangel for secretarial and accounting support.

Author contributionsJRC, NML, CPF, RLF, RM, TRB, CKK, DR, BMC, FCR, FSN, JHL and DB performed

the experiments. JRC, NML, CPF, RLF, TRB, LAV, STF and FDF analyzed and

discussed the data. DPM and FDF designed the experiments in macaques, and

DPM performed the experiments in macaques. JRC, MAS and RF performed

and analyzed the electrophysiological experiments. WLK, MJS, JBC, FML, DPM

and LAV contributed animals, reagents, materials and analysis tools. JRC, NML,

STF and FDF analyzed and discussed the results. FDF supervised the project.

FDF, STF, JRC and NML wrote the manuscript.

Conflict of interestWLK is a cofounder of Acumen Pharmaceuticals, which has been licensed by

Northwestern University to develop ADDL technology for Alzheimer’s thera-

peutics and diagnostics.

References

Arruda AP, Milanski M, Coope A, Torsoni AS, Ropelle E, Carvalho DP,

Carvalheira JB, Velloso LA (2011) Low-grade hypothalamic inflammation

leads to defective thermogenesis, insulin resistance, and impaired insulin

secretion. Endocrinology 152: 1314 – 1326

Bomfim TR, Forny-Germano L, Sathler LB, Brito-Moreira J, Houzel JC, Decker

H, Silverman MA, Kazi H, Melo HM, McClean PL et al (2012) An

anti-diabetes agent protects the mouse brain from defective insulin

signaling caused by Alzheimer’s disease- associated Abeta oligomers. J Clin

Invest 122: 1339 – 1353

Cai D, Yuan M, Frantz DF, Melendez PA, Hansen L, Lee J, Shoelson SE (2005)

Local and systemic insulin resistance resulting from hepatic activation of

IKK-beta and NF-kappaB. Nat Med 11: 183 – 190

Chhatwal JP, Sperling RA (2012) Functional MRI of mnemonic networks

across the spectrum of normal aging, mild cognitive impairment, and

Alzheimer’s disease. J Alzheimers Dis 31(Suppl 3): S155 – S167

Costa RO, Lacor PN, Ferreira IL, Resende R, Auberson YP, Klein WL, Oliveira

CR, Rego AC, Pereira CM (2012) Endoplasmic reticulum stress occurs

downstream of GluN2B subunit of N-methyl-d-aspartate receptor in

mature hippocampal cultures treated with amyloid-beta oligomers. Aging

Cell 11: 823 – 833

Craft S, Zallen G, Baker LD (1992) Glucose and memory in mild senile

dementia of the Alzheimer type. J Clin Exp Neuropsychol 14: 253 – 267

Craft S (2007) Insulin resistance and Alzheimer’s disease pathogenesis:

potential mechanisms and implications for treatment. Curr Alzheimer Res

4: 147 – 152

Craft S (2012) Alzheimer disease: insulin resistance and AD–extending the

translational path. Nat Rev Neurol 8: 360 – 362

De Felice FG (2013) Alzheimer’s disease and insulin resistance: translating

basic science into clinical applications. J Clin Invest 123: 531 – 539

De Felice FG, Ferreira ST (2014) Inflammation, defective insulin signaling, and

mitochondrial dysfunction as common molecular denominators

connecting type 2 diabetes to Alzheimer disease. Diabetes 63: 2262 – 2272

De Felice FG, Velasco PT, Lambert MP, Viola K, Fernandez SJ, Ferreira ST, Klein

WL (2007) Abeta oligomers induce neuronal oxidative stress through an

N-methyl-D-aspartate receptor-dependent mechanism that is blocked by

the Alzheimer drug memantine. J Biol Chem 282: 11590 – 11601

De Felice FG, Vieira MN, Bomfim TR, Decker H, Velasco PT, Lambert MP, Viola

KL, Zhao WQ, Ferreira ST, Klein WL (2009) Protection of synapses against

Alzheimer’s-linked toxins: insulin signaling prevents the pathogenic

binding of Abeta oligomers. Proc Natl Acad Sci USA 106: 1971 – 1976

Denis RG, Arruda AP, Romanatto T, Milanski M, Coope A, Solon C, Razolli DS,

Velloso LA (2010) TNF-alpha transiently induces endoplasmic reticulum

stress and an incomplete unfolded protein response in the hypothalamus.

