the con-a-peroxidase method for tissue localization of glucosyl and

iv

Agradecimentos

Aos Profs. Drs. membros desta banca pela contribuição à qualidade deste trabalho.

À minha orientadora, Profa. Dra. Maria Luiza Silveira Mello, pela excelente

orientação, por ter acreditado no meu trabalho ao longo desses anos, e por ter me

estimulado e incentivado, estando sempre preocupada com a minha formação e com o meu

bem estar, dentro e fora do laboratório.

Ao Prof. Dr. Benedicto de Campos Vidal por todo o companheirismo, à todos os

ensinamentos a mim dispensados, e ao incentivo à minha carreira científica.

Ao Dr. Nick Gilbert, do Cancer Research Centre da Universidade de Edimburgo,

pela co-orientação, e aos amigos do mesmo laboratório Jayne, Tom, Charlotte, Ben e

Bernie, pelo companheirismo e por toda a ajuda durante o estágio de doutorado no exterior.

Ao Dr. John Sommervile, da Universidade de St. Andrews, Escócia, pelo

financiamento concedido para que eu participasse do 20th Wilhelm Bernhard Workshop on

The Cell Nucleus.

Aos Profs. Drs. Edson Rosa Pimentel e Laurecir Gomes, do Depto. de Biologia

Celular, e aos pós-graduandos Hugo Takeda e Flávia Winck pela amizade e pelas

contribuições nos experimentos bioquímicos.

Ao grande amigo Prof. Dr. Mateus Mondin e à Profa. Dra. Margarida LR Aguiar-

Perecin pelas contribuições nos experimentos definitivos com hibridação in situ.

À Profa. Dra. Shirlei Maria Recco-Pimentel também pela amizade e por

disponibilizar seu laboratório para que os experimentos iniciais com hibridação in situ

fossem realizados.

v

Ao Programa de Pós-graduação em Biologia Celular e Estrutural, por contribuir na

manutenção da infra-estrutura para a realização do meu trabalho e pela contribuição na

minha formação.

À Líliam Panagio, por toda atenção e auxílio durante estes anos de pós-graduação.

Ao CNPq pela bolsa de estudos concedida para a parte do doutorado realizada no

Brasil, e à CAPES, pela bolsa de estudos referente ao doutorado sanduíche.

À FAPESP pelo auxílio técnico e financeiro relacionado a projetos do nosso

laboratório.

A todos os meus companheiros de laboratório e amigos: Adriana, André, Andressa,

Antonella, Elenice, Felipe, Flávia Ghiraldini, James, Juliano, Letícia, Marcela, Marina,

Martha e Murilo.

Aos técnicos de laboratório, Mário Bianchi, Martha Almeida e Adilson por todo

auxílio e dedicação.

A todos os amigos do Departamento de Bilogia Celular e da pós-graduação em

Biologia Celular e estrutural por todos os bons momentos que tivemos juntos.

Às queridas amigas Daniele Lisboa e Flávia de Paoli, pela valiosa amizade.

À minha querida Lilian Martinho, por ser sempre me trazer paz e alegria.

Aos meus irmãos, Márcio e Elaine, por serem presença constante de amor em minha

vida.

Aos meus amados pais, João e Marilena, por toda dedicação, amor e apoio em todos

os momentos.

Ao Universo, o qual muitos gostam de personificar e chamar Deus, que conspirou a

meu favor e me mostrou sempre os melhores caminhos para que eu pudesse prosperar em

termos pessoais e acadêmicos.

vii

“Não tenho vergonha de mudar de opinião

porque não tenho vergonha de pensar.”

Blaise Pascal

viii

Sumário

Abreviaturas............................................................................................................................x

Resumo...................................................................................................................................xi

Abstract................................................................................................................................xiii

Introdução

1. Organização da cromatina interfásica Composição genômica.................................................................................................1 A estrutura altamente organizada da cromatina...........................................................2 Eucromatina e heterocromatina...................................................................................4 Territórios cromossômicos........................................................................................10 A matriz nuclear.........................................................................................................12 Fibras estendidas de cromatina..................................................................................18 Glicoproteínas nucleares............................................................................................19

2. Organização da cromatina em células de camundongo Hepatócitos................................................................................................................20 Espermatozóides........................................................................................................24

3. Fatores que influenciam a organização nuclear e da cromatina Jejum..........................................................................................................................25 Desenvolvimento e envelhecimento………………………………………………..26

Objetivos................................................................................................................................28

Capítulo I

Chromatin supraorganization and extensibility in mouse hepatocytes with

development and aging..........................................................................................................29

Capítulo II

Accumulation of heterochromatin marks in aging hepatocytes influences interphase

chromatin structure………………………………………………………………………...40

Capítulo III

Effects of aging and starvation on chromatin-nuclear matrix association in mouse

hepatocytes…………………………………………………………………………………63

ix

Capítulo IV

Extended chromatin fibers in mouse testicular spermatozoa………………………90

Capítulo V

The Con A-peroxidase method for tissue localization of glucosyl and mannosyl

groups applied to mouse hepatocytes and chicken erythrocytes…………………………..97

Discussão geral....................................................................................................................103

Conclusões gerais…………………………………………………………………………106

Referências bibliográficas………………………………………………………………...107

Anexo I

Nucleus image properties assessed by video image analysis in mouse hepatocytes

under a short lysis for extended chromatin fiber formation……………………………...118

Anexo II

Extended chromatin fibers in spermatozoa of Apis mellifera (Hymenoptera,

Apoidea)…………………………………………………………………………………..127

Declaração comitê de ética em experimentação animal......................................................134

x

Abreviaturas

AcH3K9 – histona H3 acetilada na lisina 9

AcH4K16 – histona H4 acetilada na lisina 16

Con A – concanavalina A

CT – território cromossômico

ECF – fibra estendida de cromatina

FISH – hibridação in situ fluorescente

HP1 – heterochromatin protein 1

ICD – domínio intercromossomal

Met3H3K9 - histona H3 trimetilada na lisina 9

Met2H3K9 - histona H3 dimetilada na lisina 9



Met2H3K4 - histona H3 dimetilada na lisina 4

Met1H3K4 - histona H3 metilada na lisina 4

NOR - região organizadora de nucléolo

rDNA - DNA ribossomal

rRNA - RNA ribossomal

Sir – Silent information regulator

S/MAR - scaffold/matrix attachment region

xi

Resumo

Envelhecimento pode ser definido como as mudanças sofridas por um organismo ao

longo do tempo. Esse processo, em biologia, é denominado senescência. A senescência

celular é um fenômeno observado em células isoladas, e tem sido estudada tipicamente em

células em cultura. Sua ocorrência in vivo foi observada em alguns tecidos de mamíferos.

As mudanças na estrutura e organização da cromatina que ocorrem em células senescentes

incluem, aumento na resistência da cromatina à digestão por nucleases e acúmulo de

modificações de histonas e proteínas associadas à heterocromatina. Embora nem todas as

células em um organismo envelhecido estejam em estado de senescência, espera-se que

mudanças na estrutura e organização da cromatina ocorram.

A restrição calórica é a única intervenção conhecida que tem a capacidade de

estender o tempo de vida em mamíferos. Após uma dieta de restrição calórica ou jejum

muitos genes, cuja expressão encontra-se alterada em animais idosos, têm sua expressão

restabelecida aos níveis observados em animais jovens. Acredita-se que mudanças na

cromatina também possam ocorrer durante o jejum, e que induzam mudanças no nível de

expressão de diversos genes.

No presente trabalho, buscando-se alterações na organização da cromatina em

hepatócitos de camundongo ao longo do envelhecimento ou submetidos ao jejum,

observou-se um aumento das propriedades viscoelásticas da cromatina ao longo do

envelhecimento, de acordo com as mudanças na habilidade dessa cromatina em formar

fibras estendidas de cromatina. Essas diferenças foram acompanhadas por um

desempacotamento da cromatina. Observou-se também que essa viscoelasticidade da

cromatina era dependente principalmente de interações desta com a matriz nuclear, e que

cópias de genes cuja atividade transcricional não é mais requerida, ou requerida em um

nível menor em animais idosos, podem desligar-se temporariamente da matriz nuclear.

Mudanças nas propriedades viscoelásticas da cromatina e no seu grau de

compactação já haviam sido observadas previamente em animais em jejum. Apesar disso,

no presente trabalho, nenhuma diferença com relação à interação dos genes rDNA com a

xii

matriz nuclear foi encontrada em animais em jejum. Contudo, independente da condição

fisiológica, o DNA aderido à matriz nuclear parece ser rico em genes, enquanto as

seqüências heterocromáticas, pobres em genes, geralmente são encontradas tanto

associadas com a matriz nuclear quanto dissociadas desta (cuidado com essa conclusão.

Está forte). Em hepatócitos de animais idosos foi observado acúmulo de marcadores

heterocromáticos (modificações de histonas) e de outras proteínas (proteínas formadoras de

heterocromatina e glicoproteínas presentes principalmente nos cromocentros), assim como

diminuição das modificações de histonas associadas com transcrição ativa. Todas essas

modificações estão relacionadas com alterações na síntese de RNA já relatadas para

animais idosos, e são uma evidência de que o controle da expressão gênica, a organização e

a composição da cromatina estão intimamente relacionados.

Em um outro tipo celular como espermatozóides de camundongo, uma diferente

organização nuclear levou a propriedades diferenciadas de sua cromatina com relação às

suas propriedades viscoelásticas (aumentadas). Tais diferenças possivelmente estejam

relacionadas com um padrão modificado de expressão gênica, uma vez que em

espermatozóides, a atividade transcricional é nula ou quase ausente.

Palavras-chave: cromatina, envelhecimento, epigenética, jejum, matriz nuclear

xiii

Abstract

Aging may be defined as the changes that take place in an organism with time. This

process, in biology, is called senescence. Cellular senescence is observed in isolated cells,

and has been studied typically in cultured cells, but its occurrence in vivo has been shown

only in some mammalian tissues. Chromatin changes that take place with cellular

senescence include increase in the resistance of chromatin to nuclease digestion and

accumulation of histone modifications and non-histone proteins associated with

heterochromatin. Although not all cells in an aged organism are subjected to cellular

senescence, it is expected that changes in the chromatin structure and organization still

occur.

Caloric restriction is the only intervention known to extend life span in mammals. It

has been shown that many genes whose expression pattern is altered in aged animals can be

reverted to the levels observed in young animals after a caloric restriction diet or complete

food withdrawal. Changes in chromatin structure may occur during the starvation period to

induce changes in the expression level of several genes.

With the aim of screening for alterations in the chromatin organization in mouse

hepatocyte nuclei with aging or following starvation, we observed an increase in the

viscoelastic properties of chromatin with aging, in terms of changes in the ability of this

chromatin to form extended chromatin fibers after a lysis treatment in liver imprints on

histological slides. These differences were accompanied by chromatin unpackage. Most of

the viscoelasticity of the chromatin were dependent on its interactions with the nuclear

matrix, and copies of genes whose transcription are no longer required in aged animals,

tended to detach from the nuclear matrix.

Changes in the viscoelastic properties and packing degree of chromatin had been

shown previously in starved animals. However, no differences regarding this feature were

seen in the present work. Nevertheless, regardless the physiological condition, DNA

attached to the nuclear matrix seems to be gene-rich, while heterochromatic gene-poor

regions were found both attached and detached from the nuclear matrix. We observed

xiv

accumulation of heterochromatic marks (histone modifications) and non-histone proteins

(heterochromatin proteins and glycoproteins present mainly in the chromocenters), as well

as decreased histone modifications associated with transcription in hepatocyte nuclei of

aged mice. All these changes are related to altered RNA synthesis observed in aged animals

and are an evidence of the strong relationship between chromatin organization,

composition, and control of gene expression.

In another cell type, mouse sperm cells, its nuclear organization lead to different

chromatin properties regarding its viscoelastic properties (increased). These differences are

possibly related to a modified pattern of gene expression since gene transcription is almost

or completely absent in sperm cells.

Keywords: aging, chromatin, epigenetics, nuclear matrix, starvation

1

Introdução

1. Organização da cromatina interfásica

Composição genômica

A maior parte do genoma nuclear de mamíferos e aves pode ser fragmentado

artificialmente em um mosaico de segmentos muito longos de DNA (>> 300 kb)

denominados isocoros. Esses isocoros são quase homogêneos em composição de bases e

pertencem a um pequeno número de classes principais discerníveis pelas suas diferenças

em conteúdo de guanina-citosina (GC) (Bernardi, 1993).

No genoma humano, que é um modelo representativo do genoma de diversos

mamíferos, a composição dos isocoros corresponde a 30-60% de bases GC, e cinco famílias

de isocoros foram identificadas: duas famílias pobres em GC, L1 e L2, que juntas

representam 62% do genoma, e três famílias ricas em GC, H1, H2, e H3, que representam,

respectivamente, 22%, 9%, e 3% do genoma (os restantes 4% são formados pelo DNA

satélite e ribossomal) (Figura 1). A concentração gênica é incrivelmente não uniforme,

sendo mais alta na família de isocoros H3, mais baixa nas famílias L1 + L2, e intermediária

nas famílias H1 + H2. Em cromossomos metafásicos a família H3 corresponde às bandas T

(teloméricas), e as famílias L1 + L2 correspondem às bandas G (Giemsa), enquanto as

bandas R (reversas) compreendem ambos isocoros ricos e pobres em GC (Bernardi, 1993).

Dois espaços gênicos foram definidos no genoma humano. No “núcleo genômico”,

formado pelas famílias de isocoros H2 e H3 (as quais compreendem 12% do genoma), a

concentração gênica é muito alta (um gene a cada 5-15 kb) e é comparável àquelas dos

genomas compactos dos eucariotos superiores, enquanto que no “espaço vazio”, formado

pelas famílias de isocoros L e H1 (as quais compreendem 88% do genoma) a concentração

gênica é muito baixa (um gene a cada 50-150 kb). Aproximadamente 54% dos genes

humanos estão localizados no pequeno núcleo genômico, e os demais 46% estão

localizados no grande espaço vazio (Bernardi, 2000).

2

Figura 1. Esquema da organização de isocoros no genoma humano. O genoma é um mosaico de longos segmentos de DNA (>300 kb), os isocoros, que são homogêneos em composição e pertencem a um pequeno número de famílias, pobres em GC (L1 e L2), ricos em GC (H1 e H2), e muito ricos em GC (H3). Degradação enzimática e física ocorrendo durante o preparo do DNA gera grandes fragmentos de aproximadamente 100 kb (adaptado de Bernardi, 1993).

A estrutura altamente organizada da cromatina

Em eucariotos, o DNA encontra-se complexado com proteínas histônicas. Cento e

quarenta e seis pares de bases de DNA (1,75 voltas) enrolam-se num octâmero formado por

duas de cada uma das histonas H2A, H2B, H3 e H4, formando uma partícula nucleossomal.

A interação de uma histona de ligação (H1) com o DNA entre duas partículas

nucleossomais (DNA de ligação) aumenta o número de pares de bases para 165, o que

corresponde a duas voltas (Bednar et al., 1998). Dessa forma, a adição da histona de ligação

contribui para o aumento da condensação da cromatina (Horn & Peterson, 2002). Um DNA

de ligação curto também contribui para a compactação do DNA, enquanto um DNA de

ligação longo produz efeito oposto. Assim, o nível primário da estrutura da cromatina é

representado pela fibra de cromatina de 11-nm, ou conformação “colar-de-contas”, que é

um arranjo estendido de nucleossomos (Woodcock & Dimitrov, 2001) (Figura 2).

3

Figura 2. A estrutura básica da cromatina. A fibra de 11 nm consiste no DNA enrolado num octâmero de histonas (nucleossomos) à intervalos de 200 pb de DNA. O empacotamento adicional cria uma estrutura espiral, a fibra de 30 nm, muitas vezes denominada solenóide. Histonas positivamente carregadas (desacetiladas) (setas) facilitam o empacotamento, enquanto a acetilação das caudas das histonas (barras) promovem um estado descompactado correspondente à cromatina ativa. Os dois estados da cromatina podem ser visibilizados nas micrografias eletrônicas (adaptado de http://sgi.bls.umkc.edu/waterborg/chromat/chroma09.html).

O DNA B puro tem um comprimento de 2,9 kb/µm e adquire uma compactação de

7 vezes na fibra de 11 nm (Goodrich & Tweedie, 2002). A estrutura secundária da

cromatina é formada pela interação entre os nucleossomos, sendo que a mais estudada é a

fibra de 30 nm, que possui um grau de compactação de 40 a 50 vezes (Woodcock &

Dimitrov, 2001). Especula-se que a fibra de 30 nm consista de uma hélice de

nucleossomos, mas a sua estrutura exata ainda é debatida (Dorigo et al., 2004). Níveis

adicionais de organização da cromatina podem ser formados através de interações entre as

fibras de 30 nm. Um modelo clássico de tal estrutura altamente organizada é o cromonema,

no qual as fibras mais finas são arranjadas para produzir fibras mais espessas de 100 a 130

nm, produzindo um grau de compactação de aproximadamente 500 vezes (Belmont &

Bruce, 1994). Outros modelos propõem um arranjo em forma de alças de 30 nm conectadas

a um arcabouço protéico central (Cremer & Cremer, 2001). Já o modelo da afinidade

nucleossômica dispensa o arranjo das fibras de 30 nm, e assume a existência de cadeias

randômicas de nucleossomos ocupando um determinado espaço no núcleo (Müller et al.,

2004). Nenhum desses modelos foi ainda provado, mas todos são consistentes com a

observação de que a cromatina descondensada forma uma série de “contas” adjacentes e

que ativação da transcrição é necessária para a manutenção dessa cromatina descondensada,

4

algo observado repetidamente em diversos organismos (Müller et al., 2004; Wegel et al.,

2005).

Eucromatina e heterocromatina

A heterocromatina foi originalmente definida como uma cromatina altamente

condensada (Heitz, 1928). Em Arabidopsis, por exemplo, a heterocromatina do

cromossomo 4 possui uma condensação de 350 vezes, enquanto uma região eucromática

adjacente mostra um grau de condensação de 60 vezes (Fransz et al., 2002). Sabe-se que a

heterocromatina consiste de um arranjo nucleossômico regular, o qual impede o acesso de

nucleases e contém uma alta proporção de sequências repetitivas transcricionalmente

inativas, intercaladas por um número relativamente pequeno de genes (Grewal & Moazed,

2003). A heterocromatina foi tradicionalmente subdividida em heterocromatina

constitutiva, cromatina que está sempre condensada, independente da fase do ciclo celular e

estado fisiológico da célula, e heterocromatina facultativa, que pode se descondensar em

algumas circunstâncias. As regiões cromossômicas ao redor dos centrômeros e telômeros

são exemplos de heterocromatina constitutiva, enquanto os genes silenciados à partir de um

certo ponto durante o desenvolvimento podem formar a heterocromatina facultativa, que

pode ser encontrada intercalada ao longo dos braços cromossômicos. Em organismos com

grandes genomas, as regiões de heterocromatina constitutiva podem se formar também ao

longo dos braços cromossômicos. A eucromatina é considerada descondensada por causa

do espaçamento irregular observado entre os nucleossomos. Esta é relativamente rica em

genes e é potencialmente ativa transcricionalmente (Elgin & Grewal, 2003). Contudo, estas

diferenças não são sempre tão claras, como mostrou uma análise recente do genoma

humano, onde se mostrou que algumas regiões pericentroméricas são descondensadas e que

algumas regiões eucromáticas são condensadas (Gilbert et al., 2004). Cada um desses dois

estados funcionais da cromatina apresenta marcadores epigenéticos específicos que

envolvem acetilação e metilação das caudas de histonas e metilação de DNA, como

detalhado à diante.

5

Os grandes blocos de heterocromatina envolvem estruturas cromossômicas

funcionais como os centrômeros e os telômeros, enquanto domínios heterocromáticos

menores estão distribuídos ao longo do cromossomo (Grewal & Elgin, 2002). Após várias

décadas de especulação, tornou-se claro que a heterocromatina tem um papel crucial na

função dos centrômeros. Proteínas da heterocromatina estão associadas com o DNA

repetitivo ao redor dos centrômeros e são necessárias para a coesão das cromátides irmãs e

correta segregação dos cromossomos (Bernard et al., 2001; Hall et al., 2003). A

heterocromatina também estabiliza as seqüências de DNA repetitivo nos centrômeros,

telômeros, e em todo o restante do genoma através das repetições homólogas (Guarente,

2000). Adicionalmente a este papel na manutenção da estabilidade genômica, a

heterocromatina tem um papel central na regulação da expressão gênica durante o

desenvolvimento e diferenciação celular. Estruturas tipo-heterocromatina estão envolvidas

na inativação estável dos reguladores do desenvolvimento, por exemplo, os agrupamentos

de genes homeóticos em Drosophila e mamíferos, e os gene mating-type de fungos

(Cavalli, 2002). A compensação de dose em fêmeas de mamíferos placentários envolve a

inativação heterocromática de um dos dois cromossomos X nas células somáticas (Avner &

Heard, 2001). Várias propriedades da heterocromatina a tornam particularmente adequada

para processos que requerem uma manutenção estável de estados de expressão gênica por

longos períodos. Primeiramente, o estado heterocromático é herdado epigeneticamente e de

maneira estável mesmo depois de muitas divisões celulares, as quais acontecem sob

diferentes condições de desenvolvimento e influência ambiental. Segundo, o mecanismo de

montagem da heterocromatina e o espalhamento desta à partir dos sítios de nucleação para

as regiões adjacentes de DNA promovem uma transição de um controle genético

relacionado à seqüência para um controle epigenético independente da seqüência do DNA.

Embora diversos estudos nas últimas décadas tenham estabelecido as propriedades básicas

da cromatina e tenham identificado muitos dos seus blocos estruturais, somente agora

começa-se a entender como esses domínios de DNA, os quais compreendem praticamente

metade de todo o DNA genômico em alguns eucariotos, são montados e epigeneticamente

propagados.

6

A divisão citológica da cromatina em eucromatina e heterocromatina não

necessariamente nos remete a uma definição molecular direta. Entretanto, há diferentes

marcadores bioquímicos associados à cromatina, e que são importantes na distinção entre

eucromatina e heterocromatina.

Estudos em diversos organismos sugerem fortemente que as histonas e suas

modifcações pós-traducionais tenham um papel fundamental na montagem da

heterocromatina (Jenuwein, 2001). As mais importantes dentre todas são a acetilação e

metilação dos resíduos de lisina nas caudas amino-terminais altamente conservadas das

histonas H3 e H4. Acetilação aumentada invariavelmente correlaciona-se com atividade

transcricional, enquanto uma acetilação diminuída correlaciona-se com um estado

transcricional represso. Assim, em quase todos os organismos, o estado heterocromático

está associado com hipoacetilação de histonas (Moazed, 2001). Muitos dos fatores

necessários para a montagem da heterocromatina incluem tanto enzimas que diretamente

modificam as histonas, quanto fatores que se ligam às histonas modificadas ou não. Em

leveduras de brotamento, os produtos dos genes reguladores de informação silente (Sir –

silent information regulator) Sir2, Sir3 e Sir4 são necessários para a montagem da

heterocromatina nas regiões teloméricas e locus mating-type. As proteínas Sir formam um

complexo, sendo que as subunidades Sir3 e Sir4 desse complexo podem se ligar às caudas

de histonas desacetiladas in vitro (Carmen et al., 2002; Hope, 2002). A subunidade Sir2 é

uma desacetilase de histona dependente de nicotinamida adenina dinucleotídeo (NAD), e

sua atividade de desacetilase é necessária para a montagem da heterocromatina (Guarente,

2000; Moazed, 2001). Em leveduras de fissão e metazoários, várias desacetilases de

histonas, incluindo ambas uma tipo Sir2 dependente de NAD e outra tipo Hda1/Rpd3

indepentende de NAD são necessárias para o silenciamento (Shankaranarayana et al.,

2003).

Além da hipoacetilação de histonas em leveduras de fissão, Drosophila e

mamíferos, a metilação da lisina 9 da histona H3 (H3K9) correlaciona-se com a montagem

da heterocromatina. Este resíduo é metilado pelas metiltransferases conservadas Su(var)3-9

em Drosophila, SUV39H1 em humanos e Clr4 nas leveduras de fissão (Nakayama et al.,

2001). Essas metiltransferases de H3K9 estão associadas com outra proteína conservada

7

chamada Swi6 nas leveduras de fissão e HP1 em Drosophila e humanos, respectivamente

(Aagaard et al., 1999). As proteínas Swi6 e HP1 ligam-se especificamente às caudas das

histonas H3 que tenham sido metiladas na lisina 9 pelas enzimas Clr4/Suv39h em levedura

de fissão e metazoários, respectivamente (Bannister et al., 2001). As proteínas HP1

(heterochromatin protein), uma classe de proteínas multifuncionais adaptadoras associadas

à cromatina, estão presentes em blocos de heterocromatina constitutiva em diversos

eucariotos, onde acredita-se que sejam importantes na regulação do silenciamento mediado

por heterocromatina e estrutura cromossômica. Há três diferentes proteínas HP1 (α, β e γ)

em mamíferos. HP1α e β estão concentradas na heterocromatina pericentromérica, embora

HP1β também possa ser encontrada em sítios nucleoplasmáticos mais difusos, enquanto

HP1γ está localizada predominantemente na eucromatina. Este padrão de distribuição é

similar àquele encontrado em Drosophila (Eissenberg & Elgin, 2000).

Acredita-se que a metilação de DNA também contribua para a estabilidade dos

estados de cromatina silente em eucariotos superiores com genomas complexos. Evidências

à partir de plantas e fungos sugerem a existência de mecanismos de retroalimentação entre

a metilação de DNA e de histonas, sendo que uma promova a manutenção da outra (Soppe

et al., 2002). A interdependência dessas marcas epigenéticas sugere que a metilação de

DNA e os mecanismos epigenéticos mediados pelas histonas atuam em sintonia para

manter o estado silente da cromatina.

Como os domínios heterocromáticos são direcionados a um domínio cromossômico

específico? Embora o papel de sítios regulatórios específicos, como por exemplo

silenciadores e proteínas que se ligam a sequências específicas do DNA, esteja bem

documentado (Moazed, 2001), há evidência da existência de um papel dos elementos de

DNA repetitivo e RNA não codificantes no direcionamento regional dos complexos

heterocromáticos (Henikoff, 2001). Acredita-se que os transposons e o DNA satélite, os

quais compõem a principal porção das seqüências heterocromáticas, preferencialmente

recrutam a maquinaria heterocromática, levando ao silenciamento dos genes vizinhos.

Ainda não se sabe como os elementos repetitivos e os transposons atraem a

heterocromatina, mas acredita-se que a natureza repetitiva desses elementos seja importante

(Selker, 1999). Em Schizosaccharomyces pombe, o elemento repetitivo homólogo

8

centromérico é um centro de nucleação da heterocromatina, e que está envolvido no

silenciamento regional de um domínio de aproximadamente 20kb (Hall et al., 2002).

Moléculas de RNA não codificante de vários tamanhos parecem ter um papel amplo na

regulação do comportamento cromossômico. Por exemplo, tais RNAs têm um papel

importante na localização cromossômico-específica de enzimas que modificam a

cromatina, e que são necessárias para a compensação de dose em Drosophila e em

humanos (Park & Kuroda, 2001), e em alguns casos, impressão genômica (Sleutels, 2002).

A via de interferência por RNA, adicionalmente ao já bem conhecido papel no

silenciamento gênico (Hannon, 2002), está também envolvida na iniciação da montagem da

heterocromatina nas sequências repetitivas de DNA (Hall et al., 2002; Volpe et al., 2002).

Estudos genéticos e bioquímicos em S. cerevisiae e S. pombe foram muito

importantes na elucidação de como se forma a heterocromatina. Aprendeu-se que a

montagem da heterocromatina é feita em sítios regulatórios específicos, e esta então se

espalha para as seqüências vizinhas de uma forma que requer o acoplamento de enzimas

modificadoras de histonas e proteínas estruturais como Sir3, Sir4 e Swi6/HP1 (Grewal &

Elgin, 2002; Moazed, 2001). Em S. cerevisiae, proteínas que reconhecem seqüências de

DNA ligam-se a sítios de nucleação (silenciadores) e então recrutam o complexo Sir2/Sir4

para o DNA. A proteína Sir2 desacetila as caudas de histonas, criando dessa forma um sítio

de ligação para as proteínas Sir3 e Sir 4. Estas, por sua vez, podem oligomerizar e, uma vez

ligadas, recrutar outros complexos Sir2/Sir4. Ciclos seqüenciais de ligação e desacetilação

então resultam no espalhamento dessas proteínas silenciadoras ao longo da fibra de

cromatina e além do sítio original de nucleação (Hoppe et al., 2002; Rusche et al., 2002).

Em S. pombe, contrariamente ao que acontece em S. cerevisiae, sequências repetitivas

especializadas e mecanismos de interferência por RNA atuam conjuntamente para iniciar a

formação da heterocromatina. A ligação de um componente de RNA à cromatina recruta

enzimas com atividade modificadora de histonas. Este recrutamento inicia a montagem da

heterocromatina através da criação de um código de histonas que será responsável pela

ligação de fatores de silenciamento (Hall et al., 2002). Especificamente, a desacetilação,

seguida da metilação da lisina 9 da histona H3 cria um sítio para a ligação de Swi6/HP1

(Nakayama et al., 2001; Shankaranarayana et al., 2003). Além do mais, a desacetilação da

9

lisina 14 da histona H3 parece ser importante para o silenciamento e localização de Swi6

nos sítios heterocromáticos nas leveduras de fissão. Contudo, a metilação de H3K9 e

H3K14 pode ocorrer concomitantemente em alguns pontos do genoma onde estas são

necessárias para a ativação transcricional de certos loci gênicos (Grewal & Moazed, 2003).

Uma vez ligadas à cromatina, Swi6/HP1 recrutam enzimas modificadoras de histonas para

criar sítios de ligação adicionais para Swi6/HP1 nos nucleossomos adjacentes (Hall et al.,

2002), o que permite a essas modificações de histonas e Swi6/HP1 espalharem-se além dos

sítios iniciais de nucleação (Grewal & Elgin, 2002).

Devido a essa capacidade de espalhamento da heterocromatina, as células

desenvolveram mecanismos antagônicos que protegem as regiões ativas dos efeitos

repressores da heterocromatina circundante. Elementos especializados de DNA conhecidos

como elementos de barreira demarcam as fronteiras entre domínios adjacentes de cromatina

e servem como barreiras contra os efeitos dos silenciadores e enhancers das regiões

vizinhas (Labrador & Corces, 2002; West et al., 2002). Nas leveduras de brotamento, as

proteínas silenciadoras Sir2/3/4 ficam restritas aos loci mating-type por causa da presença

de elementos de barreira (Dhillon & Kamakaka, 2002). Assim, tais elementos ajudam a

separar domínios de cromatina com padrões distintos de modificação de histonas e servem

para conter a heterocromatina dentro de um domínio particular (Figura 3).

Figura 3. O silenciamento de genes dentro da heterocromatina é mediado pelo direcionamento de HP1 aos sítios heterocromáticos. Ac – cauda de histona acetilada; HP1 – heterochromatin protein 1; Me – cauda de histona metilada. (adaptado de www.biochemsoctrans.org/bst/031/0741/bst0310741f01.htm?resolution=STD)

10

Como esses elementos trabalham? Diversos estudos demonstraram que há um pico

de metilação de H3K4 e de acetilação de histonas na cromatina imediatamente circundante

a um bloco de cromatina condensada, que é rico em metilação de H3K9 (Litt et al., 2001).

Tais modificações interrompem o avanço dos complexos heterocromáticos (Dhillon &

Kamakaka, 2002). Além disso, esses elementos podem servir como sítios de partida para o

recrutamento de acetiltransferases ou enzimas que remodelam a cromatina e que atuam

inibindo ou eliminando a ligação de proteínas silenciadoras às histonas e assim bloqueando

o espalhamento das estruturas heterocromáticas. Adicionalmente, acredita-se que os

elementos de barreira dividam a cromatina em unidades funcionais através da formação das

alças de cromatina, ao interagirem entre si ou com outra estrutura nuclear, como a matriz

nuclear (Labrador & Corces, 2002).

