Alberto Hiroyuki Tomiyama

Micro RNA em adenocarcinoma de próstata : caracterização

da expressão em tumores de baixo grau, órgão-confinados

Dissertação apresentada à Faculdade de Medicina

da Universidade de São Paulo para obtenção do

título de Mestre em Ciências

Programa de Urologia

Orientadora: Prof. Dra. Kátia Moreira Ramos Leite

(Versão corrigida. Resolução CoPGr 5890, de 20 de dezembro de 2010.

A versão original está disponível na Biblioteca FMUSP)

São Paulo

2011

Dados Internacionais de Catalogação na Publicação (CIP)

Preparada pela Biblioteca da

Faculdade de Medicina da Universidade de São Paulo

reprodução autorizada pelo autor

Tomiyama, Alberto Hiroyuki

Micro RNA em adenocarcinoma de próstata : caracterização da expressão em

tumores de baixo grau, órgão-confinados / Alberto Hiroyuki Tomiyama. -- São Paulo,

2011.

Dissertação(mestrado)--Faculdade de Medicina da Universidade de São Paulo.

Programa de Urologia.

Orientadora: Kátia Moreira Ramos Leite.

Descritores: 1.Micro RNAs 2.Neoplasias próstata 3.Adenocarcinoma 4.Perfilação

da expressão gênica 5.Antígeno prostático específico

USP/FM/DBD-337/11

ii

Agradecimentos

Ao meu pai Jorge Tomiyama e a minha mãe Elza Tomiyama que me deram

todo suporte necessário para o meu desenvolvimento tanto pessoal quanto

profissional.

Aos meus queridos irmãos Bruno Tomiyama e Cláudia Tomiyama pelo

incentivo, companheirismo e paciência.

À Profa. Dra. Katia Leite que acreditou no meu potencial e me deu a

oportunidade de desenvolver o projeto. E por ter sempre me incentivado e

me cobrado durante o mestrado.

Ao Dr. Flavio Canavez pela oportunidade de fazer o mestrado e pelo

incentivo.

Aos meus queridos amigos do LIM-55 Iran Silva, Camila Piantino, Luciana

Timoszczuk, Dr. José Pontes pelo incentivo e ajuda.

A Sabrina Reis e Juliana M. S. Canavez pela enorme contribuição a este

projeto.

A Adriana Sañudo que tanto me ajudou nas análises estatísticas.

Aos queridos amigos da Genoa Biotecnologia pela amizade.

iii

Sumário

Lista de abreviaturas e siglas

Lista de Tabelas e Figuras

Resumo

Abstract

I. Introdução..................................................................................................... 1

1. Generalidades....................................................................................... 2

2. miRNA................................................................................................... 5

a. miRNA como biomarcador.............................................................. 8

b. miRNA no câncer de próstata....................................................... 11

II. Objetivo...................................................................................................... 12

III. Métodos….…............................................................................................ 14

1. Exame macroscópico.......................................................................... 16

2. Exame microscópico........................................................................... 16

3. Seleção dos pacientes........................................................................ 17

4. Isolamento do RNA total......................................................................18

5. Análise da expressão de miRNA......................................................... 20

6. Análise estatística................................................................................22

IV. Resultados................................................................................................23

V. Discussão.................................................................................................. 33

VI. Conclusão.................................................................................................40

VII. Perspectivas............................................................................................ 42

VIII. Referências…….....…............................................................................ 44

Apêndices

iv

Lista de abreviaturas e siglas

CaP câncer de próstata

Ct threshold cycle

DNA ácido desoxirribonucléico

EGF fator de crescimento da epiderme

IC intervalo de confiança

HPB hiperplasia prostática benigna

LLC leucemia linfocítica crônica

LMA leucemia mielóide aguda

miRNA micro-RNA

MMP metaloproteínase

mRNA RNA mensageiro

pri-miRNA miRNA primário

pT estadiamento patológico

PSA antígeno específico da próstata

RISC complexo silenciador induzido por RNA

RNA ácido ribonucléico

rpm rotações por minuto

PCR reação da polimerase em cadeia

PTEN homólogo da fosfatase e tensina

qPCR PCR em tempo real

RB retinoblastoma

TNM classificação dos tumores malignos

VEGF fator de crescimento vascular endotelial

v

Lista de Tabelas e Figuras

Tabela 1. Estudos de “microarray” para identificação de genes diferencialmente

expressos em câncer de próstata. Modificado de Nelson PS ........................ 5

Tabela 2. Características demográficas, clínicas e anátomo-patológicas dos

pacientes com CaP favorável e desfavorável................................................ 16

Tabela 3. miRNA a serem estudados em pacientes com carcinoma de

próstata favorável, órgão-confinados............................................................ 21

Tabela 4. Média geométrica dos níveis de expressão de 14 miRNA em

carcinomas de próstata favorável e desfavorável......................................... 24

Figura 1. Representação esquemática do modelo atual da biogênese e ação

supressora pós-transcripcional dos miRNA e RNA de interferência, segundo

Winter J et al (2009)........................................................................................ 7

Figura 2. Perfil de expressão de 14 miRNA em 45 pacientes com tumores

favoráveis...................................................................................................... 25

Figura 3. Perfil de expressão de 14 miRNA em 53 pacientes com tumores

desfavoráveis................................................................................................. 26

Figura 4. Box plot representando os valores médios e o intervalo de confiança

de expressão do miR-143 em pacientes com câncer de próstata.................. 27

Figura 5. Box plot representando os valores médios e o intervalo de confiança

de expressão do miR-145 em pacientes com câncer de próstata.................. 28

Figura 6. Box plot representando os valores médios e o intervalo de confiança

de expressão do miR-146a em pacientes com câncer de próstata................ 29

vi

Figura 7. Box plot representando os valores médios e o intervalo de confiança

de expressão do miR-191 em pacientes com câncer de próstata.................. 30

Figura 8. Box plot representando os valores médios e o intervalo de confiança

da expressão do miR-218 em pacientes com câncer de próstata.................. 31

Figura 9. Box plot representando os valores médios e o intervalo de confiança

da expressão do miR-Let7c em pacientes com câncer de próstata............... 32

vii

Resumo

Tomiyama AH. Micro RNA em adenocarcinoma de próstata : caracterização

da expressão em tumores de baixo grau, órgão-confinados [dissertação].

São Paulo: Faculdade de Medicina, Universidade de São Paulo; 2011. 56p.

Introdução: Os micro RNA (miRNA) são formados a partir de RNA

precursores de fita dupla que contém entre 60 a 110 nucleotídeos e formam

estruturas do tipo hairpin. Imediatamente após sua transcrição pela RNA

polimerase II a enzima “Dicer” promove a clivagem do RNA precursor em

seqüências menores contendo 19 a 22 nucleotídeos. Após a clivagem, o

miRNA integra-se ao complexo silenciador induzido pelo RNA (RISC) que o

conduz ao seu RNA mensageiro (mRNA) homólogo recém transcrito. Esta

associação promove a degradação do mRNA, ou interfere na tradução da

proteína caracterizando um grande mecanismo de controle da expressão

dos genes. Este mecanismo está relacionado ao desenvolvimento de órgãos

e tecidos, e está envolvido no processo de carcinogênese. Nosso objetivo é

identificar um perfil de expressão de miRNA que defina o adenocarcinoma

de próstata de prognóstico favorável e desfavorável considerando os níveis

de PSA e dados anatomopatológicos. Materiais e métodos: Foram

selecionados 53 pacientes com tumores desfavoráveis (mediana do escore

de Gleason igual a 8, 79,2% estadiados pT3, mediana de PSA 10,1 ng/mL e

mediana do volume tumoral de 23%) e 45 considerados favoráveis (mediana

do escore de Gleason igual a 5, 80% estadiados pT2, mediana de PSA de

7,8 ng/mL e mediana do volume tumoral de 11,5%). O controle foi

representado por 7 pacientes com hiperplasia prostática benigna (HPB).

Todos os pacientes foram submetidos a prostatectomia radical pelo mesmo

cirurgião. Os espécimes cirúrgicos foram examinados na sua totalidade pelo

mesmo patologista. A análise dos miRNA foi feita a partir de tecido

congelado e tecido incluído em parafina usando a técnica da reação em

cadeia da polimerase em tempo real quantitativa (qRT-PCR) utilizando

primers e sondas Taqman® específicas. O RNU43 foi usado como controle

viii

interno. Resultados: Com exceção dos miRNA 199a, 21, 15a, 16 e 25 que

se mostraram subexpressos tanto nos casos desfavoráveis como nos

favoráveis, houve uma diminuição global na expressão dos miRNA com

redução estatisticamente significativa na expressão dos miRNA 143, 145 e

146a, 191, 218 e Let7c em tumores desfavoráveis em relação aos tumores

favoráveis. Conclusão: Demonstramos que no processo de transição entre

os carcinomas favoráveis e desfavoráveis de próstata existe uma perda

global na expressão de miRNA que podem ser importantes controladores de

expressão de uma série de genes relacionados a progressão desta

neoplasia. Dados experimentais avaliando o papel desses miRNA devem ser

conduzidos para que possamos definir seu papel na evolução do câncer de

próstata.

Descritores: 1.Micro RNAs 2.Neoplasias próstata 3.Adenocarcinoma

4.Perfilação da expressão gênica 5.Antígeno prostático específico

ix

Abstract

Tomiyama AH. Micro RNA in prostate adenocarcinoma : characterization of

expression in low-grade tumors, organ-confined tumours [dissertation]. São

Paulo: "Faculdade de Medicina, Universidade de São Paulo"; 2011. 56p.

Introduction: micro RNA (miRNA) are formed from double-stranded RNA

precursors that contain between 60-110 nucleotides and form structures such

as hairpin. Immediately after their transcription by RNA polymerase II, the

enzyme Dicer promotes the cleavage of precursor RNA sequences

containing minor 19-22 nucleotides. After cleavage, the miRNA is part of the

RNA-induced silencing complex (RISC) that leads to its messenger RNA

(mRNA) newly transcribed counterpart. This association promotes the

degradation of mRNA, or interferes with the protein translation characterizing

a great mechanism for controlling gene expression. This mechanism is

related to the development of organs and tissues, and may be involved in the

process of carcinogenesis. Our goal is to identify a miRNA expression profile

that distinguishes prostate adenocarcinoma of favorable and unfavorable

prognosis considering the PSA and pathological findings. Material and

Methods: We studied 53 patients with tumors considered unfavorable

(Median of Gleason score 8, 79.2% staged pT3, median of PSA 10.1 ng/mL

and median of tumor volume of 23%) and 45 considered favorable (Median

of Gleason score 5, 80% staged pT2, median of PSA 7.8 ng/mL and median

of tumor volume of 11.5%). The control group was represented by seven

patients with benign prostatic hyperplasia (BPH). All patients underwent

radical prostatectomy by the same surgeon. The surgical specimen was

examined entirely by the same pathologist. The analysis of miRNA was made

from frozen and paraffin embedded tissue by quantitative real-time

polymerase chain reaction (qRT-PCR) using the Taqman® specific primers

and probes. The RNU43 was used as a internal control. Results: Except for

miRNAs 199a, 21, 15a, 16 e 25 that were underexpressed by both favorable

and unfavorable prostate cancer, there was a global decrease of all miRNAs

x

studied, and some differences were statistically significant as miRNAs 143,

145 e 146a, 191, 218 e Let7c that were underexpressed in unfavorable

carcinomas compared favorable tumor. Conclusion: We have demonstrated

that in the process of transition between favorable and unfavorable prostate

cancer there is a global loss of expression of miRNAs. These molecules can

be important controllers of a series of genes related to cancer progression.

Experimental studies are needed in order to comprehend the role of these

genes in prostate carcinogenesis.

Descriptors: 1.Micro RNAs 2.Prostatic neoplasms 3.Adenocarcinoma

4.Gene expression profiling 5.Prostate-specific antigen

INTRODUÇÃO

2

I. INTRODUÇÃO

I.1. Generalidades

O câncer de próstata (CaP) é o tumor mais comum do homem e a

segunda causa de óbito por câncer no Brasil (www.inca.gov.br). Nos

Estados Unidos, estima-se que um homem em cada seis será diagnosticado

com CaP durante sua vida, sendo que um homem em 33 morrerá da doença

(Jemal et al 2011). Estes altos índices estão ligados principalmente ao

aumento da expectativa de vida da população, aos hábitos de vida

principalmente alimentares e aos programas de rastreamentos.

Os tumores da próstata são bastante heterogêneos no seu

comportamento clínico, sendo os principais indicadores de prognósticos os

níveis séricos do antígeno específico da próstata (PSA), estadiamento

clínico e o grau histológico de Gleason (Graefen et al. 2002). Outros fatores

prognósticos considerados importantes são o volume tumoral, a presença e

a porcentagem dos padrões 4 e 5 de Gleason (Stamey et al. 1999 e Leite et

al. 2005). De modo geral os CaP são considerados como favoráveis se

apresentarem Gleason < 7, PSA < 10 ng/mL, e forem órgão-confinados ou

desfavoráveis se apresentarem Gleason > 7, PSA ≥ 10 ng/mL e forem não

confinados à glândula. No entanto, estes dados têm–se mostrado

insuficientes para o correto agrupamento dos tumores. Este fato é relevante

no momento em que as modalidades terapêuticas, tanto curativas quanto

paliativas apresentam graus expressivos de morbidade.

Existe uma clara necessidade por novos métodos que sejam capazes

de prever o comportamento do CaP, desde que as decisões terapêuticas

contemplam um amplo espectro desde a simples observação, nos casos

favoráveis até a castração nos tumores mais avançados e o uso de

marcadores moleculares é um dos caminhos recentemente discutidos

(Stenhenson et al. 2005). A zona intermediária entre os dois extremos de

evolução do CaP é a mais comumente observada e também nesses casos

as possibilidades de tratamento se dividem principalmente entre a

3

prostatectomia radical e a radioterapia intersticial (braquiterapia), ou externa.

Houve melhora considerável dessas metodologias, no entanto os índices de

incontinência urinária, disfunção erétil e lesões actínicas no reto e bexiga

não são desprezíveis. Infelizmente, após o tratamento curativo, cerca 40%

dos pacientes necessitarão de resgate por recidiva.

Numerosas alterações genéticas têm sido envolvidas na promoção e

progressão do CaP, sendo algumas de aparecimento precoce como a re-

expressão da telomerase (Leite et al. 2001), outras tardias como a mutação de

p53 que se relaciona com doença metastática e refratária a hormônio (Navone

et al. 1993). Demonstramos que a expressão imunohistoquímica de p53

relacionou-se com o aumento da proliferação celular, diminuição dos índices de

apoptose, maior grau de Gleason e maior estádio patológico. Propusemos

também um papel do MDM2, regulador negativo da proteína p53. Mostramos

uma expressão imunohistoquímica de MDM2 em 41% dos CaP, relacionada

com atividade proliferativa e maior volume tumoral (Leite et al. 2001).

Os estudos moleculares iniciais em CaP foram descritos com grupos

restritos de genes e só agora, com o desenvolvimento de metodologias

robustas, é possível uma avaliação mais ampla e complexa dos genes

envolvidos no desenvolvimento do CaP. A metodologia do “microarray”,

descrita por Schena et al, (1995) tem possibilitado a avaliação simultânea de

grande número de genes diferencialmente expressos entre o tecido normal e

o tumoral (Tabela 1) (Nelson 2004). Numa análise da tabela constata-se que

o gene Hepsin está presente em quase todos os trabalhos publicados.