Neuroscience 170: 1035 – 1044

El-Haschimi K, Pierroz DD, Hileman SM, Bjorbaek C, Flier JS (2000) Two

defects contribute to hypothalamic leptin resistance in mice with

diet-induced obesity. J Clin Invest 105: 1827 – 1832

The paper explained

ProblemDiabetes and Alzheimer’s disease (AD) are chronic degenerativediseases increasing in prevalence in aging populations worldwide.Although clinical and epidemiological studies have linked AD to diabe-tes, with each disease increasing the risk of developing the other, whyAD patients present increased probability of developing diabetes isunknown.

ResultsWe demonstrate that AbOs, toxins that accumulate in the AD brainand have been linked to neuronal dysfunction in brain areas relatedto learning and memory, impact the hypothalamus of mice andmacaques. The hypothalamus is a brain region that regulates glucosehomeostasis in the body. Intriguingly, infusion of AbOs in the braintriggers glucose intolerance, insulin resistance and other manifesta-tions of diabetes in mice. Similar alterations were observed in twotransgenic mouse models of AD.

ImpactOur discovery that AbOs instigate hypothalamic deregulation andglucose intolerance draws attention to a brain structure that has beenlargely ignored to date in the study of Alzheimer’s pathogenesis. Sincethere is evidence that Alzheimer’s patients present glucose intoler-ance, our results highlight the importance of recognizing Alzheimer’sas a disease of both the brain and the periphery.



207

Ferreira ST, Klein WL (2011) The Abeta oligomer hypothesis for synapse

failure and memory loss in Alzheimer’s disease. Neurobiol Learn Mem 96:

529 – 543

Figueiredo CP, Bicca MA, Latini A, Prediger RD, Medeiros R, Calixto JB (2011)

Folic acid plus alpha-tocopherol mitigates amyloid-beta-induced

neurotoxicity through modulation of mitochondrial complexes activity. J

Alzheimers Dis 24: 61 – 75

Figueiredo CP, Clarke JR, Ledo JH, Ribeiro FC, Costa CV, Melo HM, Mota-Sales

AP, Saraiva LM, Klein WL, Sebollela A et al (2013) Memantine rescues

transient cognitive impairment caused by high-molecular-weight abeta

oligomers but not the persistent impairment induced by

low-molecular-weight oligomers. J Neurosci 33: 9626 – 9634

Forny-Germano L, Lyra E, Silva NM, Batista AF, Brito-Moreira J, Gralle M,

Boehnke SE, Coe BC, Lablans A, Marques SA et al (2014) Alzheimer’s

Disease-Like Pathology Induced by Amyloid-beta Oligomers in Nonhuman

Primates. J Neurosci 34: 13629 – 13643

Frazao R, Cravo RM, Donato J Jr, Ratra DV, Clegg DJ, Elmquist JK, Zigman JM,

Williams KW, Elias CF (2013) Shift in Kiss1 cell activity requires estrogen

receptor alpha. J Neurosci 33: 2807 – 2820

Fronczek R, van GS, Frolich M, Overeem S, Roelandse FW, Lammers GJ, Swaab

DF (2012) Hypocretin (orexin) loss in Alzheimer’s disease. Neurobiol Aging

33: 1642 – 1650

Gomes S, Martins I, Fonseca AC, Oliveira CR, Resende R, Pereira CM (2014)

Protective effect of leptin and ghrelin against toxicity induced by

amyloid-beta oligomers in a hypothalamic cell line. J Neuroendocrinol 26:

176 – 185

Gregor MF, Hotamisligil GS (2011) Inflammatory mechanisms in obesity. Annu

Rev Immunol 29: 415 – 445

Hahn TM, Breininger JF, Baskin DG, Schwartz MW (1998) Coexpression of

Agrp and NPY in fasting-activated hypothalamic neurons. Nat Neurosci 1:

271 – 272

Hill JW, Elias CF, Fukuda M, Williams KW, Berglund ED, Holland WL, Cho YR,

Chuang JC, Xu Y, Choi M et al (2010) Direct insulin and leptin action on

pro-opiomelanocortin neurons is required for normal glucose homeostasis

and fertility. Cell Metab 11: 286 – 297

Ho L, Qin W, Pompl PN, Xiang Z, Wang J, Zhao Z, Peng Y, Cambareri G,

Rocher A, Mobbs CV et al (2004) Diet-induced insulin resistance promotes

amyloidosis in a transgenic mouse model of Alzheimer’s disease. FASEB J

18: 902 – 904

Hotamisligil GS, Spiegelman BM (1994) Tumor necrosis factor alpha: a key

component of the obesity-diabetes link. Diabetes 43: 1271 – 1278

Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, Spiegelman BM

(1996) IRS-1-mediated inhibition of insulin receptor tyrosine kinase

activity in TNF-alpha- and obesity-induced insulin resistance. Science 271:

665 – 668

Hotamisligil GS (2010) Endoplasmic reticulum stress and the inflammatory

basis of metabolic disease. Cell 140: 900 – 917

Huang S, Czech MP (2007) The GLUT4 glucose transporter. Cell Metab 5:

237 – 252

Jankowsky JL, Slunt HH, Ratovitski T, Jenkins NA, Copeland NG, Borchelt DR

(2001) Co-expression of multiple transgenes in mouse CNS: a comparison

of strategies. Biomol Eng 17: 157 – 165

Janson J, Laedtke T, Parisi JE, O’Brien P, Petersen RC, Butler PC (2004)

Increased risk of type 2 diabetes in Alzheimer disease. Diabetes 53:

474 – 481

Ji Y, Sun S, Xia S, Yang L, Li X, Qi L (2012) Short term high fat diet challenge

promotes alternative macrophage polarization in adipose tissue via

natural killer T cells and interleukin-4. J Biol Chem 287: 24378 – 24386

Jimenez-Palomares M, Ramos-Rodriguez JJ, Lopez-Acosta JF, Pacheco-Herrero

M, Lechuga-Sancho AM, Perdomo G, Garcia-Alloza M, Cozar-Castellano I

(2012) Increased Abeta production prompts the onset of glucose

intolerance and insulin resistance. Am J Physiol Endocrinol Metab 302:

E1373 – E1380

Jurgensen S, Antonio LL, Mussi GE, Brito-Moreira J, Bomfim TR, De Felice FG,

Garrido-Sanabria ER, Cavalheiro EA, Ferreira ST (2011) Activation of D1/D5

dopamine receptors protects neurons from synapse dysfunction induced

by amyloid-beta oligomers. J Biol Chem 286: 3270 – 3276

Koch L, Wunderlich FT, Seibler J, Konner AC, Hampel B, Irlenbusch S, Brabant

G, Kahn CR, Schwenk F, Bruning JC (2008) Central insulin action regulates

peripheral glucose and fat metabolism in mice. J Clin Invest 118:

2132 – 2147

Lacor PN, Buniel MC, Chang L, Fernandez SJ, Gong Y, Viola KL, Lambert MP,

Velasco PT, Bigio EH, Finch CE et al (2004) Synaptic targeting by

Alzheimer’s-related amyloid beta oligomers. J Neurosci 24: 10191 – 10200

Lambert MP, Barlow AK, Chromy BA, Edwards C, Freed R, Liosatos M, Morgan

TE, Rozovsky I, Trommer B, Viola KL et al (1998) Diffusible, nonfibrillar

ligands derived from Abeta1-42 are potent central nervous system

neurotoxins. Proc Natl Acad Sci USA 95: 6448 – 6453

Lambert MP, Velasco PT, Chang L, Viola KL, Fernandez S, Lacor PN, Khuon D,

Gong Y, Bigio EH, Shaw P et al (2007) Monoclonal antibodies that target

pathological assemblies of Abeta. J Neurochem 100: 23 – 35

Laursen SE, Belknap JK (1986) Intracerebroventricular injections in mice.