Territórios cromossômicos

Os modelos atuais propostos acerca da arquitetura nuclear em mamíferos mostram

que os cromossomos no núcleo celular estão organizados em regiões distintas denominadas

territórios cromossômicos (CTs - chromosome territories) (Figura 4) (Cremer et al., 1988).

Posteriormente, a descrição do corpúsculo de Barr e a análise comparativa dos CTs do

cromossomo X nas suas formas ativa e inativa (Eils et al., 1996), originaram a idéia de que

a estrutura de um CT correlaciona-se fortemente com seu estado funcional.

Os CTs são construídos à partir de uma hierarquia de fibras de cromatina de

espessuras crescentes, desde as fibras de 10 e 30 nm até o cromonema, com 100-130 nm

(Belmont & Bruce, 1994). No entanto, a forma pela qual essas fibras são empacotadas em

estruturas cromatínicas de hierarquia superior ainda é motivo de debate.

Os territórios cromossômicos podem ser vistos utilizando-se sondas específicas para

pintura cromossômica ou por hibridação in situ genômica (GISH) (Cremer & Cremer,

2001). Esses estudos mostraram a variabilidade na forma e posicionamento dos territórios

cromossômicos. Em leveduras, Drosophila e trigo, os cromossomos se espalham por todo

o núcleo, formando estruturas na forma de bastão com os dois braços de cada cromossomo

próximos uns dos outros e os centrômeros e telômeros localizados em polos opostos (confi-

Figura 4. Modelo de organização nuclear proposto por Kosak e Groudine e modificado à partir de Cremer e col. (2000). Note que este modelo implica um grande empacotamento dos CTs com um aumento concomitante na superfície destes e um aumento do domínio intercromatínico (ICD).

guração tipo Rabl) (Abranches et al., 1998).

mamíferos geralmente possuem uma distribuição radial, onde os territórios cromossômicos

estão ou perifericamente ou centralmente posicionados e possuem uma forma irregular

(Cremer & Cremer, 2001). A marcação seletiva de cromossomos individuais sugere que os

territórios cromossômicos est

(Visser et al., 2000), embora a prova definitiva disso seja difícil de se obter.

O grau de compactação de cada CT é dependente da riqueza de genes presentes e do

grau de atividade desses genes (Stadler et al., 2004). Incorporação de bromo

ativos de transcrição, seguida por imunofluorescência para detectar BrUTP, mostrou que a

transcrição ocorre em muitos focos pequenos em diversos sítios distintos. No trigo, esses

sítios de transcrição encontram

os nucléolos contém uma concentração muito maior de sítios transcricionalmente ativos e

intensamente marcados por BrUTP (Abranches et al., 1998).

nenhuma evidência foi encontrada para uma localização preferencial dos sítios de

transcrição próximos à periferia dos territórios cromossômicos, ou exclusão desses sítios do

Alça, locus ativo

IGC

Transcritos de RNA

Compartimento intercromatínico

Locus ativo

DNA intergênico

Alça, locus pronto

11

. Modelo de organização nuclear proposto por Kosak e Groudine e modificado à partir de Cremer e

(2000). Note que este modelo implica um grande empacotamento dos CTs com um aumento concomitante na superfície destes e um aumento do domínio intercromatínico (ICD).

(Abranches et al., 1998). Diferentemente, os cromossomos de

geralmente possuem uma distribuição radial, onde os territórios cromossômicos



ios cromossômicos estejam geralmente separados em espaços não sobrepostos

(Visser et al., 2000), embora a prova definitiva disso seja difícil de se obter.

grau de compactação de cada CT é dependente da riqueza de genes presentes e do

esses genes (Stadler et al., 2004). Incorporação de bromo-



s de transcrição encontram-se uniformemente distribuídos pelo nucleoplasma, enquanto


intensamente marcados por BrUTP (Abranches et al., 1998). Ainda no caso do trigo,

enhuma evidência foi encontrada para uma localização preferencial dos sítios de


DNA intergênico Lâmina nuclear

Sub-domínio de território

Locus inativo

PML-NB

Território cromossômico

Corpúsculo de Cajal

heterocromatina

Alça, locus inativo

Domínio regulatório

Alça, locus pronto

. Modelo de organização nuclear proposto por Kosak e Groudine e modificado à partir de Cremer e

(2000). Note que este modelo implica um grande empacotamento dos CTs com um aumento concomitante

, os cromossomos de

geralmente possuem uma distribuição radial, onde os territórios cromossômicos



geralmente separados em espaços não sobrepostos

grau de compactação de cada CT é dependente da riqueza de genes presentes e do

-UTP em sítios



se uniformemente distribuídos pelo nucleoplasma, enquanto


o caso do trigo,

enhuma evidência foi encontrada para uma localização preferencial dos sítios de


NB

Território cromossômico

Corpúsculo de Cajal

12

interior dos territórios. Já em mamíferos, Kurz e col. (1996), revelaram que os genes

localizam-se preferencialmente na periferia dos territórios cromossômicos. Este

posicionamento independe da atividade do gene, enquanto fragmentos que não são

expressos foram localizados randomicamente dispersos ou localizados preferencialmente

no interior de seu território cromossômico. A síntese de RNA foi localizada na periferia de

regiões de cromatina condensada, em fibrilas pericromatínicas (Cmarko et al., 1999), que

poderiam ser sítios de transcrição na cromatina das alças que se estendem além do território

cromossômico. É importante salientar que, a densidade e a atividade gênica local, ao invés

da atividade de genes individuais, são os fatores que influenciam a organização dos

cromossomos no núcleo (Mahy et al., 2002), o que está de acordo com a idéia de que há um

agrupamento, em regiões próximas do genoma, de genes relacionados com os mesmos

processos metabólicos. Existe consenso, que, exceto pelos nucléolos e grânulos

intercromatínicos, quase todo o RNA nascente co-localiza com domínios de cromatina e

que não há localização preferencial em áreas que não contenham cromatina (Sadoni &

Zink, 2004).

Como visto anteriormente, o genoma é formado por segmentos muito longos de

DNA denominados isocoros, os quais podem ser diferenciados em termos de seu conteúdo

GC. Quando os territórios cromossômicos são estudados em termos da localização de seus

isocoros, vê-se que os isocoros ricos em GC, que no caso de humanos e camundongos, são

os mais ricos em genes, estão distribuídos mais aleatoriamente pelo território, enquanto os

isocoros ricos em AT, geralmente associados a seqüências heterocromáticas e pobres em

genes, estão localizados mais para o interior dos territórios (Bernardi, 1993; Tajbakhsh et

al., 2000). Todos os dados apontam para o fato de que, mesmo que a estrutura da maioria

dos CTs seja permissível à transcrição, a organização dos genes e do DNA em cada

território não é randômica.

A matriz nuclear

Como visto anteriormente, o espaço nuclear, delimitado pela lamina nuclear, pode

ser dividido em dois compartimentos, o território cromossômico e o espaço restante, ou

13

domínio intercromatínico (ICD - interchromatin domain). Em uma interpretação moderna,

o ICD é um equivalente da matriz nuclear quando observado em núcleos não fracionados

ou em células vivas (Figura 5).

A matriz nuclear foi primeiramente isolada e caracterizada por Berezney e Coffey

em 1977. É uma estrutura essencialmente protéica, sendo que mais de 97% de seu conteúdo

corresponde a proteínas, mas também contém traços de DNA, RNA e fosfolipídeos

(Berezney & Coffey, 1977). Esses componentes formam uma rede constituída pelo

complexo lâmina-poro nuclear, estrutura residual do nucléolo, e uma rede fibrilar-globular

(Berezney, 1991). A lâmina nuclear pode ter um importante papel na organização funcional

da cromatina interfásica (Bode et al., 2003).

De fato, dois componentes podem ser identificados na matriz nuclear, uma matriz

interna, contendo polipeptídeos não histônicos e não lamínicos, e que compreendem

aproximadamente 20% das proteínas nucleares totais, as quais formam uma rede

intranuclear adicionalmente a componentes do nucléolo, e uma matriz periférica, que

compreende proteínas do envelope nuclear, como as laminas (Kaufmann & Shaper, 1984).

Cerca de 2% da matriz nuclear interna e 6% da matriz nuclear periférica são

glicoproteínas ricas em resíduos de glicose e/ou manose. Essas glicoproteínas podem ter

um papel importante na estabilidade da matriz nuclear, pois a retirada parcial de seus

resíduos de carboidratos causa uma desestruturação relativamente alta na matriz nuclear.

Assim, é possível que essas glicoproteínas estejam localizadas em pontos chave da

estrutura (Ferraro et al., 1994). De fato, muitas funções importantes são atribuídas às

glicoproteínas nucleares como será abordado posteriormente.

Entretanto, a maioria das proteínas da matriz nuclear compreendem proteínas não-

histônicas, incluindo laminas, ribonucleoproteínas, e um grupo de proteínas pouco

caracterizadas e presentes em pouca quantidade (Stuurman et al., 1990). Exceto pelas

proteínas de matriz nuclear que são constitutivamente expressas, muitas dessas proteínas

são específicas para cada tipo celular e sua produção é dependente de ação hormonal,

diferenciação ou transformação celular (Yam et al., 2002).

O processo de isolamento da matriz nuclear envolve tratamentos com nucleases e

extração com alta concentração salina, os quais removem a cromatina e outras proteínas

14

Figura 5. Representação esquemática dos compartimentos nucleares (adaptado de Tsutsui et al., 2005).

nucleares solúveis (Nickerson, 2001). Um fator importante de ser mencionado é que até

proteínas naturalmente solúveis adquirem graus variados de insolubilidade por causa dos

processos aos quais os núcleos são submetidos durante a purificação da matriz nuclear,

resultando então na presença de proteínas nos preparados. Existem diversos processos de

purificação e, dessa forma, não é surpresa que o uso desses diferentes processos resulte em

matrizes nucleares com diferentes composições. Numa visão simplista, todas as proteínas

nucleares que não aquelas associadas com a cromatina podem ser consideradas proteínas de

matriz nuclear. A copurificação de certas proteínas com a matriz nuclear pode não ser

significante por si só, a menos que outros achados corroborem essa conclusão. Se a matriz

nuclear contém ou não uma estrutura filamentosa assim como o citoesqueleto, ainda é tema

de debate. No entanto, parece plausível a idéia de que exista uma estrutura fibrilar

formando uma rede que preenche todo o núcleo e que serve como um arcabouço para a

montagem funcional de proteínas (Tsutsui et al., 2005).

O compartimento nuclear é inevitavelmente um espaço preenchido, no qual o RNA

é transcrito, processado e transportado para o citoplasma, e o DNA é replicado e reparado.

Esse compartimento é também uma parada transitória para proteínas entrando e saindo do

15

núcleo através dos complexos de poro. Adicionalmente, esse compartimento contém

corpúsculos nucleares como os corpúsculos da leucemia promielocítica (PML), corpúsculos

de Cajal, espículas nucleares e regiões transientes nas quais proteínas funcionais são

montadas dependendo do estado fisiológico celular, por exemplo, crescimento,

diferenciação ou estresse. Esta idéia é consistente com diversos relatos prévios mostrando

conexões plausíveis entre a matriz nuclear e funções nucleares múltiplas (Tsutsui et al.,

2005).

A fibra de cromatina de células eucarióticas está organizada em grandes domínios

ou alças separados através de suas interações com a matriz nuclear (Figura 5). As

seqüências específicas de DNA na base das alças são denominadas regiões de adesão à

matriz (MARs – matrix attachment regions) ou regiões de adesão ao arcabouço nuclear

(SARs – scaffold attachment regions) (Paulson & Laemmli, 1977; Mirkovitch et al., 1984).

Bode e col. (1995) criaram o termo S/MARs para unificar os dois nomes atualmente

utilizados. As alças de DNA são formadas por causa das regiões de adesão à matriz. O

comprimento das alças de DNA varia de acordo com a distância relativa entre as S/MARs,

estando entre 5 e 200 kb (Razin et al., 1995), ao passo que as S/MARs têm um

comprimento que varia de 500 a 3000 pares de bases (Vassetzky et al., 2000), mas podem

ser bem maiores em isocoros ricos em AT (Svetlova et al., 2001). Estima-se que haja em

torno de 50.000 S/MARs por núcleo (Razin, 2001). Embora se saiba que as S/MARs sejam

responsáveis pelo aparecimento das alças de cromatina, sua existência por si só não é

suficiente para tal; ou seja, a formação das alças depende de ambos, requerimentos

celulares e presença de S/MARs numa determinada região (Heng et al., 2004).

As S/MARs são fragmentos que podem se ligar especificamente a matrizes

nucleares isoladas, estruturas residuais produzidas pela extração total das histonas

nucleares. Dois tipos de S/MARs foram descritos: permanentes (constitutivas), que contêm

DNA regulatório não transcrito, e transientes (facultativas), que contêm DNA que é

replicado e transcrito (Iarovaia et al., 2005). As S/MARs estão localizadas mais

freqüentemente em regiões de DNA não codificante contendo elementos potencialmente

regulatórios e sítios de ligação à DNA topoisomerase II (Adachi et al., 1989).

16

A dinâmica de associação/dissociação das S/MARs é governada por três princípios:

1. a proporção entre o DNA associado à matriz e o presente nas alças depende tanto da

força de ligação das S/MARs quanto do número de cópias destas; 2. embora a associação

das S/MARs com a matriz nuclear seja um processo dinâmico, transgenes presentes em

poucas cópias podem ser acomodados na matriz a maior parte do tempo; 3. quando

presentes em muitas cópias, a maioria dos genes localiza-se nas alças, onde eles perdem a

competência transcricional (Bode et al., 2003).

Com relação à sua seqüência nucleotídica, não foram reveladas características

comuns entre as S/MARs, mas estudos prévios indicam que os genes para os RNAs

ribossomais estão enriquecidos no DNA associado à matriz nuclear (Pardoll & Vogelstein,

1980). Estudos mais recentes, no entanto mostraram que as S/MARs geralmente

apresentam seqüências ricas em AT, como por exemplo os motivos (AT)n e ATATTT, que

geralmente são clivadas/protegidas por proteínas que se ligam às S/MARs (Razin, 2001).

Há diferentes métodos para se isolar as S/MARs (ex.: os ensaios in vitro e in vivo

[revisado por Fiorini et al., 2005]). No ensaio in vitro os núcleos são tratados com DNase I,

e as proteínas histônicas e algumas não histônicas são extraídas com NaCl 2M. As matrizes

nucleares são então incubadas com fragmentos marcados de DNA exógeno, e que se ligam

às regiões de interesse. O DNA ligado é coletado por centrifugação, purificado e

visualizado após eletroforese em gel de agarose e autoradiografia (Cockerill & Garrard,

1986; Izaurralde et al., 1988).

A detecção in vivo (Fig. 6) separa as seqüências de DNA em fragmentos associados

e não associados à matriz nuclear. Neste ensaio, os núcleos são extraídos com 3,5-

diiodosalicilato de lítio (LIS), e o DNA liberado é digerido com uma combinação de

enzimas de restrição. As matrizes são recuperadas por centrifugação e as frações em relação

ao DNA ligado e não ligado à matriz são detectadas por “Southern blotting” ou PCR

(Gasser Laemmli, 1986; Razin, 2001; Ostermeier et al., 2003) utilizando-se sondas

específicas, ou visualizadas por eletroforese. Estima-se que cerca de 1% do DNA total

encontra-se associado à matriz nuclear (Pardoll & Vogelstein, 1980).

17

As alças de DNA e as S/MARs podem ser visualizadas por hibridação in situ

fluorescente (FISH), indicando as propriedades estruturais dessas regiões (Gerdes et al.,

1994; Iarovaia et al., 2005). Recentemente, foi relatada uma nova metodologia para

mapeamento das S/MARs em grandes segmentos de DNA, baseando-se na tecnologia de

microarranjos de DNA (Ioudinkova et al., 2005). Essa metodologia se mostrou muito

poderosa devido à facilidade de acesso à seqüência de diversos genomas nos bancos de

dados disponíveis.

A matriz nuclear está envolvida na organização da cromatina, assim como na

replicação e expressão gênica (Bode et al., 2003; Moraes et al., 2005). As S/MARs estão

envolvidas na modulação da transcrição em mamíferos e plantas (Whitelaw et al., 2000),

além de terem sido identificadas experimentalmente em diversos loci gênicos, incluindo os

genes rDNA humanos (Loc & Stratling, 1988; Fukuda & Nishikawa, 2003; Ioudinkova et

al., 2005), ou associadas a origens de replicação em eucariotos (Razin et al., 1986;

Ostermeier et al., 2003). Proteínas que se ligam às S/MARs estão diretamente envolvidas

no controle transcricional (Nepveu, 2001). Muitas delas ligam-se a seqüências ricas em AT

que possuem uma tendência ao dobramento, ligando estrutura da cromatina, transcrição e

processamento de mRNA (Razin et al., 1986; Fukuda, 2000; Yamamura & Nomura, 2001).

A correlação entre o número de S/MARs e o nível de expressão gênica ilustra a importância

Figura 6. Interação entre as S/MARs e a matriz nuclear. Procedimento experimental para separar o DNA nuclear em duas frações: DNA presente nas alças e DNA associado à matriz nuclear (modificado à partir de Tsutsui et al., 2005).

18

da interação direta com a matriz nuclear. Parece que tal interação mediada por ancoragem é

essencial para a expressão gênica (Figura 7).

Fibras estendidas de cromatina

Como descrito anteriormente, o processo de preparo de matrizes nucleares envolve

extração salina e tratamento com nucleases, os quais removem a cromatina e as proteínas

nucleares solúveis (Berezney & Coffey, 1977; Nickerson, 2001). Após o processo de

extração, os núcleos adquirem a forma de halos nucleares, nos quais o DNA “vaza” para

fora da estrutura nuclear residual (matriz nuclear) na forma de alças superenoveladas

negativamente. Se os núcleos são tratados verticalmente sob a ação da gravidade, eles

podem produzir fibras estendidas de cromatina (ECFs - extended chromatin fibers), cuja

formação pode ser fortemente afetada pelas proteínas da matriz nuclear através da sua

contribuição para a estrutura de ordem superior da cromatina (Gerdes et al., 1994; Cremer

et al., 2000; Vidal 2000). Um maior número de interações entre a cromatina e a matriz

nuclear pode alterar a formação de ECFs (Moraes et al., 2005).

Figura 7. Modelo proposto para o uso seletivo das S/MARs na regulação da transcrição/replicação. O painel da esquerda mostra um gene localizado numa alça e que contém uma S/MAR (verde). Quando a demanda funcional requer a associação específica deste gene com a maquinaria de transcrição localizada na matriz nuclear, a S/MAR move o gene para a matriz nuclear, iniciando assim a transcrição (painel central). Após a iniciação, o gene é empurrado através da maquinaria transcricional completando assim o processo (painel direito). Há dois tipos de S/MARs. As funcionais ou facultativas (verde) servem como mediadores para trazer genes até a matriz nuclear. As S/MARs estruturais ou constitutivas (vermelho) servem como âncoras, que são menos dinâmicas se comparadas com as S/MARs funcionais (adaptado de Heng et al., 2004).

A observação dos halos nucleares e das ECFs reflete a organização do DNA de

eucariotos superiores em grandes domínios ou alças de cromatina, os quais permanecem

19

aderidos à matriz nuclear através de suas bases, que contém as S/MARs (Paulson &

Laemmli, 1977; Iarovaia et al., 2004). Levando-se em consideração que o número de

S/MARs ligadas à matriz nuclear depende do número de genes transcricionalmente ativos,

a produção de ECFs pode ser utilizada como uma ferramenta para estimar diferenças na

atividade de certos genes, ou até mesmo do estado global de atividade transcricional de um

tipo celular, em diferentes condições fisiológicas.

Glicoproteínas nucleares

Acreditava-se que radicais glicosídicos em proteínas estivessem quase que

exclusivamente localizados na superfície celular de compartimentos tubulares. Atualmente,

sabe-se que essas proteínas estão largamente espalhadas pela célula, estando presentes

inclusive no núcleo celular (Hart et al., 1989 - revisão). Esses achados são baseados em

estudos que se utilizaram da capacidade ligante de carboidratos das lectinas, análises da

composição de núcleos isolados, ou radiomarcação metabólica com precursores de açúcares

seguida por fracionamento dos componentes subcelulares marcados.

Há evidências convincentes para a existência de glicoproteínas nucleoplasmáticas

desde a descoberta de alguns tipos novos de glicoconjugados. Resíduos únicos de N-

acetilglicosamina são ligados glicosidicamente às hidroxilas dos resíduos de serina ou

treonina (O-GlcNAc) nas proteínas nucleares, incluindo proteínas nas faces dos poros

nucleares (Park et al., 1987) e fatores de transcrição (Jackson & Tjian, 1988).

As lectinas, que são proteínas ligadoras de carboidratos, são sondas bastante úteis

para detectar a presença de sacarídeos nas superfícies celulares, em componentes do

interior da célula, e em glicoproteínas purificadas (Hawkes, 1982). Um exemplo de lectina

com uma especificidade de ligação cruzada é a concanavalina A (Con A), que liga-se tanto

a oligossacarídeos ricos em manose e glicose quanto contendo inositol (Wassef et al.,

1985). Utilizando lectinas marcadas radioativamente em núcleos isolados, encontrou-se

evidência de estruturas contendo manose na face citoplasmática de núcleos isolados de

bovinos (Nicolson et al., 1972).

20

Sítios ligantes de lectina foram demonstrados na cromatina isolada e preparados de

matriz nuclear. Análises da ligação de Con A na cromatina purificada de núcleos de ratos

indicam sítios de ligação a cada 1400 pb de DNA (Rizzo et al., 1977).

Electroblots de frações cromossômicas não histônicas marcadas com 125I-Con A

detectaram três polipeptídeos principais de 135, 125 e 69 kD. As histonas isoladas dos

macronúcleos do protozoário, Tetrahymena thermophila, parecem conter resíduos de

manose, baseando-se na sua especificidade pela Con A (Levy-Wilson, 1983). A Con A

também cora o nucléolo e o nucleoplasma de folículos ovarianos em Lacerga vivipara

(Seve et al., 1986). Foram encontradas glicoproteínas em núcleos das células exócrinas

colunares pancreáticas através de estudos com microscopia eletrônica de crio-fratura e

lectinas como sondas conjugadas com ouro coloidal ou peroxidase (Kan & Pinto da Silva,

1986). Neste caso, a Con A ligou-se preferencialmente à eucromatina, levantando então a

conclusão de que glicoproteínas contendo manose parecem estar mais presentes em sítios

de cromatina ativa (Kan & Pinto da Silva, 1986; Ferraro et al., 1991). Adicionalmente, as

glicoproteínas nucleares podem também ser componentes da matriz nuclear, já que elas são

encontradas também em matrizes nucleares isoladas (Ferraro et al., 1991).

Em núcleos de hepatócitos de camundongos, glicoproteínas reativas à Con A podem

ter um papel importante na formação da heterocromatina pericentromérica, já que elas

foram encontradas citologicamente concentradas nos cromocentros (Vidal et al., 1997). A

detecção citológica de glicoproteínas reativas à Con A em núcleos de hepatócitos após um

período de jejum de 48 h é drasticamente diminuída, possivelmente por causa de um

aumento no empacotamento da cromatina e degradação protéica. A realimentação quase

que restabeleceu a presença dessas glicoproteínas (Moraes et al., 2005), mostrando que o

estado fisiológico da célula pode influenciar o conteúdo glicoprotéico nuclear.

2. Organização da cromatina em células de camundongo

Hepatócitos

21

O genoma de camundongos é aproximadamente 14% menor que o genoma humano,

mas ambos possuem mais de 90% de regiões que podem ser consideradas como sintênicas,

com ambos organismos contendo mais de 30.000 genes que codificam proteínas.

Como visto anteriormente, o genoma dos mamíferos pode ser dividido em

fragmentos muito longos de DNA chamados isocoros, os quais podem ser classificados de

acordo com seu crescente teor em bases GC.

Semelhantemente ao genoma humano, em camundongos, aproximadamente 60%

dos pares de bases do DNA são ricos em AT sobrando, portanto, à porção do genoma rica

em GC cerca de 40% do DNA. Comparativamente a outros organismos, esses genomas

podem ser considerados como ricos em sequências AT. De modo interessante, 70 a 80%

dos genes residem na metade mais rica em GC do genoma. Ou seja, esses 40% do genoma

de camundongos e humanos abrigam a grande maioria dos genes, o que mostra uma

correlação bastante positiva entre conteúdo GC e densidade gênica (Waterston et al., 2002).

No caso de camundongos, a maior parte do DNA rico em AT está localizado nos

domínios heterocromáticos, sendo eles o DNA satélite da cromatina pericentromérica e os

telômeros.

A heterocromatina pericentromérica é caracterizada por baixos níveis de acetilação,

e altos níveis de metilação de H3K9, seguida pela ligação de HP1, a qual é também

dependente de um componente de RNA ainda não identificado. Metilação de DNA

geralmente é encontrada como um produto da metilação de histonas ou como uma causa

desta (Maison et al., 2002; Lehnertz, 2003).

Em camundongos, dois tipos de sequências de DNA repetitivo estão associadas com

os centrômeros, as repetições satélite principais (6 megabases de unidades com 234 pb

cada) e as repetições secundárias (600 kb de unidades com 120 pb cada). As sequências

principais estão localizadas pericentroméricamente, enquanto as secundárias coincidem

com a constrição centromérica. As repetições primárias são responsáveis pela formação dos

cromocentros (descritos a seguir), enquanto as secundárias localizam-se em torno destes na

forma de diversas entidades separadas. O acúmulo de HP1 é restrito às repetições

primárias, enquanto nas secundárias há acúmulo da proteína centromérica CENP. Embora

ambas as regiões sejam ricas em metilação de H3K9, estas apresentam diferente

22

sensibilidade à digestão enzimática, o que mostra que essas duas regiões estão associadas à

distintas organizações de alta ordem da cromatina (Guenatri et al., 2005).

Em camundongos, as fibras de cromatina centroméricas são mais condensadas que o

restante da cromatina, enquanto a cromatina pericentromérica apresenta um grau

intermediário de compactação. Foi proposto que a cromatina satélite possui uma

conformação helical regular compatível com o modelo da fibra de cromatina de 30 nm,

enquanto o restante da cromatina apresenta-se menos regularmente empacotado e

provavelmente recheado de deformações. Esta conformação de alta ordem da fibra de

cromatina nos domínios centroméricos pode ter um papel importante na formação da

heterocromatina e na determinação da identidade centromérica (Gilbert & Allan, 2001).

O DNA dos telômeros de vertebrados é bastante simples, sendo composto de

pequenas repetições em tandem da seqüência 5’(TTAGGG)3’. O comprimento do arranjo

telomérico é geneticamente regulado e depende da espécie, medindo aproximadamente 15

kb nas células germinativas humanas, e é um tanto menor que o dos telômeros de tecidos

somáticos, dependendo da idade e da história replicativa. Já camundongos possuem

telômeros consideravelmente mais longos, os quais podem se estender por mais de 50 kb

(Broccoli, 2004 – revisão).

A cromatina telomérica é organizada com um espaçamento menor entre os

nucleossomos se comparada ao restante da cromatina. Sabe-se há muito tempo que os

telômeros são heterocromáticos. A heterocromatina telomérica foi descrita em diversos

modelos como tendo a capacidade de silenciar a expressão de genes adjacentes num

fenômeno conhecido como efeito telomérico de posição (TPE –telomeric position effect).

Assim como outras regiões heterocromáticas, em camundongos, os telômeros contêm

proteínas HP1, as quais ligam-se à caudas de histonas metiladas pelas metiltransferases de

histonas SuVar(39)H1 e SuVar(39)H2 (Blasco, 2004).

Linhagens de células humanas e murinas geralmente apresentam seus telômeros

aderidos à matriz nuclear. Em mamíferos, um sítio de ligação à matriz nuclear ocorre a

cada 1000 pb do trato telomérico (Luderus et al., 1996). Isto indica que os telômeros de

mamíferos possuem múltiplos e frequentes pontos de interação com a matriz nuclear, sendo

que cerca de 50% do DNA telomérico encontra-se ancorado. Assim como a

23

heterocromatina telomérica, a heterocromatina pericentromérica também se ancora na

matriz nuclear (Pandita, 2002).

No interior do núcleo interfásico os cromossomos encontram-se não

randomicamente dispersos, mas orientados de maneira ordenada, ocupando territórios

nucleares muito bem definidos. De acordo com a hipótese de Rabl, os cromossomos

ocupam, durante a intérfase, domínios distintos, com todos os centrômeros apontados para

um pólo nuclear e todos os telômeros voltados para o pólo oposto. Este padrão básico,

denominado configuração ou orientação Rabl, foi descrito em diversos tipos celulares

(Comings, 1980). Há evidência da existência de variantes que se desviam do padrão básico

observado em núcleos interfásicos e descrito por Rabl. Em células humanas e de

camundongo os centrômeros apresentam uma distribuição espacial única (Hsu et al., 1971).

Em camundongos, e em alguns outros modelos, sabe-se que as regiões de

heterocromatina constitutiva tendem a associar-se formando cromocentros (Hochstrasser et

al., 1986). Na grande maioria dos eucariotos superiores, os domínios heterocromáticos

localizam-se próximos aos centrômeros, às regiões organizadoras de nucléolos (NORs), ou

aos telômeros. Aparentemente, o alto conteúdo de sequências de DNA repetitivo aumenta a

condensação e agregação que caracterizam as regiões heterocromáticas.

Uma relação espacial entre os cromocentros e os centrômeros e NORs foi

encontrada em células de camundongo. Os nucléolos sempre são encontrados associados

aos cromocentros, o que sugere que os cromossomos que contenham NORs estejam

participando na formação dos agregados centroméricos. Esta característica é uma

consequência da posição paracentromérica das NORs em cromossomos de camundongos

(Cerda et al., 1999).

Em células de camundongo, o número e o tamanho dos cromocentros variam de

acordo com o tecido e o tipo celular, o que sugere a existência de diferentes níveis de

organização heterocromática (Hsu et al., 1971). Sabe-se também que há mais de um

cromossomo portador de regiões NOR, e que esses cromossomos tendem a se associar

através da cromatina NOR, produzindo então um nucléolo que é compartilhado por vários

cromossomos. O número de nucléolos e a posição dos telômeros também variam bastante

24

de acordo com o tipo celular em camundongos, mas no caso dos nucléolos, seu número é

sempre o mesmo que o número de cromocentros presentes (Cerda et al., 1999).

As regiões NOR contém os genes para rRNA 45S, os quais, na maioria dos

eucariotos, são encontrados repetidos de 100 a 5000 vezes por genoma haplóide, e

localizam-se em um ou mais sítios cromossômicos denominados organizadores nucleolares

(Gaubatz et al., 1976; Long & Dawid, 1980). As regiões codificadoras de rRNA estão

organizadas como repetições todas no mesmo sentido, com as regiões transcritas separadas

por segmentos espaçadores não transcritos (Fedoroff, 1979). Em hepatócitos de ratos, após

tratamento com DNase I, a matriz nuclear permanece rica em sequências de rDNA,

indicando que a transcrição desses genes está relacionada com sua interação com a matriz

nuclear (Pardoll & Vogelstein, 1980).

Os genes para rRNA existem em duas formas distintas de estrutura cromatínica:

uma forma “aberta”, que corresponde aos genes transcricionalmente ativos, e uma forma

“fechada” representando os genes silentes. Estudos recentes indicam que uma rede

epigenética medeia o estado transcricional do rDNA. A interação entre a metilação de

DNA, modificações de histonas (por exemplo, acetilação de H3K9 nos genes ativos e

trimetilação de H3K9 nos inativos, seguida da ligação de HP1) e atividades remodeladoras

da cromatina estabelecem o silenciamento do locus de rDNA em eucariotos superiores

(Santoro, 2005 – revisão). Apesar do elevado número de cópias dos genes para rRNA no

genoma de eucariotos superiores, não mais que metade dela parece estar ativa, indicando

que nos eucariotos haja controle da dosagem efetiva desses genes.