Hepsin é uma serina protease transmembrana tipo II que têm-se encontrada

super-expressa no CaP, sendo que sua função não está bem definida na

progressão do CaP. Magee et al. (2001) analisaram amostras de 15

pacientes, 11 CaP e 4 tecidos normais, e encontraram alterações

importantes em 4 genes entre 4.712 estudados. Dhanasekaran et al. (2001)

encontraram 7 genes diferencialmente expressos entre 9.984 estudados.

Singh et al. (2002) estratificaram o comportamento dos carcinomas de

próstata através das diferenças de expressão de 5 genes. Ramaswamy et al.

4

(2003) relacionaram os níveis de expressão de 17 genes com o potencial

metastático do CaP.

Apesar dos numerosos estudos e do uso de plataformas amplas, os

achados não se sobrepõem. Isto pode dever-se ao pequeno número de

pacientes estudados, a má qualidade das amostras, ausência de avaliação

padronizada dos espécimes ou falta de acompanhamento regular dos

doentes. Esta fragilidade dos estudos é suplantada pelo nosso grupo, que

possui um banco de dados de mais de 2500 pacientes tratados

cirurgicamente para CaP pelo mesmo grupo de cirurgiões e acompanhados

regularmente por mais de 10 anos. Todas as peças cirúrgicas foram

examinadas na sua totalidade de modo padronizado pelo mesmo

patologista. E contamos ainda com um banco de tumor de próstata com

mais de 2000 amostras armazenadas a –170ºC.

A maioria dos estudos publicados busca diferenças nos níveis de

RNA e não consideram mecanismos de regulação pós-transcripcionais.

5

Tabela 1. Estudos de microarray para identificação de genes diferencialmente expressos em câncer de próstata. Modificado de Nelson OS 2004

Número de

Amostras

Nº de genes

Pesquisados

Benigno Tumor Genes

diferencialmente

expressos

Exemplos Referências

588 1 1 13 GTM1 Chetcuti et al. 2001

588 1 1 15 VEGF Chaib et al. 2001

6.112 9 16 210 Hepsin Luo et al. 2002

9.984 Pool 56 >200 Hepsin, PIM 1 Dhanasekaran et

al. 2001

4.712 4 14 4 Hepsin Magee et al. 2001

8.920 9 25 >400 Hepsin, MIC1,

FAS

Welsh et al. 2001

6.800 8 9 86 Hepsin, PSMA Stamey et al. 2001

35.000 15 15 84 Hepsin AMACR Luo et al. 2002

12.600 9 17 216 Hepsin, AMACR Ernst et al .2002

12.000 3 41 >3.400 USP13, STK11 La Tulippe et al.

2002

12.600 50 52 456 Hepsin,

Tetraspan 1

Singh et al. 2002

12.600 8 33 50 Hepsin, AMACR Vanaja et al. 2003

6.400 Pool 13 136 Hepsin, nectin3 Best et al. 2003

46.000 Não

utilizou

72 266 Trp-p8, seladin Henshall et al.

2003

I.2. miRNA

Os microRNA formam um grupo de moléculas pequenas de RNA, que

contém entre 19 a 25 nucleotídeos não codificantes de proteínas com ação

fundamental na regulação da expressão dos genes. São sintetizados pela

RNA polimerase II e modificados, após a transcrição, pela adição de uma

guanosina metilada (CAP) na região 5’ e uma cauda poli A na região 3’. Esta

molécula denominada pri-miRNA forma uma estrutura do tipo “hairpin” e no

6

núcleo sofre a ação do complexo microprocessador (Drosha-DGCR8),

enzimas do tipo Rnase-III, quando ocorre o fenômeno “cropping”. O produto

deste processamento tem aproximadamente 70 nucleotídeos, é denominado

pre-miRNA e é transportado ao citoplasma pela Exportina 5, onde sofre a

ação de Dicer, formando o complexo miRNA-miRNA duplex contendo o

miRNA maduro e sua fita complementar. Apenas uma das fitas do duplex

será montada dentro do complexo silenciador induzido por RNA (RISC), que

por sua vez atua sobre o seu RNA mensageiro (RNAm) alvo reprimindo a

tradução da proteína ou promovendo a degradação do RNAm (Figura 1).

Essa ação depende do grau de complementaridade do miRNA com o seu

RNAm alvo. Nos animais, a ligação do miRNA com a região 3’ do RNAm

alvo resulta na repressão da tradução da proteína com exceção do miRNA-196.

Nas plantas a ligação pode-se fazer nas regiões codificantes do mRNA e

resulta na degradação deste (Kim 2005). A complementaridade das

primeiras duas a oito bases é fundamental para a estabilização da ligação do

miRNA com o RNAm alvo e essa região é a mais estudada pela

bioinformática para identificação de genes alvos controlados pelos miRNA.

Um mesmo miRNA pode ter mais de 100 genes alvo com funções diversas e

acredita-se que um terço dos RNAm humanos sejam controlados por esse

mecanismo (Yoon et al. 2005).

7

Figura 1. Representação esquemática do modelo atual da biogênese e ação supressora pós-transcripcional dos miRNA e RNA de interferência, segundo Winter J et al (2009).

Os miRNA são responsáveis pela regulação de processos

fundamentais da célula, como a proliferação, a diferenciação, a resposta ao

estresse e a apoptose. Eles estão localizados nas regiões intergênicas,

assim como em íntrons, e podem fazer parte da fita codificante ou não

codificante do DNA. Metade dos miRNA situam-se nos chamados sítios

frágeis do genoma, cujas anormalidades estão associadas ao câncer. As

alterações da biogênese e expressão dos miRNA são similares àquelas

descritas nos processos carcinogenéticos e resultam em irregularidades nos

níveis de expressão tanto de genes supressores de tumor como de

oncogenes. Os mecanismos fundamentais que alteram o controle da

expressão dos miRNA são:

8

1. Alterações genéticas grosseiras como deleções, inserções,

inversões e translocações.

2. Alterações genéticas mínimas como mutações puntiformes em

elemento promotor de um gene de miRNA ou em região

codificadora de fator de transcrição.

3. Mutações na seqüência de interação miRNA:RNAm.

4. Alterações epigenéticas.

5. Inserção de um elemento viral nas proximidades de um gene de

miRNA.

6. Anormalidades do processamento e/ou transporte do miRNA por

alterações nas enzimas processadoras, Drosha, Dicer e Exportina

5 (Esquela-Kerscher et al. 2006).

I.2.a. miRNA como biomarcador

Com base nas experiências com o uso de perfis de expressão de

RNAm como assinaturas de doenças específicas, vários grupos abordaram

a questão dos perfis de expressão dos miRNA que poderiam ser úteis como

biomarcadores para o diagnóstico de câncer, determinação do prognóstico e

desenvolvimento de tratamentos específicos. Os perfis de expressão de

miRNA em tecidos neoplásicos em comparação com suas contrapartidas

normais têm se mostrado diferentes. E essas assinaturas podem ser úteis na

discriminação entre câncer e lesões reativas, na classificação de tumores

indiferenciados ou para diferenciação de diferentes tipos histológicos de

neoplasias comprometendo o mesmo órgão.

Numerosos estudos têm demonstrado que marcadores moleculares

sozinhos ou em combinação com métodos convencionais são capazes de

melhorar a discriminação do comportamento de neoplasias. Experiências

com as assinaturas de expressão gênica de mRNA em mais de uma década

na investigação do câncer da mama só neste momento levou à introdução

na prática de ensaios de expressão gênica que orientam o tratamento e

classificam o risco de recidiva (Benowitz 2008).

9

Os primeiros estudos utilizando os perfis de expressão de miRNA

foram promissores para a determinação do prognóstico, sobrevida e

resposta terapêutica de pacientes com carcinoma de cólon (Schetter et al.

2008), avaliação de resposta ao gefitinib no câncer de pulmão (Weiss et. al

2008) e resistência à quimioterapia de células de câncer de pulmão (Blower

et al. 2008). Foi recentemente demonstrado que a resistência no câncer de

pulmão ao metotrexate, até agora inexplicável, está relacionada a um

polimorfismo do miR-24. Esse polimorfismo leva a um mecanismo novo que

relaciona os miRNA a resistência às drogas e é o primeiro exemplo da

chamada "farmacogenômica miRNA" (Mishra et al. 2008).

Uma outra possibilidade de uso dos miRNA é a detecção deste livres

que são liberados das células do câncer no sangue e em fluidos biológicos.

Este fato possibilita a detecção não invasiva de células tumorais em fluidos

corpóreos ou circulantes caracterizando uma nova metodologia diagnóstica.

Os primeiros estudos sobre miRNA no sangue foram realizados em

pacientes com câncer de ovário e próstata e apóiam a viabilidade desta

abordagem não só para diagnosticar o câncer, mas também para estimar a

sua agressividade (Taylor et al. 2008 e Mitchell et al. 2008).

Na maioria dos tumores os miRNA estão subexpressos, do que se

conclui que atuem como supressores de tumor, ligando-se a oncogenes e

diminuindo sua expressão. Os primeiros miRNA supressores de tumor

foram descritos por Calin et al. (2002), que mostraram que nas leucemias

linfocíticas crônicas (LLC) havia deleção de miR-15a e miR-16-1. Ambos

estão mapeados em uma região de 30 kb de um íntron de um RNA não

codificante de proteína cuja função é desconhecida. Regulam

negativamente o BCL2, um gene com poderosa ação anti-apoptótica. Os

miR-143 e miR-145 tem sido descritos como subexpressos em câncer de

mama, câncer coloretal, adenocarcinoma de próstata e neoplasias

hematogênicas (Esquela-Kerscher et al., 2006).

Um dos miRNA mais estudados é o let-7c, que é reconhecido como

regulador do oncogene RAS. Sua subexpressão é um fenômeno constante

no câncer de pulmão, onde se correlaciona com menor sobrevida em

10

pacientes submetidos a tratamento supostamente curativo. O RAS é uma

proteína de membrana com atividade de GTPase que induz a proliferação

celular via Map-quinase. A introdução de let-7c em linhagens de câncer de

pulmão leva a diminuição dos níveis de RAS e diminuição na proliferação

celular (Takamizawa et al. 2004).

Existem certos miRNA que estão superexpressos em tumores e

promovem aumento na atividade proliferativa ou têm ação anti-apoptótica e

são denominados oncomirs. O miR-21, que está superexpresso em

glioblastomas e colangiocarcinomas tem como alvo o gene supressor de

tumor PTEN. O cluster miR-221-222 está relacionado ao desenvolvimento

do glioblastoma multiforme, atuando na repressão do p27kip1 inibidor de

quinases ciclinas-dependentes e controlador do ciclo celular (Nicoloso et al.

2008). Também é descrita a ação inibitória desse cluster sobre o receptor de

estrógeno, conferindo resistência ao tamoxifen no câncer de mama (Zhao et

al. 2008).

Algumas leucemias de comportamento agressivo apresentam

translocações t(8;17) envolvendo miR-142 e o oncogene MYC. Este

fenômeno altera o processamento e a atividade do miRNA, aumentando a

expressão de MYC, um potente controlador positivo do ciclo celular.

Outro miRNA relacionado à expressão de MYC é o miR-155, que está

superexpresso nos linfomas de Burkitt pediátricos, linfoma de Hodgkin,

linfomas de células grandes B primários do mediastino e no câncer de

mama. Um achado freqüente em linfomas é a amplificação de 13q31, porém

a única anormalidade encontrada nesses casos é a superexpressão de um

RNA não codificante de proteína denominado C13orf25. Este RNA

transcreve um grupamento de miRNA, miR-17-5p, miR-17-3p, miR-18a, miR-

19a, miR-20a, miR-19b e miR-92-1, conhecido como cluster miR-17-92, que

está super-regulado em linfomas de células B, câncer de pulmão,

rabdomiossarcoma alveolar e lipossarcomas. Demonstrou-se que a proteína

MYC liga-se ao primeiro íntron de C13orf25, regulando a transcrição deste

grupamento de miRNA. Desse modo, este cluster de miRNA poderia ser

considerado como miRNA oncogênico (Esquela-Kerscher et al. 2006).

11

I.2.b. miRNA no câncer de próstata

São poucos os estudos dedicados a avaliação dos perfis de miRNA

em câncer de próstata, principalmente em espécimes clínicos. Porkka et al.

(2007) publicaram um estudo de expressão de miRNA por técnica de

microarray em apenas 9 tumores primários, 6 linhagens celulares e 9

implantes tumorais em camundongos; demonstraram subexpressão de 37

miRNA e superexpressão de 14 correlacionando este perfil de expressão

com o comportamento dos tumores em relação aos andrógenos. Shi et al.

(2007) estudando apenas linhagens comerciais de carcinomas da próstata,

demonstraram um papel oncogênico de miR-125b que seria importante para

o crescimento andrógeno-independente das células. Este achado foi

recentemente confirmado por deVere White et al. (2009) que demonstraram

superexpressão de miR-125b de modo mais significativo em linhagens de

tumores independentes de andrógeno. Ambs et al. (2008) explorando 60

tumores primários apresentaram um painel de miRNA sub e superexpressos,

entre eles a superexpressão de miR-32. Bonci et. al. (2008) publicaram a

subexpressão de miR-15a e de miR-16-1, reguladores de expressão de

BCL2 em linhagens de células de CaP. O miR-221 e o miR-222 mostraram-

se superexpressos em linhagens de CaP andrógeno-independentes em

estudo publicado por Sun et al. (2009), sendo que a inibição dos mesmos

com antagomiR (RNA de sequência complementar ao miRNA) restaurou a

sensibilidade das células tumorais ao bloqueio hormonal, vislumbrando uma

nova possibilidade terapêutica.

Nós já determinamos uma mudança no perfil de expressão de miRNA

entre o carcinoma de próstata desfavorável e o carcinoma metastático.

Identificamos uma menor expressão de miR-100, miR-Let7c e miR-218 em

tumores metastáticos especulando um provável papel desses miRNA na

progressão da doença (Leite KRM et al. 2009).

Objetivo

13

II. Objetivo

Caracterizar a expressão de miRNA no CaP considerados favoráveis,

com escore de Gleason<7, órgão-confinado e PSA<10 ng/mL e comparar o

perfil de expressão destes com tumores desfavoráveis Gleason>7, não

órgão-confinados e com PSA≥10 ng/mL.

Métodos

15

III. Métodos

Estudamos o perfil de expressão de miRNA de 53 pacientes com CaP

tratados por prostatectomia radical entre dezembro de 1997 e agosto de

2000 no Hospital Sírio Libanês pela equipe do Prof. Dr. Miguel Srougi, cujo

diagnóstico foi de carcinoma desfavorável (mediana do escore de Gleason

igual a 8, 79,2% estadiados pT3, mediana de PSA 10,1 ng/mL e mediana do

volume tumoral de 23%) e 45 considerados favoráveis (mediana do escore

de Gleason igual a 5, 80% estadiados pT2, mediana de PSA de 7,8 ng/mL e

mediana do volume tumoral de 11,5%). Como controle, estudamos 7 tecidos

prostáticos benignos, provenientes de glândulas hiperplásicas. Nesse grupo

a idade média dos pacientes foi de 72 anos variável de 65 a 84.

Os espécimes cirúrgicos foram examinados a fresco, pelo patologista,

imediatamente após a sua ressecção, sendo seccionado fragmento de 1cm2

e congelado e armazenado a -170ºC em nitrogênio líquido.