Some methodological refinements. J Pharmacol Methods 16: 355 – 357

Le BJ, Pelletier AL, Arapis K, Hourseau M, Cluzeaud F, Descatoire V, Ducroc R,

Aparicio T, Joly F, Couvelard A et al (2014) Overexpression of gastric leptin

precedes adipocyte leptin during high-fat diet and is linked to

5HT-containing enterochromaffin cells. Int J Obes (Lond) 38:

1357 – 1364

Ledo JH, Azevedo EP, Clarke JR, Ribeiro FC, Figueiredo CP, Foguel D, De Felice

FG, Ferreira ST (2013) Amyloid-beta oligomers link depressive-like behavior

and cognitive deficits in mice. Mol Psychiatry 18: 1053 – 1054

Lee YS, Li P, Huh JY, Hwang IJ, Lu M, Kim JI, Ham M, Talukdar S, Chen A, Lu

WJ et al (2011) Inflammation is necessary for long-term but not

short-term high-fat diet-induced insulin resistance. Diabetes 60:

2474 – 2483

Loskutova N, Honea RA, Brooks WM, Burns JM (2010) Reduced limbic and

hypothalamic volumes correlate with bone density in early Alzheimer’s

disease. J Alzheimers Dis 20: 313 – 322

Lourenco MV, Clarke JR, Frozza RL, Bomfim TR, Forny-Germano L, Batista AF,

Sathler LB, Brito-Moreira J, Amaral OB, Silva CA et al (2013) TNF-alpha

mediates PKR-dependent memory impairment and brain IRS-1 inhibition

induced by Alzheimer’s beta-amyloid oligomers in mice and monkeys. Cell

Metab 18: 831 – 843

Ma QL, Yang F, Rosario ER, Ubeda OJ, Beech W, Gant DJ, Chen PP, Hudspeth

B, Chen C, Zhao Y et al (2009) Beta-amyloid oligomers induce

phosphorylation of tau and inactivation of insulin receptor substrate via

c-Jun N-terminal kinase signaling: suppression by omega-3 fatty acids and

curcumin. J Neurosci 29: 9078 – 9089

Ma T, Trinh MA, Wexler AJ, Bourbon C, Gatti E, Pierre P, Cavener DR, Klann E

(2013) Suppression of eIF2alpha kinases alleviates Alzheimer’s

disease-related plasticity and memory deficits. Nat Neurosci 16:

1299 – 1305

Matsuzaki T, Sasaki K, Tanizaki Y, Hata J, Fujimi K, Matsui Y, Sekita A, Suzuki

SO, Kanba S, Kiyohara Y et al (2010) Insulin resistance is associated with

the pathology of Alzheimer disease: the Hisayama study. Neurology 75:

764 – 770



208

Milanski M, Degasperi G, Coope A, Morari J, Denis R, Cintra DE, Tsukumo DM,

Anhe G, Amaral ME, Takahashi HK et al (2009) Saturated fatty acids

produce an inflammatory response predominantly through the activation

of TLR4 signaling in hypothalamus: implications for the pathogenesis of

obesity. J Neurosci 29: 359 – 370

Milanski M, Arruda AP, Coope A, Ignacio-Souza LM, Nunez CE, Roman EA,

Romanatto T, Pascoal LB, Caricilli AM, Torsoni MA et al (2012) Inhibition

of hypothalamic inflammation reverses diet-induced insulin resistance in

the liver. Diabetes 61: 1455 – 1462

Moloney AM, Griffin RJ, Timmons S, O’Connor R, Ravid R, O’Neill C (2010)

Defects in IGF-1 receptor, insulin receptor and IRS-1/2 in Alzheimer’s

disease indicate possible resistance to IGF-1 and insulin signalling.

Neurobiol Aging 31: 224 – 243

Mucke L, Selkoe DJ (2012) Neurotoxicity of amyloid beta-protein: synaptic

and network dysfunction. Cold Spring Harb Perspect Med 2: a006338

Oddo S, Caccamo A, Shepherd JD, Murphy MP, Golde TE, Kayed R, Metherate

R, Mattson MP, Akbari Y, LaFerla FM (2003) Triple-transgenic model of

Alzheimer’s disease with plaques and tangles: intracellular Abeta and

synaptic dysfunction. Neuron 39: 409 – 421

Ogomori K, Kitamoto T, Tateishi J, Sato Y, Suetsugu M, Abe M (1989)