Espermatozóides

Em espermatozóides, foram relatadas mudanças drásticas na composição da

cromatina quando comparados às células somáticas, sendo que as histonas somáticas são

parcialmente ou totalmente substituídas por proteínas básicas tipo-histona (Bloch, 1969;

Oliva & Dixon, 1991). Em espermatozóides contendo protaminas (ricas em arginina e

nenhuma lisina), histonas tipo-protamina (ricas em arginina, lisina pouco presente ou

ausente, cisteínas oxidadas), ou proteínas básicas intermediárias (contém histidina e/ou

25

lisina em adição à arginina), essas proteínas empacotam firmemente o DNA, enquanto em

alguns outros animais, os espermatozóides que contêm histonas tipo-somáticas e ricas em

lisina, com uma cromatina muito menos empacotada (Mello & Vidal, 1973; Oliva &

Dixon, 1991; Taboga et al., 1996; Falco & Mello, 1999). Histonas tipo-protaminas são

encontradas geralmente em mamíferos. Os espermatozóides de camundongos possuem duas

variantes tipo-protamina (Calvin, 1976).

A supraorganização da cromatina nos espermatozóides depende do tipo de proteína

básica nuclear envolvida na compactação do DNA e das interações deste com a matriz

nuclear. O DNA de espermatozóides também se organiza em alças que se aderem à matriz

nuclear através das S/MARs (Ward & Coffey, 1990). Como o DNA de espermatozóides

não está envolvido em processos transcricionais, as regiões de adesão do DNA à matriz

nuclear provavelmente possuem funções tipicamente estruturais (Santi et al., 1994).

Além da estrutura extremamente compacta da cromatina em espermatozóides de

mamíferos, diferentes hierarquias e heterogeneidade existem no empacotamento do DNA

no núcleo dessas células (Haaf & Ward, 1995). Assim, espera-se que os complexos DNA-

proteína dos espermatozóides de mamíferos apresentem um comportamento diferente

daqueles das células somáticas na mesma espécie, e que esta característica seja influenciada

pelo estado nutricional e estágio do desenvolvimento, assim como por tratamentos com

soluções com alta concentração salina e detergente, as quais desestabilizam esses

complexos.

3. Fatores que influenciam a organização nuclear e da cromatina

Jejum

O estado nutricional do organismo é capaz de induzir diversas mudanças na

estrutura da cromatina. A desnutrição ou jejum modificam o conteúdo das histonas do

núcleo octamérico e também da histona H1 e podem produzir hipometilação de DNA e

mudanças na atividade enzimática de topoisomerases e proteínas desestabilizadoras da

hélice de DNA, além de induzir um aumento generalizado no estado de agregação da

26

cromatina (Castro & Sevall, 1980; Castro et al., 1986; Amaral & Mello, 1989). O jejum

prolongado (5 dias) ou uma dieta pobre em proteínas, são capazes de alterar o conteúdo

protéico nucleoplasmático em hepatócitos (Paliga et al., 1991), e causam uma diminuição

de até 35% no conteúdo protéico total no fígado de camundongos, sendo que a

administração de uma dieta normal pós-jejum foi capaz de restaurar rapidamente as

proteínas perdidas, exceto pelas proteínas histônicas, que são sintetizadas numa velocidade

menor (Pucciarelli & Conde, 1984; Cassia & Conde, 1994). Parece que o jejum é capaz de

alterar também o conteúdo de glicoproteínas nucleares, já que a detecção citoquímica

dessas proteínas é drasticamente diminuída após um jejum de 48 h e, novamente, a

realimentação foi capaz de restabelecer a capacidade de localizar essas proteínas no núcleo

de hepatócitos (Moraes et al., 2005).

Após o jejum, a estrutura da cromatina ao redor de certas enzimas hepáticas também

é afetada, e há uma diminuição na hipersensibilidade dessas áreas à ação da DNase I, que

ocorre simultaneamente à formação de novos nucleossomos nessas áreas (Ma & Goodridge,

1992). Uma completa reorganização da cromatina em hepatócitos de camundongo, com

aumento dos níveis de compactação de áreas eucromáticas frente à heterocromatina foi

observado após jejum de 48 h. Neste caso, essa reorganização sofrida pela cromatina

modificou suas propriedades viscoelásticas, tornando essa cromatina menos fluida e mais

resistente a processos extrativos para a produção de fibras estendidas de cromatina e

isolamento da matriz nuclear sob condições específicas (Moraes et al., 2005).

Desenvolvimento e envelhecimento

Uma das marcas características do envelhecimento é a redução na expressão de uma

série de genes (Burzynski, 2005). O silenciamento genético, como visto anteriormente,

pode ser efetuado por diversos mecanismos complexos os quais incluem metilação de

DNA, modificação de histonas, e remodelação da cromatina. Essa remodelação é causada

por mudanças estruturais na cromatina ao longo do processo de envelhecimento, as quais

incluem aumento da resistência da cromatina à degradação por nucleases (Chaurasia et al.,

1996; Mahendra et al., 1999; Ranhotra & Sharma, 2001), ao mesmo tempo em que os

27

defeitos acumulados na cromatina durante a vida do organismo podem levar a uma

inadequada manutenção da integridade genômica e conseqüentes defeitos na formação de

heterocromatina (Bitterman et al., 2003).

A já bem conhecida perda de porções dos telômeros que ocorre ao longo do tempo

contribui para o processo de envelhecimento, pois as células perdem a proteção contra

rearranjos cromossômicos que podem levar ao aparecimento de doenças genéticas diversas

e câncer (Cong et al., 2002). Essa perda de material genético pode estar envolvida com uma

atividade diminuída de enzimas desacetiladoras de histonas, as quais atuam nos momentos

iniciais da formação da heterocromatina telomérica (Grunstein, 1997; Bitterman et al.,

2003). Tais enzimas são dependentes de NAD+ como um cofator, e este cofator tem sua

concentração bastante diminuída em células cujo metabolismo é baixo, o que é o caso das

células em estado de senescência.

Em animais idosos, essa diminuição do metabolismo está associada a uma

diminuição da produção de RNA, ribossômico e mensageiro, e de proteínas (Bolla &

Denckla, 1979; Ma & Nagata, 1990). De fato, quando camundongos de diferentes idades

são comparados, pode ser observado um pico de síntese de RNA no fígado de

camundongos jovens, e um decréscimo acentuado na síntese de RNA conforme o animal

envelhece.

28

Objetivos

Frente às informações apresentadas na Introdução, o presente trabalho teve os

seguintes objetivos:

1. Avaliar a influência do desenvolvimento e envelhecimento na supra-organização da

cromatina em hepatócitos de camundongo, e verificar se essas possíveis mudanças

estariam associadas com alterações nas propriedades viscoelásticas da cromatina.

2. Determinar os fatores epigenéticos ligados ao processo de envelhecimento, sua

distribuição nuclear, e como estes influenciam a estrutura da fibra de 30 nm da

cromatina e os seus níveis hierárquicos superiores de organização.

3. Avaliar se o envelhecimento induz mudanças na cromatina com respeito ao seu

conteúdo protéico, especificamente de proteínas relacionadas à formação de

heterocromatina e glicoproteínas nucleares.

4. Avaliar como o jejum e o envelhecimento, separadamente, atuam na interação de

sequências específicas de DNA à matriz nuclear.

5. Determinar como uma diferente organização da cromatina, como aquela observada

em espermatozóides de camundongos, pode influenciar as propriedades

viscoelásticas da cromatina, quando comparadas às de hepatócitos.

6. Padronizar a técnica de detecção in situ de glicoproteínas nucleares em decalques de

fígado de camundongo.

Para cumprir estes propósitos foram empregadas técnicas morfológicas

(citoquímica, imunofluorescência, hibridação in situ fluorescente e topoquímica), e

bioquímicas (eletroforese, westernblotting, imunoprecipitação de cromatina, isolamento e

fracionamento de cromatina em gradiente de sacarose), além de análise de imagem e

citofotometria. Os resultados obtidos foram divididos em cinco artigos científicos (três dos

quais já publicados), os quais estão apresentados nos próximos capítulos. Adicionalmente,

constam em anexo dois artigos publicados, nos quais o aluno teve participação como co-

autor.

29

Capítulo I

Artigo: Chromatin supraorganization and extensibility in mouse hepatocytes with

development and aging. Artigo publicado em Cytometry Part A 71A:28-37 (2007).

DOI: 10.1002/cyto.a.20356

40

Capítulo II

Artigo: Accumulation of heterochromatin marks in aging hepatocytes influences

interphase chromatin structure. Este artigo encontra-se em fase experimental.

41

Accumulation of heterochromatin marks in aging hepatocytes influences interphase

chromatin architecture

Alberto S. Moraes1, Shelagh Boyle2, Maria Luiza S. Mello1, Nick Gilbert3

1Department of Cell Biology Institute of Biology State University of Campinas (Unicamp) Campinas Sao Paulo Brazil 2MRC Human Genetics Unit Institute for Genetics and Molecular Medicine Crewe Road South Edinburgh EH4 2XU UK 3Edinburgh Cancer Research Centre Institute for Genetics and Molecular Medicine Crewe Road South Edinburgh EH4 2XR UK

Corresponding Author [email protected] Tel +44 131 332 2471

42

Abstract

The distribution and concentration of epigenetic marks and proteins associated with

heterochromatin and gene transcription was compared in hepatocyte nuclei from adult and

old mice. Even in the absence of a change in the distribution of such epigenetic marks

between adult and old mice, aged animals accumulate marks of heterochromatin and show

a decrease in histone modifications associated with gene transcription. Although we still do

not have data showing if this increase in the presence of heterochromatin markers are

associated with changes in overall chromatin structure, it seems reasonable to conclude that

in hepatocyte nuclei from old mice this increase in heterochromatin markers is associated

with a decreased metabolical rate if compared with that from adult mice, since in liver from

aged mice there is an expressive decrease in the total RNA synthesis.

43

Introduction

In the nuclei of all eukaryotic cells, genomic DNA is highly folded, constrained, and

compacted by histone and non-histone proteins in a dynamic polymer called chromatin.

This structure is formed by repeated series of octameric cores of histone proteins on which

DNA wraps around leading to various levels of folded structures.

In the interphase nucleus, two types of chromatin have been shown: the silent

heterochromatin and the active euchromatin. The heterochromatin was originally defined

cytologicaly as a highly condensed form of chromatin (Hetz, 1928). It can be responsible

for a 350 fold increase in the chromatin condensation while the euchromatic regions

generally show condensations of about 60 fold (Fransz et al., 2002). The heterochromatin is

composed of regular arrays of nucleosomes, it is packed in a manner that avoids or impairs

the access of nucleases, and also contains a high proportion of transcriptionally inactive

repetitive sequences interspersed with a very small number of genes (Elgin and Grewal,

2003; Grewal and Moazed, 2003). It is traditionally subdivided in constitutive

heterochromatin (i.e., telomeres and chromosomal regions flanking centromeres), that is

always condensed throughout the cell cycle, and facultative heterochromatin (i.e., genes

that were silenced during a given period of development), which can be unpacked under

some circumstances. The euchromatin is considered unpacked because of the irregular

spacing between nucleosomes, it is gene-rich, and is potentially active in terms of

transcription (Elgin and Grewal, 2003). However, not always these differences are so clear,

taking into account that some pericentric regions may be unpacked while some euchromatic

regions are more condensed than others (Gilbert et al., 2004).

The heterochromatin located around centromeres is organized as clearly

distinguishable chromocenters (CCs) (Müller, 1966). These are visible as dark spots with

phase-contrast microscopy or as bright, fluorescent domains after 4’,6-diamidino-2-

phenylindole staining. In mouse somatic cells, chromocenters are formed after aggregation

of pericentric heterochromatin from more than one chromosome (Hsu et al., 1971).

Typically, each chromocenter has about 4 centromeres, and all the pericentric

heterochromatin always co-localize with chromocenters (Cerda et al., 1999). In the mouse,

44

four types of repetitive DNA sequences are associated with centromeres. These are the AT-

rich major (6 megabases of 234 bp units) and minor satellite repeats (600 kb of 120 bp

units) (Hörz and Altenburger, 1981; Choo, 1997), besides the CG-rich MS3 and MS4

satellites, which localizes in the outer layer of chromocenters in interphase nuclei

(Kuznetsova et al., 2005).

The N-terminal tails of each of the four core histones are highly conserved in their

sequence, and perform crucial functions in regulating chromatin structure. Each tail is

subject to several types of covalent modifications, including acetylation and methylation of

lysines, and phosphorylation of serines (reviewed by Razin et al., 2007).

The various modifications of the histone tails have several important consequences.

Although modifications of the tails have little direct effect on the stability of an individual

nucleosome, they seem to affect the stability of the 30-nm chromatin fiber and of the

higher-order structures of the chromatin. For example, histone acetylation tends to

destabilize chromatin structure, perhaps in part because adding an acetyl group removes the

positive charge from the lysine, thereby making it more difficult for histones to neutralize

the charges on DNA as chromatin is compacted (Krajewski, 2000).

Both heterochromatin and euchromatin have their specific types of epigenetic

marks. Generally, silent chromatin is characterized by hypoacetylation, di- and tri-

methylation of lysines 9 and 27 on histone H3 and lysine 20 on histone H4, as well as DNA

methylation at cytosines (Fischle et al., 2003; Grewal and Moazed, 2003; Wu et al., 2005).

Euchromatic regions are generally associated with histone hyperacetylation and di- and tri-

methylation of lysine 4 on histone H3 (Fischle et al., 2003). If combinations of

modifications are taken into account, the number of possible distinct markings for each

histone tail can be very large. However, the most profound effect of modified histone tails

is their ability to attract specific proteins to a stretch of chromatin that has been

appropriately modified. Depending on the precise tail modifications, these additional

proteins can either cause further compaction of the chromatin or can facilitate access to the

DNA. Heterochromatin protein 1 (HP1) proteins are an example. They comprise a class of

multifunctional chromatin-associated adapter proteins, which are present at blocks of

constitutive heterochromatin in diverse eukaryotes, where they are thought to be important

45

for regulating heterochromatin-mediated silencing and chromosome structure (Ekwall et

al., 1995; Kellum et al., 1995; Yamaguchi et al., 1998). Three HP1 proteins (α,β and γ) are

known in mammals. HP1α and β are concentrated at pericentric heterochromatin, although

HP1γ can also be seen at more diffuse nucleoplasmic sites, whereas HP1γ is predominantly

localized in euchromatin (Minc et al., 2000; Nielsen et al., 2001).

Several mechanisms govern the spread of silencing proteins from heterochromatin

initiation sites, which can involve series of histone modifications. In the yeast S. cerevisiae

for example, Sir proteins are responsible for the formation of heterochromatin foci at

telomeres and mating-type loci. At yeast chromosomes, Sir proteins are present throughout

an approximately 3-kb region distal to the telomere, reflecting their ability to spread from

heterochromatin nucleation sites (Rusche et al., 2002). In Drosophila HP1 both recognizes

Met3H3K9, a marker of heterochromatin, and interacts with the modifying

methyltransferase SU(VAR)3–9 (Eskeland et al., 2007). These interactions suggest a model

for the spreading of this packaging form along the chromatin fiber, a property of

heterochromatin inferred after the observation of PEV (position effect variegation). In both

systems, however, the presence of boundary elements (Ishii and Laemmli, 2003) or

association of DNA with the nuclear matrix (Cai et al., 2003) avoid the excessive spreading

of the heterochromatic domains far beyond of their domains. When flanked by these

elements, genes are protected from the repressive or activating effects of nearby

heterochromatin or enhancer elements, respectively (Gerasimova and Corces, 2001; West et

al., 2002).

Experimental evidences involving chromatin resistance to digestion by restriction

enzymes and nucleases showed that chromatin becomes increasingly compact with aging

(Chaurasia et al., 1996; Mahendra et al., 1999; Ranhotra and Sharma, 2001), which is in

accordance with an age-related decline in the expression of some genes (Cao et al., 2001).

More recently it has been shown that senescence is often accompanied by specific

alterations of the chromatin structure, known as senescence-associated heterochromatic

foci. These foci are enriched for markers of heterochromatin, such as heterochromatin

protein 1 (HP1) and tri-methylated lysine 9 of histone H3 (which confers a docking site to

HP1), and exclude euchromatic markers, such as lysine 9 acetylation and tri-methylation of

46

lysine 4 of histone H3 (Narita et al., 2003). Nevertheless, this molecular mechanism of

senescence described in vitro is not yet clear to occur also in vivo.

In contrast, evidence exists that, despite the increased chromatin resistance to

enzymatic digestion in some models, chromatin in old animals suffers from inadequate

genomic integrity maintenance in many regions (Bitterman et al., 2003).

Genomic instability, which can be expressed as aberrant chromosome number,

chromosome deletions, rearrangements, loss or duplication beyond the normal diploid

number, has been implicated as a major cause of both cancer and aging. It may be caused

by several factors including defects in the DNA-repair pathway, shortening of telomeres by

a decreased activity of telomerase, and by an enhanced genomic plasticity which occurs

upon loss of tumor suppression gene function (Almasan et al., 1995). The genomic

plasticity contributes to the adaptability of senescent and tumor cells to their changing

environment, but may allow for accumulation of more genomic defects.

More recently, by using image analysis data from Feulgen-stained liver imprints, it

was observed a chromatin unpacking in hepatocyte nuclei from old mice when compared to

young/adult ones (Moraes et al., 2007), thus contrasting with the previous findings of an

increase in chromatin condensation with aging. However, two different systems are being

compared: one testing for the accessibility of chromatin to nucleases, and that is generally

focused in some genomic regions or single genes; the other, using liver imprints, which

does not allow to look at changes in the chromatin structure at the 30 nm level, but at the

same time makes possible to see cytological differences in the organization of chromatin in

the nucleus as hole. It might be a paradox, but considering that the two methodologies are

looking at chromatin at different resolutions, it seems reasonable to think that these

contrasting results may not be necessarily self-excluding.

In order to understand what molecular mechanisms may be involved with these

changes in chromatin organization during aging in vivo, we proposed to look for some

epigenetic marks in hepatocyte nuclei from adult and old mice, and at the same time,

evaluate how this changes have altered the chromatin structure. These results can help bring

light on the mechanisms responsible for the different levels of gene transcription observed

along the aging process.

47

Materials and methods

Tissue

Mouse liver was obtained from culled adult (10-15 weeks old) and old (8-9 months

old) mice.

Preparation of nuclei and chromatin

Hepatocyte nuclei from mouse were extracted according to Blobel and Potter (1966)

with modifications. The livers from adult, starved, and old mice were removed quickly and

chilled immediately in several volumes of ice-cold 0.25 M sucrose in TKM (0.05 M Tris-

HCl, pH 7.5, at 20ºC; 0.025 M KCl; and 0.005 M MgCl2). All subsequent operations were

performed at temperatures near 0ºC, and all buffers contained 0.1 M PMSF. Livers were

blotted, weighted, and minced with scissors in two volumes of ice-cold 0.25 M sucrose in

TKM. They were homogenized in a motor-driven, glass-Teflon homogenizer (Potter S; B.

Braun, Allentown, PA) using twenty up and down strokes at 1500 rpm. The homogenate

was filtered through four layers of cheese cloth. 2.0 mL of the filtered homogenate was

pipetted into a polyallomer tube that fit the Beckman 80Ti rotor; 4.0 mL of 2.3 M sucrose

in TKM was then added by means of a syringe and 13-gauge needle and thoroughly mixed

with the 0.25 M sucrose homogenate by inversion. The mixture was then underlayed by 2.0

mL 2.3 M sucrose in TKM with a syringe and 13-gauge needle: tip of the needle was

placed at the bottom of the tube and the heavy sucrose solution introduced, forcing the

lighter homogenate upward. After centrifugation for 30 min at 124,000g at 0º to 4ºC, the

supernatant was poured off. Material adhering to the wall of the tube was removed with a

spatula and added to the supernatant; the tube wall was then wiped dry with tissue paper

wrapped around a spatula. The white nuclear pellet was taken up in 4.0 mL, and 3.6 mL of

this pooled solution was mixed with 0.4 mL of 0.5 % Triton X-100 in 0.25 M sucrose in

TKM. Sample was centrifuged for 5 min at 800g, and pellets were stored at -70ºC.

Antibodies

48

The antibodies used were: mAb HP1α, β, and γ [Chemicon; western blotting (wb)

1:1000, immunofluorescence (immuno) 1:1000]; Met3H3K9 (Upstate; wb 1:1000, immuno

1:1000); Met2H3K9 (Upstate; wb 1:1000, immuno 1:1000); AcH3K9 (Upstate; wb 1:2000,

immuno 1:1000); Met3H4K20 (Upstate; wb 1:1000, immuno 1:1000); Met3H3K4 (Abcam;

wb 1:5000); Met2H3K4 (Abcam; wb 1:5000); AcH4K16 (Upstate; wb 1:1000); Met3H3K4

(Abcam; wb 1:5000); Met2H3K4 (Abcam; wb 1:5000); Met1H3K4 (Abcam; wb 1:1000);

C-terminal H3 (wb 1:30 000); H4 (wb 1:7500).

Western blotting

Nuclei were lysed in 1xSDS sample buffer (62.5 mM Tris-HCl, pH 6.75, 2% SDS,

5% ß-mercaptoethanol, 10% glycerol, and bromophenol blue), sonicated, and ~5 µg of total

protein was fractionated on a 12% SDS polyacrylamide gel. Proteins were transferred to

PVDF membrane by electroblotting, and the membranes were probed with antibodies using

standard techniques and detected by enhanced chemiluminescence.

Immunofluorescence

The animals were decapitated and their livers immediately removed and placed in

cold PBS. Liver slices were imprinted on histological slides. Briefly, the liver was cut into

pieces, and, by using a forceps, the exposed surface was imprinted on the histological

slides, thus, disrupting the cell membrane, and allowing the whole nuclei to adhere to the

slide surface. The material was treated 10 min room temperature in 1% Triton X-100 in

KCM buffer, and fixed 10 min using 4% pFa in PBS. The nuclei were permeabilized using

Triton X-100 in PBS, blocked with 4% horse serum in PBS, and were sequentially

incubated with the primary and secondary antibodies (Jackson Laboratories).

Two-color DNA FISH

Plasmids pCR4 Maj9-2 and pCR4 Min5-1 (Lehnertz et al., 2003) contain major and

minor satellite DNA (provided by T. Jenuwein, Research Institute of Molecular Pathology,

The Vienna Biocenter, Vienna, Austria). DNA fragments were labeled by nick translation

(Life Technologies) with digoxigenin-11-dUTP or biotin-16-dUTP (Boehringer). Liver

49

imprints fixed in ethanol/acetic acid 3:1 (v/v) for 1 min and then washed for 5 min in 70%

ethanol, followed by denaturation with 2xSSC/50% formamide 30 min at 80°C. Heat-

denatured probes were hybridized overnight at 37°C. After hybridization and washes with

2xSSC/50% formamide and 2xSSC at 42°C, probe detection used a three-step procedure

for amplification (Manuelidis et al., 1982). Biotin was revealed using Texas red–conjugated

streptavidin and biotinylated antistreptavidin antibody (Vector Laboratories), followed by

Texas red–conjugated streptavidin. Digoxigenin was detected using a sheep FITC-

conjugated anti-digoxigenin serum (Roche), rabbit FITC-conjugated anti–sheep antibodies

(Jackson ImmunoResearch Laboratories), and goat FITC-conjugated anti–rabbit antibodies

(Jackson ImmunoResearch Laboratories). Coverslips were mounted in Vectashield

containing 0.5 µg/mL DAPI (Vector Laboratories).

For immuno-DNA FISH the slides were first processed for immuno. Coordinates of

nuclei were taken and the slide was then processed for two-color DNA FISH.

Micrococcal Nuclease Assay

Isolated nuclei were permeabilized with 0.01% L-α-lysophosphatidylcholine

(Sigma) in 150 mM sucrose, 80 mM KCl, 35 mM HEPES (pH 7.4), 5 mM K2HPO4, 5 mM

Mg2Cl, and 0.5 mM CaCl2 for 90 s, followed by digestion with 2 U/ml micrococcal

nuclease (Sigma) in 20 mM sucrose, 50 mM Tris (pH 7.5), 50 mM NaCl, and 2 mM CaCl2

at room temperature for various times. DNA was isolated and subjected to 0.8% agarose gel

electrophoresis.

Chromatin Immunoprecipitation Assay

ChIP was performed as described previously (Chambeyron and Bickmore, 2004)

with antibodies recognizing the aforementioned Met2H3K9, Met3H3K9, and AcH3K9.

Immunoprecipitated chromatin was dot blotted onto Hybond N+ (GE Healthcare).

Membranes were probed with minor and major satellite repeats and the 5S rDNA, and the

blots were analyzed on a phosphorimager (FLA5100; Fuji).

RNA and DNA extraction

50

Total RNA and DNA were simultaneously isolated from freshly obtained mouse

liver using Tri® Reagent (Sigma). The protocols followed the instructions of the

manufacturer. Total DNA and RNA were quantified by spectrophotometry and the

RNA/DNA ratio between adult and old mice was compared. Three livers were used for

each age.

51

Results

Distribution of heterochromatin proteins and epigenetic marks does not change with

the aging process

Immunofluorescence in liver imprints from adult and old mice showed no

differences regarding the distribution of heterochromatin proteins and epigenetic marks

(Fig. 1). In both cases, HP1α proteins were found concentrated in the chromocenters, and

co-localized with the sites of Met3H3K9. The chromocenters were also enriched in

Met3H4K20. HP1β was found in a more diffuse manner throughout nuclei as well the

Met2H3K9. Interestingly, HP1γ, contrarily to previous results with chicken erythrocytes

(Gilbert et al., 2003), was found primarily in chromocenters, but also in a diffuse manner in

the nucleus. However, it was thought that this protein was mainly localized in the

euchromatin. As expected, the acetylated H3K9 was almost completely excluded from the

chromocenters, localizing practically in the euchromatic areas.

Genome-wide accumulation of heterochromatin proteins and epigenetic marks during

aging

The aging process in hepatocyte nuclei came accompanied by an accumulation of

epigenetic marks associated with heterochromatin such as, increase in the three HP1

proteins (α, β, and γ), as well increase in Met3H3K9, Met2H3K9, and Met3H4K20. At the

same time, the epigenetic markers associated with gene transcription (acetylation of H3K9

and H4K16, and also mono-, di- or tri-methylation of H3K4) decreased with the aging

process. But certainly, the most profound difference was that of the histone modifications

associated with heterochromatin formation (Fig. 2).

Decrease of the hepatocyte production of RNAs

As documented elsewhere, the aging process is characterized by a decrease in the

metabolical rate of the organism. In the liver, it was reported a steady decrease in the RNA

production as the animal becomes older. In order to confirm that information, we extracted

total RNA and DNA from mouse liver and compared the ratio RNA/DNA for both adult

52

and aged mice. The comparison using the ratios was used instead of comparing only the

total RNA extracted because of the well known polyploidy in aged liver, as a consequence

of the increase in DNA content. So, an absolute comparison could give for old animals a

very higher RNA content. But the true is that the RNA levels droped in aged animals, thus

evidencing the decrease in the metabolical rate in the liver of aged animals (Fig. 3).

Spreading of epigenetic repressive marks, preservation of minor and major satellite

architecture, nucleosomal repeat length, compaction of major satellite chromatin

structure

To test the hypothesis that this increase in epigenetic marks associated with

heterochromatin formation are associated with a decrease in gene transcription through a

spreading of the heterochromatic domains to euchromatic areas, several approaches were

used as stated in Materials and Methods. Additionally, we looked for differences in the

structure of the 30 nm fiber testing the sensitivity of the chromatin to nuclease digestion.

All this experiments are actually in course, and final results will be available in the next

weeks.

Discussion

The organization of DNA into heterochromatin contributes to nuclear organization,

chromosome structure, and gene silencing. Constitutive heterochromatin primarily

encompasses the pericentric regions of chromosomes and is important for chromosome

segregation and the silencing of repetitive elements. Facultative heterochromatin is

developmentally controlled and contributes to gene regulation during differentiation and

influences dosage compensation. Here, we observed an accumulation of epigenetic marks

associated with chromatin in cells from aged animals. It’s possible that this increase in

heterochromatic marks modifies the chromatin overall organization and the resistance of

the DNA to nuclease digestion.

Changes in nuclear architecture do not appear to be restricted to defects in the

structural components of the nucleus. An age-related loss of epigenetic silencing at certain

repetitive elements was reported previously. Specifically, the major satellite repeats that

53

form heterochromatic chromatin structures around the centromeres of every chromosome

were shown to be more transcriptionally active in aged cardiac tissue, which suggests a

progressive loss of silencing of these elements (Gaubatz and Cutler, 1990). Given the

number of repetitive elements in mammalian genomes, a reduction in repeat-associated

heterochromatin would be consistent with significant changes in nuclear architecture. A

possible mechanistic link between mammalian aging and changes in heterochromatin exists

(Shen et al., 2006). Older individuals show altered activity in their histone-modifying

enzymes, which causes a loss of perinuclear heterochromatin and concomitant changes in

gene expression. These observations are reminiscent of the chromatin changes that occur

during yeast aging and in Hutchinson–Gilford progeria syndrome (HGPS), and raise the

possibility that changes in perinuclear architecture contribute to normal aging in mammals.

Numerous other epigenetic changes in nuclear architecture and gene expression

have been associated with aging. More than a decade ago, it was showed that collagenase, a

gene associated with cellular aging, is differentially regulated during cellular senescence —

a phenomenon that is often referred to as cellular aging (Imai et al., 1997). This effect

appears to be due to changes in the subnuclear localization of the collagenase gene as cells

undergo senescence. In young cells, the collagenase gene is repressed by the transcription

factor OCT1. A considerable proportion of OCT1 was found in the heterochromatic nuclear

periphery, where it colocalized with lamin B, a component of the nuclear membrane. This

interaction was abrogated in senescent cells and, concomitantly, collagenase repression was

lost (Imai et al., 1997). On the basis of these findings, the authors proposed a model of age-

associated heterochromatin reorganization that would account for such transcriptional

changes in a global manner (Imai and Kitano, 1998).

This idea gained support from recent studies of cellular senescence, most notably by

Lowe and colleagues, who found that senescence is associated with an overall increase in

non-pericentromeric, facultative heterochromatin domains, known as senescence-associated

heterochromatin foci (SAHFs) (Narita et al., 2003). SAHFs form repressive chromatin

structures that can be found at, but are not limited to, promoter regions of certain cell-cycle

regulators, in particular target promoters of the cell-cycle regulator E2F. This finding led to

the hypothesis that SAHFs promote senescence through direct repression of growth-

54

promoting genes. Although the repression of cell-cycle regulators is an important function

of SAHFs during cellular senescence, the frequency and distribution of these foci suggests

a much broader impact of SAHFs. This notion is further supported by the finding that the

formation of SAHFs appears to rely on the recruitment of proteins from promyelocytic

leukaemia nuclear bodies (Zhang et al., 2005), which have been implicated in numerous

cellular processes including transcriptional regulation, apoptosis and cellular defense in

response to stress (most notably to DNA damage [Dallaire and Bazett-Jones, 2004]).

A connection between senescence-associated heterochromatin formation and

mammalian aging has recently been made using baboon skin fibroblasts (Herbig et al.,

2006). Tissue from older individuals accumulates cells containing heterochromatic foci that

are reminiscent of SAHFs in senescent cells. These foci were found in >15% of the total

cell population in aged tissues, which suggests that a significant fraction of aged tissue may

be expressing markers of senescence and undergoing large-scale heterochromatic changes.

Importantly, the emergence of heterochromatin foci occurred simultaneously with telomere

shortening, which points to a shift from stable, perinuclear heterochromatin to induced, or

facultative, heterochromatin. These studies raise the intriguing possibility that the age-

associated loss of genomic silencing detected in previous studies may be linked to, or

caused by, the formation of SAHF-like heterochromatic foci, a phenomenon that occurs in

yeast in response to DNA damage and aging. However, it is important to keep in mind that

cellular senescence - although it is a likely contributor to cancer and organismal aging -

does not equal the complex physiological processes that, together, define what is called

aging. The relevance of the aforementioned findings to the functional decline of higher

organisms remains to be elucidated.

55

References

Almasan A, Linke SP, Paulson TG, Huang LC, Wahl GM (1995) Genetic instability as a

consequence of inappropriate entry into and progression through S-phase. Cancer

Metastasis Rev 14:59-73.

Bitterman KJ, Medvedik O, Sinclair DA (2003) Longevity regulation in Saccharomyces

cerevisiae: linking metabolism, genome stability, and heterochromatin. Microbiol

Mol Biol Rev 67:376-99.