Os pacientes foram informados sobre o uso do material em pesquisa

científica, e assinaram o termo de consentimento. As características

demográficas, clínicas e anatomopatológicas dos grupos estão expostas na

Tabela 2.

O estudo foi aprovado pelo comitê de ética em pesquisa do Hospital

das Clínicas da Faculdade de Medicina da Universidade de São Paulo

(Protocolo nº0216/8).

16

Tabela 2. Características demográficas, clínicas e anatomopatológicas dos

pacientes com CaP favorável e desfavorável

CaP – Favorável (45)

CaP - Desfavorável (53)

p

Idade (anos) 0,107* Mediana 61 65 Variação 47 – 76 50 – 79

PSA (ng/mL) 0,011*

Mediana 7,8 10,1 Variação 2,1 – 37 4,2 – 40

Escore de Gleason

<0,001**

Mediana 5 8 Variação 4 – 6 8 – 10

Volume do tumor

(%) <0,001**

Mediana 11,5 23 Variação 1 - 43 3 – 88

Estadio <0,001***

pT2 36 (80%) 11 (20,8%) pT3 9 (20%) 42 (79,2%)

*Mann-Whitney, ** Teste T, *** Qui-Quadrado

III.1. Exame macroscópico

Os espécimes cirúrgicos foram fixados em formalina tamponada 10%

por um período de 4 a 16 h. Toda a glândula foi submetida a estudo

histológico seguindo recomendações previamente descritas (Henshall et al.

2003). Toda a glândula foi incluída para estudo após suas margens serem

pintadas com tinta nanquim. Os lobos direito e esquerdo foram separados,

sendo realizados cortes transversais seqüenciais a cada 3 mm, designados

da região proximal em direção a distal. De 10 a 15 cortes de cada lobo foram

incluídos para estudo histológico.

17

III.2. Exame Microscópico

Os espécimes foram submetidos a processamento habitual com

inclusão em parafina. Toda a próstata foi submetida a exame histológico.

Cortes de 4 a 6 µm foram corados pela Hematoxilina e Eosina e analisados

em microscópio óptico por um único patologista (KRML).

Foram considerados para análise os seguintes parâmetros:

a. Grau Histológico – O grau histológico de Gleason foi utilizado

para avaliação de diferenciação histológica. Foram considerados os dois

padrões predominantes de Gleason, variáveis de 1 a 5, que somados

resultam no escore de Gleason, variável de 2 a 10 (Epstein et al. 2005).

b. Estadiamento – Utilizamos o estadiamento TNM 2002. A

infiltração do tecido extra prostático que caracteriza a doença pT3a,

carcinoma não-órgão confinado foi considerado na presença de invasão do

tecido adiposo e do plexo vásculo-nervoso periprostático. A infiltração das

vesículas seminais, que caracteriza a doença pT3b somente foi considerada

positiva quando havia invasão de sua parede muscular.

III.3. Seleção dos pacientes

Os pacientes foram divididos em 2 grupos distintos:

a. Portadores de adenocarcinoma desfavorável: Gleason > 7,

não órgão-confinados (pT3) e PSA ≥ 10 ng/ml;

b. Portadores de adenocarcinoma favorável: Gleason < 7,

órgão-confinados (pT2) e PSA < 10 ng/ml.

III.4. Isolamento do RNA total

A partir de 53 espécimes congelados foram extraídos o miRNA (25

casos favoráveis e 28 desfavoraveis). Essas amostras foram maceradas e

colocadas em tubo de microcentrífuga de 1,5 ml estéril. O miRNA total foi

18

isolado com o kit mirVana (Ambion) seguindo o seguinte protocolo. O material

fresco foi macerado através do congelamento com nitrogênio com o uso do

pistilo e cadinho. O material foi macerado até virar um pó bem fino. Essa

material foi colocado em um tubo de 1,5ml e adicionou-se 300ul de tampão

de lise. Esse material foi homogeneizado e adicionou-se 30ul de tampão

aditivo. O tubo foi colocado em repouso no gelo por 10 minutos. Após esse

intervalo adicionou-se 300ul de fenol-clorofórmio. O material foi “vortexado” e

centrifugado a 14.000 rpm por 5 minutos. Após a centrifugação, foi coletado a

fase superior com o auxílio de uma pipeta em outro tubo. A esse material foi

adicionado 200ul de etanol 100%. Essa solução foi transferida para um tubo

contendo uma coluna de sílica. O material foi centrifugado a 1000rpm por 15

segundos. A coluna foi descartada e ao material obtido foi adicionado 400ul

de etanol 100%. A solução foi novamente transferida para outra coluna de

sílica. Centrifugou-se novamente o material a 10.000rpm por 15 segundos.

Descartou-se o material obtido e adicionou-se 700 de tampão de lavagem I a

coluna de sílica. Centrifugou-se novamente a coluna a 10.000rpm por 15

segundos. Descartou-se o sobrenadante e adicionou-se 500ul de tampão de

lavagem 2/3. Centrifugou-se a 10000rpm por 15 segundos. Descartou-se o

sobrenadante e adicionou-se 500ul de tampão de lise 2/3. Centrifugou-se a

14.000 por 1 minuto e descartou-se o sobrenadante. Centrifugou-se

novamente a 14.000rpm por 3 minutos e descartou-se o sobrenadante.

Adicionou-se 50ul de tampão de eluição (previamente aquecido) o centro da

sílica. Deixou-se em repouso por 1 minuto na bancada e centrifgou-se a

14.000rpm por 1 minuto. Ao final foi obtido o miRNA eluido em 50ul de

tampão de eluição. A integridade de cada amostra de miRNA foi determinada

no bioanalisador (2100 Bioanalyzer – Agilent Technologies). Paralelamente, a

concentração e pureza foram estimadas em espectrofotômetro (260/280 nM).

Para extração de miRNA dos 45 espécimes parafinados (20 casos

favoráveis e 25 desfavoraveis) foram feitos dez cortes de 10µm do bloco de

parafina correspondente ao tumor. Para a extração do miRNA foi utilizado o

kit Recover All (Ambion). Foram colocados em tubo de microcentrífuga e

incubados com 1ml de xilol a 50ºC por 3 h para remoção da parafina. Uma

19

lâmina corada com Hematoxilina e Eosina foi examinada para garantia de

representatividade de pelo menos 75% de tumor. Após centrifugação por 2

min a 14.000 rpm o xilol foi descartado e feitos dois banhos com etanol

100% (1ml), sendo realizadas duas centrifugações por 2 min a 14.000 rpm

após cada banho. Adicionou-se então protease (8µl) e tampão de digestão

(800µl), deixando em incubação por 3 h a 50ºC. Adicionou-se 480µl de

solução aditiva, seguida por agitação em vortex. Adicionou-se 1,1ml de

etanol a 100%, após mistura a solução foi transferida para a coluna GFX em

um volume máximo de 700µl. Centrifugou-se a 10.000 rpm por 1 min. A

coluna foi lavada com 700µl de solução de lavagem 1. O sobrenadante foi

descartado e lavou-se a coluna com 500µl da solução de lavagem 2/3.

Houve nova centrifugação a 10.000 rpm por 30 seg, o sobrenadante foi

descartado, centrifugado novamente a 10.000 rpm por 30 seg e

ressuspendeu-se o “pellet” em 60µl de DNAse mix, constituído por 6µl de

10X tampão DNAse, 4µl de DNAse e 50µl de H2O livre de nucleases, a

95ºC. O material foi deixado em repouso em temperatura ambiente por 30

min, posteriormente foi lavado com 700µl de solução de lavagem 1,

mantendo incubado por 1 min a temperatura ambiente. O material foi

centrifugado a 10.000 rpm por 30 seg, descartou-se o sobrenadante e foi

lavado por duas vezes com 500µl de solução de lavagem 2/3, seguindo após

cada lavagem uma centrifugação a 10.000 rpm por 30 seg. O sobrenadante

foi descartado e submetido a nova centrifugação com adição de 60µl de

solução de eluição a 95ºC. O material foi deixado em repouso por 1 min a

temperatura ambiente, centrifugado a 14.000 rpm por 1 min e armazenado a

-20ºC. Do mesmo modo as amostras foram quantificadas em

espectrofotômetro (Nanodrop ND-1000, Wilmington, EUA), e a integridade

verificada em Agilent 2100 Bioanalyzer (Agilent technologies, CA, EUA).

20

III.5. Análise da expressão de miRNA

Todas as amostras com coeficiente 28S/18S superior a 1.8 foram

utilizadas para a determinação dos níveis de expressão de miRNA. A

Reação em cadeia da Polimerase (PCR) em tempo real quantitativa (qRT-

PCR) foi realizada usando sistema ABI 7500 Fast RT-PCR no modo

standard com utilização de Master Mix PCR Taqman Universal (Applied

Biosystems, Foster City, CA, EUA). A expressão de miRNA individuais

especificados na Tabela 2 foi analisada com primers seqüencia-específicos.

(http://microrna.sanger.ac.uk/ sequences).

Este protocolo TaqMan® utiliza dois iniciadores não fluorescentes e uma

sonda com dupla marcação que se anela à região localizada entre os

iniciadores. Esta marcação dupla é formada por um fluoróforo que emite luz

quando excitado e um quencher que absorve a luz emitida pelo fluoróforo.

Durante os ciclos da PCR, a sonda é quebrada pela Taq polimerase na etapa

de extensão do iniciador anelado. Esta quebra da sonda elimina a absorção

pelo quencher da fluorescência emitida que pode ser então medida através de

uma câmera situada na parte superior do equipamento. A quantificação da

emissão absorvida pela câmera após quebra da sonda permite a quantificação

indireta do miRNA alvo contido na reação após cada ciclo da PCR.

As reações foram feitas com 0,5µl de uma solução contendo um par

de primers e a sonda, 5µl do TaqMan master mix e 3,8µl de água livre de

nuclease e 1 µl de cDNA. As condições da reação foram: 2 min a 50ºC, 10

min a 95ºC, e 40 ciclos de 15 seg a 95ºC e 1 min a 60ºC. O nível de

expressão dos miRNA foi obtido pela quantificação relativa e dos níveis de

expressão em vezes determinado pelo método 2-∆∆CT. Todas as reações

foram realizadas em duplicata e o pequeno RNA nucleolar RNU43 usado

como controle endógeno. Para análise de expressão, além do controle

endógeno, analisamos a expressão dos níveis dos mesmos miRNA em

tecido prostático normal não neoplásico. Para este grupo utilizamos

espécimes de pacientes que sofreram ressecção prostática retropúbica para

tratamento de hiperplasia prostática benigna.

21

Tabela 3. miRNA estudados em pacientes com carcinoma de próstata favoráveis e desfavoráveis

miRNA Gene alvo Referência

Let-7c RAS, MYCC, HMGA2, BUB1 Johnson et al. 2005 Johnson et al. 2007

Mayr et al. 2007 Sampson et al. 2007

15a BCL2, MCL1, CCND1, WNT3A Calin et al. 2002 Aqeilan et al. 2009

16 BCL2, MCL1, CCND1, WNT3A Calin et al. 2002 Aqeilan et al. 2009

21 PTEN, Tropomiosina 1, PDCD4, STAT-3, TPM1, MASPIN Cheng et al. 2005 Frankel et al. 2008

Zhu et al. 2008 25 Bim Petrocca et at. 2008

32 Bim Petrocca et at. 2008

100 THAP2, KBTBD8, C4orf16, SMARCA5, BAZ2A, VLDLR Bessière et al. 2008

143 ERK5, KRAS Michael et al. 2003 Akao et al. 2006 Faber et al. 2009

145 MAP3K3, MAP4K4 Michael et al. 2003 Akao et al. 2006

Sempere et al. 2007 146a NFkB, ROCK1 Lin et al. 2008

191 Precursor de VEGF, Proteína associada a metástase MTA1, TIMP2

Volinia et al. 2006 Cheng et al. 2005

199a MET, ERK2, Ciclina L1, Fator de transcrição 8 (NIL-2-A Zinc finger protein) (Regulador negativo de IL2)

Volinia et al. 2006 Kalscheuer et al. 2008

Kim et al. 2005 206 CALM1, CALM2, CITED2, CTNND2, DAAM1, EDN1, EID1,

FGF14, FN1, GAS2L1, GPR125, GPR158, IGF1, IGFBP5, ITGB3, MTSS1, NCOA1, NCOA3, NGFR, PAFAH1B1, PDCD10, PDCD4, PDGFA, RASA1, SDCBP, SOX6, STC2, TAGLN2, TGFBR3, TGIF1, TGIF2, THRB, ESR1

Kondo et al. 2008 Adams et al. 2009

218 Coativador 3 do receptor nuclear, Proteína tirosina quinase FRK, Proteína treonina/serina quinase DCAMKL1, antígeno nuclear induzido por MYC

Volinia et al. 2006

22

III.6. Análise estatística

Como a distribuição dos níveis de expressão dos miRNA apresentou

grande variabilidade os dados foram transformados em logaritmo para

análise. Os resultados foram apresentados através da média geométrica, ou

seja, o anti-log do nível de expressão dos miRNA, e seu respectivo intervalo

de 95% de confiança.

A comparação das expressões de miRNA entre os grupos (CaP

favorável e CaP desfavorável) foi realizada com base nos dados log-

transformados através de uma análise de variância com um fator (ONEWAY-

ANOVA).

Para as análises das características demográficas e clínicas foram

utilizados os testes de Mann-Whitney, Teste T de Sudent e Qui-Quadrado.

Foi adotado um nível de significância foi de 5%. Toda análise

estatística foi realizada no software SPSS 16.0 para Windows com exceção

do cálculo da média geométrica e intervalo de 95% de confiança que foi

realizada no software STATA 8.0 para Windows.

Resultados

24

IV. Resultados

O perfil de expressão dos 14 miRNA dos casos de CaP favorável e

desfavorável estão expostos nas Figuras 2 e 3.

Foram encontradas diferenças significativas na expressão dos miR-

143, miR-145, miR-146a, miR-191, miR-218 e miR-Let7c entre os grupos

(Tabela 4). Nos demais miRNA não foram encontradas diferenças

significativas.

Tabela 4. Média geométrica dos níveis de expressão de 14 miRNA em carcinomas de próstata favoráveis e desfavoráveis

miRNA CaP favoráveis (n=45) CaP desfavoráveis (n=53)

Média 95% IC Média 95% IC p valor

100 40,47 19,90 82,32 16,84 9,39 30,19 0,079

143 9,25 5,10 16,76 2,97 1,60 5,52 <0,001

145 5,40 2,98 9,81 2,65 1,33 5,26 <0,001

146ª 1,06 0,58 1,95 0,55 0,34 0,87 0,001

15ª 0,86 0,46 1,59 0,38 0,21 0,66 0,083

16 0,78 0,41 1,51 0,38 0,22 0,65 0,214

191 1,13 0,59 2,19 0,56 0,36 0,87 0,043

199ª 0,39 0,21 0,70 0,80 0,41 1,58 0,308

206 2,30 1,28 4,13 1,32 0,80 2,20 0,207

21 0,12 0,06 0,25 0,21 0,12 0,34 0,244

218 26,03 12,19 55,62 14,97 8,32 26,95 0,005

25 0,97 0,56 1,66 0,65 0,37 1,14 0,566

32 0,87 0,42 1,81 1,30 0,81 2,11 0,098

Let7c 6,46 4,00 10,45 5,98 3,57 10,02 0,048

25

Figura 2. Perfil de expressão de 14 miRNA em 45 pacientes com tumores favoráveis

26

Figura 3. Perfil de expressão de 14 miRNA em 53 pacientes com tumores desfavoráveis

27

Observamos que o miR-143 foi expresso diferentemente entre os

grupos de favorável e desfavorável onde a média foi de 9,247 para os tumores

favoráveis e 2,970 para os tumores desfavoráveis (p<0,001) (Figura 4).