Beta-protein amyloid is widely distributed in the central nervous

system of patients with Alzheimer’s disease. Am J Pathol 134:

243 – 251

Ott A, Stolk RP, van HF, Pols HA, Hofman A, Breteler MM (1999) Diabetes

mellitus and the risk of dementia: the Rotterdam Study. Neurology 53:

1937 – 1942

Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, Ozdelen E, Tuncman G,

Gorgun C, Glimcher LH, Hotamisligil GS (2004) Endoplasmic reticulum

stress links obesity, insulin action, and type 2 diabetes. Science 306:

457 – 461

Ozcan U, Yilmaz E, Ozcan L, Furuhashi M, Vaillancourt E, Smith RO, Gorgun

CZ, Hotamisligil GS (2006) Chemical chaperones reduce ER stress and

restore glucose homeostasis in a mouse model of type 2 diabetes. Science

313: 1137 – 1140

Purkayastha S, Zhang H, Zhang G, Ahmed Z, Wang Y, Cai D (2011) Neural

dysregulation of peripheral insulin action and blood pressure by

brain endoplasmic reticulum stress. Proc Natl Acad Sci USA 108:

2939 – 2944

Romanatto T, Roman EA, Arruda AP, Denis RG, Solon C, Milanski M, Moraes

JC, Bonfleur ML, Degasperi GR, Picardi PK et al (2009) Deletion of tumor

necrosis factor-alpha receptor 1 (TNFR1) protects against diet-induced

obesity by means of increased thermogenesis. J Biol Chem 284:

36213 – 36222

Ropelle ER, Flores MB, Cintra DE, Rocha GZ, Pauli JR, Morari J, de Souza CT,

Moraes JC, Prada PO, Guadagnini D et al (2010) IL-6 and IL-10

anti-inflammatory activity links exercise to hypothalamic insulin and

leptin sensitivity through IKKbeta and ER stress inhibition. PLoS Biol 8:

e1000465

Rossmeisl M, Rim JS, Koza RA, Kozak LP (2003) Variation in type 2 diabetes–

related traits in mouse strains susceptible to diet-induced obesity.

Diabetes 52: 1958 – 1966

Sanchez-Lasheras C, Konner AC, Bruning JC (2010) Integrative neurobiology of

energy homeostasis-neurocircuits, signals and mediators. Front

Neuroendocrinol 31: 4 – 15

Schwartz MW, Woods SC, Porte D Jr, Seeley RJ, Baskin DG (2000) Central

nervous system control of food intake. Nature 404: 661 – 671

Schwartz MW, Porte D Jr (2005) Diabetes, obesity, and the brain. Science 307:

375 – 379

Sebollela A, Freitas-Correa L, Oliveira FF, Paula-Lima AC, Saraiva LM, Martins

SM, Mota LD, Torres C, Alves-Leon S, de Souza JM et al (2012)

Amyloid-beta oligomers induce differential gene expression in adult

human brain slices. J Biol Chem 287: 7436 – 7445

Selkoe DJ (2011) Resolving controversies on the path to Alzheimer’s

therapeutics. Nat Med 17: 1060 – 1065

Selkoe DJ (2012) Preventing Alzheimer’s disease. Science 337: 1488 – 1492

Shi JQ, Shen W, Chen J, Wang BR, Zhong LL, Zhu YW, Zhu HQ, Zhang QQ,

Zhang YD, Xu J (2011a) Anti-TNF-alpha reduces amyloid plaques and tau

phosphorylation and induces CD11c-positive dendritic-like cell in the APP/

PS1 transgenic mouse brains. Brain Res 1368: 239 – 247

Shi JQ, Wang BR, Jiang WW, Chen J, Zhu YW, Zhong LL, Zhang YD, Xu J (2011b)

Cognitive improvement with intrathecal administration of infliximab in a

woman with Alzheimer’s disease. J Am Geriatr Soc 59: 1142 – 1144

Sims-Robinson C, Kim B, Rosko A, Feldman EL (2010) How does diabetes

accelerate Alzheimer disease pathology? Nat Rev Neurol 6: 551 – 559

Standaert DG, Lee VM, Greenberg BD, Lowery DE, Trojanowski JQ (1991)