Blobel G, Potter VR. 1966. Nuclei from rat liver: isolation method that combines purity

with high yield. Science 154:1662-1665.

Cai S, Han HJ, Kohwi-Shigematsu T (2003) Tissue-specific nuclear architecture and gene

expression regulated by SATB1. Nat Genet 34:42-51.

Cao SX, Dhahbi JM, Mote PL, Spindler SR (2001) Genomic profiling of short- and long-

term caloric restriction effects in the liver of aging mice. Proc Natl Acad Sci USA

98:10630-10635.

Cerda MC, Berrios S, Fernandez-Donoso R, Garagna S, Redi C (1999) Organisation of

complex nuclear domains in somatic mouse cells. Biol Cell 91:55-65.

Chambeyron, S, Bickmore WA (2004) Chromatin decondensation and nuclear

reorganization of the HoxB locus upon induction of transcription. Genes Dev

18:1119–1130.

Chaurasia P, Mukherjee S, Thakur MK (1996) Age-related analysis of EcoRI generated

satellite DNA-containing chromatin of rat liver. Biochem Mol Biol Int 40:1261-1270.

Choo KH (1997) Centromere DNA dynamics: latent centromeres and neocentromere

formation. Am J Hum Genet 61:1225-1233.

Dellaire G, Bazett-Jones DP (2004) PML nuclear bodies: dynamic sensors of DNA damage

and cellular stress. Bioessays 26:963–977.

Herbig U, Ferreira M, Condel L, Carey D, Sedivy JM (2006) Cellular senescence in aging

primates. Science 311:1257.

56

Ekwall K, Javerzat JP, Lorentz A, Schmidt H, Cranston G, Allshire R (1995) The

chromodomain protein Swi6: a key component at fission yeast centromeres. Science

269:1429-1431.

Elgin SC, Grewal SI (2003) Heterochromatin: silence is golden. Curr Biol 13:R895-R898.

Eskeland R, Eberharter A, Imhof A (2007) HP1 binding to chromatin methylated at H3K9

is enhanced by auxiliary factors. Mol Cell Biol 27:453-465.

Fischle W, Wang Y, Allis CD (2003) Histone and chromatin cross-talk. Curr Opin Cell

Biol 15:172-183.

Fransz P, De Jong JH, Lysak M, Castiglione MR, Schubert I (2002) Interphase

chromosomes in Arabidopsis are organized as well defined chromocenters from

which euchromatin loops emanate. Proc Natl Acad Sci USA 99:14584-14589.

Gaubatz JW, Cutler RG (1990) Mouse satellite DNA is transcribed in senescent cardiac

muscle. J Biol Chem 265:17753–17758.

Gerasimova TI, Corces VG (2001) Chromatin insulators and boundaries: effects on

transcription and nuclear organization. Annu Rev Genet 35:193-208.

Gilbert N, Boyle S, Fiegler H, Woodfine K, Carter NP, Bickmore WA (2004) Chromatin

architecture of the human genome: gene-rich domains are enriched in open chromatin

fibers. Cell 118:555-566.

Grewal SI, Moazed D (2003) Heterochromatin and epigenetic control of gene expression.

Science 301:798-802.

Heitz E (1928) Das Heterochromatin der Moose. Jahrb Wiss Bot 69:762-818.

Horz W, Altenburger W (1981) Nucleotide sequence of mouse satellite DNA. Nucleic

Acids Res 9:683-696.

Hsu TC, Cooper JE, Mace ML Jr, Brinkley BR (1971) Arrangement of centromeres in

mouse cells. Chromosoma 34:73-87.

Imai S, Kitano H (1998) Heterochromatin islands and their dynamic reorganization: a

hypothesis for three distinctive features of cellular aging. Exp Gerontol 33:555–570.

Imai S, Nishibayashi S, Takao K, Tomifuji M, Fujino T, Hasegawa M, Takano T (1997)

Dissociation of Oct-1 from the nuclear peripheral structure induces the cellular aging-

associated collagenase gene expression. Mol Biol Cell 8:2407–2419.

57

Ishii K, Laemmli UK (2003) Structural and dynamic functions establish chromatin

domains. Mol Cell 11:237–248.

Kellum R, Raff JW, Alberts BM (1995) Heterochromatin protein 1 distribution during

development and during the cell cycle in Drosophila embryos. J Cell Sci 108:1407-

1418.

Krajewski WA (2000) Histone hyperacetylation facilitates chromatin remodelling in a

Drosophila embryo cell-free system. Mol Gen Genet 263:38-47.

Kuznetsova IS, Prusov AN, Enukashvily NI, Podgornaya OI (2005) New types of mouse

centromeric satellite DNAs. Chromosome Res 13:9-25.

Mahendra G, Gupta S, Kanungo MS (1999) Effect of 17beta estradiol and progesterone on

the conformation of the chromatin of the liver of female Japanese quail during aging.

Arch Gerontol Geriatr 28:149-158.

Minc E, Courvalin JC, Buendia B (2000) HP1gamma associates with euchromatin and

heterochromatin in mammalian nuclei and chromosomes. Cytogenet Cell Genet

90:279-284.

Moraes AS, Guaraldo AM, Mello ML (2007) Chromatin supraorganization and

extensibility in mouse hepatocytes with development and aging. Cytometry A 71:28-

37.

Müller HA (1966) Chromocenters in liver cell nuclei of the mouse under normal and

pathological conditions. Ergeb Allg Pathol Pathol Anat 47:144-185.

Narita M, Nunez S, Heard E, Narita M, Lin AW, Hearn SA, Spector DL, Hannon GJ, Lowe

SW (2003) Rb-mediated heterochromatin formation and silencing of E2F target genes

during cellular senescence. Cell 113:703-716.

Nielsen AL, Oulad-Abdelghani M, Ortiz JA, Remboutsika E, Chambon P, Losson R (2001)

Heterochromatin formation in mammalian cells: interaction between histones and

HP1 proteins. Mol Cell 7:729-739.

Ranhotra HS, Sharma R (2001) Modulation of hepatic and renal glucocorticoid receptors

during aging of mice. Biogerontology 2:245-251.

58

Razin SV, Iarovaia OV, Sjakste N, Sjakste T, Bagdoniene L, Rynditch AV, Eivazova ER,

Lipinski M, Vassetzky YS (2007) Chromatin domains and regulation of transcription.

J Mol Biol 369:597-607.

Rusche LN, Kirchmaier AL, Rine J (2002) Ordered nucleation and spreading of silenced

chromatin in Saccharomyces cerevisiae. Mol Biol Cell 13:2207-2222.

Shen, S., Liu, A., Li, J., Wolubah, C. & Casaccia-Bonnefil, P. Epigenetic memory loss in

aging oligodendrocytes in the corpus callosum. Neurobiol. Aging 19 December 2006

(doi: 10.1016/j.neurobiolaging.2006.10.026).

West AG, Gaszner M, Felsenfeld G (2002) Insulators: many functions, many mechanisms.

Genes Dev 16:271-288.

Wu R, Terry AV, Singh PB, Gilbert DM (2005) Differential subnuclear localization and

replication timing of histone H3 lysine 9 methylation states. Mol Biol Cell 16:2872-

2881.

Yamaguchi K, Hidema S, Mizuno S (1998) Chicken chromobox proteins: cDNA cloning of

CHCB1, -2, -3 and their relation to W-heterochromatin. Exp Cell Res 242:303-314.

Zhang, R. et al. Formation of macroH2A-containing senescence-associated

heterochromatin foci and senescence driven by ASF1a and HIRA. Dev. Cell 8, 19–30

(2005).

59

Figure 1

A

B

60

Figure 2

A B

61

Figure 3

RN

A/D

NA

62

Figure captions

Figure 1. Distribution of HP1 isoforms and histone modifications in hepatocyte nuclei. Co-

immunostaining of hepatocyte nuclei from adult (A) and old (B) mice fixed with pFa

using antibodies that detect HP1α, β and γ (green), and met3H3K9, met2H3K9,

AcH3K9 and met3H4K20 (red). DNA was counterstained with DAPI to highlight the

foci of heterochromatin.

Figure 2. Genome-wide accumulation of heterochromatin proteins and epigenetic marks

during aging. (A) Western blot of mouse hepatocyte isolated nuclear proteins with

antibodies detecting HP1 isoforms, di- (met2H3K9) and tri- (met3H3K9) methylated

H3 K9, mono- (met1H3K4), di- (met2H3K4), and tri- (met3H3K4) methylated H3 K4,

trimethylated H4K20, and acetylated H3 K9 and H4 K16 (AcH3K9, AcH4K16,

respectively). Antibody detecting H3 and H4 were used as a loading controls. (B)

Quantification of the bands evidenced by these antibodies. Signals were normalized in

respect to total histone H3 (for HP1 proteins and H4 modified histones) or histone H4

(when the primary antibodies detected modified H3). Calculations were done using

the average values of three experiments for each antibody. The error bars indicate

standard errors.

Figure 3. Decrease of RNA production in hepatocyte from aged mice. The proportion

between total RNA and total DNA was compared in hepatocyte nuclei from adult and

old mice. The decrease in RNA production in liver from old mice is more than half of

that found in adult animals.

63

Capítulo III

Artigo: Effects of aging and starvation on chromatin-nuclear matrix association in mouse

hepatocytes.

64

Effects of aging and starvation on chromatin-nuclear matrix association in mouse

hepatocytes

Alberto S. Moraes,1 Mateus Mondin,3 Margarida L. R. Aguiar-Pericin,3 Ana M. A.

Guaraldo,2 and Maria Luiza S. Mello1

1Departamento de Biologia Celular, and 2Departamento de Parasitologia, Instituto de

Biologia, Universidade Estadual de Campinas (UNICAMP), 13083-863 Campinas, São

Paulo, Brazil 3Departamento de Genetica, Escola Superior de Agricultura Luiz de Queiroz, Universidade

de São Paulo, 13418-900, Piracicaba, SP, Brazil

Correspondence

Maria Luiza S. Mello, State University of Campinas, Institute of Biology, Department of

Cell Biology, Campinas, São Paulo, Brazil, 13087-930. Tel.: +55 19 35216122; fax: +55 19

35216111; e-mail: [email protected]

E-mails:

Alberto S. Moraes: [email protected]; Mateus Mondin: [email protected];

Margarida L.R. Aguiar-Perecin: [email protected]; Ana M.A. Guaraldo:

[email protected]

Alberto S. Moraes and Mateus Mondin contributed equally to this project.

Keywords: aging; chromatin extensibility; isochors; nuclear matrix; rDNA; starvation.

65

Summary

Active genes have been suggested to be associated with the nuclear matrix while non-

transcribing sequences have been found in DNA loops. However, although dozen of genes

have their expression level affected by starvation and aging, no data exist on chromatin-

nuclear matrix interactions under these physiological conditions. Liver imprints from well-

fed young, adult, and old, as well as from starved adult mice, were subjected to sequential

staining with fluorochromes, and DNA FISH with 45S rDNA and telomeric DNA, both

after a lysis treatment under the gravity action. We observed that, irrespective of the

physiological condition, gene-rich regions remained attached to the nuclear matrix while

gene-poor regions associated with DNA loops. Aging, but not starvation, increased the

amount of 45S rDNA sequences in the DNA loops. We assume that transcribing genes, but

not non-coding repetitive sequences, associate with the nuclear matrix. Transcription of

each 45S rDNA repeat unit is suggested to be dependent on the interaction of this DNA

with the nuclear matrix.

66

Introduction

Most of the nuclear genome of warm-blooded vertebrates can be partitioned in very long

DNA segments (more than 200 kilobases) termed isochors (from the Greek isos, equal; and

choros, region). The isochors are fairly homogeneous in base composition and belong to a

small number of major classes distinguished by differences in cytosine-guanine (CG)

content (Bernardi et al., 1985; Salinas et al., 1986). The CG sequences of genes are linearly

related to those of the isochors in which they are located, with a preferential distribution of

genes in CG-rich isochors (Bernardi et al., 1985). In mammalian genomes there is a

positive correlation between gene density and CG content. In mice, 75-80% of genes reside

in the CG-richest part of the genome (Waterston et al., 2002). However, the assertion that

the expression of a given gene depends on the CG-content of the regions where it is situated

has not been corroborated (Semon et al., 2005). It has been shown that the CG content of

rodent genomes, especially in murids, is less heterogeneous than that of other mammals

(ie., CG-rich genes are less CG-rich in murids than in other mammals, and, conversely, for

CG-poor genes) (Mouchiroud & Gautier, 1988; Mouchiroud et al., 1988). All the

characteristic features of the CG-richest isochors (gene density, expression level, lengths of

coding and non-coding regions and hydrophobicity of the corresponding encoded proteins)

are tightly linked to each other and are under the same selective pressure. All this data

converge to the conclusion that the CG-richest isochors are characterized by a high level of

transcription (Arhondakis et al., 2004).

AT-rich isochors, on the contrary, contain only 20-25% of all nuclear genes,

but comprise 58% of the mouse genome (Waterston et al., 2002). Most of these gene-poor

sequences, which are permanently condensed in a form known as constitutive

heterochromatin, is transcriptionally inert, rich in repetitive DNA (frequently satellite,

telomeric and transposable DNA) and capable of silencing genes of adjacent euchromatin

(Choo, 1997; Cerda et al., 1999; Fransz et al., 2002; Grewal & Elgin, 2002; Sharma et al.,

2003; Garcia-Cao et al., 2004; Gilbert et al., 2004). A second type of heterochromatin,

commonly known as facultative, involves chromosome regions that, in specific cells,

67

become compact and transcriptionally inactive during remodeling of chromatin (Strahl &

Allis, 2000; Kurdistani et al., 2004; Zardo et al., 2005; Nishida et al., 2006).

Both euchromatic and heterochromatic sequences are restricted to a nuclear

compartment termed chromosome domain or chromosome territory. The remaining space,

denominated the interchromatin domain (ICD), is equivalent to the nuclear matrix when

observed in non-fractionated nuclei or in living cells (Cremer et al., 2000).

The nuclear matrix is an operationally defined nuclear skeletal structure that is

believed to be involved in many nuclear functions including DNA replication, transcription,

repair and pre-mRNA processing/transport. It is also a passenger concourse for proteins

going in and out of the nucleus through the nuclear pore complex (Tsutsui et al., 2005). It

comprises a protein framework formed by the nuclear pore/nuclear lamina complex, a

residual nucleolus and a residual internal fibrilar-globular mesh (Berezney & Coffey, 1974;

Gasser & Laemmli, 1986; Berezney, 1991). Until relatively recently, the nuclear matrix

was thought to be a rigid and static structure, but it is now thought to be dynamic (Tsutsui

et al., 2005).

The preparative procedure for the nuclear matrix involves nuclease and high salt

treatments that remove chromatin and other soluble nuclear proteins (Nickerson, 2001). A

common method for matrix preparation is the extraction of nuclei with highly concentrated

solutions with monovalent salts such 2 M NaCl (Berezney & Coffey, 1977). After the

extraction procedure, the nuclei adopt a so-called nuclear halo configuration, in which

nuclear DNA sticks out from the residual structure (nuclear matrix) in the form of

negatively supercoiled loops. If nuclei are extracted vertically and under the gravity action,

they can produce extended chromatin fibers (ECFs), whose formation can strongly be

affected by nuclear matrix proteins through their contribution to the higher-order chromatin

structure (Gerdes et al., 1994; Haaf & Ward, 1994, 1995; Davie, 1995; Cremer et al., 2000;

Vidal, 2000). An increased number of interactions between chromatin and nuclear matrix

can render the ECF formation or even prevent it (Moraes et al., 2005, 2007). After this

extraction, nuclear DNA can be fractionated into loop DNA and matrix-associated DNA,

which counts for 1% of total nuclear DNA, and is enriched in rRNA sequences (Pardoll &

Vogelstein, 1980). When this treatment is followed by FISH analysis, transcriptionally

68

inactive sequences produce long strings of signal extending out onto the DNA halo or loop,

whereas active sequences remain tightly condensed as single spots within the residual

nucleus, thus showing that the association of genes with the nuclear matrix may be a factor

responsible for transcription activation (Gerdes et al., 1994).

The observation of nuclear halos or ECFs reflects the organization of the DNA of

higher eukaryotic cells into large domains, or chromatin loops, restrained at their bases by

interaction with the nuclear scaffold or matrix (Iarovaia et al., 2004). This interaction is

provided by DNA sequences named matrix attachment regions (MARs) or scaffold

adhesion regions (SARs) (Paulson & Laemmli, 1977; Mirkovitch et al., 1984). MARs have

possibly a structural role in chromosome organization, serving to bring the promoter

regions of active genes physically close together, thus forming a subcompartment in the

nucleus of reduced dimensionality and rich in RNA polymerase and regulatory factors

(Gasser & Laemmli, 1986). In addition, it has been argued that the presence of MARs in a

gene is not related to transcription activation or inactivation, but determines the level of

transcription of that gene (Allen et al., 2000; Whitelaw et al., 2000). The correlation

between the number of anchored MARs and the gene expression level illustrates the

importance of direct interaction with the nuclear matrix. It seems that such anchor-mediated

interaction is essential for gene expression (Heng et al., 2004).

The key process in mouse aging involves reduced expression of a number of genes

(Burzynski, 2005), which could be associated with a decrease in the total nuclear amount of

interactions between the chromatin and the nuclear matrix (Moraes et al., 2007). rRNA

synthesis in mouse liver was observed to be decreased after starvation and along

development and aging (Sollner-Webb & Tower, 1986; Ma & Nagata, 1990). It is thus

possible that rDNA sequences show reduced degrees of interaction with the nuclear matrix

under these processes.

The rRNA genes, which in most eukaryotic species, are repeated ~100-5,000 times

per haploid genome, are located at one or a few chromosomal sites called nucleolar

organizers (Gaubatz et al., 1976; Long & Dawid, 1980). The rRNA coding regions are

organized as head-to-tail repeats, with the transcribed regions separated by segments of

nontranscribed spacer (Fedoroff, 1979). In rat hepatocytes these sequences are enriched in

69

the nuclear matrix after DNase I digestion, thus indicating that the high transcription rates

of rRNA genes and nuclear matrix interaction are linked (Pardoll & Vogelstein, 1980).

To test the hypothesis that regions of the mouse genome with different transcription

levels are differently associated with the nuclear matrix and suspecting that starvation and

aging processes could affect this interaction, the nuclear localization of CG and AT-rich

regions was studied in mouse hepatocyte nuclei with or without treatment for ECF

formation. Also, the 45S rDNA repeats, representing highly transcribed regions, and the

gene/transcription poor telomeric DNA, were assessed by FISH in fixed non-lysed and

lysed hepatocyte nuclei.

The liver is the central organ for the regulation of glucose homeostasis, xenobiotic

metabolism and detoxification, as well for steroid hormone biosynthesis and degradation.

This organ also has a major impact on health and homeostasis through its control of serum

protein composition. While differentiated hepatic functions are generally well maintained

with age, changes do occur. The resilience of the liver to aging and starvation and its

central role in the maintenance of the whole body health and homeostasis make it an

intriguing target for genome-wide analysis of the DNA/nuclear matrix interaction during

development, aging and starvation.

70

Results

AT- and CG-rich DNA are differently anchored onto the nuclear matrix

Figure 1 shows the fluorescence patterns observed in fixed non-lysed interphase nuclei

from well-fed and starved adults, and well-fed old mice using CMA3 and DAPI dyes. Since

hepatocyte nuclei from young mice do not form ECFs (Moraes et al., 2007), they are not

presented here. In DAPI-stained nuclei, the brighter points correspond to the

chromocenters. In comparison with nuclei from well-fed adult mice (Fig. 1a), the contrast

between chromocenters and the euchromatic regions was lower in nuclei from starved mice

(Fig. 1b) and higher in nuclei from well-fed old mice (Fig. 1c). The intensity of staining

with CMA3 in nuclei from the starved adult mice was also lower (Fig. 1b) and in nuclei

from well-fed old mice it was higher (Fig. 1c), if compared to that of nuclei from well-fed

adult mice (Fig. 1a).

In fixed and lysed preparations (Fig. 2), the intranuclear intensity of DAPI staining

was higher than that in controls (non-lysed nuclei). ECF formation was observed in nuclei

from well-fed adult (Fig. 2a) and old mice (Fig. 2c) after DAPI staining. In contrast, only

nuclear halos were observed in most nuclei from the starved mice (Fig. 2b). In contrast to

the DAPI-stained nuclei, no ECF or nuclear halos could be clearly observed in the same

nuclei stained with CMA3 (Fig. 2).

rDNA genes detach from the nuclear matrix along development and aging but not

after starvation

In the fixed non-lysed hepatocyte nuclei, 45S rDNA was observed as large spots in all

nuclei except those for young mice, where it appeared as little dots (Fig. 3a). In polyploid

nuclei from well-fed and starved adults, and well-fed old mice, 45S rDNA was always seen

in the periphery of the negative images of the nucleoli (n), and in association with DAPI-

dense clusters (chromocenters) (Fig. 3b,c,d).

71

In fixed and lysed hepatocytes from well-fed young mice, which do not form ECFs,

the small dots of 45S rDNA detected by FISH remain inside the nuclei after lysis treatment

(Fig. 4a), whereas in fixed and lysed preparations from well-fed and starved adults, and

well-fed old mice, the FISH signals spread throughout and outside the nuclei, where many

small spots can be seen in the halos and ECFs (Fig. 4b,c,d). The 45S rDNA signal intensity

outside the nuclei was higher in old mice and weaker in well-fed and starved adult mice,

but the intensity of the signal inside the nuclei was higher in the preparations from starved

mice (Fig. 4c,d).

Telomeric sequences are always located in chromatin loops

In the mouse hepatocyte nuclei, telomeres were also detected as small dots (well-fed young

and adult mice) (Fig. 5a,c). In well-fed adult (Fig. 5c), and old mice (not shown), and from

starved adults (not shown), telomeric sequences were seen spread in all cell nuclei, whereas

in young mice these sequences were detected only in erythroblast nuclei (e) (Fig. 5a).

After lysis treatment, the telomeric sequences extended outside the nuclei under all

conditions studied, but with less intensity in nuclei from well-fed young mice (Fig. 5b,d).

72

Discussion

The different patterns of staining observed among hepatocyte nuclei from well-fed and

starved adults, and well-fed old mice after DAPI and CMA3 staining were due to their

different levels of chromatin packing. An increase in chromatin packing levels, with a

decrease in the contrast between condensed and non-condensed areas, has been reported in

hepatocyte nuclei from starved mice in comparison to well-fed adult mice (Moraes et al.,

2005). In addition, a decrease in the level of chromatin packing associated with an increase

in contrast between the states of condensed and non-condensed areas was reported for

hepatocyte nuclei from aged mice (Moraes et al., 2007). While DAPI stained the nuclei,

and particularly the chromocenters, CMA3 did not reveal any brighter points coincident

with the chromocenter position. Since DAPI attaches preferentially to AT-rich clusters

(Kapuscinski & Szer, 1979), it seemed that chromocenter areas in the analyzed cells

contained AT-rich DNA, which is in agreement with published data (Hörz & Altenburger,

1981; Choo, 1997). In fixed and lysed nuclei, the increase in brightness of DAPI stained

nuclei was a result of the chromatin reorganization and unpacking observed after the lysis

treatment (Vidal et al., 2006; Moraes et al., 2005, 2007), thus allowing the binding of more

dye molecules to the DNA strands.

The affinity of ECFs for DAPI but not for CMA3 showed that many AT-rich regions

went preferentially outside the nuclei after the lysis treatment, while the CMA3-stained

regions (CG-rich isochors) remained inside the nuclei. The high-salt extraction procedure

used for production of ECFs in the present work is part of the process for isolation of

nuclear matrices (Berezney & Coffey, 1977). After the high-salt extraction, which removes

mainly histone proteins, the histone-depleted chromatin becomes more fluid and can flow

outside the nuclei. This extended chromatin is composed of looped DNA, the bases of

which remain attached to the nuclear matrix (Iarovaia et al., 2004). Therefore, it can be

assumed that in mice the bases of the DNA loops are composed of both AT- and CG-rich

isochors, while the loop DNA properly is composed practically of AT-rich sequences.

Taking into account that the mouse has 75-80% of its genes residing in the CG-richest half

of the genome (Waterston et al., 2002) and that the CG-richest isochors are characterized

73

by a high level of transcription (Arhondakis et al., 2004), it can be concluded that most of

the transcribing genes are bound to the nuclear matrix, while the loop DNA contain mainly

non-coding regions.

This hypothesis can be confirmed by the present results using 45S rDNA and

telomeric sequences. In fact, when 45S rDNA genes are being transcribed, as is the case in

young mice, they are found completely attached to the nuclear matrix, while along

development and aging, they are progressively switched off and detached from it. At the

same time, the non-coding telomeric sequences were always found detached from the

nuclear matrix despite the animal age or the physiological condition employed, thus

supporting the idea that non-coding regions do not associate with the nuclear matrix.

It was expected that 45S rDNA repeats, which are responsible for production of a

large amount of rRNAs (Wellauer et al., 1974; Perry, 1976; Kominami et al., 1978), could

show different levels of binding to the nuclear matrix along development and aging, and

following starvation, when decreased transcription levels have been observed (Ma &

Nagata, 1990). Indeed, the present results indicate that, as the animal ages, there is an

increase in the amount of 45S rDNA repeats non-associated with the nuclear matrix and,

present in the nuclear halos or ECFs, thus linking the nuclear matrix with gene

transcription. At the same time, although it has earlier been proposed that a decrease in

gene expression could exist in liver from starved mice (Sollner-Webb & Tower, 1986;

Moraes et al., 2005), no great differences were seen in the amount of 45S rDNA that goes

outside the nuclei in preparations from starved and from well-fed adult mice. Consequently

indicating that the amount of transcribing sequences attached to the nuclear matrix in adult

well-fed and in starved mice seems not to vary significantly. Microarray studies using

mouse liver have shown that while nearly 50 genes have their level of expression increased,

practically the same amount of genes showed decreased expression levels after starvation

(Bauer et al., 2004). In fact, the present study suggests that a global decrease in gene

expression in the liver of starved adult mice appears not to occur, at least regarding the 45S

rDNA genes.

Additionally, the results concerning the 45S rDNA genes imply that coding

repetitive sequences indicate a nuclear matrix-dependent control of gene expression

74

different from that of single copy genes. In single copy genes, the adhesion to the nuclear

matrix determines their transcription level (Allen et al., 2000; Whitelaw et al., 2000), while

in the case of 45S rDNA, this interaction may determine if each gene copy will be

transcribed or not.

In conclusion, it was showed here that the transcription of rDNA genes may depend on

their interaction with the nuclear matrix, while non-coding sequences do not associate with

it. Consequently, it seems that this type of gene expression control surpasses that

dependence on heterochromatin formation, since nuclei from starved adult mice, with

chromatin more packed than those in well-fed adult mice (Moraes et al., 2005), showed no

differences regarding the amount of 45S rDNA genes associated or not with the nuclear

matrix when compared to nuclei from well-fed adult mice. In addition, hepatocyte nuclei

from well-fed old mice contain less packed chromatin than nuclei from well-fed adult mice

(Moraes et al., 2007) and show a decrease in transcription levels (Ma & Nagata, 1990),

which is associated with an increase in the number of 45S rDNA copies not associated with

the nuclear matrix.

75

Materials and methods

Male mice from the inbred strain A/Uni, obtained from the Multidisciplinary Center for

Biological Investigation (CEMIB) of the State University of Campinas (Brazil), were

reared under normal conditions and fed extruded chow (Purina®, Paulinia, Brazil) ad

libitum. The animals used were 1-2 (young), 15-20 (adult), and 61-100 (aged) weeks old.

Additionally, a group of adult mice was subsequently deprived of food for 48 h but

receiving water ad libitum. At least 3 animals of each group were used. The animals were

decapitated and their livers immediately removed and placed in cold 0.9% NaCl solution.

Liver slices were imprinted on histological slides. All protocols involving animal care and

use were approved by the Institutional Committee for Ethics in Animal Experimentation

(CEEA/IB/UNICAMP) and met the guidelines of the Canadian Council on Animal Care.

Treatments

Freshly prepared imprints were fixed in a mixture of absolute ethanol and acetic acid (3:1,

v/v) for 1 min and then rinsed in 70% ethanol for 5 min. The preparations were positioned

vertically using a protocol previously described for ECF investigation (Moraes et al.,

2005). Briefly, the lysis solution consisted of 2 M NaCl plus 1% Triton X-100 in Tris-HCl

buffer (25 mM, pH 7.4). Treatment lasted for 5 h at 25ºC after which the volume of

solution was completed with absolute ethanol to a final concentration of 50% and treatment

was prolonged for another 10 min. The slides were then carefully removed from the lysis

solution and transferred to 70% ethanol for 30 min. Fixed preparations not subjected to the

lysis protocol were used as controls.

Fluorochrome staining

The fluorochrome staining followed Schweizer’s (1980) technique, with modifications. The

fluorochromes employed were 4-6 diamine-2 phenylindole (DAPI), chromomycin A3

76

(CMA3), and distamycin (DA) (Sigma-Aldrich Co., St Louis, MO, USA). The dyes were

dissolved in McIlvaine’s citric acid/ Na2HPO4 buffer at pH 7.0, except for CMA3, which

was dissolved in diluted (1:1) buffer (pH 7.0) containing 5 mM MgCl2. The preparations

were stained with CMA3 (0.5 mg/mL) for 60 min, then briefly rinsed with distilled water

followed by buffer, stained with DA (0.3 mg/mL) for 15 min, again briefly washed, and

then stained with DAPI (0.3 µg/mL) for 15 min. The preparations were next rinsed with

buffer, air dried, mounted in a well-stirred mixture of glycerol and McIlvaine’s buffer at pH

7.0 (1:1).

DNA FISH

The plasmid HM 456 which contains part of the 18S and 28S rDNA of Xenopus laevis

(Meunier-Rotival et al., 1979) was kindly provided by Dr. A. M. Vianna-Morgante (IB-

USP, São Paulo, Brazil) and the telomere-like oligonucleotide DNA sequence was

synthesized by Invitrogen (Carlsbad, CA, USA). DNA was labeled by nick-translation with

biotin-16dUTP (Bionick Labeling Kit, Life Technologies, Inc., France). The preparations

were treated with ribonuclease A (Sigma), followed by treatment with 0.01 M HCl for 2

min, washed in 2 X SSC, treated with 5 µg/mL pepsin for 10 min at 37ºC, and fixed in 4%

paraformaldehyde for 5 min, followed by denaturation with 2 X SSC/50% formamide 10

min at 80ºC. Heat-denatured probes were hybridized at 37ºC for 20 h. After hybridization

and washes with 2 X SSC/50% formamide and 2 X SSC at 42ºC, probe detection used a

two-step procedure for amplification. Biotin was revealed using mouse antibiotin and rabbit

antimouse TRITC conjugated. Coverslips were mounted in Vectashield (Vector

Laboratories, Burlingame, CA, USA) containing 0.5 g/mL DAPI.

Microscopy

The nuclear fluorescence after staining with fluorochromes was observed with a Zeiss

Axiophot 2 microscope equipped for epifluorescence with an HBO-100W stabilized

mercury lamp as the light source. Operating conditions were as follows: Zeiss Plan

77

Neofluar 25X/0.50 and 40X/0.75 objectives, condenser adequate for fluorescence, and filter

set 01 (G365 – BP exciting filter; FT 395 chromatic splitter; LP420 barrier filter).

Chromomycin fluorescence was selectively obtained by exciting with light in the

wavelength range of 440 to 480 nm. Fluorescence predominantly due to DAPI was

obtained by exciting with light of 360-400 nm. Photomicrographs were obtained with

Kodak Ultra 400 film. The sequence for photograph obeyed a sequence in which the

excitation λ for CMA3 preceded that for DA-DAPI. Photographs were digitalized and

subsequently processed (cropping, rotating, contrast, and intensity adjustments) using

Adobe Photoshop® 7.0 for Windows. Selected images were assembled using

CorelDraw®12 software.