O miR-143 apresentou-se superexpresso em cerca de 93% das amostras

favoráveis enquanto nas amostras de pacientes com CaP desfavorável

estava superexpresso em cerca de 72% dos casos.

Figura 4. Box plot representando os valores médios e o intervalo de confiança de expressão do miR-143 em pacientes com câncer de próstata

Para o miR-145 também houve uma diferença entre os grupos

(p>0,001) com níveis de expressão de 5,404 no CaP favorável, 2,646 no

CaP desfavorável (Figura 5). O miR-145 esteve superexpresso em cerca de

86% dos tumores favoráveis, enquanto que nos tumores desfavoráveis

encontrava-se super-expresso em 70% dos casos.

28

Figura 5. Box plot representando os valores médios e o intervalo de confiança de expressão do miR-145 em pacientes com câncer de próstata

A análise de miR-146a também foi diferente entre os grupos. A média

dos níveis de expressão foi de 1,061 nos tumores favoráveis e de apenas

0,546 nos tumores desfavoráveis (p=0,001) (Figura 6). Verificamos que o

miR-146a esteve superexpresso em 42,2% dos pacientes com tumores

favoráveis enquanto que nos tumores desfavoráveis esse numero foi

de 35,8%.

29

Figura 6. Box plot representando os valores médios e o intervalo de confiança de expressão do miR-146a em pacientes com câncer de próstata

O miR-191 foi significativamente diferente entre os grupos, sendo os

níveis de expressão respectivamente 1,132 e 0,355 para os tumores

favoráveis de desfavoráveis (p=0,043) (Figura 7).

30

Figura 7. Box plot representando os valores médios e o intervalo de confiança de expressão do miR-191 em pacientes com câncer de próstata

O miR-218 também mostrou-se diferentemente expresso com níveis de

26,037 nos tumores favoráveis e 14,978 nos CaP desfavoráveis (p=0,005)

(Figura 8).

31

Figura 8. Box plot representando os valores médios e o intervalo de confiança da expressão do miR-218 em pacientes com câncer de próstata

Finalmente a avaliação da expressão de miR-let7c mostrou uma

diferença estatística (p=0,048) entre os grupos de com média dos níveis de

expressão de 6,464989 e 5,975 para os tumores favoráveis e desfavoráveis

(Figura 9).

32

Figura 9. Box plot representando os valores médios e o intervalo de confiança da expressão do miR-Let7c em pacientes com câncer de próstata

Discussão

34

V. Discussão

Neste estudo demonstramos perda de expressão de quase todos os

miRNA na transição do casos favoráveis para os desfavoráveis, sendo que

para alguns miRNA a diferença foi estatisticamente significativa.

Identificamos uma subexpressão dos miRNA 199a, 21, 15a, 16 e 25 já

nos tumores favoráveis e uma redução na expressão dos miRNA 143, 145,

146a, 191, 218 e Let-7c na transição carcinoma favorável para desfavorável.

Este achado é consistente com os dados da literatura onde a perda

de expressão de miRNA é uma característica de uma série de tumores

humanos, podendo assim ser considerados preferencialmente como

supressores de tumor (Lu et al 2005).

Os miR-191, miR-199a, miR-15 e miR-16 encontravam-se

subexpressos em quase todos casos estudados. Apesar de termos

encontrado diferença estatística nos níveis de expressão apenas do

primeiro, as demais constatações também são interessantes do ponto de

vista biológico. Os primeiros estudos sobre a anormalidade de expressão de

miRNA em neoplasias foram descritos por Calin et al. em 2002, onde foi

demonstrado a subexpressão de miR-15a e miR-16 relacionada a perda da

região 13q14 em 65% de LLC, em 50% dos linfomas de células do manto e

em até 40% dos casos de mieloma múltiplo. Na próstata, esta anormalidade

foi descrita em 60% dos casos (Esquela-Kerscher et al. 2006) relacionada ao

pior prognóstico e morte pela doença (Krajewska et al. 2008). Ambos miRNA

tem como gene alvo o BCL2 que tem ação anti-apoptótica e está

superexpresso em uma série de neoplasias principalmente leucemias e

linfomas.

Mutações de BCL2 tem sido descritas apenas tardiamente nos casos

de CaP, assim acreditamos que a subexpressão desses miRNA seja

importante na patogênese do CaP permitindo a ação desta proteína anti-

apoptótica. Outros alvos do miR-15a e miR-16 são os genes CCND1 e

WNT3A, estes genes causam um aumento da proliferação celular através do

aumento da fosforilação do Rb (via CCND1) e da expressão de β−catenina

35

(via WNT3) (Bonci et al. 2008). Os miRNA15a e 16 além de não deixarem a

célula entrar em apoptose, podem atuar também na proliferação e invasão

celular.

MiR-191 está localizado no cromossomo humano 3p21.31, e é gerado

a partir de uma unidade intrônica transcricional. Existem trabalhos citando a

sua superexpressão no câncer de pâncreas (Roldo et al. 2006), cólon,

pulmão e próstata (Volinia et at. 2006) e LMA com cariótipo anormal (Garzon

et al 2008). Outros relatam sub-expressão em pacientes com LLC (Calin et

al. 2004) e melanoma (Caramuta et al 2010). Este miRNA parece agir de

forma diferente, dependendo do tipo celular. Estudos experimentais

demonstram que a inibição de miR-191 causa um aumento da proliferação

em cultura de células de câncer cervical HeLa, mas diminui a proliferação

em linhagens de câncer de pulmão A549 (Cheng et al 2005). Este miRNA

também pode agir em diferentes alvos (mRNA). Dependendo do estágio da

doença ele pode estar atuando como supressor de tumor (subexpressão) ou

como oncogênico (superexpresso).

Estudos em tumores sólidos demonstraram uma subexpressão dos

miR-143 e miR-145 no adenocarcinoma colo-retal (Volinia et al. 2006). Um

estudo demonstrou a subexpressão de miR-143 na transformação do epitélio

cólico normal para adenoma e foi atribuído à falta deste miRNA uma

superatividade de K-RAS nessa fase da carcinogênese (Faber et al. 2009).

Clapé et al. (2009) encontraram a sub-expressão do miR-143 em tumores de

próstata favorável em relação ao escore Gleason e demonstraram que além

do K-RAS o gene ERK-5 pode ser alvo do miR-143. O ERK-5 atua na

regulação da proliferação celular e regula vários genes entre eles: MEF2, c-

Fos e Fra1, Sap-1, c-Myc e NF-kappaB que têm papel importante durante o

processo de carcinogênese (Zhou et al. 1995).

Em carcinomas de mama, estudos demonstram uma progressiva

regulação negativa de miR-145 desde a mama normal até o câncer com alta

atividade proliferativa (Sempere et al. 2007). Estudos em CaP mostram que

ele encontra-se subexpresso, sendo um de seus possíveis alvo o gene Myc

(Wang et al. 2009). O gene Myc é um fator de transcrição que ativa a

36

expressão de um grande número de genes por meio da ligação de

seqüências consenso e recrutamento de histonas acetil-transferases. Ele

também pode atuar como repressor da transcrição, ao ligar-se ao fator de

transcrição Miz-1 e deslocando o p300 co-ativador, o que causaria a inibição

da expressão dos genes alvos de Miz-1. Além disso, Myc tem um papel

direto no controle da replicação do DNA (Dominguez-Sola et al 2007). O Myc

é ativado mediante diversos sinais mitogênicos como a Wnt, Shh e EGF

(através da via ERK/MAPK). Ao modificar a expressão de seus genes-alvo,

os resultados de ativação do Myc têm inúmeros efeitos biológicos. O

primeiro a ser descoberto foi a sua capacidade de induzir a proliferação

celular atuando sobre as ciclinas, mas também desempenhando um papel

muito importante na regulação do crescimento celular através do controle da

apoptose, diferenciação e auto-renovação de células-tronco. Myc é proto-

oncogene forte, e tem se mostrado superregulado em muitos tipos de

cânceres (Nathalie et al 1991). Estes dados da literatura reforçam a nossa

hipótese do papel dos miR-143 e 145 na transformação dos tumores

favoráveis para desfavoráveis atuando possivelmente sobre o Myc.

Recentemente, demonstramos que em carcinoma de próstata

metastático e linhagens de câncer avançado existe uma subexpressão dos

miR-143 e miR-145a, o que nos leva a acreditar que o seu papel esteja

relacionado não somente com a agressividade mas também com o processo

de disseminação sistêmica dessa neoplasia (Leite et al. 2011).

A mudança do perfil de expressão do miR-146a entre tumores

favoráveis e desfavoráveis, pode estar relacionada ao aumento da

expressão do gene ROCK1. ROCK1 é um dos genes alvo de miR-146a, que

está envolvido no remodelamento da matriz extracelular, na transformação

dos tumores de próstata em hormônios refratários e no processo de

metástase (Lin et al. 2008). ROCK1 pode sinalizar de 3 modos no CaP: a)

fosforilando MLC (myosin light chain) o que promoveria um aumento da

migração e invasão celular; b) ativando a via PI3K-Akt/TOR/eIF4E que

aumenta proliferação da célula cancerosa e diminui a apoptose; c)

aumentando M-CSF para a produção de citocinas que facilitariam o

37

processo de metastatização óssea, característica do CaP (Lin et al. 2007).

Assim, a subexpressão do miR-146a seria responsável por um aumento de

expressão de ROCK1 ocasionando um aumento da proliferação celular,

diminuição da apoptose, promovendo a invasão, e o surgimento de

metástases, desempenhando um papel importante na progressão desta

neoplasia.

Estudos mostraram a subexpressão de miR-218 em linhagens de

carcinoma de cérvix uterina, relacionada principalmente a infecção pelo

papiloma vírus humano (HPV) de alto risco tipo 16 (Martinez et al. 2008). Um

dos alvos do miR-218 é a Laminina 5 β3 (LAMB3), que tem ação sobre a

migração e tumorigênese em estudos experimentais em camundongos. Em

estudos, in vitro demonstrou-se que o miR-218 colabora com o seu ligante,

pró-laminina promovendo tumorigênese em queratinócitos humanos

(Martinez et al. 2008). A diminuição da expressão do miR-218 dos tumores

desfavoráveis em relação aos tumores favoráveis, provavelmente esteja

relacionada a facilitação da migração de células neoplásicas nos tumores

desfavoráveis. Isto nos faz supor que este miRNA tenha um papel

importante durante o processo de progressão do CaP, aumentando a

migração celular e sendo um dos responsáveis pelo processo de

metastização.

O miR-let7c tem como alvo os genes RAS e c-MYC (Johnson et al.

2005). É importante ressaltar que o ganho do cromossomo 8q, onde o c-

MYC está localizado é frequente no CaP avançado (Nupponen et al. 1998).

Como previamente discutido este fator de transcrição está envolvido no

controle do crescimento, diferenciação e apoptose celular e está amplificado

em uma parte dos tumores avançados, particularmente metastáticos

(Buttyan et al. 2005). RAS é outro proto-oncogene envolvido na patogênese

de vários tumores humanos, mas diferente de c-MYC, suas anormalidades

são raras no CaP (Gumerlock et al. 1991). Cada miRNA tem centenas de

genes alvos e entre os preditos para miR-let7c está o BUB1. Este gene tem

papel específico no “checkpoint” mitótico e suas anormalidades podem

causar instabilidade cromossômica e aneuploidia em fibroblastos de

38

camundongo e linhagens celulares de carcinomas humanos. A

subexpressão de BUB1 tem sido descrita em carcinomas coloretais

aneuplóides (Grady 2004) e tumor de Wilms (Haruta et al. 2008).

Aneuploidia é um fator prognóstico importante no CaP e está relacionada à

progressão da neoplasia independente do escore de Gleason (Lerner et al.

1996). Apesar dos valores de expressão do miR-let7c serem muito próximos

nos casos favoráveis em relação aos desfavoráveis, podemos supor que a

maior expressão do miR-Let-7-c nos casos favoráveis provavelmente esteja

relacionado ao gene BUB1, ocasionando aneuploidia e instabilidade

cromossômica. E a sua menor expressão nos tumores desfavoráveis esteja

relacionada ao crescimento e proliferação celular

Em relação aos demais miRNA estudados notamos que o miR-100

manteve-se superexpresso entre os tumores de baixo e alto grau. A

superexpressão do miR-100 têm sido descrita em tumores de ovário (Dahiya

et al. 2008) e tem como possíveis alvos os genes THAP2, SMARCA5 e

BAZ2A. THAP2 é membro de uma família de proteínas recentemente

descrita representada por RB/E2F que regula a proliferação celular,

modulando genes com importante ação no controle celular (Bessière et al.

2008). SMARCA5, também conhecida como SNF2h, é um membro da

família ISWI envolvida no remodelamento da cromatina, com papel na

replicação do DNA, facilitando a sua síntese (Bozhenok et al. 2002). Estudos

mostraram que a superexpressão de BAZ2A, também conhecida como TTF1

interacting protein 5 tem um efeito repressivo sobre a transcrição do DNA

(Zhou et al. 2002). Podemos especular que a superexpressão de miR-100

tem papel no processo de carcinogênese e progressão do carcinoma da

próstata influenciando a proliferação celular e transcrição do DNA, e sua

menor expressão pode representar um marcador de doença de melhor

prognóstico.

O nível de miR-32 mostrou-se crescente em relação a progressão do

tumores favoráveis para os desfavoráveis. Estudos demonstraram que o

miR-32 estava super-expresso em tumores de cólon e do pâncreas (Volinia

et al. 2006). Em próstata foi encontrada a superexpressão do miR-32, tendo

39

como alvo o Bim (Ambs et al. 2008). Bim é membro pró-apoptoptico da

família do BCL2. Estudos mostraram a superexpressão do miR-25 em

tumores gástricos, e esta superexpressão está relacionada a menor

expressão do Bim (Petrocca et al. 2008). Esses achados confirmam os

nossos resultados e nos fazem supor que a superexpressão desses miRNA

seja importante durante a transição entre os tumores favoráveis e

desfavoráveis.

Em contraste com a literatura, o miR-21 mostrou-se subexpresso na

maioria dos tumores da próstata, independente de seu grau de malignidade.

O miR-21 aparece superexpresso em uma série de tumores sólidos, como o

carcinoma de mama, o glioblastoma multiforme, o câncer de pâncreas e da

próstata (Volinia et al. 2006). Age deprimindo a tradução de PTEN, que tem

ação negativa sobre a sobrevivência e crescimento celular promovidos pela

via PI-3 quinase-Akt (Meng et al. 2006). O gene PTEN codifica uma proteína

que desfosforila o fosfatidilinositol 3,4,5-trifosfato (PIP3), que por sua vez

antagoniza a atividade de fosfatidilinositol 3-quinase (PI3K). A perda de

função de PTEN em modelos animais promove o desenvolvimento de PIN

com progressão ao carcinoma invasivo da próstata (Ma et al. 2005).

Estudos in vitro demonstraram que na linhagem HeLa a inibição do

miR-21 promove um aumento importante na proliferação celular (Cheng et

al. 2005). Talvez, também no CaP, a subexpressão de miR-21 esteja

envolvida com outras vias de regulação, que não o PTEN, e esta

subexpressão promova aumento de proliferação. Essa hipótese deve ser

comprovada experimentalmente em estudos futuros.