Molecular features of hypothalamic plaques in Alzheimer’s disease. Am J

Pathol 139: 681 – 691

Steinberg GR, Michell BJ, van Denderen BJ, Watt MJ, Carey AL, Fam BC,

Andrikopoulos S, Proietto J, Gorgun CZ, Carling D et al (2006) Tumor

necrosis factor alpha-induced skeletal muscle insulin resistance

involves suppression of AMP-kinase signaling. Cell Metab 4:

465 –474

Takeda S, Sato N, Uchio-Yamada K, Sawada K, Kunieda T, Takeuchi D,

Kurinami H, Shinohara M, Rakugi H, Morishita R (2010)

Diabetes-accelerated memory dysfunction via cerebrovascular

inflammation and Abeta deposition in an Alzheimer mouse model with

diabetes. Proc Natl Acad Sci USA 107: 7036 – 7041

Talbot K, Wang HY, Kazi H, Han LY, Bakshi KP, Stucky A, Fuino RL, Kawaguchi KR,

Samoyedny AJ, Wilson RS et al (2012) Demonstrated brain insulin

resistance in Alzheimer’s disease patients is associated with IGF-1

resistance, IRS-1 dysregulation, and cognitive decline. J Clin Invest 122:

1316– 1338

Thaler JP, Yi CX, Schur EA, Guyenet SJ, Hwang BH, Dietrich MO, Zhao X,

Sarruf DA, Izgur V, Maravilla KR et al (2012) Obesity is associated with

hypothalamic injury in rodents and humans. J Clin Invest 122:

153 – 162

Thaler JP, Guyenet SJ, Dorfman MD, Wisse BE, Schwartz MW (2013)

Hypothalamic inflammation: marker or mechanism of obesity

pathogenesis? Diabetes 62: 2629 – 2634

Walsh DM, Klyubin I, Fadeeva JV, Cullen WK, Anwyl R, Wolfe MS, Rowan MJ,

Selkoe DJ (2002) Naturally secreted oligomers of amyloid beta protein

potently inhibit hippocampal long-term potentiation in vivo. Nature 416:

535 – 539

Wang KC, Woung LC, Tsai MT, Liu CC, Su YH, Li CY (2012) Risk of Alzheimer’s

disease in relation to diabetes: a population-based cohort study.

Neuroepidemiology 38: 237 – 244

Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols A, Ross JS,

Tartaglia LA et al (2003) Chronic inflammation in fat plays a crucial role

in the development of obesity-related insulin resistance. J Clin Invest 112:

1821 – 1830

Zhang G, Li J, Purkayastha S, Tang Y, Zhang H, Yin Y, Li B, Liu G, Cai D (2013)

Hypothalamic programming of systemic ageing involving IKK-beta,

NF-kappaB and GnRH. Nature 497: 211 – 216

Zhang X, Zhang G, Zhang H, Karin M, Bai H, Cai D (2008) Hypothalamic

IKKbeta/NF-kappaB and ER stress link overnutrition to energy imbalance

and obesity. Cell 135: 61 – 73



209

Zhang Y, Zhou B, Zhang F, Wu J, Hu Y, Liu Y, Zhai Q (2012) Amyloid-beta

induces hepatic insulin resistance by activating JAK2/STAT3/SOCS-1

signaling pathway. Diabetes 61: 1434 – 1443

Zhao WQ, De Felice FG, Fernandez S, Chen H, Lambert MP, Quon MJ, Krafft

GA, Klein WL (2008) Amyloid beta oligomers induce impairment of

neuronal insulin receptors. FASEB J 22: 246 – 260

Zussy C, Brureau A, Delair B, Marchal S, Keller E, Ixart G, Naert G, Meunier J,

Chevallier N, Maurice T et al (2011) Time-course and regional analyses of

the physiopathological changes induced after cerebral injection of an

amyloid b fragment in rats. Am J Pathol 179: 315 – 334

License: This is an open access article under the

terms of the Creative Commons Attribution 4.0

License, which permits use, distribution and reproduc-

tion in any medium, provided the original work is

properly cited.