Image acquisition of FISH stained preparations was performed with the same device

as mentioned above using Neofluar 25X/0.50, 40X/0.75, and 100X/1.4 oil-immersion

objectives, and filter set 15 (BP 546/12, exciting filter; FT 580, beamsplitter; LP 590

emission filter). TRITC fluorescence was selectively obtained by exciting with light in the

wavelength range of 565 to 605 nm. Fluorescence predominantly due to DAPI was

obtained by exciting with 360-400 nm wavelength light. Instrument settings (laser

intensity, gain) were kept constant for each experiment. Digital images were acquired using

a Sony CCD-IRIS/RGB Hyper HAD color video camera and ISIS (MetaSystems,

Altlussheim) acquisition software. Subsequent image processing and selection of images

was performed as cited above.

78

Acknowledgements

We are grateful to Nick Gilbert from the Cancer Research Centre (Edinburgh) for the

critical reading of the manuscript. This work was supported by grants from the Brazilian

National Council for Research and Development (CNPq, grant n°. 470587/03-2), the São

Paulo State Research Foundation (FAPESP, grant n°. 2006/00066-8), and CAPES. A.S.M

and M.L.S.M. received PhD and research fellowships, respectively, from CNPq. M.Mondin

received a Pro-Doc fellowship from CAPES.

79

References

Allen GC, Spiker S, Thompson WF (2000) Use of matrix attachment regions (MARs) to

minimize transgene silencing. Plant Mol. Biol. 43, 361-376.

Arhondakis S, Auletta F, Torelli G, D’Onofrio G (2004) Base composition and expression

level of human genes. Gene 325, 165-169.

Bauer M, Hamm A, Pankratz MJ (2004) Linking nutrition to genomics. Biol. Chem. 385,

593-596.

Berezney R (1991) The nuclear matrix: a heuristic model for investigating genomic

organization and function in the cell nucleus. J. Cell Biochem. 47, 109-123.

Berezney R, Coffey DS (1974) Identification of a nuclear protein matrix. Biochem.

Biophys. Res. Commun. 60, 1410-1417.

Berezney R, Coffey DS (1977) Nuclear matrix. Isolation and characterization of a

framework structure from rat liver nuclei. J. Cell Biol. 73, 616-637.

Bernardi G, Olofsson B, Filipski J, Zerial M, Salinas J, Cuny G, Meunier-Rotival M,

Rodier F (1985) The mosaic genome of warm-blooded vertebrates. Science 228,

953-958.

Burzynski SR (2005) Aging: gene silencing or gene activation? Med. Hypotheses 64, 201-

208.

Cerda MC, Berrios S, Fernandez-Donoso R, Garagna S, Redi C (1999) Organisation of

complex nuclear domains in somatic mouse cells. Biol. Cell. 91, 55-65.

Choo KH (1997) Centromere DNA dynamics: latent centromeres and neocentromere

formation. Am. J. Hum. Genet. 61, 1225-1233.

Cremer T, Kreth G, Koester H, Fink RH, Heintzmann R, Cremer M, Solovei I, Zink D,

Cremer C (2000) Chromosome territories, interchromatin domain compartment, and

nuclear matrix: an integrated view of the functional nuclear architecture. Crit. Rev.

Eukaryot. Gene Expr. 10, 179-212.

Davie JR (1995) The nuclear matrix and the regulation of chromatin organization and

function. Int. Rev. Cytol. 162A, 191-250.

80

Fransz P, Jong JH, Lysak M, Castiglioni MR, Schubert I (2002) Interphase chromosomes in

Arabidopsis are organized as well defined chromocenters from which euchromatin

loops emanate. Proc. Natl. Acad. Sci. USA 99, 14584-14589.

Fedoroff NV (1979) On spacers. Cell 16, 697-710.

Garcia-Cao M, O’Sullivan R, Peters AHFM, Jenuwein T, Blasco MA (2004) Epigenetic

regulation of telomere length in mammalian cells by the Suv39h1 and Suv39h2

histone methyltransferases. Nat. Genet. 36, 94-99.

Gasser SM, Laemmli UK (1986) Cohabitation of scaffold binding regions with

upstream/enhancer elements of three developmentally regulated genes of D.

melanogaster. Cell 46, 521-530.

Gaubatz J, Prashad N, Cutler RG (1976) Ribosomal RNA gene dosage as a function of

tissue and age for mouse and human. Biochim. Biophys. Acta 418, 358-375.

Gerdes MG, Carter KC, Moen PT, Lawrence JB (1994) Dynamic changes in the higher-

level chromatin organization of specific sequences revealed by in situ hybridization

to nuclear halos. J. Cell Biol. 126, 289-304.

Gilbert N, Boyle S, Fiegler H, Woodfine K, Carter NP, Bickmore WA (2004) Chromatin

architecture of the human genome: gene-rich domains are enriched in open

chromatin fibers. Cell 118, 555-566.

Grewal SIS, Elgin SCR (2002) Heterochromatin: new possibilities for the inheritance of

structure. Curr. Opin. Genet. Dev. 12, 178-187.

Haaf T, Ward DC (1994) Structural analysis of alpha-satellite DNA and centromere

proteins using extended chromatin and chromosomes. Hum. Mol. Genet. 3, 697-

709.

Haaf T, Ward DC (1995) Higher order nuclear structure in mammalian sperm revealed by

in situ hybridization and extended chromatin fibers. Exp. Cell Res. 219, 604-611.

Heng HH, Goetze S, Ye CJ, Liu G, Stevens JB, Bremer SW, Wykes SM, Bode J, Krawetz

SA (2004) Chromatin loops are selectively anchored using scaffold/matrix-

attachment regions. J. Cell Sci. 117, 999-1008.

Horz W, Altenburger W (1981) Nucleotide sequence of mouse satellite DNA. Nucleic

Acids Res. 9, 683-696.

81

Iarovaia OV, Bystritskiy A, Ravcheev D, Hancock R, Razin SV (2004) Visualization of

individual DNA loops and a map of loop domains in the human dystrophin gene.

Nucleic Acids Res. 32, 2079-2086.

Kapuscinski J, Szer W (1979) Interactions of 4’, 6-diamidine-2-phenylindole with synthetic

polynucleotides. Nucleic Acids Res. 6, 3519-3534.

Kominami R, Hamada H, Fujii-Kuriyama Y, Muramatsu M (1978) 5’-Terminal processing

of ribosomal 28S RNA. Biochemistry 17, 3965-3970.

Kurdistani SK, Tavazoie S, Grunstein M (2004) Mapping global histone acetylation

patterns to gene expression. Cell 117, 721-733.

Long EO, Dawid IB (1980) Repeated genes in eukaryotes. Annu Rev Biochem 49, 727-

764.

Ma H, Nagata T (1990) Study on RNA synthesis in the liver of aging mice by means of

electron microscopic radioautography. Cell. Mol. Biol. 36, 589-600.

Meunier-Rotival M, Cortadas J, Macaya G, Bernardi G (1979) Isolation and organization of

calf ribosomal DNA. Nucleic Acids Res. 6, 2109–2123.

Mirkovitch J, Mirault ME, Laemmli UK (1984) Organization of the higher-order chromatin

loop: specific DNA attachment sites on nuclear scaffold. Cell 39, 223-232.

Moraes AS, de Campos Vidal B, Guaraldo AM, Mello ML (2005) Chromatin

supraorganization and extensibility in mouse hepatocytes following starvation and

refeeding. Cytometry A 63, 94-107.

Moraes AS, Guaraldo AM, Mello ML (2007) Chromatin supraorganization and

extensibility in mouse hepatocytes with development and aging. Cytometry A 71,

28-37.

Mouchiroud D, Gautier C (1988) High codon-usage changes in mammalian genes. Mol.

Biol. Evol. 5, 192-194.

Mouchiroud D, Gautier C, Bernardi G (1988) The compositional distribution of coding

sequences and DNA molecules in humans and murids. J. Mol. Evol. 27, 311-320.

Nickerson J (2001) Experimental observations of a nuclear matrix. J. Cell Sci. 114, 463-

474.

82

Nishida H, Suzuki T, Kondo S, Miura H, Fujimura Y, Hayashizaki Y (2006) Histone H3

acetylated at lysine 9 in promoter is associated with low nucleosome density in the

vicinity of transcription start site in human cell. Chromosome Res. 14, 203-211.

Pardoll DM, Vogelstein B (1980) Sequence analysis of nuclear matrix associated DNA

from rat liver. Exp. Cell Res. 128, 466-470.

Paulson JR, Laemmli UK (1977) The structure of histone-depleted metaphase

chromosomes. Cell 12, 817-828.

Perry RP (1976) Processing of RNA. Annu Rev Biochem 45, 605-629.

Salinas J, Zerial M, Filipski J Bernard G (1986) Gene distribution and nucleotide sequence

organization in the mouse genome. Eur. J. Biochem. 160, 469-478.

Semon M, Mouchiroud D, Duret L (2005) Relationship between gene expression and CG-

content in mammals: statistical significance and biological relevance. Hum. Mol.

Genet. 14, 421-427.

Schweizer D (1980) Simultaneous fluorescent staining of R bands and specific

heterochromatic regions (DA-DAPI bands) in human chromosomes. Cytogenet.

Cell. Genet. 27, 190-193.

Sharma GG, Hwang K, Pandita RK, Gupta A, Dhar S, Parenteau J, Agarwal M, Worman

HJ, Wellinger RJ, Pandita TK (2003) Human heterochromatin protein 1 isoforms

HP1Hsα and HP1Hsß interfere with hTERT-telomere interactions and correlate

with changes in cell growth and response to ionizing radiation. Mol. Cell. Biol. 23,

8363-8376.

Sollner-Webb B, Tower J (1986) Transcription of cloned eukaryotic ribosomal RNA genes.

Annu. Rev. Biochem. 55, 801-830.

Strahl BD, Allis CD (2000) The language of covalent histone modifications. Nature 403,

41-45.

Tsutsui KM, Sano K, Tsutsui K (2005) Dynamic view of the nuclear matrix. Acta. Med.

Okayama 59, 113-120.

Vidal BC (2000) Extended chromatin fibres: crystallinity, molecular order and reactivity to

concanavalin-A. Cell Biol. Int. 24, 723-728.

83

Vidal BC, Moraes AS, Mello MLS (2006) Nucleus image properties assessed by video

image analysis in mouse hepatocytes under a short lysis for extended chromatin

fiber formation. Cytometry A 69, 1106-1113.

Waterston RH, et al. (2002) Initial sequence and comparative analysis of the mouse

genome. Nature 420, 520-562.

Wellauer PK, Reeder RH, Carroll D, Brown DD, Deutch A, Higashinakagawa T, Dawid IB

(1974) Amplified ribosomal DNA from Xenopus laevis has heterogeneous spacer

lengths. Proc. Natl. Acad. Sci. USA 71, 2823-2827.

Whitelaw CBA, Grolli S, Accornero P, Donofrio G, Farini E, Webster J (2000) Matrix

attachment region regulates basal b-lactoglobulin transgene expression. Gene 244,

73-80.

Zardo G, Fazi F, Travaglini L, Nervi C (2005) Dynamic and reversibility of

heterochromatic gene silencing in human disease. Cell Res. 15, 679-690.

89

Figure legends

Fig. 1 Chromocenters are enriched in AT-rich but not in CG-rich sequences. Fixed

hepatocyte nuclei from well-fed (a) and starved adult (b), and well-fed old (c) mice. Scale

bar: 22 µm

Fig. 2 (Verificar) CG-rich sequences preferentially attach to the nuclear matrix. Fixed and

vertically lysed hepatocyte nuclei from well-fed (a), starved adult (b), and well-fed old (c)

mice. Arrowheads: extended chromatin fibers. Arrows indicate direction of gravity. Scale

bar: 22 µm

Fig. 3 DNA FISH of 45S rDNA on fixed hepatocyte nuclei from well-fed young (a), well-

fed (b) and starved (c) adults, and well-fed old (d) mice. DNA is visualized with DAPI.

rDNA repeats are seen around the nucleoli (n). Scale bars: 12 µm

Fig. 4 rDNA 45S differently associates with the nuclear matrix in different physiological

conditions. DNA FISH of 45S rDNA on fixed and vertically lysed hepatocyte nuclei from

well-fed young (a), well-fed (b) and starved (c) adult, and well-fed old (d) mice. DNA is

visualized with DAPI. Arrows indicate direction of gravity. Scale bars: 30 µm

Fig. 5 Regardless the physiological condition, telomeres do not associate with the nuclear

matrix. DNA FISH of telomeric DNA on fixed non-lysed (a,c) and fixed and vertically

lysed (b,d) hepatocyte nuclei from well-fed young (a,b) and adult (c,d) mice. DNA was

counterstained with DAPI. e, erithroblast. h, hepatocyte. Arrow indicates direction of

gravity. Scale bars: 12 µm

90

Capítulo IV

Artigo: Extended chromatin fibers in mouse testicular spermatozoa. Artigo publicado em

Brazilian Journal of Morphological Sciences 2(2):91-96 (2005).

91Braz. J. morphol. Sci. (2005) 22(2), 91-96

ISSN- 0102-9010

Correspondence to: Dr. Maria Luiza S. MelloDepartamento de Biologia Celular, Instituto de Biologia, Universidade Estadual de Campinas (UNICAMP), CP 6109, CEP 13083-863, Campinas, SP, Brazil. Tel.: (55)(19) 3788-6122, Fax: (55) (19) 3788-6111. E-mail: [email protected]

*Dedicated to Professor Benedicto de Campos Vidal on the occasion of his 75th birthday.

EXTENDED CHROMATIN FIBERS IN MOUSE TESTICULAR SPERMATOZOA*

Alberto da Silva Moraes and Maria Luiza Silveira Mello

Department of Cell Biology, Institute of Biology, State University of Campinas (UNICAMP), Campinas, SP, Brazil.

ABSTRACT

Since in mouse spermatozoa the somatic histones are replaced by other basic proteins and there are changes

in the chromatin supraorganization, different patterns of extended chromatin fiber (ECF) formation would

be expected compared with those formed by somatic cells that were previously studied. In this study, we

investigated the formation of ECF in mouse testicular spermatozoa after lysis with 2 M NaCl plus 1% Triton

X-100, and under the action of gravity. ECFs were observed under polarized light in fixed and unfixed

spermatozoa subjected to lysis in a vertical position and stained with toluidine blue at pH 4.0. In unfixed

preparations, all of the sperm nuclei showed ECFs, whereas in fixed preparations 60% of the cells had ECF.

The latter frequency was much higher than that previously reported for mouse hepatocytes. Even in cells

that did not produce ECFs in vertically and horizontally lysed preparations, an ordered reorganization of the

chromatin was observed after lysis. The faint positive response to acid fast green at the nuclear periphery

in spermatozoa that did not develop ECF after lysis was assumed to represent residual protamine and

nuclear matrix proteins. The high frequency of mouse sperm cell nuclei with ECF probably reflected the

extraction of protamines from the DNA-protein complexes of sperm cell nuclei facilitated by the specific

lysis protocol.

Key words: Chromatin extensibility, mouse, optical anisotropy, testicular spermatozoa, topochemistry

INTRODUCTION

Chromatin and DNA flow from cell nuclei

following treatment with lysis solutions that remove

RNA, histones, and some of the non-histone proteins

from chromatin, as well as part of the lipids from the

nuclear envelope. As a result, extended chromatin

fibers (ECF) can be formed under the action of

gravity or during manual mechanical stretching

[5-7,14,21]. ECF formation depends on the

rheological properties of the DNA [13,18] and also

on proteins that contribute to higher-order chromatin

structure, as in the case of nuclear matrix proteins

[11,17,19,20,22]. Chromatin extensibility under

gravity in adult mouse hepatocyte nuclei is affected

by the organism’s nutritional state, such that the

frequency of nuclei that form ECF under the same

treatment conditions decreases drastically following

starvation and is accompanied by an increase in

chromatin condensation [11].

In sperm cells, drastic changes in chromatin

composition compared to somatic cells have been

reported, with somatic histones being totally or par-

tially replaced by histone-like basic proteins [1,12].

In spermatozoa-containing protamines (arginine-

rich; no lysine), protamine-like histones (arginine-

rich, little or no lysine, but oxidized cysteine also

being present), or intermediate basic sperm proteins

(containing histidine and/or lysine in addition to ar-

ginine), these proteins tightly pack the DNA in most

animal species, whereas in spermatozoa-containing

somatic-like lysine-rich histones, such as honey bees,

sea-urchins and Rana, the DNA-protein complex is

much less tightly packed [1,4,9,12]. Protamine-like

histones are usually found in eutherian mammals,

with mice having two protamine-like variants [2].

The supraorganization of chromatin in

mammalian sperm cells depends on the basic nuclear

protein types and on a nuclear matrix. Sperm cell

DNA organizes in loop domains that are attached to

the nuclear matrix by their matrix attachment regions

(MARs) [22]. Since the DNA in sperm cells is not

involved in transcriptional processes, the regions of

DNA attachment to the nuclear matrix probably has

only structural functions [16].

REGULAR PAPER

92

A. S. Moraes and M. L. S. Mello92

Braz. J. morphol. Sci. (2005) 22(2), 91-96

In addition to the extremely compact structure

of chromatin in mammalian sperm, different

hierarchies and heterogeneity exist in the packaging

of DNA in the nucleus of these cells [6]. Hence, the

DNA-protein complexes of mammalian sperm cells

are expected to behave differently from those of

somatic cells in the same species, and this behavior

is likely to be influenced by the nutritional state and

developmental stage, as well as by treatment with

high salt and detergent solutions that destabilize

these complexes.

In the present study, the formation of ECF

was investigated in mouse testicular spermatozoa

and the findings were compared with data for

mouse hepatocytes [11] under the same treatment

conditions.

MATERIAL AND METHODS

Male mice of the inbred strain A/Uni, obtained from

the Multidisciplinary Center for Biological Investigation

of the State University of Campinas (CEMIB/UNICAMP)

were reared under normal conditions and fed standard

extruded chow (Purina®) ad libitum until 15 weeks old.

Four specimens were killed by decapitation and their

testes were immediately removed, placed in cold saline

solution (0.9% NaCl in distilled water), and used for

the preparation of imprints on glass slides. All of the

protocols involving animal care and use were approved by

the Committee for Ethics in Animal Experimentation of

the State University of Campinas (CEEA/IB) and met the

guidelines of the Canadian Council on Animal Care.

Treatments

Freshly prepared imprints were fixed in absolute

ethanol-glacial acetic acid (3:1, v/v) for 1 min and

then rinsed in 70% ethanol for 5 min. The slides were

positioned vertically and horizontally and the imprints

were immediately lysed in 2 M NaCl plus 1% Triton X-

100 in Tris-HCl buffer (25 mM, pH 7.4) for 5 h at 25oC

to obtain ECFs, after which the volume of solution was

completed with absolute ethanol to a final concentration of

50% for 10 min. Following this, the slides were removed

from the lysis solution and transferred to 70% ethanol for

30 min [5,7,21]. Unfixed preparations subjected to the

same lysis protocol but incubated for a shorter period of

time (10 min) [11] were also used. Fixed preparations that

had not been subjected to the lysis protocol were used as

treatment controls.

Topochemistry and optical anisotropy

Staining was done with a 0.025% toluidine blue

(Merck, Darmstadt, Germany) solution in McIlvaine

buffer at pH 4.0 for 15 min [20,21]. The preparations

were then rapidly (5 s) rinsed in distilled water, air dried,

cleared in xylene, and mounted in natural Canada balsam.

This staining procedure was used because: 1) it is a classic

method in cytochemical and biochemical assay of nucleic

acids; 2) the molecules of this dye bind electrostatically to

available DNA and RNA phosphates; 3) it is possible to

detect the optical anisotropic characteristics (birefringence

and linear dichroism) of stained DNA or DNA-protein

complexes, inclusive in ECFs [11,19,21], under suitable

conditions [8,10,20]. Birefringence and linear dichroism

(selective absorption of polarized light) in toluidine blue-

stained chromatin reveals the availability and proximity

of free DNA phosphates suitable for dye binding and the

oriented arrangement of the dye binding sites in chromatin

[8,10,19]. The frequency of nuclei showing ECFs was

estimated under polarized light. Birefringence was

investigated in toluidine blue-stained cells under polarized

light using a Zeiss polarizing microscope equipped with a

16/0.32 Planachromatic objective, and was photographed

in a Zeiss Axiophot 2 microscope equipped with 40/0.75

and 100/1.30 Pol-Neofluar objectives, optovar 2, and a

1.4 condenser. Kodak Gold 100 film was used for the

photomicrographs.

Some preparations were also subjected to the Feulgen

reaction, a highly specific, stoichiometric reaction to detect

DNA. Acid hydrolysis pertinent to this assay was done in

4 M HCl at 25oC for 60 min. Some preparations were also

screened for the presence of total nuclear proteins using

acid fast green (Sigma, St. Louis, USA), in which the

anionic dye binds electrostatically to the available cationic

groups of proteins [3,8].

RESULTS

Sperm cell nuclei that were not subjected to lysis

stained light blue (unfixed preparations) or deep

violet (fixed preparations) with toluidine blue (Figs.

1 and 2). The sperm cell nuclei also stained intensely

with the Feulgen reaction and by the acid fast green

method (Figs. 3-5). The toluidine blue-stained

spermatozoal nuclei showed no birefringence when

examined with a crossed analyzer and polarizer.

When the fixed preparations were subjected

to lysis horizontally or vertically and then stained

with toluidine blue, weak metachromatic staining

was seen in the center of the sperm cell nuclei, and

was surrounded by stronger staining at the nuclear

periphery; a deeply birefringent nuclear image

was also observed (Figs. 6-8). In slides positioned

vertically, toluidine blue-stained filaments were

seen flowing from some nuclei in the form of ECFs

(Figs. 8-10). Approximately 60 % of the sperm cells

showed ECFs but no ECFs were seen in cells other

than spermatozoa (data not shown).

The ECFs were stained in the Feulgen reaction

but not by the acid fast green method (Fig. 11). Fixed

93

Chromatin extensibility in spermatozoa 93


Figures 1-5. Unfixed (1, 3) and fixed (2, 4, 5) mouse testicular imprints (control) stained with toluidine blue (1, 2), the Feulgen reaction (3, 4) and acid fast green (5). s, spermatozoa. Figures 6-10. Fixed and vertically lysed mouse sperm cells treated with toluidine blue. 7 is a polarized light view of 6, showing spermatozoa that did not form prominent ECFs. 8-10. General view (8) and details (9, 10) of ECFs as seen un-der polarized light. The arrows indicate the direction of gravity. Bars = 20 m.

94



sperm cells that were subjected to lysis showed

nuclear images that stained faintly with acid fast

green, whereas the sperm cell tails were deeply

stained (Fig. 12). Compared to the sperm cells, the

nuclei of the spermatogonia showed filamentous,

granular elements and a nucleolar-like body that

stained very well with acid fast green (Fig. 13).

In unfixed preparations treated vertically

with the short lysis protocol, practically all of the

sperm cell nuclei produced long ECFs that stained

metachromatically with toluidine blue and showed

birefringence (Figs. 14 and 15), and responded

positively to the Feulgen reaction (Fig. 16). In

the slides positioned horizontally, a halo of thin

birefringent filaments was present at the nuclear

periphery of the spermatozoa and in meiotic

chromosomes (Figs. 17 and 18).

DISCUSSION

As shown here, ECFs were produced in mouse

sperm cells by the action of gravity after some

components of the chromatin, nuclear matrix and

nuclear envelope had been disrupted and removed

by lysis. ECF formation results from increased

chromatin fluidity as a consequence of the breakdown

of most DNA/nuclear protein interactions by saline

treatment plus the disruption of components of the

nuclear envelope by the non-ionic detergent Triton

X-100 [5]. The ECFs formed in mouse sperm cells

consisted mostly of DNA, as shown by the Feulgen

reaction and toluidine blue staining, and by the

optical anisotropy following toluidine blue binding

[20,21], as well as the practically negative response

to acid fast green [3]. These findings differed from

those for mouse hepatocytes in which nuclear protein

granules were well-identified throughout the ECFs

[11]. The anisotropical images seen in ECFs are

a function of the highly ordered organization and

alignment of the DNA in these fibers [21].

Lysis was more effective in unfixed preparations,

although fixation relaxed or removed part of the

nuclear protein components from the nuclei [11,15]

sufficiently to generate additional free phosphate

groups suitable for binding toluidine blue. This

binding is responsible for metachromasy (violet

color) [8,10,20] in the DNA of mouse spermatozoa.

In fixed, vertically lysed mouse testicular

preparations, only part of the spermatozoal popu-

lation formed ECFs. Even so, the frequency of

spermatozoa ECF formation was much higher (60%)

than that for hepatocytes under same experimental

conditions (maximum, 22%) [11], or even for cells

other than spermatozoa present in the testicular

preparations. The apparent ease of ECF formation in

mouse spermatozoa occurred despite the extremely

compact chromatin structure present in these

mammalian cells [6]. Although a tightly packed

chromatin organization is assumed for DNA-protein

complexes containing protamines and protamine-

like proteins [1,12], the protocol used here for

ECF formation in the presence of a reducing agent

is proposed to solubilize protamines for studies of

nuclear matrix proteins in rat spermatozoa [16]. This

could explain the high frequency of ECF formation

in mouse spermatozoa, even in the absence of a

reducing agent in the lysis protocol, but could also

account for the finding that some of the sperm cells

showed no formation of typical ECF, perhaps because

the nuclei of these cells contained some protamine-

like proteins, as well as a nuclear matrix [6,16]. Even

in the sperm cells that did not produce ECFs in fixed

and vertically or horizontally lysed preparations, a

change in the distribution of the sites responsible for

toluidine blue binding, and the presence of a deep

birefringent image were detected in the cell nuclei

and chromosomes, thus demonstrating an ordered

reorganization of the chromatin in response to the

different treatments.

The lack of a staining with acid fast green in

the center of treated sperm cell nuclei rather than

at the nuclear periphery suggests the presence of

discrete protamine remnants and nuclear matrix

proteins at this location. These observations agree

with the consideration that different hierarchies and

heterogeneity exist in the packaging of DNA by

proteins at different sites in the mammalian sperm

cell nucleus [6]. The abundant and prominent

reactivity of the nuclear matrix and nucleolar

proteins of spermatogonia with acid fast green after

lysis and the action of gravity, and the faint staining

with acid fast green at the nuclear periphery of

spermatozoa, indicate that nuclear matrix proteins

change in nature and/or composition during

spermatogenesis in the mouse.

ACKNOWLEDGMENTS

The authors thank Dr. Ana Maria A. Guaraldo (CEMIB, UNICAMP) for supplying the mice and to the Brazilian National Research and Development Council (CNPq) for financial support.

95

Chromatin extensibility in spermatozoa 95


Figures 11-13. Fixed and vertically lysed cell nuclei treated with acid fast green, showing an essentially negative response in sperm ECFs (11), and well-stained sperm tails (t, 12), and filaments, granules and a nucleolar-like body (nu) inside spermatogonial nuclei (13). s, spermatozoa.Figures 14-16. Unfixed and vertically lysed sperm cell nuclei stained with toluidine blue (14, 15) and the Feulgen reac-tion (16). 15 is a polarized light view of 14. Figures 17 and 18. Unfixed and horizontally lysed sperm cells treated with toluidine blue. 18 is a polarized light view of 17, showing a halo of thin birefringent filaments encircling the sperm cell nuclei and meiotic chromosomes (ch). The arrows indicate the direction of gravity. Bars = 20 m.

96



REFERENCES

1. Bloch DP (1969) A catalog of sperm histones. Genet-ics 61, 93-111.

2. Calvin HI (1976) Comparative analysis of the nucle-

ar basic proteins in rat, human, guinea pig, mouse,

and rabbit spermatozoa. Biochim. Biophys. Acta 434,

377-389.

3. Deitch AD (1966) Cytochemistry of nucleic acids. In:

Introduction to Quantitative Cytochemistry (Wied GL,

ed). pp. 451-468. Academic Press: New York.

4. Falco JRP, Mello MLS (1999) Critical electrolyte con-

centration of spermatozoal chromatin containing his-

tone H1 variants. Genet. Mol. Biol. 22, 197-200.

5. Haaf T, Ward DC (1994) Structural analysis of -sat-

ellite DNA and centromere proteins using extended

chromatin and chromosomes. Hum. Mol. Genet. 3,

697-709.

6. Haaf T, Ward DC (1995) Higher order nuclear struc-

ture in mammalian sperm revealed by in situ hybrid-

ization and extended chromatin fibers. Exp. Cell Res. 219, 604-611.

7. Heng HHQ, Squire J, Tsui LC (1992) High-resolution

mapping of mammalian genes by in situ hybridiza-

tion to free chromatin. Proc. Natl. Acad. Sci. USA 89,

9509-9513.

8. Mello MLS (1997) Cytochemistry of DNA, RNA and

nuclear proteins. Braz. J. Genet. 20, 257-264.

9. Mello MLS, Falco JRP (1996) Critical electrolyte con-

centration of DNA-protein complexes in spermatozo-

al and somatic cell nuclei of the honey bee, Apis mel-lifera. Insect Biochem. Mol. Biol. 26, 793-795.

10. Mello ML, Vidal BC (1973) Linear dichroism and

anomalous dispersion of birefringence on bee sperm

heads. Acta Histochem. 45, 109-114.

11. Moraes AS, Vidal BC, Guaraldo AMA, Mello MLS

(2005) Chromatin supraorganization and extensibility

in mouse hepatocytes following starvation and refeed-

ing. Cytometry A 63, 94-107.

12. Oliva R, Dixon GH (1991) Vertebrate protamine genes and the histone-to-protamine replacement reaction. Progr. Nuclei. Acid Res. Mol. Biol. 40, 25-94.

13. Perkins TT, Smith DE, Chu S (1997) Single polymer dynamics in an elongational flow. Science 276, 2016-2021.

14. Poirier M, Eroglu S, Chatenay D, Marko JF (2000) Reversible and irreversible unfolding of mitotic newt chromosomes by applied force. Mol. Biol. Cell 11, 269-276.

15. Retief AE, Rüchel R (1977) Histones removed by fix-ation: their role in the mechanism of chromosomal banding. Exp. Cell Res. 106, 233-237.

16. Santi S, Rubbini S, Cinti C, Squarzoni S, Matteucci A, Caramelli E, Guidotti L, Maraldi NM (1994) Nuclear matrix involvement in sperm head structural organization. Biol. Cell 81, 47-57.

17. Sivak A, Wolman SR (1974) Chromosomal proteins in fixed metaphase cells. Histochemistry 42, 345-349.

18. Strick TR, Allemand JF, Bensimon D, Bensimon A, Croquette V (1996) The elasticity of a single super-coiled DNA molecule. Science 271, 1835-1837.

19. Vidal BC (1972) Doppelbrechungsdispersion und Lin-eardichroismus von Eu- und Heterochromatin nach Färbung mit Toluidinblau. Nachweis eines Cotton-Ef-fektes. Beitr. Path. 145, 269-285.

20. Vidal BC (1987) Métodos em Biologia Celular. In: Bi-ologia Celular (Vidal BC, Mello MLS, eds). pp. 5-39. Livraria Atheneu: Rio de Janeiro.

21. Vidal BC (2000) Extended chromatin fibers: crystal-linity, molecular order and reactivity to concanavalin A. Cell Biol. Int. 24, 723-728.

22. Ward WS, Coffey DS (1990) Specific organization of genes in relation to the sperm nuclear matrix. Bio-chem. Biophys. Res. Commun. 173, 20-25.

Received: November 29, 2004Accepted: April 6, 2005

97

Capítulo V

Artigo: The Con A-peroxidase method for tissue localization of glucosyl and mannosyl

groups applied to mouse hepatocytes and chicken erythrocytes. Artigo publicado em Acta

Histochemica 108:475-479 (2006).

DOI:10.1016/j.acthis.2006.09.001

98

Acta histochemica 108 (2006) 475—479

The Con-A-peroxidase method for tissue

localization of glucosyl and mannosyl groups

applied to mouse hepatocytes and

chicken erythrocytes

Alberto S. Moraes, Maria Luiza S. Mello�

Department of Cell Biology, Institute of Biology, State University of Campinas (Unicamp),

P.O. Box 6019, 13083-863, Campinas, Sao Paulo, Brazil

KEYWORDS

Chicken erythro-cytes;Concanavalin A;Con-A-peroxidase;Lectin;Mouse hepatocytes;Nuclear glycopro-teins;Nuclear matrix

SummaryA variation of the Concanavalin A (Con-A)-peroxidase labelling method originallydescribed by Kiernan [Localization of alpha-D-glucosyl and alpha-D-mannosyl groupsof mucosubstances with Concanavalin A and horseradish peroxidase. Histochemistry1975;44:39–45] was applied to unsectioned cell preparations, with an emphasis onthe nuclear localization of glycoproteins. Mouse liver imprints and chicken bloodsmears fixed in acetic acid-ethanol solution were studied. Modifications of themethod included using increased Con-A concentration, and a range of pH values forthe Con-A solutions. The strongest Con-A labelling of both erythrocytes andhepatocytes was obtained after incubation with Con-A at pH 6.5 and with Con-Aconcentrations at least two-fold greater than those used for tissue sections. Theseconditions may alter the Con-A conformation, enabling the lectin molecule to enterthe cell nucleus and bind to nuclear glycoproteins, thus allowing their localizationand quantification.& 2006 Elsevier GmbH. All rights reserved.