Conclusão

41

VI. Conclusão

Identificamos uma mudança no perfil de expressão de miRNA

comparando adenocarcinomas favoráveis e desfavoráveis de próstata,

sendo as mais significativas a diminuição na expressão de miR-145, 143,

146-a, 218, 191 e Let7c.

Perspectivas

43

VII. Perpectivas

A análise da expressão das proteínas reguladas pelos miRNA

estudados podem ajudar num futuro próximo a desvendar alguns dos

mecanismos relacionados à progressão do CaP podendo criar novos grupos

prognósticos e de tratamento.

Referências

45

VIII. Referências

Adams BD, Cowee DM, White BA. The role of miR-206 in the epidermal

growth factor (EGF) induced repression of estrogen receptor-alpha

(ER{alpha}) signaling and a luminal phenotype in MCF-7 breast cancer cells.

Mol Endocrinol. 2009 May 7. doi:10.1210/me.2009-0062.

Akao Y, Nakagawa Y, Naoe T. let-7 MicroRNA functions as a potential

growth suppressor in human colon cancer cells. Biological & Pharmaceutical

Bulletin 2006; 29(5903):903-906.

Ambs S, Prueitt RL, Yi M. Genomic profiling of miRNA and messenger RNA

reveals deregulated microRNA, expression in prostate cancer. Cancer Res.

2008; 68:6162-70.

Aqeilan RI, Calin GA, Croce CM. miR-15a and miR-16-1 in cancer:

discovery, function and future perspectives. Cell Death Differ. 2009; doi:

10.1038/cdd.2009.69.

Benowitz S. Revised guidelines signal that gene expression profiles are

coming of age. J Natl Cancer Inst 2008; 100:916 –7.

Bessière D, Lacroix C, Campagne S, Ecochard V, Guillet V, Mourey L, Lopez

F, Czaplicki J, Demange P, Milon A, Girard JP, Gervais V. Structure-function

analysis of the THAP zinc finger of THAP1, a large C2CH DNA-binding

module linked to Rb/E2F pathways. J Biol Chem. 2008; 283:4352-63.

Best CJ, Leiva IM, Chuaqui RF et al. Molecular differentiation of high and

moderate grade human prostate cancer by cDNA microarray analysis. Diag

Mol Pathol 2003; 12:63-70.

Blower PE, Chung JH, Verducci JS, et al. MicroRNAs modulate the

chemosensitivity of tumor cells. Mol Cancer Ther 2008;7:1–9.

46

Bonci D, Coppola V, Musumeci M, Addario A, Giuffrida R, Memeo L, D'Urso

L, Pagliuca A, Biffoni M, Labbaye C, Bartucci M, Muto G, Peschle C, De

Maria R The miR-15a, miR-16-1 cluster controls prostate cancer by targeting

multiple oncogenic activities. Nat Med. 2008;14:1271-7.

Bozhenok L, Wade PA, Varga-Weisz P. STF-ISWI chromatin remodeling

complex targets heterochromatic replication foci. EMBO J. 2002; 21:2231-41.

Buttyan R, Sawczuk IS, Benson MC, Siegal JD, Olsson CA. Enhanced

expression of the c-myc protooncogene in high-grade, human prostate

cancers. Prostate. 1987; 11:327-37.

Calin GA, Dumitru CD, Shimizu M, Bichi R, Zupo S, Noch E, Aldler H, Rattan

S, Keating M, Rai K, Rassenti L, Kipps T, Negrini M, Bullrich F, Croce CM.

Frequent deletions and down-regulation of micro- RNA genes miR15 and

miR16 at 13q14 in chronic lymphocytic leukemia. PNAS 2002; 99(24):15524-9.

Calin GA, Liu CG, Sevignani C, Ferracin M, Felli N, et al. MicroRNA profiling

reveals distinct signatures in B cell chronic lymphocytic leukemias. Proc Natl

Acad Sci USA 2004, 101: 11755–11760.

Caramuta S, Egyha´zi S, Rodolfo M, Witten D, Hansson J, Larsson C, Onn Lui

W. MicroRNA expression profiles associated with mutational status and survival

in malignant melanoma, Journal of Investigative Dermatology 2010, 130.

Chaib H, Cockrell EK, Rubin MA et al. Profiling and verification of gene

expression patterns in normal and malignant human prostate tissues by

cDNA microarray analysis. Neoplasia 2001; 3:43-52

Cheng AM, Byrom MW, Shelton J, Ford LP. Antisense inhibition of human

miRNAs and indications for an involvement of miRNA in cell growth and

apoptosis. Nucl Acids Res. 2005; 33:1290-7.

47

Chetcuti A, Margan S, Mann S et al. Identification of differentially expressed

genes in organ-condined prostate cancer by gene expression array. Prostate

2001; 47:132-40.

Clapé C, Fritz V, Henriquet C, Apparailly F, Fernandez PL, Iborra F, Avancès

C, Villalba M, Culine S, Fajas L. miR-143 interferes with ERK5 signaling, and

abrogates prostate cancer progression in mice, PLoS One 2009; 4(10):

e7542.

Dahiya N, Sherman-Baust CA, Wang TL, Davidson B, Shih IM, Zhang Y,

Wood W 3rd, Becker KG, Morin PJ. MicroRNA expression and identification

of putative miRNA targets in ovarian cancer. PLoS One. 2008; 3:2436.

deVere White RW, Vinall RL, Tepper CG, Shi ZB. MicroRNAs and their

potential for translation in prostate cancer. Urol Oncol Sem Orig Invest. 2009;

27:307-11.

Dhanasekaran SM, Barrette TR, Ghosh D, Shah R, Varambally S, Kurachi K,

Pienta KJ, Rubin MA, Chinnaiyan AM. Delineation of prognostic biomarkers

in prostate cancer. Nature 2001; 412:822-6.

Dominguez-Sola D, Ying CY, Grandori C, Ruggiero L, Chen B, Li M,

Galloway DA, Gu W, Gautier J, Dalla-Favera R. Non-transcriptional control of

DNA replication by c-Myc. Nature 2007; 448 (7152): 445–51.

Epstein JI, Allsbrook Jr WC; Amin MB, Egevad LL. The 2005 International

Society of Urological Pathology (ISUP) consensus conference on Gleason

grading of prostatic carcinoma. Am J Surg Pathol 2005; 29:1228-42.

Ernst T, Hengenhahn M, Kenzelmann M et al. Decrease and gain of gene

expression are equally discriminatory markers for prostate carcinoma: a gene

48

expression analysis on total and microdissecte prostatic tissue. Am J Pathol

2002; 160:2169-80.

Esquela-Kerscher A, Slack FJ. Oncomirs – microRNAs with role in cancer.

Nat Rev Cancer. 2006; 6:259-69.

Faber C, Kirchner T, Hlubek F. The impact of microRNAs on colorectal

cancer. Virchows Arch. 2009; 454:359–67.

Frankel LB, Christoffersen NR, Jacobsen A, Lindow M, Krofh A, Lund AH.

Programmed cell death 4 (PDCD4) is an important functional target of the

microRNA miR-21 in breast cancer cells. J Biochem. 2008; 283:1026-33.

Garzon R, Volinia S, Liu CG et al. MicroRNA signatures associated with

cytogenetics and prognosis in acute myeloid leukemia. Blood 2008,

111:3183–9.

Grady WM. Genomic instability and colon cancer. Cancer Metastasis Rev.

2004; 23:11-27.

Graefen M, Karakiewicz PI, Cagiannos I, Quinn DI, Henshall SM, Grygiel JJ,

Sutherland RL, Stricker PD, Klein E, Kupelian P, Skinner DG, Lieskovsky G,

Bochner B, Huland H, Hammerer PG, Haese A, Erbersdobler A, Eastham JA,

Kernion J, Cangiano T, Schröder FH, Wildhagen MF, van der Kwast TH,

Scardino PT, Kattan MW. International validation of a preoperative

nomogram for prostate cancer recurrence after radical prostatectomy. J Clin

Oncol 2002; 20:3206-12.

Gumerlock PH, Poonamallee UR, Meyers FJ, deVere White RW. Activated

ras alleles in human carcinoma of the prostate are rare. Cancer Res 1991;

51:1632-7.

49

Haruta M, Matsumoto Y, Izumi H, Watanabe N, Fukuzawa M, Matsuura S,

Kaneko Y. Combined BubR1 protein down-regulation and RASSF1A

hypermethylation in Wilms tumors with diverse cytogenetic changes. Mol

Carcinog. 2008; 47:660-6.

Henshall SM, Afar DEH, Hiller J, Horvath LG, Quinn DI, Rasiah KK, Gish K,

Willhite D, Kench JG, Gardiner-Garden M, Stricker PD, Scher HI, Grygiel JJ,

Agus DB, Mack DH, Sutherland RL. Survival analysis of genoma-wide gene

expression profiles of prostate cancer identifies new prognostic targets of

disease relapse. Cancer Res 2003; 63:4196-203.

Humphrey PA, Vollmer RT: Intraglandular tumor extent and prognosis in

prostatic carcinoma: aplication of a grid method to prostatectomy specimens.

Hum Pathol 1990; 21:799-804.

Jemal A, Bray F, Center MM, Ferlay J, Ward El, Forman D. Global cancer

statistics. CA: A Cancer Journal for Clinicians 2011; 61 (2):69–90.

Johnson SM, Grosshans H, Shingara J, Byrom M, Jarvis R, Cheng A,

Labourier E, Reinert K, Brown D, Slack F. Is regulated by the MicroRNA

family. Cell 2005; 120(5):635–647.

Johnson CD, Esquela-Kerscher A, Stefani G, Byrom M, Kelnar K,

Ovcharenko D, Wilson M, Wand X, Shelton J, Shingara J, Chin L, Brown D,

Slack FJ. The Let-7: microRNA represses cell proliferation pathways in

human cells. Cancer Res. 2007; 67:7713-22.

Kalscheuer S, Zhang X, Zeng Y, Upadhyaya P. Differential expression of

microRNAs in early-stage neoplastic transformation in the lungs of F344 rats

chronically treated with the tobacco carcinogen 4-(methylnitrosamino)-1-(3-

pyridyl)-1-butanone. Carcinogenesis. 2008;29:2394-99.

50

Kim VN. MicroRNA biogenesis: Coordinated cropping and dicing. Nature

Reviews Mol Cel Biol. 2005; 6:376-85.

Kondo N, Toyama T, Sugiura H, Fujii Y, Yamashita H. miR-206 Expression is

down-regulated in estrogen receptor alpha-positive human breast cancer.

Cancer Res. 2008;68:5004-8.

Krajewska M, Kitada S, Winter JN, Variakojis D, Lichtenstein A, Zhai D,

Cuddy M, Huang X, Luciano F, Baker CH, Kim H, Shin E, Kennedy S, Olson

AH, Badzio A, Jassem J, Meinhold I. Bcl-B expression in human epithelial

and nonepithelial malignancies. Clin Ca Res. 2008; 14:3011-21.

La Tulippe E, Satagopan J, Smith A et al. Comprehensive gene expression

analysis of prostate cancer reveals distinct transcriptional programs

associated with metastatic disease. Cancer Res 2002; 62:4499-506.

Leite KRM, Srougi M, Darini E, Carvalho CM, Camara-Lopes LH.

Telomerase activity in localized prostate cancer: correlation with histological

parameters. Int Braz J Urol 2001; 27:341-7.

Leite KRM, Franco MF, Srougi M, Nesrallah LJ, Nesrallah A, Bevilacqua RG,

Darini E, Carvalho CM, Meirelles MI, Santana I, Camara-Lopes LH.

Abnormal expression of MDM2 in prostate carcinoma. Mod Pathol 2001;

14:428-36.

Leite KRM, Srougi M, Kauffmann JR, Bevilacqua RG, Nesrallah AJ,

Nesrallah LJ, Camara-Lopes LH. Well differentiated localized prostate

carcinoma: prognostic relevance of tertiary Gleason pattern 4 and tumor

volume. Rev Assoc Med Bras 2005; 51(6):329-33.

Leite KRM, Sousa-Canavez JM, Reis ST, Tomiyama AH, Camara-Lopes LH,

Sañudo A, Antunes AA, Srougi M, Change in expression of miR-let7c, miR-

51

100, and miR-218 from high grade localized prostate cancer to metastasis.

Urol Oncol 2011;29:265-9.

Lerner SE, Blute ML, Bergstralh EJ, Bostwick DG, Eickholt JT, Zincke H.

Analysis of risk factors for progression in patients with pathologically confined

prostate cancers after radical retropubic prostatectomy. J Urol. 1996;

156:137-43.

Lin SL, Chang D, Ying SY. Hyaluronan stimulates transformation of

androgen-independent prostate cancer, Carcinogenesis 2007; 28:310–320.

Lin SL, Chiang A, Chang D, Ying SY. Loss of mir-146a function in hormone-

refractory prostate cancer. RNA. 2008; 14:417-24.

Luo J, Dunn T, Ewing C et al. Gene expression signature of benign prostatic

hyperplasia revealed by cDNA microarray analysis. Prostate 2002; 51:189-200.

Luo JH, Yu YP, Cieply K et al. Gene expression analysis of prostate cancers.

Mol Carcinog 2002; 33:25-35.

Ma L, Teruya-Felstein J, Behrendt N, Chen Z, Noda T, Hino O, Cordon-

Cardo C, Pandolgi PP. Genetic analysis of Pten and Tsc2 functional

interations in the mouse reveals asymmetrical haploinsufficiency in tumor

suppression. Genes and Development. 2005; 19:1779-86.

Magee JA, Araki T, Patil S, Ehrig T, True L, Humphrey PA, Catalona WJ,

Watson MA, Milbrandt J. Expression profiling reveals hepsin overexpression

in prostate cancer. Cancer Res 2001; 61:5692-6.

Martinez I, Gardiner AS, Board KF, Monzon FA, Edwards RP, Khan AS.

Human papillomavirus type 16 reduces the expression of microRNA-218 in

cervical carcinoma cells. Oncogene. 2008; 27:2575-82.

52

Mayr C, Hermann MT, Bartel DP. Disrupting the pairing between let-7 and

HMGA2 enhances oncogenic transformation. Science. 2007;315:1576-79.

Michael MZ, O’Connor SM, van Holst Pellekaan NG, YoungGP, James RJ.

Reduced accumulation of specific microRNAs in colorectal neoplasia. Mol Ca

Res. 2003; 1:882–91.

Mishra PJ, Banerjee D, et al. MiRSNPs or MiR-polymorphisms, new players

in microRNA mediated regulation of the cell: Introducing microRNA

pharmacogenomics. Cell Cycle 2008; 7:853–8.

Mitchell PS, Parkin RK, Kroh EM, et al. Circulating microRNAs as stable

blood-based markers for cancer detection. Proc Natl Acad Sci

USA 2008; 105:10513– 8.

Nathalie D, Kitzis A, Kruh J, Dautry F, Corcos D. Stimulation of methotrexate

resistance and dihydrofolate reductase gene amplification by c-myc.

Oncogene 1991, 6 (8): 1453–7.

Navone NM, Troncoso P, Pisters LL, Goodrow TL, Palmer JL, Nichols WW, von

Eschenbach AC, Conti CJ. p53 accumulation and gene mutation in the

progression of human prostate carcinoma. J Natl Cancer Inst 1993; 85:1657-69.