210

1"

"

Clarke et al. - Supplementary Information

Contents

Supplementary Figure 1. Injection of blue dye into the lateral cerebral ventricle of

mice…………………………………………………………………………………............

2

Supplementary Figure 2. Effects of i.c.v.-injected AβOs or scrambled Aβ peptide versus

short-term high-fat diet on peripheral glucose tolerance in mice…………………………...

3

Supplementary Figure 3. AβOs do not induce changes in hypothalamic levels of several

ER stress markers and of phosphorylated JNK and PKR…………………...........................

4

Supplementary Figure 4. Cytoarchitecture of monkey hypothalamus……………............... 5

Supplementary Figure 5. AβOs do not affect electrophysiological properties of NPY-

neurons………………………………………………………………………………………

6

Supplementary Figure 6. I.c.v. treatment with infliximab prevents AβO-induced glucose

intolerance in mice…………………………………………………………………………..

7

Supplementary Table 1. Primer sequences used for qPCR reactions………………………. 8

Supplementary References 9

2"

"

Supplementary Figure 1. Injection of blue dye into the lateral cerebral ventricle of Swiss

mice. A, Scheme of a coronal section of the mouse brain (adapted from Paxinos and Franklin,

1997). Blue labeling represents the right lateral ventricle targeted by i.c.v. injection. Cx, cortex;

CC, corpus callosum; Cpu, caudate putamen; LV, lateral ventricle; 3V, third ventricle; D3V,

dorsal third ventricle. (B, C) bottom-up and sagital (along the midline) views, respectively, of a

mouse brain after injection of 3 µl of blue dye into the lateral ventricle. Note the hypothalamus is

completely surrounded by the dye (B, C). Arrow in C indicates the site of injection. Scale bar =

0.5 cm.

3"

"

Supplementary Figure 2. Effects of i.c.v.-injected AβOs or scrambled Aβ peptide versus

short-term high-fat diet on peripheral glucose tolerance in mice. A, Adult Swiss mice that

received a single i.c.v. injection of vehicle, scrambled Aβ peptide (scrAβOs) or AβOs (10 pmol)

were tested on a glucose tolerance test 7 days post-injection (2g glucose/kg body weight, i.p.; n =

13 Veh; 12 AβOs; 10 scrAβOs). B, Swiss mice fed a high-fat diet for 7 days were submitted to a

glucose tolerance test (2g glucose/kg body weight, i.p.) and compared to Swiss mice that

received a single i.c.v. injection of vehicle or AβOs (10 pmol; 7 days post-injection, n = 8

animals/group). C-E, Glucose tolerance tests (2 g glucose/kg body weight, i.p.) performed 12 h

(C; n = 6 Veh; 9 AβOs), 36 h (D; n = 6 Veh; 9 AβOs) or 14 days (E; n = 6 animals/group) after

i.c.v. injection of vehicle or AβOs (10 pmol). Bar graphs represent areas under the curves in the

4"

"

time-course plots of the glucose tolerance test. Data are expressed as means ± S.E.M. A, * p =

0.0155; B, * p = 0.0061 and 0.0037; D, * p = 0.0247; E, # p = 0.0665) one-way ANOVA

followed by Bonferroni.

Supplementary Figure 3. AβOs do not induce changes in hypothalamic levels of several ER

stress markers and of phosphorylated JNK and PKR. (A-H) Western blot analysis of

hypothalamic levels of PERKpTHr980 (A; n = 5 Veh; 8 AβOs), ATF6 (B; n = 7 Veh; 8 AβOs),

pIRE-1α (C; n = 7 animals/group), spliced (D; n = 5 Veh; 8 AβOs) and unspliced Xbp1 (E; n = 6

Veh; 8 AβOs), GRP78 (F; n = 7 Veh; 8 AβOs), pJNK (G; n = 4 Veh; 5 AβOs) and pPKR (H; n =

5"

"

5 Veh; 4 AβOs) 4 h after a single i.c.v. injection of vehicle or 10 pmol AβOs in mice. (I, J),

Western blot analysis of hypothalamic levels of p-JNK (I; n = 5 animals/group) and pPKR (J; n

= 4 Veh; 5 AβOs) 7 days after i.c.v. injection of vehicle or AβOs in mice. Graphs show

densitometric data normalized by actin (B, D-F) or by the levels of total PERK (A), IRE-1 (C),

JNK (G and I) or PKR (H and J). Data are expressed as mean ± S.E.M.