Introduction

Lectins are proteins that bind to sugar moieties incell walls or membranes and change the physiologyof the cell membrane to cause agglutination,

mitosis, and other intracellular biochemical altera-tions (Goldstein et al., 1980). Concanavalin A (Con-A), a protein extracted from the seeds of Canavaliaensiformis (jack bean), is a member of this sugar-binding protein family. Con-A has affinity for a-D-glucosyl and a-D-mannosyl residues (Lis and Sharon,1973).

The conformation of Con-A is pH-dependent. AtpH below 5.6, the protein exists as a dimer, while at

ARTICLE IN PRESS

www.elsevier.de/acthis

0065-1281/$ - see front matter & 2006 Elsevier GmbH. All rights reserved.doi:10.1016/j.acthis.2006.09.001

�Corresponding author. Tel.: +551937886122,fax: +551937886111.

E-mail address: [email protected] (M.L.S. Mello).

99

pH 5.6–7.0, the main Con-A conformation is atetramer, and at pH47.0, higher aggregates areformed (Kanellopoulos et al., 1996). Con-A boundto fluorescent probes and metals has been used influorescence and transmission electron microscopy,respectively, for the cytochemical localization ofsugar residues in tissues and cells. Kiernan (1975)described a technique for light microscopy in whichCon-A was used in association with horseradishperoxidase, a heavily mannosylated enzyme.According to Kiernan (1975), at pH 7.2 Con-A hasmore than one site of attachment to carbohy-drates. When applied to tissue sections, Con-Abinds to available carbohydrate residues andperoxidase binds to Con-A through its mannosylresidues. The diaminobenzidine oxidation reactioncatalyzed by peroxidase in the presence of hydro-gen peroxide allows localization of the bound sugarresidues (Kiernan, 1990).

Raw extracts of Con-A in solution at pH 6.5provide strong labelling of the plasma and nuclearmembrane regions of chicken erythrocytes andstrong intranuclear labelling of chromocenter areasin mouse hepatocyte nuclei (Vidal et al., 1997;Moraes et al., 2005). Since Kiernan’s method(Kiernan, 1975, 1990) was developed for tissuesections using a pH at which Con-A formshigh aggregates, in this study we examined thepattern of nuclear labelling when the Con-A-peroxidase method was applied to unsectionedtissues, and also assessed the influence of pH andconcentration of Con-A on this response. Improve-ments of the original methodology proposed byKiernan (1975) may be useful for the in situlocalization and quantification of nuclear glycopro-teins in different materials, since these glycopro-teins appear to be involved in important nuclearprocesses (Hawkes, 1982; Kan and da Silva, 1986;Hart et al., 1989; Monnier, 1989, 1990; Ferraroet al., 1991; Comer and Hart, 2000; Iyer andHart, 2003).

Material and methods

Male mice from the inbred strain A/Uni, obtainedfrom the Multidisciplinary Center for BiologicalInvestigation (CEMIB) of the State University ofCampinas, Brazil, were reared under standardconditions and fed extruded chow (Purinas, Pau-linia, Brazil) with access to water ad libitum. 21-day-old male chickens were maintained in 0.5m2

cages under standard conditions and fed a balancedchicken diet (Purina, Brazil), with access to waterad libitum. At least three animals of each species

were used. All protocols involving animal care anduse were approved by the Institutional CommitteeFor Ethics in Animal Experimentation (CEEA/IB/UNICAMP) and met the guidelines of the CanadianCouncil on Animal Care.

The mice were decapitated and their liversimmediately removed and placed in cold 0.9% w/vNaCl solution. Liver slices were imprinted onhistological slides (4–5 slides per animal). Hepar-inized whole blood (5 U heparin/1mL blood)collected from the branchial vein of chickenswas used to make blood smears on histologicalslides (4–5 slides per animal). Both materials werefixed in 3:1 v/v absolute ethanol/glacial acetic acidfor 1min and rinsed in 70% ethanol for 5min.Endogenous peroxidase activity was quenched bypretreatment with absolute ethanol plus 2% con-centrated hydrochloric acid for 10min. The slideswere then incubated at room temperature withCon-A solutions (Con-A type IV, Sigma Chemical Co.,St. Louis, MO, USA) at concentrations of 0.5 and1mg/mL, in 0.06M sodium phosphate buffers at pH5.0, 6.5, and 7.2 for 10min, or with Con-A solutionsplus 400mg/mL sucrose as negative controls. Thiswas followed by treatment with horseradish perox-idase (HRP) solution (0.04mg/mL in 0.06M sodiumphosphate buffer, pH 7.2) for 10min. Slides werethen incubated with peroxidase substrate solutioncontaining 1mg/mL 3,30-diaminobenzidine (FlukaChemie, Switzerland), and 0.01% H2O2. All Con-Asolutions were supplemented with 5mM MgCl2,MnCl2, and CaCl2, since Con-A is a metalloproteinand requires a transition metal ion, such asmanganese and calcium ions for binding (Sumnerand Howell, 1936). To confirm that the localizationof the Con-A labelling was nuclear, the slides werecounterstained in 0.01% purified aqueous methylgreen (Merck, Darmstadt, Germany) for 30min.Slides were then air dried, cleared in xyleneand mounted in Canada balsam. The imageswere obtained using a digital imaging system witha light microscope (Eclipse E800, Nikon, Japan), aCool Snap-Pro Color camera (Media Cybernetics)and capture software (Image-Pro Plus, MediaCybernetics).

Results

Chicken erythrocytes did not label with Con-A inthe presence of sucrose (Fig. 1A), but labeled afterincubation with Con-A solutions without sucrose, ata concentration of 0.5mg/mL at pH 7.2 and 6.5(Fig. 1B and C). Nuclear labelling was stronger atpH 6.5, and mainly localized in the nuclear

ARTICLE IN PRESS

A.S. Moraes, M.L.S. Mello476

100

envelope region, while labelling of the glycocalixwas stronger at pH 7.2. Weak labelling wasdetected at pH 5.0 (Fig. 1D).

In mouse liver imprints, labelling with Con-A wasalso inhibited in the presence of sucrose (Fig. 1E).In the absence of sucrose, labelling with Con-A at aconcentration of 0.5mg/mL was restricted to

cytoplasmic remnants (not shown). When the lectinwas used at the higher concentration of 1mg/mL,stronger labelling was restricted to the cytoplasmicremnants, mainly glycogen, at pH 7.2 (Fig. 1F).There was strongly positive intranuclear Con-Alabelling only at pH 6.5 (Fig. 1G), and the labellingwas almost absent at pH 5.0 (Fig. 1(H)).

ARTICLE IN PRESS

Fig. 1. Con-A labelling of fixed chicken erythrocyte (A-D) and hepatocyte (E-H) nuclei. A and E are negative controls.Green shows nuclei counterstained with methyl green. Con-A positive areas label brown. n ¼ nuclear envelope areaand g ¼ glycocalix. Asterisks indicate glycogen and other cytoplasmic remnants. Arrows indicate the Con-A intranuclearpositive granules. Bars ¼ 25 mm.

Con-A-peroxidase method applied to whole nuclei 477

101

Discussion

In tissue sections, sugar residues are highlyavailable for lectin binding. The effects of the pHat which Con-A is used and, consequently, thespatial configuration of the lectin, are unimpor-tant. However, when cultured cells, cell smears orimprints are used, the pH and concentration of theCon-A solution may be relevant. High aggregates ofCon-A in solutions at pH47.0 can cross the plasmamembrane but not the nuclear envelope, assuggested by the results of the study reportedhere. Hence, if intranuclear labelling of glycopro-teins is required, it is necessary to decrease the pHof the Con-A solution to between 6.0 and 7.0. Inthis pH range, Con-A exists mainly as tetramersthat can enter the cell nucleus and bind tointranuclear sugar residues. Indeed, Con-A tetra-mers have a molecular weight of 104 kDa, withdimensions that allow the protein complex to passthrough the nuclear pore complex (Greer et al.,1970; Pante and Kann, 2002), thus explaining thestronger reactivity seen in hepatocyte nuclei aftertreatment with Con-A at pH 6.5 compared with pH7.2. Another advantage of using Con-A at pH 6.5 isthat the background staining is weaker than thatseen at pH 7.2, thus allowing a clearer visualizationof the specific nuclear labelling. High backgroundstaining is a particular problem of using liver tissuebecause of the large amount of glycogen in thistissue. At pHo5.6, there was almost no Con-Alabelling, even when, at this pH, the Con-A formsdimers that could enter the nucleus more easilythan tetramers. Since the interaction between Con-A and carbohydrates is dependent on the pH of thesolution, a low pH may decrease, or completelyabolish, this interaction (Olson and Liener, 1967;Yariv et al., 1968). In the study reported here, theeffects of pH on carbohydrate availability of thesubstrate can be ignored since the slides were pre-fixed. Additionally, dimers of Con-A bind only oneHRP molecule while bound to the substrate, thusproviding a very faint labelling when compared tothat of Con-A tetramers, which can bind three HRPmolecules after binding to the substrate.

Ferraro et al. (1994) stated that there is only asmall quantity of glycoprotein in the nuclear matrixof chicken hepatocytes. A similar situation inchicken erythrocyte nuclei could explain the weakintranuclear Con-A labelling reported here. Thestrong positivity seen in the nuclear enveloperegion of these cells after incubation with Con-Aat pH 6.5 may have been caused by glycoproteinsnormally present on the lumenal surface of thenuclear membrane (da Silva et al., 1981), while thestronger cell membrane positivity observed after

incubation of the lectin at pH 7.2 is a consequenceof the binding of high aggregates of Con-A to theglycocalyx. According to Kiernan (1975), weaklabelling may be improved by including osmiumtetroxide in the procedure. However, when rawextracts of Con-A were employed at a concentra-tion much higher than that of purified Con-A asused by Kiernan (1975), more intense labelling wasreported (Vidal et al., 1997; Moraes et al., 2005). Inthe present study, the addition of osmium tetr-oxide, a toxic compound, was avoided by simplyincreasing the Con-A concentration.

The modifications to the Con-A-peroxidase meth-od originally reported by Kiernan (1975) describedin this study may prove useful for studying altera-tions in the concentration and composition ofnuclear glucose/mannose-rich glycoproteins. Thesemoieties are of interest as they are associated withmany physiological processes (Moraes et al., 2005).

Acknowledgments

The authors thank Dr. Ana Maria A. Guaraldo(CEMIB, UNICAMP) for supplying the mice, Dr. AureoTatsumi Yamada for allowing the use of the imagecapture system in his laboratory, and to theBrazilian National Research and DevelopmentCouncil (CNPq), FAEPEX/UNICAMP and FAPESP (SaoPaulo, Brazil) for financial support.

References

Comer FI, Hart GW. O-glycosylation of nuclear andcytosolic proteins. Dynamic interplay between O-GlcNAc and O-phosphate. J Biol Chem 2000;38:29179–82.

da Silva PP, Torrisi MR, Kachar B. Freeze-fracturecytochemistry: replicas of critical point-dried cellsand tissues after fracture-label. J Cell Biol 1981;91:361–72.

Ferraro A, Grandi P, Eufemi M, Altieri F, Cervoni L, TuranoC. The presence of N-glycosylated proteins in cellnuclei. Biochem Biophys Res Commun 1991;178:1365–70.

Ferraro A, Eufemi M, Altieri F, Cervoni L, Turano C.Glycoproteins of the nuclear matrix from chicken livercells. Cell Biol Int 1994;18:655–91.

Goldstein IJ, Hughes RC, Monsigny M, Osawa T, Sharon N.What should be called a lectin? Nature 1980;285:66.

Greer J, Kaufman HW, Kalb AJ. An X-ray crystallographicstudy of Concanavalin A. J Mol Biol 1970;48:365–6.

Hart GW, Haltiwanger RS, Holt GD, Kelly WG. Glycosyla-tion in the nucleus and cytoplasm. Annu Rev Biochem1989;58:841–74.

ARTICLE IN PRESS

A.S. Moraes, M.L.S. Mello478

102

Hawkes R. Identification of Concanavalin A-bindingproteins after dodecyl sulfate-gel electro-phoresis and protein blotting. Anal Biochem 1982;123:143–6.

Iyer SPN, Hart GW. Dynamic nuclear and cytoplasmicglycosylation: enzymes of O-GlcNAc cycling. Biochem-istry 2003;42:2493–9.

Kan FW, da Silva PP. Preferential association of glycopro-teins to the euchromatin regions of cross-fracturednuclei is revealed by fracture-label. J Cell Biol1986;102:576–86.

Kanellopoulos PN, Pavlou K, Perrakis A, Agianian B,Vorgias CE, Mavrommatis C, et al. The crystalstructure of the complexes of Concanavalin A with40-nitrophenyl-alpha-D-mannopyranoside and 40-nitro-phenyl-alpha-D-glucopyranoside. J Struct Biol 1996;116:345–55.

Kiernan JA. Localization of alpha-D-glucosyl and alpha-D-mannosyl groups of mucosubstances with Concanava-lin A and horseradish peroxidase. Histochemistry1975;44:39–45.

Kiernan JA. Histological and histochemical methods:theory and practice, 2nd ed. Oxford: Pergamon; 1990.

Lis H, Sharon N. The biochemistry of plant lectins(phytohemagglutinins). Annu Rev Biochem 1973;42:541–74.

Monnier VM. Mailard theory of ageing. In: Baynes JW,Monnier VM, editors. The Mailard reaction in aging,diabetes, and nutrition. New York: Alan R Liss; 1989.p. 1–22.

Monnier VM. Nonenzymatic glycosylation, the Mailiardreaction and the aging process. J Gerontol 1990;45:B105–11.

Moraes AS, Vidal BC, Guaraldo AMA, Mello MLS. Chroma-tin supraorganization and extensibility in mousehepatocytes following starvation and refeeding. Cyto-metry A 2005;63A:94–107.

Olson MO, Liener IE. Some physical and chemicalproperties of Concanavalin A, the phytohemagglutininof the jack bean. Biochemistry 1967;6:105–11.

Pante N, Kann M. Nuclear pore complex is able totransport macromolecules with diameters of about39 nm. Mol Biol Cell 2002;13:425–34.

Sumner JB, Howell SF. The role of divalent metals in thereversible inactivation of jack bean hemagglutinin.J Biol Chem 1936;115:583.

Vidal BC, Maria SS, Klaczko LB. Concanavalin A-reactivenuclear matrix glycoproteins. Braz J Genet 1997;20:631–8.

Yariv J, Kalb AJ, Levitzki A. The interaction of Con-canavalin A with methyl alpha-D-glucopyranoside.Biochim Biophys Acta 1968;165:303–5.

ARTICLE IN PRESS

Con-A-peroxidase method applied to whole nuclei 479

103

Discussão geral

Diversas alterações na organização e composição nuclear e cromatínica foram

observadas neste trabalho com o desenvolvimento e o envelhecimento em hepatócitos de

camundongos. A análise de imagem revelou que tanto em animais jovens quanto idosos, a

estrutura global da cromatina interfásica encontra-se menos compactada que em animais

adultos. No caso dos hepatócitos de animais jovens, a descompactação da cromatina pode

ter uma relação com a alta atividade transcricional já observada previamente. No caso do

fígado de animais idosos, onde a taxa metabólica é bastante baixa, assim como a atividade

genética, poder-se-ia esperar um aumento no grau de compactação da cromatina.

Entretanto, o processo de envelhecimento é acompanhado por uma série de defeitos na

manutenção da integridade do genoma, principalmente no que diz respeito à manutenção

das pontas dos cromossomos, os telômeros. Com uma baixa atividade da telomerase,

enzima responsável pela manutenção do comprimento dos telômeros, provavelmente haja

um encurtamento dessas estruturas após ciclos repetidos de replicação o que levaria a uma

perda de material genético, mutações, e instabilidade genômica. Em leveduras, os

telômeros são o depósito nuclear de domínios heterocromáticos, os quais são responsáveis

pelo espalhamento da heterocromatina para regiões eucromáticas, quando a atividade

genética dessas não é mais necessária. Possivelmente um mecanismo semelhante ocorra em

mamíferos, e a perda dos telômeros, além de causar uma perda de proteção nas pontas dos

cromossomos, possa causar uma perda de sequências heterocromáticas e portanto uma

descompactação global da cromatina, apesar de uma atividade transcricional reduzida.

O equilíbrio entre o nível de compactação da cromatina e da associação desta com a

matriz nuclear (e este parece ser o fator principal) dita as propriedades viscoelásticas da

cromatina. Assim, em animais jovens, apesar de ocorrer uma cromatina mais

descompactada, esta exerça maior interação com a matriz nuclear, decorrente de uma alta

atividade transcricional, o que pode ser evidenciado pela ausência de formação de fibras

estendidas de cromatina nesses núcleos, como observado neste trabalho. Em animais

idosos, uma cromatina mais descompactada e menos aderida à matriz nuclear, graças a uma

104

atividade genética diminuída, levaria a um aumento na viscoelasticidade da cromatina e

consequentemente na sua capacidade em formar fibras estendidas (ECF). Corroborando

essas conclusões: 1 – os resultados para formação de ECF em espermatozóides de

camundongos, onde a transcrição é baixa ou nula, mostram uma grande facilidade para a

formação dessas fibras em tais células, apesar do altíssimo grau de compactação de sua

cromatina, e pouca adesão a uma matriz nuclear; 2 – os resultados de FISH para seqüências

de rDNA 45S e telômeros mostram que, genes com atividade transcricional diminuída ao

longo do envelhecimento, tendem a se desligar da matriz nuclear, enquanto seqüências

normalmente não codificantes associam-se pouco à matriz nuclear, independente do estado

fisiológico do organismo e consequentemente, da célula; 3 – seqüências ricas em GC que,

no caso do camundongo, possuem de 70 a 80% dos genes, estão muito aderidas à matriz

nuclear, enquanto as seqüências ricas em AT, que são em sua maioria heterocromáticas e

pobres em genes, localizam-se tanto aderidas à matriz nuclear quanto nos halos nucleares e

fibras estendidas de cromatina após um tratamento de lise.

Como a organização da fibra de cromatina na sua estrutura mais íntima estaria

relacionada com as modificações vistas em nível mais global? A resposta a esta questão

veio dos experimentos bioquímicos envolvendo a quantificação de proteínas associadas

com a formação de heterocromatina, das modificações pós-traducionais das caudas das

histonas associadas com cromatina compactada ou descompactada, e do estudo das

glicoproteínas nucleares ricas em glicose e/ou manose. A presença destas nas regiões de

heterocromatina pericentromérica aponta para uma função de tais proteínas na compactação

da cromatina. Observou-se aumento das três proteínas HP1 responsáveis pela formação de

cromatina compacta, aumento das metilações de caudas de histonas relacionadas com

heterocromatização, diminuição das metilações e acetilações relacionadas com transcrição,

e aumento na quantidade de glicoproteínas ricas em glicose e/ou manose, com

espalhamento destas para as regiões eucromáticas. Estes achados apontam para uma fibra

de cromatina mais compacta em animais idosos, o que seria uma contradição com relação

aos dados de análise de imagem. Entretanto, os resultados com espectrometria de massa

para histonas totais apontaram para uma dimunuição no total de modificações pós-

traducionais de histonas com o envelhecimento, o que estaria de acordo com uma possível

105

descompactação na estrutura global da cromatina. Dados bioquímicos referentes aos

animais jovens ainda são necessários. Estes poderão ser obtidos tão logo um protocolo

adequado para isolar núcleos de hepatócitos de fígados de camundongos jovens venha a ser

obtido, o que vem sendo dificultado, uma vez que o fígado nos animais jovens apresenta

uma grande quantidade de eritroblastos, já que é ainda um órgão hematopoiético.

Dados prévios nossos já haviam sido publicados com relação à influência do jejum

na organização da cromatina. Os dados aqui descritos mostraram que em termos de adesão

de genes para rRNA 45S à matriz nuclear nenhuma clara diferença pode ser notada entre

animais em jejum e animais bem alimentados, apesar do já descrito aumento no grau de

compactação da cromatina após jejum de 48 h. Esse fato pode estar relacionado com uma

mudança muito pequena nos níveis de transcrição no fígado de animais em jejum, uma vez

que, ao mesmo tempo em que uma série de genes têm sua atividade diminuída, o que

poderia levar a uma compactação de seus domínios, um número semelhante de genes

possui atividade aumentada. Portanto, em termos de produção de rRNAs, nenhuma ou

pouca diferença seria observada. Entretanto, dados quantitativos do RNA total produzido

tanto em fígado de animais bem alimentados quanto de animais em jejum são ainda

necessários para confirmar se existem diferenças.

�

��

�

Conclusões gerais

1. Os presentes dados mostram que o jejum e o envelhecimento induzem mudanças na

estrutura e organização da cromatina em hepatócitos de camundongo. Em ambos,

camundongos jovens e idosos, houve uma descompactação da cromatina, mas que

só é condizente com uma alta atividade transcricional no caso dos camundongos

jovens, já que em animais idosos seria de se esperar um aumento no grau de

empacotamento da cromatina em resposta a uma reduzida atividade transcricional.

2. Análise da estrutura íntima da cromatina em hepatócitos de camudongos idosos

mostrou, entretanto, acúmulo de marcadores epigenéticos e proteínas para formação

de heterocromatina, o que condiz com a baixa taxa transcricional de hepatócitos de

camundongos idosos. Esta contradição com os resultados de análise de imagem, que

mostram descompactação da cromatina em animais idosos, pode ser explicada

possivelmente por um aumento na instabilidade genômica nesses animais, que é

causada pela perda de porções telôméricas comuns em células senescentes.

3. As diferenças na atividade transcricional de genes para rRNA observadas entre

camundongos jovens, adultos e idosos refletem os diferentes níveis de interação

destes com a matriz nuclear, mas diferenças na interação da cromatina como um

todo com a matriz nuclear não foram encontradas entre esses modelos ou em

camundongos adultos após jejum.

4. Alterações no conteúdo de glicoproteínas nucleares associadas à cromatina foram

encontradas ao longo do envelhecimento ou após jejum. Entretanto, a função dessas

glicoproteínas na organização e funcionamento da cromatina ainda é incerta e para

ser definida depende da identificação e caracterização destas.

5. Em espermatozóides de camundongo, o aumento nas propriedades viscoelásticas da

cromatina são o reflexo de uma diferente organização dessa cromatina, no que diz

respeito ao seu grau de compactação e de uma reduzida interação dessa com a

matriz nuclear.

��

�

Referências bibliográficas

Aagaard L, Laible G, Selenko P, Schmid M, Dorn R, Schotta G, Kuhfittig S, Wolf A,

Lebersorger A, Singh PB, Reuter G, Jenuwein T. 1999. Functional mammalian

homologues of the Drosophila PEV-modifier Su(var)3-9 encode centromere-associated

proteins which complex with the heterochromatin component M31. EMBO J, 18:1923-

1938.

Abranches R, Beven AF, Aragón-Alcaide L, Shaw PJ. 1998. Transcription sites are not

correlated with chromosome territories in wheat nuclei. J Cell Biol, 143:5-12.

Adachi Y, Kas E, Laemmli UK. 1989. Preferential, cooperative binding of DNA

topoisomerase II to scaffold-associated regions. EMBO J, 8:3997-4006.

Amaral MJLV, Mello MLS. 1989. Critical electrolyte concentration of heterochromatin and

euchromatin in cells of starved animals. Cytobios, 59:159-165.

Avner P, Heard E. 2001. X-chromosome inactivation: counting, choice and initiation.

Nature Rev Genet, 2:59-67.

Bannister AJ, Zegerman P, Partridge JF, Miska EA, Thomas JO, Allshire RC, Kouzarides

T. 2001. Selective recognition of methylated lysine 9 on histone H3 by the HP1

chromo domain. Nature, 410:120-124.

Bednar J, Horowitz RA, Grigoryev SA, Carruthers LM, Hansen JC, Koster AJ, Woodcock

CL. 1998. Nucleosomes, linker DNA, and linker histone form a unique structural motif

that directs the higher-order folding and compaction of chromatin. Proc Natl Acad Sci

USA, 95:14173-14178.

Belmont AS, Bruce K. 1994. Visualization of G1 chromosomes: a folded, twisted,

supercoiled chromonema model of interphase chromatid structure. J Cell Biol,

127:287-302.

Berezney R. 1991. The nuclear matrix: a heuristic model for investigating genomic

organization and function in the cell nucleus. J Cell Biochem, 47:109-123.

Berezney R, Coffey DS. 1977. Nuclear matrix: isolation and characterization of a

framework structure from rat liver nuclei. J Cell Biol, 73:616-637.

Bernard P, Maure JF, Partridge JF, Genier S, Javerzat JP, Allshire RC. 2001. Requirement

of heterochromatin for cohesion at centromeres. Science, 294:2539-2542.

Bernardi G. 1993. The vertebrate genome: isochores and evolution. Mol Biol Evol, 10:186-

204.

��

�

Bernardi, G. 2000. Isochores and the evolutionary genomics of vertebrates. Gene, 241:3-17.

Bitterman KJ, Medvedik O, Sinclair D. 2003. Longevity regulation in Saccharomyces

cerevisiae: Linking metabolism, genome stability, and heterochromatin. Microbiol Mol

Biol Rev, 67:376-399.

Blasco MA. 2004. Telomere epigenetics: a higher–order control of telomere length in

mammalian cells. Carcinogenesis, 25:1083-1087.

Bloch DP. 1969. A catalog of sperm histones. Genetics, 61:93-111.

Bode J, Schlake T, Rios-Ramirez M, Mielke C, Stengert M, Kay V, Klehr-Wirth D. 1995.

Scaffold/matrix-attached regions: structural properties creating transcriptionally active

loci. Int Rev Cyto, 162A:389-454.

Bode J, Goetze S, Heng H, Krawetz SA, Benham C. 2003. From DNA structure to gene

expression: mediators of nuclear compartmentalization and dynamics. Chromosome

Res, 11:435-445.

Bolla R, Denckla WD. 1979. Effect of hypophysectomy on liver nuclear ribonucleic acid

synthesis in aging rats. Biochem J, 184:669-674.

Broccoli D. 2004. Function, replication and structure of the mammalian telomere.

Cytotechnology, 45:3-12.

Burzynski SR. 2005. Aging: Gene silencing or gene activation? Med Hypotheses, 64:201-

208.

Calvin HI. 1976. Comparative analysis of the nuclear basic proteins in rat, human, guinea

pig, mouse, and rabbit spermatozoa. Biochim Biophys Acta, 434:377-389.

Carmen A, Milne L, Grunstein M. 2002. Acetylation of the yeast histone H4 N terminus

regulates its binding to heterochromatin protein SIR3. J Biol Chem, 277:4778-4781.

Cassia RO, Conde RD. 1994. Influence of protein nutrition on proteolytic activity and

histone content in isolated mouse liver nuclei. Horm Metab Res, 26:137-140.

Castro CE, Armstrong-Major J, Ramirez ME. 1986. Diet-mediated alteration of chromatin

structure. Fed Proc, 45:2394-2398.

Castro CE, Sevall JS. 1980. Alteration of higher order structure of rat liver chromatin by

dietary composition. J Nutr, 110:105-116.

Cavalli G. 2002. Chromatin as a eukaryotic template of genetic information. Curr Opin

Cell Biol, 14:269-278.

Cerda MC, Berrios S, Fernandez-Donoso R, Garagna S, Redi C. 1999. Organisation of

complex nuclear domains in somatic mouse cells. Biol Cell, 91:55-65.

��

�

Chaurasia P, Mukherjee S, Thakur MK. 1996. Age-related analysis of EcoRI generated

satellite DNA-containing chromatin of rat liver. Biochem Mol Biol Int, 40:1261-1270.

Cmarko D, Verschure PJ, Martin TE, Dahmus ME, Krause S, Fu XD, van Driel R, Fakan

S. 1999. Ultrastructural analysis of transcription and splicing in the cell nucleus after

bromo-UTP microinjection. Mol Biol Cell, 10:211-223.

Cockerill PN, Garrard WT. 1986. Chromosomal loop anchorage sites appear to be

evolutionarily conserved. FEBS Lett, 204:5-7.

Comings D. 1980. Arrangement of chromatin in the interphase nucleus. Hum Genet,

53:131-143.

Cong YS, Wright WE, Shay JW. 2002. Human telomerase and its regulation. Microbiol

Mol Biol Rev, 66:407-425.

Cremer T, Cremer C. 2001. Chromosome territories, nuclear architecture and gene

regulation in mammalian cells. Nat Rev Genet, 2:292-301.

Cremer T, Kreth G, Koester H, Fink RH, Heintzmann R, Cremer M, Solovei I, Zink D,

Cremer C. 2000. Chromosome territories, interchromatin domain compartment, and

nuclear matrix: an integrated view of the functional nuclear architecture. Crit Rev

Eukaryot Gene Expr, 10:179-212.

Cremer T, Lichter P, Borden J, Ward DC, Manuelidis L. 1988. Detection of chromosome

aberrations in metaphase and interphase tumor cells by in situ hybridization using

chromosome specific library probes. Hum Genet, 80:235-246.

Dhillon N, Kamakaka R. 2002. Breaking through to the other side: silencers and barriers.

Curr Opin Genet Dev, 12:188-192.

Dorigo B, Schalch T, Kulangara A, Duda S, Schroeder RR, Richmond TJ. 2004.

Nucleosome arrays reveal the two-start organization of the chromatin fiber. Science,

306:1571-1573.

Eils R, Dietzel S, Bertin E, Schröck E, Speicher MR, Ried T, Robert-Nicoud M, Cremer C,

Cremer T. 1996. Three-dimensional reconstruction of painted human interphase

chromosomes: active and inactive X-chromosome territories have similar volumes but

differ in shape and surface structure. J Cell Biol, 135:1427-1440.

Eissenberg JC, Elgin SC. 2000. The HP1 protein family: getting a grip on chromatin. Curr

Opin Genet Dev, 10:204-210.

Elgin SC, Grewal SI. 2003. Heterochromatin: silence is golden. Curr Biol, 13:R895-R898.

Falco JRP, Mello MLS. 1999. Critical electrolyte concentration of spermatozoal chromatin

containing histone H1 variants. Genet Mol Biol, 22:197-200.

��

�

Fedoroff NV. 1979. On spacers. Cell, 16, 697-710.

Ferraro A, Eufemi M, Altieri F, Cervoni L, Turano C. 1994. Glycoproteins of the nuclear

matrix from chicken liver cells. Cell Biol Int, 18:655-691.

Fiorini A, Gouveia FS, Fernandez MA. 2005. Scaffold/Matrix attachment regions and

intrinsic DNA curvature. Biochemistry, 71:481-488.

Fransz P, De Jong JH, Lysak M, Castiglione MR, Schubert I. 2002. Interphase

chromosomes in Arabidopsis are organized as well defined chromocenters from which

euchromatin loops emanate. Proc Natl Acad Sci USA, 99:14584-14589.

Fukuda Y. 2000. Interaction of nuclear proteins with intrinsically curved DNA in a matrix

attachment region of a tobacco gene. Plant Mol Biol, 44:91-98.

Fukuda Y, Nishikawa S. 2003. Matrix attachment regions enhance transcription of a

downstream transgene and the accessibility of its promoter region to micrococcal

nuclease. Plant Mol Biol, 51:665-675.

Gasser, SM, Laemmli, UK. 1986. Cohabitation of scaffold binding regions with

upstream/enhancer elements of three developmentally regulated genes of D.

Melanogaster. Cell, 46:521-530.

Gaubatz J, Prashad N, Cutler RG. 1976. Ribosomal RNA gene dosage as a function of

tissue and age for mouse and human. Biochim Biophys Acta, 418:358-375.