Nelson PS. Predicting prostate cancer behavior using transcript profiles. J

Urol 2004; 172:S28-S33.

Nicoloso MS, Calin GA. MicroRNA involvement in brain tumors: from bench

to bedside. Brain Pathol. 2008;18:122-9.

Nupponen NN, Kakkola L, Koivisto P, Visakorpi T. Genetic alterations in

hormone-refractary recurrent prostate carcinomas. Am J Pathol. 1998;

153:141-48.

53

O’Donnell KA, Wentzel EA, Zeller KI, Dang CV, Mendell JT c-Myc regulated

microRNAs modulate E2F1 expression. Nature. 2005;435:839-43.

Petrocca F, Visone R, Onelli MR, Shah MH, Nicoloso MS, de Martino I,

Iliopoulos D, Pilozzi E, Liu C-G, Negrini M, Cavazzini L, Volinia S, Alder H,

Ruco LP, Baldassarre G, Croce CM, Vecchione A. E2F1-regulated

microRNAs impair TGFbeta-dependent cell-Cycle arrest and apoptosis in

gastric cancer. Cancer Cell 2008; 13:272–286.

Porkka KP, Pfeiffer MJ, Waltering KK, Vessella RL, Tammela TL, Visakorpi

T. MiRNA expression profiling in prostate cancer. Cancer Res. 2007;

67:6130-5.

Ramaswamy S, Ross KN, Lander ES, Golub TR. A molecular signature of

metastasis in primary solid tumors. Nat Genet 2003; 33:49-54.

Roldo C, Missiaglia E, Hagan JP, Falconi M, Capelli P, et al. MicroRNA

expression abnormalities in pancreatic endocrine and acinar tumors are

associated with distinctive pathologic features and clinical behavior. J Clin

Oncol 2006, 24: 4677–4684.

Sampson VB, Rong NH, Han J, Yang Q, ARis V, Soteropoulos P, Petrelli NJ,

Dunn SP, Krueger LJ. MicroRNA let-7c downregulates MYC and reverts MYC-

induced growth in Burkitt lymphoma cells. Cancer Res. 2007;67:9762-70.

Schetter AJ, Leung SY, Sohn JJ, et al. MicroRNA expression profiles

associated with prognosis and therapeutic outcome in colon

adenocarcinoma. JAMA 2008; 299:425–36.

Sempere LF, Christensen M, Silahtaroglu A, Bak M, Heath CV, Schwartz G,

Wells W, Kauppinen S, Cole CN. Altered MicroRNA expression confined to

54

specific epithelial cell subpopulations in breast cancer. Cancer Res 2007;

67:11612-20.

Shah O, Melamed J, Lepor H. Analysis ofapical soft tissue margins during

radical retropubic prostatectomy. J Urol 2001;165(6 Pt 1):1943-9.

Shi X-B, Xue L, Yang J, Ma A-H, Zhao J, Xu M, Tepper CG, Evans CP,

Kung H-J, deVere White RW. An androgen-regulated miRNA suppresses

Bak1 expression and induces androgen-independent growth of prostate

cancer cells. PNAS. 2007; 104:19983-8.

Singh D, Febbo P, Ross K, Jackson D, Manola J, Ladd C, Tamayo P,

Renshaw A, D’Amico A, Richie J. Gene expression correlates of clinical

prostate cancer behavior. Cancer cell 2002; 1:203-9.

Stamey TA, McNeal JE, Yemoto CM; Sigal BM, Johnstone IM. Biological

determinants of cancer progression in men with prostate cancer. JAMA 1999;

281:1395-400.

Stamey TA, Warrington JÁ, Caldwell MC et al. Molecular genetic profiling of

Gleason grade 4/5 prostate cancer compared to benign prostatic hyperplasia.

J Urol 2001; 166:2171-7.

Stephenson AJ, Smith A, Kattan MW, Satagopan J, Reuter VE, Scardino PT,

Gerald WL. Integration of gene expression profiling and clinical variables to

predict prostate carcinoma recurrence after radical prostatectomy. Cancer

2005; 104:290-8.

Sun T, Wang Q, Balk S, Brown M, Lee GS, Kantoff P. The role of microRNA-

221 and microRNA-222 in androgen-independent prostate cancer cell lines.

Cancer Res. 2009;69:3356-63

Takamizawa J, Konish H, Yanagisawa K, Tomida S, Osada H, Endoh H,

Harano T, Yatabe Y, Nagino M, Nimura Y, Mitsudomi T, Takahashi T.

55

Reduced expression of the let-7 microRNAs in human lung cancers in

association with shortened postoperative survival. Cancer Res.

2004;64:3753-6.

Taylor DD, Gercel-Taylor C. MicroRNA signatures of tumor-derived

exosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol 2008;

110:13–21.

Vanaja DK, Cheville JC, Iturria SJ et al. Transcriptional silencing of zinc

finger protein 185 identified by expression profiling is associated with

prostate cancer progression. Cancer Res 2003; 63:3877-82.

Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, Iorio

M, Roldo C, Ferracin RL, Prueittt RL, Yanaihara N, Lanza G, Scarpa A,

Vecchione A, Negrini M, Harris CC, Croce CM. A miRNA expression

signature of human solid tumors defines cancer gene targets. Proc Natl Acad

Sci USA. 2006; 103:2257-61.

Yoon S, De Micheli G. Prediction of regulatory modules comprising

microRNAs and target genes. Bioinformatics. 2005; 21(Suppl 2):ii93-100.

Wang L, Tang H, Thayanithy V, Subramanian S, Oberg AL, Cunningham JM,

Cerhan JR, Steer CJ, Thibodeau SN. Gene networks and microRNAs

implicated in aggressive prostate cancer, Cancer Research 2009, 69:9490.

Welsh JB, Sapinoso LM, Su AI et al. Analysis of gene expression identifies

candidate markers and pharmacological targets in prostate cancer. Cancer

Res 2001; 61:5974-8.

Weiss GJ, Bemis LT, Nakajima E, et al. EGFR regulation by microRNA in

lung cancer: Correlation with clinical response and survival to gefitinib and

EGFR expression in cell lines. Ann Oncol 2008; 19:1053–9.

56

Winter J, Jung S, Keller S, Gregory RI, Diederichs S. Many roads to maturity:

microRNA biogenesis pathways and their regulation. Nature Cell Biology

2009; 11:228 – 234.

Zhao JJ, Lin J, Yang H, Kong W, He L, Ma X, Coppola D, Cheng JO.

MicroRNA-221/222 negatively regulates ERand associates with tamoxifen

resistance in breast cancer, JBiol Chem. 2008.

Zhou G, Bao ZQ, Dixon JE. Components of a new human protein kinase

signal transduction pathway. J Biol Chem. 1995; 270:12665–12669.

Zhou Y, Santoro R, Grummt I. The chromatin remodeling complex NoRC

targets HDAC1 to the ribosomal gene promoter and represses RNA

polymerase I transcription. EMBO J. 2002; 21:4632-40.

Zhu S, Wu H, Wu F, Nie D, Sheng S, Mo YY. MicroRNA-21 targets tumor

suppressor genes in invasion and metastasis. Cell Res. 2008; 18:350-9.

Apêndices

Aprovação conselho de Ética – Faculdade de Medicina da USP

Tabela de resultados obtidos

Publicações

1- Leite KRM, Souza-Canavez JM, Reis ST, Tomiyama AH, Camara-Lopes

LH, Sãnudo A, Antunes AA, Srougi M. Change in expression of miR-let7c,

miR-100, and miR-218 from high grade localized prostate cancer to

metastasis. Urologic Oncology, v. x, p. x-x, 2009.

2- Leite KRM, Souza-Canavez JM, Reis ST, Tomiyama AH, Piantino CB,

Sãnudo A, Camara-Lopes LH, Srougi M. miRNA analysis of prostate cancer

by quantitative real time PCR: Comparison between formalin-fixed paraffin

embedded and freshfrozen tissue. Urologic Oncology, v. x, p. xxx-xxx, 2009.

3 - Leite KRM, Tomiyama AH, Reis ST, Sousa-Canavez JM, Sañudo A,

Dall’Oglio MF, Camara-Lopes LH, Srougi M. MicroRNA-100 Expression is

Independently Related to Biochemical Recurrence of Prostate Cancer, The

Journal of Urology 2011, 185:1118-1122.

4 - Leite KRM, Tomiyama A, Reis ST, Sousa-Canavez JM, Sañudo A,

Camara-Lopes LH, Srougi M. MicroRNA expression profiles in the

progression of prostate cancer – From high-grade prostate intraepithelial

neoplasia to metastasis. Urologic Oncology 2011;

doi:10.1016/j.urolonc.2011.07.002.

Participações em congressos

1- Tomiyama AH ; Souza-Canavez JM ; Reis ST ; Sãnudo A ; Camara-Lopes

LH ; Srougi M ; Leite KRM . MicroRNA em adenocarcinoma de próstata:

caracterização da expressão em tumores de alto e baixo grau. 2009, Búzios RJ.

2- Tomiyama AH ; Souza-Canavez JM ; Reis ST ; Sãnudo A ; Camara-Lopes

LH; Srougi M ; Leite KRM . MicroRNA em adenocarcinoma de

próstata:caracterização da expressão em pacientes com e sem recidiva.

2009, Búzios RJ.

3- Leite KRM ; Souza-Canavez JM ; Reis ST ; Tomiyama AH; Piantino CB;

Camara-Lopes LH ; Srougi M . Comparação entre tecido fixado em formalina

embebido em parafina e congelado. 2009, Búzios RJ.

Tabela contendo os valores de expressão (2-∆∆CT) de 14 miRNA em 53 pacientes com CaP com prognóstico desfavorável:

Código miR-100 miR-143 miR-145 miR-146a

miR-15a miR-16

miR-191

miR-199a miR-206

miR-21 miR-218 miR-25 miR-32

miR-Let7c

31 0,2177 1,7192 2,1255 2,3946 3,6944 3,8310 4,6255 2,7408 0,1617 0,5834 270,8781 1,6748 0,0908 4,1690

55 36,0340 36,7256 6,3021 0,4979 1,5925 0,5111 0,4511 1,5917 2,1223 0,8684 28,4923 1,2100 3,5607 6,1081

76 7,8206 29,5834 1,6131 4,3044 0,5603 3,0329 1,5174 0,1947 1,9354 0,6371 70,0105 0,0674 1,1120 1,3507

83 16,0365 15,1970 0,3403 1,0452 0,3822 0,4920 0,4995 0,2009 0,1935 0,3431 36,8221 0,2630 3,4370 1,3535

87 148,9089 2,9439 29,5039 0,2964 3,8009 1,1509 0,9704 8,7641 2,5256 1,1635 65,4581 3,1777 4,2023 13,0295

118 180,1788 47,4953 2,1109 9,2394 1,8691 1,6174 2,1267 0,9990 0,6019 0,8517 185,4003 0,6242 17,5472 7,7099

148 95,4244 8,2518 2,0277 1,6008 0,6604 1,0775 6,5550 0,6015 188,2623 0,1116 61,4568 1,9413 5,5912 16,0079

182 39,4044 8,7042 1,4802 2,0846 0,5599 1,5941 2,1208 0,9737 1,1766 0,3564 67,3450 0,4948 1,6692 3,8336

183 15,8157 3,4912 0,7683 0,0440 0,1372 0,0557 0,1572 0,0443 0,1092 0,0522 0,4369 0,1819 0,2838 1,3954

186 56,7003 10,5613 4,0751 0,8321 0,5765 0,5482 0,6948 0,6318 2,2777 0,4074 12,0629 0,5366 0,5423 6,7119

187 55,3797 5,5317 3,3725 0,9868 0,5709 0,7678 1,3695 0,8542 1,8707 0,4635 41,2265 0,6281 1,1536 11,1945

201 14,3751 8,1558 0,7865 0,0316 0,2018 0,0521 0,0548 1,5038 12,6547 0,0302 4,0208 1,3826 0,9329 2,0186

206 36,3349 7,7260 2,2499 1,0459 0,4145 0,6749 1,3028 2,1223 0,2677 0,3576 21,4588 1,0943 0,7995 5,7626

216 29,4723 1,5645 2,9584 0,8531 0,3774 0,7710 1,5017 0,5136 0,4792 0,1329 9,2375 0,5303 0,3516 4,0354

230 412,2243 91,3108 1,3387 4,9754 4,6439 2,1416 4,6415 3,7963 0,9432 1,3136 166,6295 1,5105 16,3154 27,6789

232 0,0126 0,0213 0,0009 0,0177 0,0027 0,0966 0,0010 1,1750 62,7960 0,0003 0,0100 2,7245 0,0866 29,7273

243 0,8023 51,0456 3,0395 0,7769 1,2819 0,7203 0,9710 0,0871 8,9296 0,4492 134,9705 1,4732 68,2694 2,8614

254 6,2053 1,8530 0,3621 0,0151 0,0371 0,0123 0,0257 * * 0,0518 1,0436 0,0298 0,2657 1,9525

293 75,2438 0,4588 3,0138 0,3076 0,8440 0,3095 0,3515 9,5573 0,3279 0,0912 7,5868 2,0596 0,0962 129,5471

303 22,8526 1,5313 0,6168 0,2122 0,2813 0,0137 0,2600 0,1618 0,1953 0,2048 3,9627 0,8604 1,2262 14,4172

421 150,2567 1,3705 18,4153 5,8881 0,4179 3,5621 4,6965 0,2475 3,5470 0,1969 144,4574 0,5212 4,5070 0,4508

459 151,7445 77,5381 8,9173 1,6495 1,4264 0,6217 0,6267 8,4187 1,2408 1,6322 31,9446 7,1223 1,7181 53,4950

463 181,3064 27,1845 30,2494 4,6454 2,7536 1,4267 5,2839 31,0732 1,4975 1,6206 27,3127 15,6273 1,3615 97,1885

1341 17,9174 0,6953 0,1648 1,1319 0,0605 0,0525 0,2605 0,0477 4,2181 0,0263 1,5708 0,0887 0,6970 1,3563

1618 29,3543 0,0170 33,8621 2,6408 21,1756 9,8574 2,4567 48,4223 0,5638 0,3863 355,4493 23,1403 31,6646 6,4636

1624 14,0306 0,1490 13,4413 1,0920 4,0093 3,3454 1,8122 13,9443 7,2363 1,3084 34,8842 3,8459 11,1024 6,6315

1643 132,9279 0,2826 342,6519 9,4939 0,8428 34,6360 9,7727 8,4071 47,5130 1,1438 630,1275 17,3556 14,0335 340,2222

1649 158,0788 0,0383 1482,2777 18,3156 43,1214 10,2902 13,0118 212,3941 12,5642 1,7312 340,2615 79,0303 72,4945 115,5468

1660 21,1928 0,0514 133,4911 2,5263 8,2668 16,7629 2,7506 42,4179 13,5597 0,4869 325,4969 21,8163 10,4962 116,6735

1664 13,4497 0,0901 23,2249 1,4290 4,8748 2,8151 0,9329 14,2966 0,3057 1,5040 254,4961 5,4844 30,0397 14,7162

1693 43,0603 18,8137 8,0324 1,2859 1,7009 0,3360 0,7158 * 0,6128 1,3779 13,7228 1,4241 1,6440 11,9979

1700 395,2167 24,7067 57,4247 1,1623 0,3163 1,3662 0,0766 9,0480 0,9574 0,1096 19,5148 3,1152 5,1615 319,2036