Supplementary Figure 4. Cytoarchitecture of monkey hypothalamus. Cresyl violet staining

of a sham-operated monkey brain. 3V, third ventricle; VMH, ventromedial hypothalamus; PME,

posterior medial eminence; opt, optic tract; Hy, hypothalamus; ArcH, arcuate nucleus of

hypothalamus (Martin and Bowden, 1996). Scale bar = 200 µm in A and 100 µm in B.

6"

"

Supplementary Figure 5. AβOs do not affect electrophysiological properties of NPY-

neurons. Swiss mice received a single i.c.v. injection of vehicle (Veh) or 10 pmol AβOs and

their brains were used for whole-cell patch clamp recordings seven days after. (A-E) Coronal

slices were used for recording of spontaneous activity of NPY neurons in the arcuate nucleus. No

changes in amplitude or frequency of spontaneous excitatory (sEPSC) or inhibitory post-synaptic

currents (sIPSC) were detected between groups. E, resting membrane potential of recorded cells.

N = 20-21 slices obtained from 7 mice in each group, 3-4 cells recorded per slice. Data are

expressed as mean ± S.E.M.

7"

"

Supplementary Figure 6. I.c.v. treatment with infliximab prevents AβO-induced glucose

intolerance in mice. Adult Swiss mice received a single i.c.v. injection of PBS or infliximab

(0.2 µg) 20 min before receiving an i.c.v. injection of vehicle (Veh) or AβOs (10 pmol). Seven

days post injection, mice were submitted to a glucose tolerance test (2g glucose/kg body weight,

i.p.). Bar graphs represent areas under the curve in the time course plots of the glucose tolerance

test. Data are expressed as means ± S.E.M. * p = 0.0494, Student’s t-test (Veh vs AβOs) (n = 12

Veh; 11 AβOs; 14 Veh+Inflix; 14 AβOs+Inflix).

8"

"

Supplementary Table 1. Primer sequences used for qPCR reactions.

Target"gene" Forward"primer" Reverse"primer"

NPY" ATG"CTA"GGT"AAC"AAG"CGA"ATG"G" TGT"CGC"AGA"GCG"GAG"TAG"TAT"

POMC" ATG"CCG"AGA"TTC"TGC"TAC"AGT" TCC"AGC"GAG"AGG"TCG"AGT"TT"

AgRP" ATG"CTG"ACT"GCA"ATG"TTG"CTG" CAG"ACT"TAG"ACC"TGG"GAA"CTC"T"

IL6" TTC"TTG"GGA"CTG"ATG"CTG"GTG" CAG"AAT"TGC"CAT"TGC"ACA"ACT"C"

TNFα" CCC"TCA"CAC"TCA"GAT"CAT"CTT"CT" GCT"ACG"ACG"TGG"GCT"ACA"G"

Leptin" TGA"GCT"ATC"TGC"AGC"ACG"TT" TTC"ACA"CAC"GCA"GTC"GGT"AT"

GLUT4" AAA"AGT"GCC"TGA"AAC"CAG"AG" TCA"CCT"CCT"GCT"CTA"AAA"GG"

Actin" GCC"CTG"AGG"CTC"TTT"TCC"AG" TGC"CAC"AGG"ATT"CCA"TAC"CC"

GAPDH" AGG"TCG"GTG"TGA"TGA"ACG"GAT"TTG" TGT"AGA"CCA"TGT"AGT"TGA"GGT"CA"

9"

"

Supplementary References

Martin RF, Bowden DM (1996) A stereotaxic template atlas of the macaque

brain for digital imaging and quantitative neuroanatomy. NeuroImage 4:

119 – 150

Paxinos G, Franklin K (1997) The Mouse Brain in Stereotaxic Coordinates. San

Diego, USA: Academic Press

Documents

Disfunção da sinalização por insulina hipocampal na Doença de Alzheimer · 2016-01-19 · entenderá o que é temer ao Senhor e achará o conhecimento de Deus. Pois o Senhor