Gerdes MG, Carter KC, Moen PT Jr, Lawrence JB. 1994. Dynamic changes in the higher-

level chromatin organization of specific sequences revealed by in situ hybridization to

nuclear halos. J Cell Biol, 126:289-304.

Gilbert N, Allan J. 2001. Distinctive higher-order chromatin structure at mammalian

centromeres. Proc Natl Acad Sci USA, 98:11949-11954.

Gilbert N, Boyle S, Fiegler H, Woodfine K, Carter NP, Bickmore WA. 2004. Chromatin

architecture of the human genome: gene-rich domains are enriched in open chromatin

fibers. Cell, 118:555-566.

Goodrich J, Tweedie S. 2002. Remembrance of things past: chromatin remodeling in plant

development. Annu Rev Cell Dev Biol, 18:707-746.

Grewal SI, Elgin SC. 2002. Heterochromatin: new possibilities for the inheritance of

structure. Curr Opin Genet Dev, 12:178-187.

Grewal SI, Moazed D. 2003. Heterochromatin and epigenetic control of gene expression.

Science, 301:798-802.

Grunstein M. 1997. Molecular model for telomeric heterochromatin in yeast. Curr Opin

Cell Biol, 9:383-387.

��

�

Guarente L. 2000. Sir2 links chromatin silencing, metabolism, and aging. Genes Dev,

14:1021-1026.

Guenatri M, Bailly D, Maison C, Almouzni G. 2005. Mouse centric and pericentric satellite

repeats form distinct functional heterochromatin. J Cell Biol, 166:493-505.

Haaf T, Ward DC. 1995. Higher order nuclear structure in mammalian sperm revealed by

in situ hybridization and extended chromatin fibers. Exp Cell Res, 219:604-611.

Hall IM, Noma K, Grewal SI. 2003. RNA interference machinery regulates chromosome

dynamics during mitosis and meiosis in fission yeast. Proc Natl Acad Sci USA,

100:193-198.

Hall IM, Shankaranarayana GD, Noma K, Ayoub N, Cohen A, Grewal SI. 2002.

Establishment and maintenance of a heterochromatin domain. Science, 297:2232-2237.

Hannon G. 2002. RNA interference. Nature, 418:244-251.

Hart GW, Haltiwanger RS, Holt GD, Kelly WG. 1989. Glycosylation in the nucleus and

cytoplasm. Annu Rev Biochem, 58:841-874.

Hawkes R. 1982. Identification of concanavalin A-binding proteins after sodium dodecyl

sulfate-gel electrophoresis and protein blotting. Anal Biochem, 123:143-146.

Heitz E. 1928. Das Heterochromatin der Moose. Jahrb Wiss Botanik, 69:762-818.

Heng HH, Goetze S, Ye CJ, Liu G, Stevens JB, Bremer SW, Wykes SM, Bode J, Krawetz

SA. 2004. Chromatin loops are selectively anchored using scaffold/matrix-attachment

regions. J Cell Sci, 117:999-1008.

Henikoff S. 2000. Heterochromatin function in complex genomes. Biochim Biophys Acta,

1470:1-8.

Hochstrasser M, Mathog D, Gruenbaum Y, Saumweber H, Sedat JW. 1986. Spatial

organization of chromosomes in the salivary gland nuclei of Drosophila melanogaster.

J Cell Biol, 102:112-123.

Hoppe GJ, Tanny JC, Rudner AD, Gerber SA, Danaie S, Gygi SP, Moazed D. 2002. Steps

in assembly of silent chromatin in yeast: Sir3-independent binding of a Sir2/Sir4

complex to silencers and role for Sir2-dependent deacetylation. Mol Cell Biol,

22:4167-4180.

Horn PJ, Peterson CL. 2002. Molecular biology. Chromatin higher order folding--wrapping

up transcription. Science, 297:1824-1827.

Hsu TC, Cooper JE, Mace ML Jr, Brinkley BR. 1971. Arrangement of centromeres in

mouse cells. Chromosoma, 34:73-87.

��

�

Iarovaia OV, Akopov SB, Nikolaev LG, Sverdlov ED, Razin SV. 2005. Induction of

transcription within chromosomal DNA loops flanked by MAR elements causes an

association of loop DNA with the nuclear matrix. Nucleic Acids Res, 33:4157-4163.

Iarovaia OV, Bystritskiy A, Ravcheev D, Hancock R, Razin SV. 2004. Visualization of

individual DNA loops and a map of loop domains in the human dytrophin gene.

Nucleic Acids Res, 32:2079-2086.

Ioudinkova E, Petrov A, Razin SV, Vassetzky YS. 2005. Mapping long-range chromatin

organization within the chicken alpha-globin gene domain using oligonucleotide DNA

arrays. Genomics, 85:143-151.

Izaurralde E, Mirkovitch J, Laemmli UK. 1988. Interaction of DNA with nuclear scaffolds

in vitro. J Mol Biol, 200:111-125.

Jackson SP, Tjian R. 1988. O-glycosylation of eukaryotic transcription factors: implications

for mechanisms of transcriptional regulation. Cell, 55:125-133.

Jenuwein T, Allis CD. 2001. Translating the histone code. Science, 293:1074-1080.

Kan FW, da Silva PP. 1986. Preferential association of glycoproteins to the euchromatin

regions of cross-fractured nuclei is revealed by fracture-label. J Cell Biol, 102:576-586.

Kaufmann SH, Shaper JH. 1984. A subset of non-histone nuclear proteins reversibly

stabilized by the sulfhydryl cross-linking reagent tetrathionate. Exp Cell Res, 155:477-

495.

Kurz A, Lampel S, Nickolenko JE, Bradl J, Benner A, Zirbel RM, Cremer T, Lichter P.

1996. Active and inactive genes localize preferentially in the periphery of chromosome

territories. J Cell Biol, 135:1195-1205.

Labrador M, Corces VG. 2002. Setting the boundaries of chromatin domains and nuclear

organization. Cell, 111:151-154.

Lehnertz B, Ueda Y, Derijck AA, Braunschweig U, Perez-Burgos L, Kubicek S, Chen T, Li

E, Jenuwein T, Peters AH. 2003. Suv39h-Mediated histone H3 lysine 9 methylation

directs DNA methylation to major satellite repeats at pericentric heterochromatin. Curr

Biol, 13:1192-1200.

Levy-Wilson B. 1983. Glycosylation, ADP-ribosylation, and methylation of Tetrahymena

histones. Biochemistry, 22:484-489.

Litt M, Simpson M, Gaszner M, Allis CD, Felsenfeld G. 2001. Correlation between histone

lysine methylation and developmental changes at the chicken beta-globin locus.

Science, 293:2453-2455.

Loc PV, Stratling WH. 1988. The matrix attachment regions of the chicken lysozyme gene

co-map with the boundaries of the chromatin domain. EMBO J, 7:655-664.

��

�

Long EO, Dawid IB. 1980. Repeated genes in eukaryotes. Annu Rev Biochem, 49, 727-764.

Ludérus ME, van Steensel B, Chong L, Sibon OC, Cremers FF, de Lange T. 1996.

Structure, subnuclear distribution, and nuclear matrix association of the mammalian

telomeric complex. J Cell Biol, 135:867-881.

Ma HJ, Nagata T. 1990. Study of RNA synthesis in the livers of aging mice by means of

electron microscopic radioautography. Cel Mol Biol, 36:589-600.

Ma XJ, Goodridge AG. 1992. Nutritional regulation of nucleosomal structure at the chicken

malic enzyme promoter in liver. Nucleic Acids Res, 20:4997-5002.

Mahendra G,Gupta S,Kanungo MS. 1999. Effect of 17 estradiol and progesterone on the

conformation of the chromatin of the liver of female Japanese quail during aging. Arch

Gerontol Geriatr, 28:149-158.

Mahy NL, Perry PE, Bickmore WA. 2002. Gene density and transcription influence the

localization of chromatin outside of chromosome territories detectable by FISH. J Cell

Biol, 159:753-763.

Maison C, Bailly D, Peters AHFM, Quivy JP, Roche D, Taddei A, Lachner M, Jenuwein T,

Almouzni G. 2002. Higher-order structure in pericentric heterochromatin involves a

distinct pattern of histone modification and an RNA component. Nature Genet, 30:329-

334.

Mello MLS, Vidal BC. 1973. Linear dichroism and anomalous dispersion of birefringence

on sperm heads. Acta Histochem, 45:109-114.

Mirkovitch J, Miralt ME, Laemmli UK. 1984. Organization of the higher-order chromatin

loop: specific DNA attachment sites on nuclear scaffold. Cell, 39:223-232.

Moazed D. 2001. Common themes in mechanisms of gene silencing. Mol Cell, 8:489-498.

Moraes AS, Vidal BC, Guaraldo AMA, Mello MLS. 2005. Chromatin supraorganization

and extensibility in mouse hepatocytes following starvation and refeeding. Cytometry

A, 63A:94-107.

Müller WG, Rieder D, Kreth G, Cremer C, Trajanoski Z, McNally JG. 2004. Generic

features of tertiary chromatin structure as detected in natural chromosomes. Mol Cell

Biol, 24:9359-9370.

Nakayama J, Rice J, Strahl B, Allis CD, Grewal SI. 2001. Role of histone H3 lysine 9

methylation in epigenetic control of heterochromatin assembly. Science, 292:110-113.

Nepveu A. 2001. Role of the multifunctional CDP/Cut/Cux homeodomain transcription

factor in regulating differentiation, cell growth and development. Gene, 271:1-15.

Nickerson J. 2001. Experimental observations of a nuclear matrix. J Cell Sci, 114:463-474.

��

�

Nicolson G, Lacorbiere M, Delmonte P. 1972. Outer membrane terminal saccharides of

bovine liver nuclei and mitochondria. Exp Cell Res, 71:468-473.

Oliva R, Dixon GH. 1991. Vertebrate protamine genes and the histone-to-protamine

replacement reaction. Progr Nuclei Acid Res Mol Biol, 40:25-94.

Ostermeier GC, Liu Z, Martins RP, Bharadwaj RR, Ellis J, Draghici S, Krawetz SA. 2003.

Nuclear matrix association of the human beta-globin locus utilizing a novel approach

to quantitative real-time PCR. Nucleic Acids Res, 31:3257-3266.

Palyga J, Witkowski A, Kowalski A, Sylwestrzak M, Zieba J. 1991. Non-histone nuclear

proteins in liver of Japanese quail divergently selected for relative susceptibility or

resistance to starvation. Comp Biochem Physiol, 100B:627-630.

Pandita TK. 2002. ATM function and telomere stability. Oncogene, 21:611-618.

Pardoll DM, Vogelstein B. 1980. Sequence analysis of nuclear matrix associated DNA

from rat liver. Exp Cell Res, 128:466-470.

Park MK, D’Onofrio M, Willingham MC, Hanover JA. 1987. A monoclonal antibody

against a family of nuclear pore proteins. nucleoporins.: O-linked N-acetylglucosamine

is part of the immunodeterminant. Proc Natl Acad Sci USA, 84:6462-6466.

Park Y, Kuroda M. 2001. Epigenetic aspects of X-chromosome dosage compensation.

Science, 293:1083-1085.

Paulson JR, Laemmli UK. 1977. The structure of histone-depleted metaphase

chromosomes. Cell, 12:817-828.

Pucciarelli MG, Conde RD. 1984. Breakdown of proteins from mouse liver subcellular

fractions. Effect of nutritional changes. Acta Physiol Pharmacol Latinoam, 34:185-

191.

Ranhotra HS, Sharma R. 2001. Modulation of hepatic and renal glucocorticoid receptors

during aging of mice. Biogerontology, 2:245-251.

Razin SV. 2001. The nuclear matrix and chromosomal DNA loops: is their any correlation

between partitioning of the genome into loops and functional domains? Cell Mol Biol

Lett, 6:59-69.

Razin SV, Gromova II, Iarovaia OV. 1995. Specificity and functional significance of DNA

interaction with the nuclear matrix: new approaches to clarify the old questions. Int Rev

Cytol, 162B:405-448.

Razin SV, Kekelidze MG, Lukanidin EM, Scherrer K, Georgiev GP. 1986. Replication

origins are attached to the nuclear skeleton. Nuclei Acids Res, 14:9189-8207.

��

�

Rizzo WB, Bustin M. 1977. Lectins as probes of chromatin structure. Binding of

concanavalin A to purified rat liver chromatin. J Biol Chem, 252:7062-7067.

Rusche L, Kirchmaier A, Rine J. 2002. Ordered nucleation and spreading of silenced

chromatin in Saccharomyces cerevisiae. Mol Biol Cell, 13:2207-2222.

Sadoni N, Zink D. 2004. Nascent RNA synthesis in the context of chromatin architecture.

Chromosome Res, 12:439-451.

Santi S, Rubbini S, Cinti C, Squarzoni S, Matteucci A, Caramelli E, Guidotti L, Maraldi

NM. 1994. Nuclear matrix involvement in sperm head structural organization. Biol

Cell, 81:47-57.

Santoro R. 2005. The silence of the ribosomal RNA genes. Cell Mol Life Sci, 62:2067-

2079.

Selker EU. 1999. Gene silencing: repeats that count. Cell, 97:157-160.

Seve AP, Hubert J, Bouvier D, Masson C, Geraud G, Bouteille M. 1984. In situ distribution

in different cell types of nuclear glycoconjugates detected by two lectins. J Submicrosc

Cytol, 16:631-641.

Shankaranarayana G, Motamedi M, Moazed D, Grewal SI. 2003. Sir2 regulates histone H3

lysine 9 methylation and heterochromatin assembly in fission yeast. Curr Biol,

13:1240-1246.

Sleutels F, Zwart R, Barlow D. 2002. The non-coding Air RNA is required for silencing

autosomal imprinted genes. Nature, 415:810-813.

Soppe WJ, Jasencakova Z, Houben A, Kakutani T, Meister A, Huang MS, Jacobsen SE,

Schubert I, Fransz PF. 2002. DNA methylation controls histone H3 lysine 9

methylation and heterochromatin assembly in Arabidopsis. EMBO J, 21:6549-6559.

Stadler S, Schnapp V, Mayer R, Stein S, Cremer C, Bonifer C, Cremer T, Dietzel S. 2004.

The architecture of chicken chromosome territories changes during differentiation.

BMC Cell Biol, 22:44.

Stuurman N, Meijne AML, van der Pol AJ, Jong L, van Driel R, van Renswoude J. 1990.

The nuclear matrix from cells of different origin. J Biol Chem, 265: 5460-5465.

Svetlova EY, Razin SV, Debatisse M. 2001. Mammalian recombination hot spot in a DNA

loop anchorage region: a model for the study of common fragile sites. J Cell Biochem,

81:170-178.

Taboga SR, Mello MLS, Falco JRP, Vidal BC. 1996. Cytochemistry and polarization

microscopy of amphibian sperm cell nuclei presumed to differ in basic protein

composition. Acta Histochem Cytochem, 29:129-134.

��

�

Tajbakhsh J, Luz H, Bornfleth H, Lampel S, Cremer C, Lichter P. 2000. Spatial distribution

of GC- and AT-rich DNA sequences within human chromosome territories. Exp Cell

Res, 255:229-237.

Tsukamoto T, Hashiguchi N, Janicki SM, Tumbar T, Belmont AS, Spector DL. 2000.

Visualization of gene activity in living cells. Nat Cell Biol, 2:871-878.

Tsutsui KM, Sano K, Tsutsui K. 2005. Dynamic view of the nuclear matrix. Acta Med

Okayama, 59:113-120.

Vassetzky YS, Hair A, Razin SV. 2000. Rearrangement of chromatin domains in cancer

and development. J Cell Biochem, S35:54-60.

Vidal BC. 2000. Extended chromatin fibres: crystallinity, molecular order and reactivity to

concanavalin-A. Cell Biol Int, 24:723-728.

Vidal BC, Maria SS, Klaczko LB. 1997. Concanavalin A-reactive nuclear matrix

glycoprotein. B J Genet, 20:39-45.

Volpe TA, Kidner C, Hall IM, Teng G, Grewal SI, Martienssen RA. 2002. Regulation of

heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science,

297:1833-1837.

Ward WS, Coffey DS. 1990. Specific organization of genes in relation to the sperm nuclear

matrix. Biochem Biophys Res Commun, 173:20-25.

Wassef NM, Richardson EC, Alving CR. 1985. Specific binding of concanavalin A to free

inositol and liposomes containing phosphatidylinositol. Biochem Biophys Res Commun

130:76-83.

Waterston RH, et al. 2002. Initial sequence and comparative analysis of the mouse genome.

Nature, 420:520-562.

Wegel E, Vallejos RH, Christou P, Stoger E, Shaw P. 2005. Largescale chromatin

decondensation induced in a developmentally activated transgene locus. J Cell Sci,

118:1021–1031.

West A, Gaszner M, Felsenfeld G. 2002. Insulators: many functions, many mechanisms.

Genes Dev, 16:271-288.

Whitelaw CBA, Grolli S, Accornero P, Donofrio G, Farini E, Webster J. 2000. Matrix

attachment region regulates basal �-lactoglobulin transgene expression. Gene 244:73-

80.

Woodcock CL, Dimitrov S. 2001. Higher-order structure of chromatin and chromosomes.

Curr Opin Genet Dev, 11:130-135.

��

�

Yam H, Lam DS, Ophth FRC, Pang C, Phil D. 2002. Changes of nuclear matrix in long-

term culture of limbal epithelial cells. Cornea, 21:215-219.

Yamamura J, Nomura K. 2001. Analysis of sequence-dependent curvature in matrix

attachment regions. FEBS Lett, 489:166-170.

�

��

�

Anexo I

Artigo: Nucleus image properties assessed by video image analysis in mouse hepatocytes

under a short lysis for extended chromatin fiber formation. Artigo publicado em Cytometry

Part A 69A:1106-1113 (2006).

DOI: 10.1002/cyto.a.20339

Nucleus Image Properties Assessed by VideoImage Analysis in Mouse HepatocytesUnder a Short Lysis for ExtendedChromatin Fiber Formation

Benedicto C. Vidal, Alberto S. Moraes, and Maria Luiza S. Mello*

Department of Cell Biology, Institute of Biology, State University of Campinas (UNICAMP),Campinas, S~ao Paulo, Brazil

Received 19 March 2006; Revision Received 9 July 2006; Accepted 1 August 2006

Background: How much DNA remains in mouse hepato-cyte nuclei after extended chromatin fiber (ECF) forma-tion or whether this content varies within the nuclearpopulation is not known. This information could be rele-vant to understanding chromatin extensibility as related tochromatin organization, possibly associated with variablenuclear activities in hepatocytes.Methods: A protocol for ECF formation under the gravityaction, image analysis of Feulgen-stained unfixed mousehepatocyte remnants, and DAPI fluorescence were used.Results: Areas, shape, Feulgen-DNA amounts, and chro-matin texture were affected in unfixed, lysed nuclei. TheFeulgen-DNA values in nuclear remnants represented 37% of the content in fixed, nonlysed nuclei in terms ofmedian values; the coefficient of variation of Feulgen-DNAvalues in the nuclear remnants was much higher thanthose in controls. Enhancement in DAPI fluorescence was

evident in chromocenters of the fixed nuclei and inremnants and some ECF granules of the unfixed, lysednuclei.Conclusions: The DNA content of the nuclear remnantswas much more variable than that assumed from knownvariability in hepatocyte ploidy degrees. The variable con-straint to chromatin extrusion from hepatocyte nuclei ishypothesized to depend on variable chromatin organiza-tion with possible involvement of nuclear matrix associa-tion, transcriptional activities, and AT-rich DNA-containingheterochromatin. q 2006 International Society for Analytical

Cytology

Key terms: extended chromatin fibers; chromatin supra-organization; nuclear remnants; Feulgen-DNA amount-hepatocytes

Extended chromatin fibers (ECF) flow from cell nucleisubjected to treatment with concentrated saline and deter-gent solutions, under the action of gravity (1–5). ECF for-mation and DNA flow are affected not only by rheologicproperties of the DNA (6,7) but also by the presenceof nuclear protein molecules. Consequently, they areaffected by nuclear physiology and by the extrinsic assayfactors involved, such as cell fixation and the lysis treat-ment protocol (2,3,5).In mouse hepatocytes, a short (5–10 min) lysis protocol

as applied vertically to unfixed preparations results in for-mation of extremely long ECFs in all of the nuclei (5).Even so, Feulgen- and toluidine blue-stained materials stillremain inside the nuclear remnants, indicating that part ofthe DNA does not flow from the nuclei, possibly becauseof some structural constraint. Con-A binding sites assumedto pertain to nuclear glycoproteins are detected at least partlyin the cell nucleus remnants under this lysis protocol (5).It is not known whether the DNA content which

remains in the cell nuclei under the mentioned conditions

varies among the cell nuclei nor which part of the wholeDNA it represents. Should a variation occurs, it could evenmean that the putative structural constraint to DNA flowmight be caused by differences in nuclear architecturepossibly in association to differences in nuclear matrixinteractions and transcriptional activities (5,8,9) in the he-patocyte nuclei. In addition, the presence of especial rich-ness in DNA AT bases in the nuclear remnants of mousehepatocytes could be associated with chromocenters gen-erated by condensed centromere regions originally rich inrepetitive AT-containing DNA (10–12).

*Correspondence to: Maria Luiza S. Mello, Department of Cell Biology,

Institute of Biology, UNICAMP, 13083-863 Campinas, S~ao Paulo, Brazil.

E-mail: [email protected]

Grant sponsor: Brazilian National Council for Research and Develop-

ment (CNPq); Grant sponsor: S~ao Paulo State Research Foundation

(FAPESP); Grant number: 99/02547-806/66-8.

Published online in Wiley InterScience (www.interscience.wiley.com).

DOI: 10.1002/cyto.a.20339

q 2006 International Society for Analytical Cytology Cytometry Part A 69A:1106–1113 (2006)

119

In the present investigation, Feulgen-DNA amounts andother image properties were estimated in nuclear rem-nants of mouse hepatocytes under a short lysis protocolfor ECF formation.

MATERIALS AND METHODS

Three female mice of the inbred strain Swiss/Uni ob-tained from the Multidisciplinary Center for BiologicalInvestigation of the State University of Campinas (Cemib/UNICAMP), reared under normal conditions, and fedstandard extruded chow (Purina�, Paulina, Brazil) ad libi-tum until 15 weeks old were used. The mice were killedby decapitation and their livers immediately removed andplaced in cold 0.9% NaCl physiological solution. Liverslices were used to prepare imprints on glass slides. Allthe protocols involving animal care and use wereapproved by the Committee for Ethics in Animal Experi-mentation of the State University of Campinas and met theguidelines of the Canadian Council on Animal Care.

Treatments

Fresh unfixed imprints and imprints fixed in an absoluteethanol-glacial acetic acid mixture (3:1, v/v) for 1 min andthen rinsed in 70% ethanol for 5 min were used. Thefreshly prepared unfixed imprints were positioned verti-cally and immediately lysed in 2 M NaCl plus 1% Triton X-100 in Tris-HCl buffer (25 mM, pH 7.4) for 10 min, afterwhich the volume of solution was completed with abso-lute ethanol to a final concentration of 50%; treatmentlasted another 10 min. Following this step, the slides werecarefully removed from the lysis solution and transferredto 70% ethanol for 30 min (1,2,4). It is assumed that underthis lysis protocol the ECFs contain long well-orientedDNA double-helix molecules, since they display intenseoptical anisotropical phenomena such as birefringenceand linear dichroism under polarized light (4,5). Somefixed and unfixed preparations not subjected to the lysisprotocol were used as controls. The unfixed, nonlysedpreparations were subjected to the topochemical tests im-mediately after the imprinting.

Topochemistry and Image Analysis

The preparations were subjected to the Feulgen reac-tion (hydrolysis done in 4 M HCl at 25 C for 60 min), airdried, cleared in xylene, and mounted in natural Canadabalsam (nD 5 1.54).Carl Zeiss/Kontron equipment and Kontron KS400 soft-

ware (Oberkochen/Munich, Germany) were used forvideo image analysis of the Feulgen-stained material. Themicroscopic images were obtained with a Zeiss Axiophot2 microscope equipped with a Neofluar 40/0.75 objective,optovar factor 2, 0.90 condenser, and light of wavelengthequal to 546 nm provided by a Schott interference filter. A100-W halogen illuminator, a voltage regulator for light in-tensity maintained constant at point 4, filter wheels 1 and2 rotated into position 100 (open position), and a lumi-nous-filter diaphragm (transmitted light) at its maximalopening were used. These illumination conditions were

maintained constant for all nuclei investigated. The imagesto be processed were fed from the microscope into a Pen-tium computer through a Sony CCD-IRIS/RGB Hyper HADcolor video camera. Segmentation was used to generate bi-nary regions needed for measurement. Two thresholdvalues determine which gray value range of the imageInput is retained or deleted in the image Output. Thresh-old values (gray values) named Low (L) and High (H) weredetermined by moving the borders in the gray value histo-gram such that nuclear images appeared well segmentedfrom each other and from the background. Since ‘‘green’’was selected as the parameter ‘‘color,’’ the pixels insidethe gray value range [L, H] were displayed in this color. Inthe present study, L and H threshold values were equal to0 and 140, respectively. Under the earlier-mentioned oper-ating conditions, 1 lm corresponded to 7.23 pixels. Theminimum area possible to be measured with this appara-tus corresponded to 4 pixels. The software providedquantitative information on several geometric, densitomet-ric, and textural parameters. Nuclear area (lm2), nuclearperimeter (lm), nuclear feret ratio (minimal feret/maxi-mal feret, as an indication of the ellipticity of the absorb-ing image), and nuclear roundness factor [5 perimeter/(2p(p 3 area)), considering a value of 1 for a circle] (13)

were used as geometric parameters which inform on sizeand shape of the Feulgen-stained measured images. Opti-cal density or absorbances (OD), and integrated opticaldensity (IOD) were used as densitometric parameters fordetection of Feulgen-DNA amounts and variability of chro-matin condensation in the nuclear population (OD).Standard deviation of total densitometric values per nu-cleus (SDtd) and entropy (number of bits needed to storedensitometric values per nucleus image) were used as tex-tural parameters. SDtd reflects the absorbance variabilityper nucleus or the variability in the degree of chromatinpacking per nucleus or Feulgen-stained image (14). En-tropy reflects the amount of pixel variability in the nuclei.Since it measures the complexity of an image (15), a highvalue for entropy in Feulgen-stained nuclei represents ahigh contrast between condensed and noncondensedchromatin, whereas a low value is an indicator of an imagewith fairly constant gray levels (homogeneous image)(14,16). For the definition of nucleus images in the lysedpreparations, extended chromatin tails flowing from thecell nuclei were excluded during the segmentation step ofthe image analysis procedure (Fig. 1).

Topochemistry and Fluorescence Microscopy

Some of the fixed, nonlysed and unfixed, lysed prepara-tions were stained with a 40-6-diamidino-2-phenylindole(DAPI) (Sigma) solution known to show a fluorescencespecificity for AT base pairs, a condition in which fluores-cence is strongly enhanced (17–19). Since mouse hepato-cyte nuclei were analyzed here, this procedure was usedjust as a qualitative approach for identification of the chro-mocenter regions known to contain centromeric/pericen-tromeric AT-rich DNA (10–12) and possible changes intheir distribution during ECF formation. A protocol modi-

1107CHROMATIN EXTENSIBILITY AND NUCLEUS IMAGE

Cytometry Part A DOI 10.1002/cyto.a

120

fied from Schweizer’s (19) method was used. Briefly, thepreparations were stained in the dark with a 0.3 lg/mlDAPI solution in McIlvaine buffer at pH 7.0 for 15 min,rinsed several times in distilled water, rinsed in McIlvaine

buffer at pH 7.0 for 15 min, air dried, and mounted in glyc-erol–McIlvaine buffer at pH 7.0 (1:1).The stained preparations were examined in a Zeiss

Axiophot 2 microscope equipped for epifluorescence

FIG. 1. Feulgen-stained fixed, nonlysed (a, b) and unfixed, lysed (c–f) hepatocyte nuclei from adult mice as photographed from the image analyzer moni-tor. (a) and (d) images pseudocolorized green represent the images evaluated for Feulgen-DNA values and other nuclear features. Images of gray level seg-mentation for nuclei (b), ECF (e), and nuclear remnants (f) are shown. E: ECF. Arrows indicate the direction of gravity.

1108 VIDAL ET AL.


121

Table1

Nucl

earG

eom

etr

icPara

mete

rsofFeulg

en-S

tain

ed

Mouse

Hep

ato

cyte

s

Hepatocyte

nucleargroups

nECF

formation

Nuclear

area

(lm2)

Nuclear

perimeter

(lm)

NFR

NRF

XS

Median

XS

Median

XS

Median

XS

Median

Fixed,nonlysed

607

No

110.43

55.48

101.53

42.94

13.19

41.25

0.87

0.10

0.90

1.19

0.18

1.14

Unfixed,nonlysed

240

No

87.85

34.70

80.14

39.12

9.56

37.55

0.89

0.05

0.90

1.20

0.14

1.16

Unfixed,lysed

151

Yes

53.69

28.79

53.17

49.38

25.25

42.19

0.60

0.25

0.67

1.95

0.81

1.74

ECF,extendedchromatinfibers;NFR,nuclearferetratio;NRF,nuclearroundnessfactor;

S,standarddeviation;X,arithmeticmean.

Table2

Densi

tom

etr

icand

Tex

tura

lPara

mete

rsofFeulg

en-S

tain

ed

Mouse

Hep

ato

cyte

s

Hepatocyte

nuclear

groups

n

OD

IOD

SDtd

Entropy

XS

XS

Median

Variability(%)

XS

XS

Fixed,nonlysed

607

0.35

0.06

38.20

19.80

36.42

51

16.47

3.74

5.94

0.35

Unfixed,nonlysed

240

0.32

0.05

28.05

11.86

25.26

42

12.26

3.08

5.50

0.33

Unfixed,lysed

151

0.33

0.09

17.21

11.11

13.40

64

7.83

2.65

4.82

0.48

IOD,Feulgen-DNAvaluesinarbitraryunits;OD,absorbances;

S,standarddeviation;SDtd,standarddeviationoftotaldensitometricvaluespernucleus;X,arithmeticmean.



122

with an HBO-100W stabilized mercury lamp as the lightsource. Operating conditions were as follows: Zeiss PlanNeofluar 203/0.50 and 403/0.75 objectives, condenseradequate for fluorescence, and filter set 01 (G365 – BPexciting filter; FT 395 chromatic splitter; LP420 barrier fil-ter). Photomicrographs were obtained with Kodak Gold400 film.

Statistics

Calculations were done using Minitab 12TM software(State College, PA). Applied tests consisted of one-wayanalysis of variance for unstacked data and Mann–WhitneyU test.