1707 24,4117 27,9124 198,8587 1,0563 2,7365 0,9384 0,6492 16,6748 11,2843 0,0000 29,0910 3,4806 0,5377 115,1470

1710 281,9900 70,5627 160,6303 0,2635 1,4433 1,7341 0,6670 6,8952 21,8756 2,4619 33,8832 2,9800 1,1179 130,0870

1724 73,5313 10,0960 9,0998 0,4169 0,6339 0,0928 1,1131 4,0017 0,0498 0,3876 13,4310 3,5603 0,9396 12,5683

1727 282,9690 178,7542 35,1049 1,7291 2,2320 0,8933 0,5292 12,2831 0,7491 2,6588 33,7426 4,1708 2,2312 80,4674

40227A2 25,7998 1,1311 0,9632 0,4140 0,0555 0,0748 0,2289 0,3757 0,4995 0,0556 1,7907 0,0598 0,3764 2,9378

41597A1 7,9960 1,2039 0,1182 0,0912 0,0263 0,0216 0,0873 0,0205 1,0081 0,0083 1,3923 0,0248 0,2689 0,8795

41625A2 10,7502 0,2102 0,0213 0,0390 0,0094 0,0200 0,0969 0,0200 0,1235 0,0207 0,9016 0,0457 0,0758 0,5691

41960A1 6,5491 0,4410 0,2879 0,2672 0,1170 0,0785 0,2040 0,0529 2,7926 0,0452 4,1829 0,0514 0,4548 3,1182

43643A2 0,1654 0,1687 0,0911 0,0150 0,0077 0,0081 0,0594 0,0166 0,0590 0,0058 0,6442 0,0188 0,1070 0,3179

44416A1 0,8750 0,5687 0,2202 0,0357 0,0146 0,0413 0,1472 0,0446 0,1459 0,0298 1,0436 0,0284 0,0564 1,2403

46722A2 41,5647 2,4635 5,7272 0,6003 0,0425 0,1261 0,5519 * 10,0047 0,1675 6,6323 0,2840 0,3620 6,8719

48165A1 1,8369 0,4900 0,3215 0,1096 0,0354 0,0331 0,2147 0,0855 0,1929 0,0328 0,9710 0,0336 0,2069 1,0047

48269A1 0,3227 1,3479 0,9011 0,1458 0,0384 0,0567 0,2096 0,0879 0,3697 0,0210 2,2230 0,0880 0,3178 0,7228

49406A1 2,8964 2,9953 1,7998 0,2172 0,2384 0,1793 0,4716 0,2306 0,6852 0,0674 4,5583 0,1433 0,5423 1,7144

50809A2F 7,5542 3,9688 1,8098 0,2500 0,0461 0,1018 0,3733 0,3884 8,0423 0,0796 4,2678 0,1636 0,5008 4,1374

51645A1E 6,1359 2,3420 1,5485 0,4600 0,0690 0,1148 0,2818 0,1556 1,8108 0,1516 9,3470 0,1525 1,6023 2,5275

52229A2E 9,7627 2,1446 0,5176 0,2801 0,1009 0,1380 0,4125 0,1031 0,3562 0,0644 3,4907 0,4104 0,8346 0,2964

61837A1D 15,7938 5,2806 1,6008 0,4295 0,0973 0,3105 0,7410 0,2428 1,4237 0,1309 12,1468 0,0256 1,4663 1,4987

71224A2H 9,8785 17,3362 1,5367 0,6801 0,2120 0,3698 0,5240 0,3433 0,7431 0,8842 56,5126 0,3368 1,2211 0,8478

217 1,4931 13,7916 0,2854 0,0978 0,1510 0,0831 0,1535 0,0225 0,2156 1,2540 14,4550 0,2048 1,4134 0,3518

223 4,1707 24,6876 1,1398 0,3133 0,4529 0,3043 0,5354 0,0544 1,4316 0,7909 54,8531 0,6413 2,7802 1,0159

Tabela contendo os valores de expressão (2-∆∆CT) de 14 miRNA em 45 pacientes com CaP com prognóstico favorável:

Caso miR-100 miR-143 miR-145 miR-146a

miR-15a miR-16 miR-191

miR-199a miR-206

miR-21 miR-218 miR-25 miR-32

miR-Let7c

35 80,1864 8,6561 3,8128 1,6515 0,9736 1,1070 1,3228 0,7576 0,9737 0,2915 74,2079 0,5370 2,3427 5,8552

44 136,2664 34,9620 3,2894 2,8104 3,9213 1,3993 1,4158 2,4960 1,0053 0,7503 103,7883 0,8305 10,5500 11,8082

49 4,2850 8,4194 0,6867 4,5340 0,1579 6,3191 19,7338 0,1372 6,6419 0,0768 49,3679 0,1839 0,5836 2,1821

15 14902,6959 2,6221 711,5559 144,5876 1,1434 73,4512 120,1423 1,2602 273,1595 0,0642 2980,8333 0,3229 887,8466 3,8098

26 30,3218 86,6252 4,3919 0,9093 0,7837 1,2939 1,9448 0,2155 0,3884 0,4388 14,8100 1,2675 0,4915 5,1434

47 46,9250 18,1855 1,2824 0,5807 1,9471 0,3212 0,2853 0,7248 0,7817 0,2048 25,6787 0,5934 16,0797 4,8390

54 55,4181 18,9577 2,7969 0,6978 3,8513 0,4992 0,2952 1,4365 152,7046 0,4203 32,9796 0,7397 3,1213 9,3354

61 39,3771 6,6933 5,0660 0,5172 0,4953 0,4940 0,5982 0,3001 3,2123 0,1688 33,1859 0,3408 1,2321 3,2416

63 29,3093 7,1143 2,6737 0,6537 0,6014 0,9915 1,2437 0,7909 1,2198 0,3604 45,6170 0,8833 0,6148 2,6753

73 34,2323 7,3448 5,9912 0,1587 0,3507 0,1393 0,2831 0,7203 0,0878 0,1532 2,7085 0,4563 0,2661 1,5776

88 134,9504 9,7656 13,2492 4,1693 0,6005 3,8230 5,5736 0,5121 1,8837 0,9457 148,3129 0,8724 1,3160 8,2062

102 51,9224 16,7224 5,3883 0,7721 3,9267 0,5543 0,4657 2,6714 2,1941 1,9126 37,8836 1,9386 6,3342 5,2333

108 270,2753 181,2346 11,0719 2,6625 12,8914 3,3887 1,3714 7,2581 4,8085 3,8119 120,2172 8,4737 25,7262 45,3715

109 9,4629 4,0946 3,2940 0,0435 0,2206 0,0400 0,0981 0,1157 0,2254 0,0618 1,6330 0,1093 0,4164 1,2106

115 28,1153 5,7109 3,7890 1,6086 0,4345 0,6810 1,2716 0,2134 0,6420 0,2863 25,3427 0,4299 2,7995 2,0990

167 7,4296 1,6208 0,4140 1,2233 0,0673 0,0300 1,2777 0,0426 1,0774 0,0230 11,5474 1,1243 0,7863 0,9895

320 1051,5383 17,1450 51,4762 100,5521 7,8317 24,8421 27,2390 2,3893 9,9149 1,2299 1707,2990 6,2723 78,9664 46,8739

325 53,0502 17,5295 6,4883 0,9245 1,8117 0,7380 0,7359 0,6244 0,4200 0,6455 18,6681 1,6099 1,3795 5,0340

342 270,0880 7,3346 95,9914 6,1084 0,4498 6,2321 14,5062 0,6933 2,5644 0,1220 147,5950 0,8055 2,0451 24,6363

79 7,4399 8,6982 5,4825 1,2985 0,4130 0,6982 2,3127 0,1639 2,7791 0,2366 41,4270 0,3336 2,2519 1,8084

121 13,9026 2,3387 0,4944 0,1793 0,0667 0,0962 0,2552 0,1917 0,2954 0,1663 1,0235 0,0941 0,0823 0,6193

122 81,1931 3,6571 9,6053 0,4301 1,1953 0,5949 1,2030 1,3525 0,3611 0,2044 10,7592 2,3086 1,4331 10,3943

296 216,5094 376,2934 15,4425 1,4160 6,2607 2,6568 0,9458 12,6372 2,0274 5,6353 155,6865 10,8830 33,1323 31,7287

327 210,1515 34,7686 5,5436 2,0559 0,9736 1,3535 1,2394 1,9678 1,1193 1,6569 27,1240 1,3177 4,1473 9,9432

40348A2 29,2890 3,5179 3,3054 0,2176 0,4946 0,1315 0,2580 0,5858 20,1625 0,1023 3,1525 0,4836 0,6046 4,2272

46588A1 2,5745 0,3029 0,0829 0,0590 0,0072 0,0149 0,2952 0,0425 0,3353 0,0071 0,0053 0,0196 0,1358 0,4549

46850A 12,2634 1,1533 1,1703 0,0648 0,1023 0,0376 0,6795 0,1482 13,3023 0,0310 2,4108 0,3025 0,1884 1,6412

57221A2B 10,4780 0,0643 4,9377 0,1297 0,0281 0,1561 0,4700 0,2127 9,0543 0,0050 6,3533 1,8725 0,1270 9,6646

22 1,1625 0,5197 0,1311 0,0296 0,0179 0,0297 0,2738 0,0203 0,0506 0,0044 0,5554 0,0254 0,1085 0,7360

99 9,3391 0,8206 0,2121 0,1260 0,0267 0,0406 0,4685 0,0138 0,5037 0,0486 2,0799 0,0646 0,4494 1,7687

125 60,1831 9,8472 8,7716 0,2491 0,8617 0,2361 0,3700 2,5912 1,9774 0,2298 5,7737 0,9183 0,6178 13,2114

155 26,5067 4,0187 3,5698 0,1254 0,3178 0,0876 0,1830 0,6843 0,8442 0,0859 2,8471 0,6259 0,3962 4,8592

168 29,5541 5,1469 8,7594 0,0599 0,3648 0,1202 0,2582 0,4906 3,7570 0,0911 4,1282 0,6101 0,2365 9,1623

75 1253,0999 321,5088 115,3467 22,8449 41,8552 36,4719 102,9357 23,0166 18,0709 0,0001 2820,0402 78,8479 0,0136 681,0960

100 1134,8498 375,5118 141,3211 129,4095 73,4318 50,0297 41,0298 0,6940 70,5513 0,0076 2252,7945 51,9481 5,3080 0,7208

195 771,9083 250,3347 79,8286 13,4525 24,0895 30,8824 28,7124 0,4550 137,8160 0,0001 2047,2903 47,5048 1,6179 261,6899

221 66,9163 18,5676 13,7450 0,7463 0,6761 1,1997 0,0031 1,4990 4,5460 0,0000 16,0277 1,5411 0,0088 9,1560

252 455,8090 126,6510 38,8767 13,1391 12,2723 15,6026 0,0472 7,3411 4,9232 0,0000 382,2837 9,2087 0,1630 69,2487

334 76,4946 11,1326 1,6956 0,7070 0,2380 0,6142 0,0052 1,9931 6,6511 * 11,1928 1,1282 0,0025 6,2667

376 80,5206 18,1100 8,4845 2,6904 0,6594 1,6018 2,0974 0,0373 4,7129 1,1475 32,5255 3,3242 * 18,1603

399 412,5101 97,4584 0,0388 5,6915 19,2040 12,6119 16,4681 0,2367 0,1413 5,8422 408,0211 6,6484 0,0079 89,7422

425 15,7719 1,0019 8,2240 0,3010 0,2755 0,0443 1,0531 0,0033 0,3780 0,0507 2,6491 0,0579 0,2574 0,5090

437 0,0275 5,7148 11,1258 0,8321 2,7214 1,0089 2,3695 0,0118 3,6215 0,8500 36,2146 0,6579 1,4401 5,5432

454 0,0882 22,4511 104,2449 3,2418 7,7884 2,1655 10,4078 0,0398 25,7161 0,0147 90,6038 2,0237 2,3541 37,0324

471 2,7746 0,1428 9,2268 0,7791 1,9057 0,4100 2,8081 0,0050 0,3925 0,0118 14,2956 1,0733 1,3615 7,7046

Original article

Change in expression of miR-let7c, miR-100, and miR-218 from highgrade localized prostate cancer to metastasis

Katia R.M. Leite, M.D., Ph.D.a,b,*, Juliana M. Sousa-Canavez, Ph.D.b,Sabrina T. Reis, Ph.D.a, Alberto H. Tomiyama, B.Sc.a, Luiz H. Camara-Lopes, M.D.b,

Adriana Sañudo, Ph.D.a, Alberto A. Antunes, M.D., Ph.D.a, Miguel Srougi, M.D., Ph.D.a

a Laboratory of Medical Investigation, LIM55, Urology Department, University of Sao Paulo Medical School, Sao Paulo, Brazilb Genoa Biotechnology, Sao Paulo, Brazil

Received 21 January 2009; received in revised form 4 February 2009; accepted 5 February 2009

Abstract

Objective: MicroRNAs (miRNAs) are small noncoding regulatory RNAs (19–25 nucleotides) that play a major role in regulation of geneexpression. They are responsible for the control of fundamental cellular processes that has been reported to be involved in humantumorigenesis. The characterization of miRNA profiles in human tumors is crucial for the understanding of carcinogenesis processes, findingof new tumor markers, and discovering of specific targets for the development of innovative therapies. The aim of this study is to findmiRNAs involved in prostate cancer progression comparing the profile of miRNA expressed by localized high grade carcinoma and bonemetastasis.

Material and methods: Two groups of tumors where submitted to analyses. The first is characterized by 18 patients who underwentradical prostatectomy for treatment of localized high grade prostate carcinoma (PC) with mean Gleason score 8.6, all staged pT3. The secondgroup is composed of 4 patients with metastatic, androgen-independent prostate carcinoma, and 2 PC cell lines. LNCaP derived from ametastatic PC to a lymph node, and another derived from an obstructive, androgen-independent PC (PcBRA1). Expression analysis of 14miRNAs was carried out using quantitative RT-PCR.

Results: miR-let7c, miR-100, and miR-218 were significantly overexpressed by all localized high GS, pT3 PC in comparison withmetastatic carcinoma. (35.065 vs. 0.996 P � 0.001), (55.550 vs. 8.314, P � 0.010), and (33.549 vs. 2.748, P � 0.001), respectively.

Conclusion: We hypothesize that miR-let7c, miR-100, and miR-218 may be involved in the process of metastasization of PC, and theirrole as controllers of the expression of RAS, c-myc, Laminin 5 �3, THAP2, SMARCA5, and BAZ2A should be matter of additionalstudies. © 2011 Elsevier Inc. All rights reserved.

Keywords: miRNA; Prostate cancer; Prognosis; Gleason score; LNCaP; Tumor progression; Metastasis

1. Introduction

MicroRNAs (miRNAs) are a class of small noncodingregulatory RNAs (19–25 nucleotides) expressed by plantsand animals involved in regulation of gene expression. Theyexert their function by binding to the 3=-untranslated regionof a subset of mRNAs resulting in their degradation orrepression of translation [1,2].

They are responsible for the control of fundamental cel-lular processes including differentiation and timing of de-velopment of the organism [3,4]. Studies demonstrate thatmore than 50% of the human miRNAs are located in so-called “fragile sites,” on chromosomes that are frequentlydeleted, amplified, or rearranged in case of cancer. miRNAshave been shown to interfere in major cellular functionssuch as cell proliferation, cell differentiation, and apoptosis,abnormalities that are hallmarks of cancer suggesting thatmiRNA might be a new class of genes involved in humantumorigenesis [5].