RESULTS

Unfixed imprint preparations not subjected to the lysisprotocol showed decrease in hepatocyte nuclear sizes(21.1% in nuclear absorbing areas) and Feulgen-DNAamounts (30.7%), but indistinguishable difference in nu-clear feret ratio and a slight departure from round shape(median values considered) in comparison with those ofthe fixed cells (Tables 1–3).After the short lysis protocol under the action of gravity

and disregarding the tail that consisted of ECFs flowingfrom the cell nuclei (Figs. 1a–1f), the Feulgen-stained ma-terial in unfixed nuclei revealed decrease in area and Feul-

gen-DNA values, and a significant increase in values of thenuclear roundness factor and decrease in values of the nu-clear feret ratio, which signify induction of deformity incontours, with departure from the round shape (13), incomparison with the nonlysed nuclei (Tables 1–3).The Feulgen-stained material that remained in the hepa-

tocyte cell nuclei after the lysis treatment represented 53% of that of the unfixed, nonlysed nuclei and 37% ofthat of the fixed, nonlysed nuclei, when considering re-spective median values. Nearly 63% of the DNA is thusreleased from the hepatocyte nuclei because of the lack offixation plus formation of ECF flow, 47% being released bythe lysis treatment applied vertically (entailing action ofgravity) to unfixed hepatocytes. The coefficient of varia-tion of the Feulgen-DNA values for unfixed, lysed nucleiwas much higher than that for fixed, nonlysed andunfixed, nonlysed nuclei (Table 2).Lack of fixation also induced decrease in nuclear absor-

bances, and in SDtd and entropy values (Table 2). NuclearSDtd and entropy values significantly decreased with thelysis treatment, although nuclear absorbances were notaffected (Tables 2 and 3).In the DAPI-stained preparations, a blueish violet fluo-

rescence was seen on nuclei, nuclear remnants, and ECFs(Figs. 2a–2c). Fluorescence enhancement (bright silvercolor) was revealed in the chromocenter areas (fixed, non-lysed nuclei) (Fig. 2a), in a homogeneously bright chroma-

Table 3Statistical Comparison of the Image Analysis Parameters of Feulgen-Stained Mouse Hepatocytes

Parameters Comparisons Test F P Decision

Nuclear area Fixed, nonlysed vs. unfixed, nonlysed Mann–Whitney 0.000 SSUnfixed, nonlysed vs. unfixed, lysed Mann–Whitney 0.000 SSFixed, nonlysed vs. unfixed, lysed Mann–Whitney 0.000 SS

Nuclearperimeter

Fixed, nonlysed vs. unfixed, nonlysed Mann–Whitney 0.000 SS

Unfixed, nonlysed vs. unfixed, lysed Mann–Whitney 0.000 SSFixed, nonlysed vs. unfixed, lysed Mann–Whitney 0.000 SS

Nuclearferet ratio

Fixed, nonlysed vs. unfixed, nonlysed Mann–Whitney 0.089 NS


Nuclear roundnessfactor

Fixed, nonlysed vs. unfixed, nonlysed Mann–Whitney 0.016 S


Absorbances (OD) Fixed, nonlysed vs. unfixed,nonlysed vs. unfixed, lysed

ANOVA 20.61 0.000 SS

Unfixed, nonlysed vs. unfixed, lysed ANOVA 0.70 0.403 NSFixed, nonlysed vs. unfixed, lysed ANOVA 13.77 0.000 SS

IOD Fixed, nonlysed vs. unfixed, nonlysed Mann–Whitney 0.000 SSUnfixed, nonlysed vs. unfixed, lysed Mann–Whitney 0.000 SSFixed, nonlysed vs. unfixed, lysed Mann–Whitney 0.000 SS

SDtd Fixed, nonlysed vs. unfixed, nonlysed ANOVA 132.12 0.000 SSUnfixed, nonlysed vs. unfixed, lysed ANOVA 256.64 0.000 SSFixed, nonlysed vs. unfixed, lysed ANOVA 450.37 0.000 SS

Entropy Fixed, nonlysed vs. unfixed, nonlysed ANOVA 46.12 0.000 SSUnfixed, nonlysed vs. unfixed, lysed ANOVA 653.49 0.000 SSFixed, nonlysed vs. unfixed, lysed ANOVA 1064.84 0.000 SS

NS, nonsignificant; S, significant; SS, highly significant.

1110 VIDAL ET AL.


123

tin region inside the nuclear remnants (unfixed, lysednuclei) and in a few small points on the ECFs (Figs. 2band 2c).

DISCUSSION

Present results reveal that the unfixed mouse hepatocytenuclei subjected to a protocol for ECF formation retainsome Feulgen-DNA content in their remnants, within 37% of that for fixed, nonlysed nuclei in terms of medianvalues considered. However, the DNA content present inthe nuclear remnants varied, although all the observed

nuclei formed ECFs. The variability in the DNA amountretained in the nuclear remnants (64%) exceeded that(51%) found in the fixed, nonlysed nuclei. The variability inFeulgen-DNA values found in the fixed, nonlysed nuclei isassumed to occur because of the well-known variability inDNA content due to various ploidy degrees occurring inthe adult liver (20–26). Indeed, the results obtained hererevealed for the fixed nuclei a very high standard deviationeven surpassing the half the IOD arithmetic means (X 538.20; S 5 19.80) (Table 2), which is in agreement withmore than one ploidy degree being present (20–26).

FIG. 2. DAPI-stained fixed, nonlysed (a) and unfixed, lysed (b, c) hepatocyte from adult mice. Fluorescence enhancement (silver color) appears in thechromocenters of the nonlysed nuclei (a) and in the remnants (r) of lysed nuclei (b, c); some few brighter granules on ECFs (E) are pointed out (arrows).Bars 5 20 lm.



124

The variation in DNA content retained in the hepato-cyte nuclear remnants may be due to a structural con-straint imposed by part of the chromatin organizationpossibly because the examined nuclei were engaged intodifferent functional activities (8,9). Supporting this hy-pothesis is the previously reported frequency of adult he-patocytes forming ECFs by the action of gravity in fixedpreparations subjected to a long lysis protocol, whichhas been found not to exceed 22% in fully nourishedmice, and which significantly decreases with starvation(5) and increases with aging (27). The decrease in fre-quency of ECF formation with starvation has been relatedto an increase in chromatin higher-order packing states(5), a situation that favors gene silencing (28,29).Increase in frequency of ECF formation with aging hasbeen hypothesized to be associated in mice with defectsin the heterochromatin formation and loss of telomeres(27), events reported to occur with advancing age inother cell systems (30,31). Simultaneous to decrease andincrease in frequency of ECF formation in mouse hepato-cytes with starvation and aging, respectively, a change inthe amount of Con-A reactive glycoproteins from the nu-clear matrix has been demonstrated (5,27). Thus, varia-tions in functional activities associated to chromatinsupraorganization, nuclear matrix interactions, and ECFformation in cell nuclei of adult mice, which have beenreported under the long lysis protocol (5), may also beresponsible in the present case, for the variable amountof DNA extruded in the ECFs or retained in the nuclearremnants.It bears mention that enhancement in DAPI fluores-

cence, assumed to represent chromocenter areas knownto contain centromeric and pericentromeric heterochro-matin AT-rich DNA in mice (10–12), is still present in thenuclear remnants of the lysed preparations. This indicatesthat at least part of this centromeric/pericentromericchromatin remained in the nuclear remnants. On theother hand, since enhancement in DAPI fluorescence wasalso found in some ECFs, it is assumed that part of the cen-tromeric/pericentromeric DNA is capable to flow fromthe cell nuclei. The finding of a unique area showing silverfluorescent color in the DAPI-stained nuclear remnantsunder this experimental condition suggests coalescenceof at least part of the centromeric/pericentromeric chro-matin areas responsible for such fluorescence characteris-tic in the unfixed, lysed nuclei. This is in agreement withthe resulting homogeneity in chromatin packing states inthe Feulgen-stained nuclear remnants (decreased SDtdand nuclear entropy values). The AT-rich DNA-containingchromatin present in the centromeric/pericentromericheterochromatin may also be involved with the constrainton ECF formation from the mouse hepatocytes.The decrease in Feulgen-DNA amounts in unfixed nuclei

not subjected to the lysis protocol is assumed to be due to acertain DNA loss during depurination under lack of fixation,because of lessened resistance to apurinic acid breakdownby acid hydrolysis pertinent to the Feulgen reaction, in com-parison with the fixed nuclei condition (32). A DNA lossmay also have occurred by DNase activity. Since no DNase

inhibitor was used in unfixed nuclei, the part played byDNase activity in these results is not to be overlooked.The morphometric changes observed in the unfixed non-

lysed cell nuclei and in the nuclear remnants of unfixed andlysed preparations are in agreement with some DNA lossduring acid hydrolysis and with induced ECF flow.

ACKNOWLEDGMENTS

The authors were recipients of research fellowships(BCV and MLSM) and a PhD fellowship (ASM) from CNPq.

LITERATURE CITED1. Heng HHQ, Squire J, Tsui LC. High-resolution mapping of mammaliangenes by in situ hybridization to free chromatin. Proc Natl Acad SciUSA 1992;89:9509–9513.

2. Haaf T, Ward DC. Structural analysis of a-satellite DNA and centro-mere proteins using extended chromatin and chromosomes. HumMol Genet 1994;3:697–709.

3. Fidlerov!a H, Senger G, Kost M, Sanseau P, Sheer D. Two simple proce-dures for releasing chromatin from fixed cells for fluorescence in situhybridization. Cytogenet Cell Genet 1994;65:203–205.

4. Vidal BC. Extended chromatin fibres: Crystallinity, molecular orderand reactivity to concanavalin A. Cell Biol Int 2000;24:723–728.

5. Moraes AS, Vidal BC, Guaraldo AMA, Mello MLS. Chromatin supraor-ganization and extensibility in mouse hepatocytes following starva-tion and refeeding. Cytometry A 2005;63:94–107.

6. Strick TR, Allemand JF, Bensimon D, Bensimon A, Croquette V. Theelasticity of a single supercoiled DNA molecule. Science 1996;271:1835–1837.

7. Perkins TT, Smith DE, Chu S. Single polymer dynamics in an elonga-tional flow. Science 1997;276:2016–2021.

8. Cremer T, Cremer C. Chromosome territories, nuclear architecture andgene regulation in mammalian cells. Nat Rev Genet 2001;2:292–301.

9. Gilbert N, Gilchrist S, Bickmore WA. Chromatin organization in themammalian nucleus. Int Rev Cytol 2005;242:283–336.

10. H€orz W, Altenburger W. Nucleotide sequence of mouse satellite DNA.Nucleic Acids Res 1981;9:683–696.

11. Guenatri M, Bailly D, Maison C, Almouzni G. Mouse centric and peri-centric satellite repeats form distinct functional heterochromatin. J CellBiol 2004;166:493–505.

12. Kuznetsova IS, Prusov AN, Enukashvily NI, Podgornaya OI. New types ofmouse centromeric satellite DNAs. Chromosome Res 2005;13:683–696.

13. Marchevsky AM, Erler BS. Morphometry in pathology. In: Marchevsky AM,Bartels PH, editors. Image Analysis. New York: Raven; 1994. pp 125–206.

14. Mello MLS, Vidal BC, Lareef MH, Hu YF, Yang X, Russo J. DNA content,chromatin texture and nuclear morphology in benzo[a]pyrene-trans-formed human breast epithelial cells after microcell-mediated transferof chromosomes 11 and 17. Cytometry 2003;52:70–76.

15. Barba J, Jeanty H, Fenster P, Gil J. The use of local entropy measuresin edge detection for cytological image analysis. J Microsc 1988;156:125–134.

16. Oberholzer M, €Ostreicher M, Christen H, Br€uhlmann M. Methods inquantitative image analysis. Histochem Cell Biol 1996;105:333–355.

17. Polysciences, Inc. 40,6-Diamidino-2-phenylindole dihydrochloride (DAPI),Technical Data Sheet 315, Polysciences, Inc., Warrington, PA, 1999.Available at www.polysciences.com.

18. http://www.celldeath.de/apometh/dapi.html.19. Schweizer D. Simultaneous fluorescent staining of R bands and speci-

fic heterochromatic regions (DA/DAPI-bands) in human chromo-somes. Cytogenet Cell Genet 1980;27:190–193.

20. Wilson JW, Leduc EH. The occurrence and formation of binucleateand multinucleate cells and polyploid nuclei in the mouse liver. Am JAnat 1948;82:353–391.

21. Brodsky WY, Uryvaeva IV. Cell polyploidy: Its relation to tissue growthand function. Int Rev Cytol 1977;50:275–332.

22. Uryvaeva IV. Biological significance of liver-cell polyploidy—An hy-pothesis. J Theor Biol 1981;89:557–571.

23. Bohm N, Noltemeyer N. Development of binuclearity and DNA-poly-ploidization in the growing-mouse liver. Histochemistry 1981;72:55–61.

24. Gupta S. Hepatic polyploidy and liver growth control. Semin CancerBiol 2000;10:161–171.

1112 VIDAL ET AL.


125

25. Vinogradov AE, Anatskaya OV, Kudryavtsev BN. Relationship of hepato-cyte ploidy levels with body size and growth rate in mammals. Genome2001;44:350–360.

26. Anatskaya OV, Vinogradov AE. Heart and liver as developmental bot-tlenecks of mammal design: Evidence from cell polyploidization. BiolJ Linnean Soc 2004;83:175–186.

27. Moraes AS, Mello MLS. Chromatin supraorganization and extensibilityin mouse hepatocytes with aging. Mol Biol Cell 2005;16:137a, 138a.

28. Castro CE, Armstrong-Major J, Ramirez ME. Diet-mediated alterationof chromatin structure. Fed Proc 1986;45:2394–2398.

29. Grewal SIS, Moazed D. Heterochromatin and epigenetic control ofgene expression. Science 2003;301:798–802.

30. Cong YS, Wright WE, Shay JW. Human telomerase and its regulation.Microbiol Mol Biol Rev 2002;66:407–425.

31. Bitterman KJ, Medvedik O, Sinclair D. Longevity regulation inSaccharomyces cerevisiae: Linking metabolism, genome stabil-ity, and heterochromatin. Microbiol Mol Biol Rev 2003;67:376–399.

32. Mello MLS. Cytochemistry of DNA, RNA and nuclear proteins. BrazJ Genet 1997;20:257–264.



126

��

�

Anexo II

Artigo: Extended chromatin fibers in spermatozoa of Apis mellifera (Hymenoptera,

Apoidea). Artigo publicado em Brazilian Journal of Morphological Sciences 23(3-4):363-

368 (2006).

Braz. J. morphol. Sci. (2006) 23(3-4), 363-368

_______________

Correspondence to: Dr. Maria Luiza S. MelloDepartamento de Biologia Celular, Instituto de Biologia, Universidade Estadual de Campinas (UNICAMP), CP 6109, CEP 13083-863, Campinas, SP, Brazil. Tel.: (55) (19) 3788-6122, Fax: (55) (19) 3788-6111, E-mail: [email protected]*Dedicated to Professor Carminda da Cruz-Landim on the occasion of her retirement.

REGULAR PAPER

EXTENDED CHROMATIN FIBERS IN SPERMATOZOA

OF Apis mellifera (HYMENOPTERA, APOIDEA)*

Raphael de Souza Mattos1, Alberto da Silva Moraes1,

José Chaud Netto2 and Maria Luiza Silveira Mello1

1Department of Cell Biology, Institute of Biology, State University of Campinas (UNICAMP), Campinas, SP, Brazil, 2Department of Biology, Institute of Biological Sciences, Paulista State University (UNESP), Rio Claro, SP, Brazil.

ABSTRACT

The flow of chromatin from the nuclei of mouse liver cells and spermatozoa after treatment with concentrated

saline and detergent solutions under the simultaneous action of gravity results in the formation of extended

chromatin fibers (ECF). In mouse somatic nuclei, the increase in chromatin condensation is accompanied

by a decrease in the frequency of ECF formation. Since tightly packed chromatin with a very lysine-rich

histone variant that resembles somatic H1 histones occurs in honey bee spermatozoa, we examined the

formation of ECF in sperm cells of Apis mellifera, and compared the findings with data for mouse cells.

Freshly prepared smears of fixed and unfixed semen from A. mellifera were lysed under the action of gravity,

stained with toluidine blue at pH 4.0, and examined with polarized and unpolarized light. A protocol using

unfixed preparations and a short lysis period that resulted in abundant ECF production in mouse hepatocytes

(which contain loosely-packed chromatin) and sperm cells produced ECF in only a few spermatozoa of A.

mellifera. In contrast, a protocol using fixed preparations and a long lysis period produced fewer ECFs in

the former two cell types and no ECF formation in honey bee spermatozoa. The limited chromatin fluidity

in A. mellifera spermatozoa may reflect their special DNA-protein composition and organization in the cell

nuclei, the participation of nuclear matrix elements, a less effective disruption of the nuclear envelope and

plasmalemmal components during lysis, and/or cytoplasmic spatial constraints resulting from particularities

in the acrosomal complex.

Key words: Chromatin extensibility, histone H1 variant, honey bee, optical anisotropy, spermatozoa

The flow of chromatin from the nuclei of mouse

liver cells and spermatozoa, as well as other cell

types, after treatment with concentrated saline and

detergent solution under the simultaneous action

of gravity results in the formation of extended

chromatin fibers (ECFs) [5-7,13,14,17,22]. Changes

in the chromatin packing states and chromatin

viscoelasticity have been found to affect the

formation of ECF in mouse liver cells following

alterations in cell physiology associated with

conditions such as starvation and aging [12,14].

These changes are characterized by an increase in

chromatin condensation accompanied by a decrease

in the frequency of ECF formation [12,14]. A high

frequency of nuclei with ECFs has also been reported

in similarly treated mouse sperm cells, and probably

results from the extraction of nuclear proteins (in

this case, two protamine-like variants) from the

DNA-protein complexes of these cells [13].

The DNA-protein complex of honey bee

spermatozoa contains a somatic-like, lysine-rich

histone H1 variant that differs from the DNA-

protein complex of spermatozoa from many

other species, the somatic histones of which

are partially or totally replaced by protamines

(arginine-rich; no lysine), protamine-like

variants (arginine-rich, little or no lysine, but

oxidized cysteine also present) or intermediate

basic sperm nuclear proteins (histidine and/or

lysine, in addition to arginine) [1,3,4,9,15,19].

The different chromatin composition and

supraorganization of A. mellifera spermatozoa

128

R. S. Mattos et al.364


suggests these cells may behave differently

from mouse liver cells and spermatozoa when

lysed under the action of gravity to induce ECF

formation [14]. In this study, ECF formation was

investigated in spermatozoa from the semen

of drones and from the spermatheca of early-

inseminated queens of A. mellifera.

MATERIAL AND METHODS

Insects and collection of spermatozoa

Drones and early-inseminated queens of A. mellifera L.

(Hymenoptera, Apoidea) from colonies maintained at the

Institute of Biosciences of Paulista State University at Rio

Claro, SP, Brazil, were used. Samples of semen collected

from drones immediately after ejaculation following

manual manipulation, or from the dissected spermathecae

of queens were diluted in cold saline solution and used to

prepare smears on glass slides.

Treatments

Freshly prepared smears were fixed in absolute

ethanol-glacial acetic acid (3:1, v/v) for 1 min and then

rinsed in 70% ethanol for 5 min. The slides were positioned

vertically (to expose them to the action of gravity) and

horizontally (a negative control for the action of gravity)

and the preparations were immediately lysed in 2 M NaCl

plus 1% Triton X-100 in Tris-HCl buffer (25 mM, pH 7.4)

for 5 h at 25oC, after which the volume of solution was

completed with absolute ethanol to a final concentration

of 50%, followed by a 10 min incubation. The slides were

then removed from the lysis solution and transferred to

70% ethanol for 30 min [6,7,22]. Unfixed preparations

subjected to the same lysis protocol but incubated for

shorter periods of time (10, 20 and 30 min) were also

used. Fixed preparations that had not been subjected to

the lysis protocol were used as controls.

Topochemistry and optical anisotropy

Staining was done with 0.025% toluidine blue (Merck,

Darmstadt, Germany) in McIlvaine buffer at pH 4.0 for

15 min [21,22]. The preparations were then rapidly (5 s)

rinsed in distilled water, air dried, cleared in xylene, and

mounted in natural Canada balsam. In addition to being

a classic cytochemical method for studying nucleic acids,

staining with toluidine blue also allows the investigation of

optical anisotropy (birefringence and selective absorption

of polarized light, i.e., linear dichroism) of DNA and

DNA-protein complexes [10,11,20-22], including ECFs

[14,20,22]. These anisotropic characteristics provide

information on the suitability and proximity of free DNA

phosphates available as binding sites for toluidine blue

[10,11,20,21].

The presence of ECFs in the stained nuclei of

spermatozoa was assessed with polarized and unpolarized

light using a Zeiss Axiophot 2 microscope equipped

with a 40/0.75 Pol-Neofluar objective, 1.4 condenser, a

compensator and polychromatic light. Birefringence was

investigated with a crossed polarizer and analyser by

orientating the long axis of the sperm heads at 45º relative

to the polarizing azimuths of the analyser and polarizer.

Linear dichroism was investigated in the same nuclei

using a polarizer and orientating the sperm heads relative

to the east-west azimuth of the electrical vector of the

polarized light.

RESULTS

The nuclei of A. mellifera spermatozoa had a

characteristic rod-like shape. Intense metachromasy

was seen in control nuclei and in nuclei that were

lysed horizontally or vertically and that did not show

ECF formation; this metachromasy was more evident

after lysis in a horizontal position. The stained

nuclei showed intense birefringence with typical

yellow/green interference colors (Fig. 1A) and linear

dichroism (Fig. 1B,C). The linear dichroism was

characterized by a negative sign since the absorption

of the stained nuclei positioned perpendicularly to

the azimuth of the electrical vector of the polarized

light was higher than that of nuclei positioned parallel

to the same azimuth (Fig. 1B,C).

There was no ECF formation in fixed preparations

subjected to the different lysis protocols. However,

toluidine blue-stained chromatin was distributed in

a helical pattern within the nuclei and this produced

a “banded” image that was especially evident under

polarized light (crossed polarizer and analyser with

or without a slight compensation for birefringence)

(Fig. 2A,B).

ECF formation was seen only in unfixed

preparations that were lysed vertically, and involved

only some of the spermatozoa (<1%, regardless

of the duration of lysis) (Fig. 3A-E). Although the

ECFs in these cells consisted of very thin filaments,

they were distinguishable from sperm tails by their

negative birefringence sign, which is typical of DNA

or DNA-protein complexes under polychromatic

light (Fig. 3A-C), and by the fact that they stained

with toluidine blue at pH 4.0 and showed negative

linear dichroism (Fig. 3D,E). The birefringence sign

mentioned here is easily demonstrated in Figure

3A-C by considering that when the birefringence in

ECFs or in sperm nuclei without ECFs positioned

at 45º with respect to the crossed polarizer-analyser

129

Extended chromatin fibers in A. mellifera 365


Figure 1. Dispersion of birefringence (A) and linear dichroism (B,C) in nuclei of fixed, toluidine blue-stained spermatozoa of Apis mellifera. All images are from the same microscopic field. Absorbances were higher in nuclei positioned perpendicularly (arrow) to the azimuth of the electrical vector of polarized light (↔). Bar = 10 μm.

is compensated, the birefringence of nuclei or ECFs

positioned at 90º with respect to the former is partially

or totally intensified. In some cases, heterogenously

distributed, granular, toluidine blue-stained material

was seen in ECFs (Fig. 3C).

There were no differences in the nuclear

morphology, metachromasy and optical anisotropy

among spermatozoa from drones and the spermathecae

of queens.

DISCUSSION

The metachromasy and optical anisotropic

characteristics of A. mellifera spermatozoa under

control conditions (linear dichroism and birefringence)

agreed with those of a previous report [10]. The present

results show that under the experimental conditions used

to form ECFs in mouse hepatocytes and sperm cells

[13,14,22], ECFs were rarely produced in A. mellifera

spermatozoa. Indeed, ECF formation was seen only in

unfixed semen preparations that were lysed in a vertical

position, as previously described [5,7,14,22]. This

treatment involves the breakdown of most DNA-histone

interactions by the saline components of the solution

and disruption of the nuclear envelope components by

Triton X-100 [5].

One explanation for the finding that most of the

A. mellifera spermatozoa did not form ECFs could

be that the chromatin involved was very resistant

130



Figure 3. Extended chromatin fibers (ECFs) formed in unfixed spermatozoal nuclei subjected to vertical lysis and toluidine blue staining. The arrow in B indicates compensation of the slight birefringence of the ECFs seen in A. Total compensation of the birefringence in ECFs (black arrowhead) and partial compensation in sperm heads highlighted the birefringence in ECFs (white arrowhead) or in part of the sperm heads positioned perpendicularly to the former (C). A fine granular distribution of stained material is seen in an ECF (arrow) (C). Negative linear dichroism was observed in ECFs (arrow) and sperm heads positioned differently relative to the azimuth of the electrical vector of polarized light (↔) (D,E). Bar = 10 μm.

Figure 2. Dispersion of birefringence in nuclei of fixed, toluidine blue-stained spermatozoa of Apis mellifera subjected to vertical lysis (A,B). Both images were from the same microscopic field. Note the helical distribution of the birefringent stained substrate. In B, a slight compensation of the birefringence was introduced to reinforce the “banded” appearance of the birefringent chromatin. Bar = 10 μm.

131

Extended chromatin fibers in A. mellifera 367


to lysis because of its protein component and its

packing state [9,10]. Indeed, the tight packing state of

the DNA-protein complex in honey bee spermatozoa

could account for the deep electron opacity of the

homogeneously distributed chromatin in these

cells [2,8]. However, at least part of the histone

H1 variant present in A. mellifera spermatozoa is

removed from the DNA-protein complex during

lysis, even when done horizontally, as shown by the

increase in nuclear metachromasy and the “banded”

distribution of toluidine blue-stained material in the

cell nuclei. The “banded” image seen in cell nuclei

may represent the helical distribution of chromatin

highlighted after lysis. A tandem, end-to-end

arrangement of chromosomes, as hypothesized for

A. mellifera spermatids [Kerr WE, 1969 – personal

communication] and observed in Drosophila and

some Orthoptera [18], could facilitate the production

of these “banded” structures after lysis.

Nuclear proteins that could adversely affect

ECF fluidity may also be present in the nuclei of A.

mellifera spermatozoa. Nuclear matrix proteins have

been reported in the spermatozoa of other animals

[23]. The very fine, granular material that stained with

toluidine blue at pH 4.0 in the ECF of A. mellifera

spermatozoa may correspond to small sites at which

the DNA is still not fully stretched, possibly because of

some retained nuclear proteins, as also seen in mouse

hepatocytes [14]. In the case of mouse hepatocytes,

nuclear matrix glycoproteins have been reported to

occur in the cell nuclei and granular material of ECFs

[14], but there has been no such report for honey bee

spermatozoa. Electron microscopy of the nuclei of Apis

spermatozoa treated with lysis solution in addition to

DNA extraction could provide data on the presence of

a nuclear matrix. The presence of nuclear glycoproteins

could be assessed immunocytochemically by detection

with Con-A. Detailed identification of putative nuclear

matrix proteins would require extraction of the proteins

followed by bidimensional electrophoresis.

An additional explanation for the lack of ECF

formation in most A. mellifera spermatozoa could be that

disruption of the nuclear envelope and plasmalemmal

components of these cells was ineffective because of

their composition and organization. A cytoplasmic

spatial constraint involving the acrosomal complex that,

in this species, shows some ultrastructural peculiarities

[16], may also be involved.

The presence of DNA in the ECFs of A. mellifera

spermatozoa was demonstrated by staining with

toluidine blue and by the anisotropic characteristics

typical for DNA, such as the negative birefringence

and linear dichroism in polychromatic light (visible

spectrum) following staining with toluidine blue [20,21].

DNA birefringence in the presence of toluidine blue is

caused by the orientation of the purine and pyrimidine

rings that overlie each other at right-angles to the axial

backbone of the macromolecule and by the orientation

of the DNA-bound toluidine blue molecules that follow

the DNA conformation [10,11,20,21]. The resulting

ordered arrangement of toluidine blue molecules

attached to the DNA phosphates allows the selective

absorption of polarized light [10,11].

Although molecular changes are expected to occur

in spermatozoa when they enter the female genital

tract, there were no detectable changes in the chromatin

fluidity under the experimental conditions used here.

In conclusion, the altered patterns of chromatin

fluidity in A. mellifera spermatozoa showed here,

demonstrate that the ECF formation is a complex

process, and it depends on the type of protein the

DNA is bound to, and how these macromolecules

are complexed, giving to each cell type a specific

chromatin organization, according to their physiological

functions.

ACKNOWLEDGMENTS

The authors thank Dr. Heidi Dolder (Department of

Cell Biology, Institute of Biology, UNICAMP, Brazil)

for helpful discussions. This work was supported by the

Brazilian National Research and Development Council

(CNPq) and FAPESP (grant no. 99/02547-8).

REFERENCES

1. Bloch DP (1969) A catalog of sperm histones. Genetics 61, 93-111.

2. Cruz-Höfling MA, Cruz-Landim C, Kitajima EW

(1970) The fine structure of spermatozoa from the

honey bee. An. Acad. Bras. Ciênc. 42, 69-78.

3. Falco JRP, Carvalho HF, Mello MLS (1996)

Cytochemical and biochemical characteristics of the

nuclear basic protein of the honey bee spermatozoa.

In: Programa e Resumos da XXVa. Reunião Anual SBBq, Caxambu (RS), Brazil. p. 25.

4. Falco JRP, Mello MLS (1999) Critical electrolyte

concentration of spermatozoal chromatin containing

histone H1 variants. Genet. Mol. Biol. 22, 197-200.

5. Haaf T, Ward DC (1994) Structural analysis of alpha-

satellite DNA and centromere proteins using extended

chromatin and chromosomes. Hum. Mol. Genet. 3,

697-709.

6. Haaf T, Ward DC (1995) Higher order nuclear structure

in mammalian sperm revealed by in situ hybridization

132



and extended chromatin fibers. Exp. Cell Res. 219, 604-611.

7. Heng HHQ, Squire J, Tsui LC (1992) High-resolution mapping of mammalian genes by in situ hybridization to free chromatin. Proc. Natl. Acad. Sci. USA 89, 9509-9513.

8. Lino-Neto J, Bao SN, Dolder H (2000) Sperm ultrastructure of the honey bee (Apis mellifera) (L) (Hymenoptera, Apidae) with emphasis on the nucleus-flagellum transition zone. Tissue Cell 32, 322-327.

9. Mello MLS, Falco JRP (1996) Critical electrolyte concentration of DNA-protein complexes in spermatozoal and somatic cell nuclei of the honey bee, Apis mellifera. Insect Biochem. Mol. Biol. 26, 793-795.

10. Mello ML, Vidal BC (1973) Linear dichroism and anomalous dispersion of birefringence on bee sperm heads. Acta Histochem. 45, 109-114.

11. Mello MLS, Vidal BC (1977) Changes in anisotropic properties and nuclear stainability during spermatogenesis in the grasshopper, Staurorhectus longicornis Giglio-Toss. In: Advances in Invertebrate Reproduction. Vol. 1. (Adiyodi KG, Adiyodi RG, eds). pp. 75-83. Peralam-Kenoth: Kerala.

12. Moraes AS, Mello MLS (2005) Chromatin supra-organization and extensibility in mouse hepatocytes with aging. 45th ASCB Annual Meeting, San Francisco 2005. Abstracts, CD-ROM.

13. Moraes AS, Mello MLS (2005) Extended chromatin fibers in mouse testicular spermatozoa. Braz. J. Morphol. Sci. 22, 91-96.

14. Moraes AS, Vidal BC, Guaraldo AMA, Mello MLS (2005) Chromatin supraorganization and extensibility in mouse hepatocytes following starvation and refeeding. Cytometry A 63, 94-107.

15. Oliva R, Dixon GH (1991) Vertebrate protamine genes

and the histone-to-protamine replacement reaction.

Prog. Nucleic Acid Res. Mol. Biol. 40, 25-94.

16. Peng CYS, Yin CM, Yin LRS (1993) Ultrastructure of

honey-bee, Apis mellifera, sperm with special emphasis

on the acrosomal complex following high-pressure

freezing fixation. Physiol. Entom. 18, 93-101.

17. Poirier M, Eroglu S, Chatenay D, Marko JF (2000)

Reversible and irreversible unfolding of mitotic newt

chromosomes by applied force. Mol. Biol. Cell 11,

269-276.

18. Taylor JH (1964) The arrangement of chromosomes in

the mature sperm of the grasshopper. J. Cell Biol. 21,

286-289.

19. Verma LR (1972) Cytochemical analysis of nuclear

histones in honey-bee (Apis mellifera L.) spermatozoa.

Can. J. Zool. 50, 1241-1242.

20. Vidal BC (1972) Doppelbrechungsdispersion und

Lineardichroismus von Eu- und Heterochromatin nach

Färbung mit Toluidinblau. Nachweis eines Cotton-

Effektes. Beitr. Path. 145, 269-285.

21. Vidal BC (1987) Métodos em Biologia Celular. In:

Biologia Celular (Vidal BC, Mello MLS, eds). pp. 5-

39. Livraria Atheneu: Rio de Janeiro.

22. Vidal BC (2000) Extended chromatin fibers:

crystallinity, molecular order and reactivity to

concanavalin A. Cell Biol. Int. 24, 723-728.

23. Ward WS, Coffey DS (1990) Specific organization of

genes in relation to the sperm nuclear matrix. Biochem.

Biophys. Res. Commun. 173, 20-25.

__________________

Received: November 11, 2005

Accepted: January 25, 2006

133

the con-a-peroxidase method for tissue localization of glucosyl and

Documents