Down-regulation of miRNAs is the most frequently ob-served phenomenon in cancer, which suggests that they

* Corresponding author. Tel.: �55-11-31550249; fax: �55-11-32312249.

E-mail address: katiaramos@uol.com.br (K.R.M. Leite).

Urologic Oncology: Seminars and Original Investigations 29 (2011) 265–269

1078-1439/$ – see front matter © 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.urolonc.2009.02.002

function as tumor suppressor genes. Lost of miR-15 andmiR-16 are related to B-cell chronic lymphocytic leukemia(CLL) [6], miR-143 and miR-145 are consistently reducedin colon [5] and lung [7] cancer. miR-let7 family was shownto negatively regulate RAS [8], miR-16 represses Bcl2 [9],miR-125a and miR-125b suppress ERBB2 and ERBB3,respectively [10].

On the other hand, up-regulation of miRNAs has alsobeen observed frequently. The function of miR-155 is un-determined and has been shown to be overexpressed inBurkitt lymphoma [11], Hodgkin lymphoma [12], and lungcancer [7].

It is known that a given miRNA may have many mes-senger RNA (mRNA) targets, so that the biological effectsof changes in miRNA expression are likely to be dependenton the cellular context. Different sets of miRNAs have beendescribed to be up-regulated or down-regulated in cancersof different cellular origins [13].

Prostate is the most frequent cancer in men and thesecond cause of death in western countries. Current prog-nostic markers of high grade tumors after radical prostatec-tomy are limited, and therapy with androgen-deprivationleads to important side effects, mainly metabolic syndromeand osteoporosis being the adjuvant therapy for high gradetumors, a subject of controversies [14].

The search for miRNA is an open field in prostate cancerand the aim of this study is to identify a profile of miRNAexpression in localized high grade and metastatic prostatecarcinoma that could be related to tumor progression, andused as a prognostic marker or even a potential target fornew therapies.

2. Methods

2.1. Prostate tissue samples and cell lines

Two groups of patients were studied. The first is char-acterized by prostate tissue of 18 patients that underwentradical prostatectomy for treatment of localized prostatecarcinoma. The median of age was 65 years old (50–79)and the median PSA was 10 ng/mL, variable from 4.2 to 22ng/mL. All the specimens were examined in toto by thesame pathologist. Tumor volume was expressed in percent-age and stage followed TNM 2002. All tumors had highGleason score, median 9, variable from 8 to 10 and themedian of tumor volume was 20%, ranging from 3% to88%. Ten (55.6%) were staged pT3aN0, 7 (38.9%) pT3bN0,and 1 (5.5%) pT3bN1. Even though all tumors were highgrade, nonorgan confined, they were still localized, non-metastatic, which led to the decision of a curative treatment.In order to compare the profiles of these tumors with moreadvanced PC, we also submitted to analysis a second groupof patients consisting of 4 men, mean age 63.3 years, me-dian 63, variable from 59 to 68, with metastatic, androgen-independent prostate cancer. Also, LNCaP (FGC clone), a

well known PC cell line, and another lineage developed inour laboratory derived from high grade local recurrent ad-enocarcinoma, were submitted to analysis. LNCaP is a cellline derived from a metastatic, androgen-dependent prostateadenocarcinoma to a lymph node of a 50-year-old man. ThePcBra1 cell line developed by us originates from an ad-vanced, obstructive, Gleason score 9 (4 � 5), androgen-independent PC from a 62-year-old man submitted to trans-urethral resection.

As control, nonneoplastic tissue was obtained from sur-gical specimens of 6 patients who underwent retropubicprostatectomy for the treatment of benign prostate hyper-plasia. The mean age of patients was 68.5 years, median 67,variable from 61 to 80 years.

2.2. RNA isolation

Small RNA fractions were isolated and enriched usingmirVana miRNA isolation kit (Ambion, Austin, TX) andthe cDNA obtained using TaqMan miRNA Reverse Tran-scription kit (Applied Biosystems, Foster City, CA). Briefly,10 ng of miRNA was reverse transcribed using sequence-specific stem-loop primers to the following miRNAs: hsa-miR-let7c, hsa-miR-15a, hsa-miR-16, hsa-miR-21, hsa-miR-25, hsa-miR-32, hsa-miR-100, hsa-miR-143, hsa-miR-145,hsa-miR-146a, hsa-miR-191, hsa-miR-199a, hsa-miR-206,and hsa-miR-218 selected based on their predicted targetgenes. The reaction was performed on 9600 emulation modewith the following parameter values: 30 minutes at 16°C, 5minutes at 42°C, 5 minutes at 85°C, and 4°C until use.

2.3. miRNA expression analysis

Quantitative RT-PCR was carried out using the ABI7500 Fast Real-Time PCR System and a TaqMan UniversalPCR Master Mix (Applied Biosystems, Foster City, CA).Expression of individual miRNAs mentioned above wasanalyzed using miRNA sequence-specific primers. ThesemiRNAs were selected for verification based on their pred-icated target genes listed in the Sanger miRBase database(http://microrna.sanger.ac.uk/sequences) [15].

miRNA expression levels were accessed by relativequantification and the fold expression changes determinedby 2-��CT method [16]. All RT-PCR were performed induplicate, and small nucleolar RNA RNU43 was used asendogenous control.

2.4. Statistical analysis

As the distribution of the levels of expression of miRNAswere skewed, these data were log transformed for analyses;results are presented as geometric means, the mean value ofthe log transformed variable transformed back into the orig-inal units, and their 95% confidence interval (95% CI). FormiR-let7c, miR-100 and miR-218 expression analysis box-plot graphics were constructed also in logarithmic scale for

266 K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 265–269

better visualization of the results. The comparison betweenthe 2 groups for each expression level of miRNA was doneusing Student’s t-test based on transformed log data. For allstatistical analysis we considered the level of significance of5% (P � 0.05).

3. Results

The mean levels of miRNA expression are presented inTable 1 and Fig. 1 The majority of miRNAs were shown tobe overexpressed in unfavorable, localized high grade pros-tate adenocarcinoma compared with metastatic tumors.miR-let7c, miR-100, and miR-218 were constantly highlyexpressed in all samples of localized high grade, nonorganconfined with more than 100-fold change. For metastatictumors and PC cell lines all 3 were significantly and per-sistently underexpressed (Fig. 2). MiR-145, miR146a andmiR-199a were also significantly overexpressed in local-ized, unfavorable tumors compared with metastatic, butthey did not follow a standard behavior of miR-let7c, miR-

100, and miR-218 as shown in Fig. 1. miR-145 was under-expressed by all metastatic tumors, including cell lines, but3 localized tumor showed the same behavior. miR-199a wasoverexpressed in 17 of 18 localized prostate carcinomas, butalso was overexpressed in 3 of 6 metastatic and PC cell lines.miR-146 was shown to be overexpressed in 12 from 18 local-ized adenocarcinomas, but was overexpressed by 1 metasta-tic prostate carcinoma. The level of expression of all othersmiRNAs did not show a well distinguished pattern of behavior,and did not show differences in statistical analysis.

4. Discussion

Knowledge about molecular steps related to PC progres-sion is essential for the discovery of new prognostic markers

Table 1Geometric mean [95% CI] expression level of miRNAs by group,high-grade unfavorable, and metastatic prostate carcinoma

High grade(n � 18)

Metastatic PC(n � 6)

P value

Mir145 20.69 [6.75–63.40] 0.05 [0.00–0.88] �0.001Mir16 0.87 [0.28–2.71] 0.53 [0.12–2.30] 0.620Mir206 1.95 [0.71–5.33] 0.71 [0.05–9.76] 0.324Mir218 33.55 [13.81–81.48] 0.86 [0.04–20.26] 0.001MirLet7c 35.07 [15.54–79.12] 1.00 [0.21–4.73] �0.001Mir100 55.55 [29.69–103.93] 8.31 [1.12–61.78] 0.010Mir143 2.35 [0.54–10.27] 0.21 [0.01–2.91] 0.103Mir146a 0.89 [0.37–2.17] 0.04 [0.00–0.92] 0.005Mir21 0.30 [0.07–1.20] 0.27 [0.08–0.98] 0.934Mir15a 1.55 [0.67–3.59] 1.27 [0.42–3.84] 0.789Mir191 0.77 [0.33–1.77] 1.91 [0.04–0.89] 0.084Mir199a 10.53 [4.54–24.40] 0.55 [0.02–20.25] 0.011Mir25 4.04 [1.74–9.35] 0.99 [0.26–3.80] 0.077Mir32 2.75 [1.14–6.64] 6.21 [0.38–101.74] 0.400

Fig. 1. Expression level of selected 14 miRNAs in 18 high grade, localized, unfavorable prostate carcinomas (L), 4 metastatic PC (M), and 2 PC cell lines(LNCaP and PcBRA1). Fold change in expression was calculated using the 2-��CT method.

Metastatic PC High Grade PC

-0.4-0.1

403.2

1095.8 Mir218MirLet7cMir100

Fig. 2. Expression level of miR-let7c, miR-100, and miR-218 in 18 unfa-vorable localized PC compared with metastatic and PC cell lines.

267K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 265–269

and development of target therapy. There are few studies ofmiRNA expression in prostate cancer, specially related totumor behavior. Wu et al. [17] studied 480 miRNAs in 16prostate cancer specimens and showed 76 miRNA down-regulated compared with normal tissue. Porkka et al. [18]identified different expression of 51 miRNA between be-nign tissue and prostate cancer being 37 down-regulated and14 up-regulated in carcinomas. They were also able todemonstrate accurate classification of prostate cancer be-havior using hierarchical clustering based in miRNA ex-pression. Ozen et al. [19], studying 30 prostate cancer tis-sues by qRT-PCR, found down-regulation of miR-let7c,miR-125b, and miR-145 in all specimens.

Studying localized high grade, unfavorable and meta-static prostate carcinomas, we have shown that while themajority of the miRNAs was overexpressed by tumors orig-inated from radical prostatectomies, the opposite occurredwith PC cell lines and metastatic adenocarcinoma. We canhypothesize that there is a shift in the miRNA profile, fromlocalized high grade to metastatic prostate carcinoma. miR-let7c, miR-100, and miR-218 were constantly and signifi-cantly overexpressed in all primary prostate tumor tissuesamples compared with metastases.

While the geometric mean level of miR-let7c expressionamong the 18 cases was 35.065, for metastasis and cell linesit was only 0.997 (P � 0.001). miR-let7c has as knowntarget oncogenes RAS and c-myc [8]. It is important to notethat gain of chromosome 8q, where c-myc is located, is acommon feature in advanced PC [20]. This transcriptionfactor, involved in the control of growth, differentiation,and apoptosis, has also been shown to be amplified in asubset of advanced tumors, particularly metastatic ones[21]. Ras is another proto-oncogene involved in the patho-genesis of many human tumors, but different from c-myc,its abnormalities seem not to be related to the developmentof prostate neoplasia [22]. miRNAs have hundreds of targetgenes, and among the many predict for miR-let7c is BUB1.BUB1 plays a specific role in the mitotic checkpoint, and itsdefects may cause chromosome instability or aneuploidy inmouse fibroblasts and human cancer cell lines. Underex-pression of BUB1 has been reported in aneuploid coloncancer [23] and Wilms tumor [24]. Aneuploidy is an im-portant prognostic factor in PC, and has been related withtumor progression independent of Gleason score [25]. Futureinvestigation could explore our hypothesis that miR-let7cmay be expressed during the first stages of development ofprostate cancer, inhibiting BUB1 allowing chromosomalinstability. The underexpression of miR-let7c in metastasiscould explain the reported role of c-myc in advanced PCacting as a mitogen.

Although authors have consistently shown underexpres-sion of miR-let7c in tumors, the profile of few cases ob-served until now are not completely homogeneous. In coloncancer for example, Akao et al. [26] showed underexpres-sion of miR-let7c only in 2 out of 6 cases studied. Butsimilar to our findings when they analyzed colon cancer cell

lines there was persistent underexpression of this specificmiRNA.

miR-218 was also overexpressed in localized high gradePC comparing to metastases and cell lines, 33.550 and 0.864respectively (P � 0.001). Cheng et al. [27] reported thatinhibition of miR-218 in HeLa cells resulted in decrease ofcell growth, showing that miR-218 may be important in cellproliferation. miR-218 is known to be underexpressed inseveral cancers and the DNA encoding miR-218 is deletedin ovarian and breast cancer and in melanoma [28,29]. Onthe other hand miR-218 reduces the levels of Laminin 5 �3(LAMB3) mRNA as well as the protein. LAMB3 increasescell migration and tumorigenicity in SCID mice, and incollaboration with its ligand �6�4-integrin promotes tumor-igenesis in human keratinocytes [30]. As hypothesized formiR-let7c, miR-218 can act differently in each step ofcarcinogenesis, activating cell cycle and inducing prolifer-ation in the beginning and allowing LAMB3 to act duringthe metastatic stage.

Authors have reported overexpression of miR-100 inovarian tumors [31]. There was significantly overexpres-sion of miR-100 in localized high grade PC comparing tometastases and cell lines, 55.550 and 8.314, respectively(P � 0.010). Possible targets of miR-100 are THAP2,SMARCA5, and BAZ2A. THAP2 is a member of a recentlydescribed family of proteins that regulates cell proliferationthrough modulation of pRb/E2F cell cycle target gene [32].SMARCA5, also known as SNF2h, is a member of theISWI family of chromatin-remodeling proteins that hasbeen implicated in cellular DNA replication, by facilitatingDNA synthesis of higher-ordered heterochromatin in mam-malian cells [33]. Zhou et al. [34] have shown that theoverexpression of BAZ2A, also known as TTF1 interactingprotein 5 (TIP5) has a repressive effect over the DNAtranscription. We can speculate that the overexpression ofmiR-100, identified in localized high grade PC, has a role inprostate carcinogenesis, influencing DNA transcription andproliferation in different levels, but other unknown, impor-tant phenomenon must occur in order to allow the develop-ment of metastasis.

One question that was only partially answered by thisstudy is whether the anti-androgen therapy received bypatients to treat metastatic disease could be the reason forthe shift in the miRNA expression we observed. However,the miRNA profile of LNCaP, a hormone-sensitive tumor,was similar to the metastatic ones, which turns improbablethe influence of treatment in the change of miRNA expres-sion. This data should be confirmed by a larger study.

In conclusion, we have shown a change in the profile ofmiRNA expression in high Gleason score and nonorganconfined PC different from that expressed by metastatic PCand PC cell lines. These data might be the subject of furtherstudies in order to prove their role in the process of metas-tasization and, moreover, may possibly be used as prognos-tic markers or even as targets for the development of newtherapies.

268 K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 265–269

References

[1] Voorhoeve PM, Agami R. Classifying microRNAs in cancer: The good,the bad, and the ugly. Biochim Biophys Acta 2007;1775:274–82.

[2] Shi XB, Tepper CG, DeVere et al. MicroRNAs and prostate cancer.Int J Cancer 2007;120:953–60.

[3] Lee RC, Ambros V. An extensive class of small RNAs in Caeno-rhabditis elegans. Science 2001;294:862–4.

[4] Reinhart BJ, Slack FJ, Basson M, et al. The 21-nucleotide let-7 RNAregulates developmental timing in Caenorhabditis elegans. Nature2000;403:901–6.

[5] Calin GA, Sevignani C, Dumitru CD, et al. Human microRNA genesare frequently located at fragile sites and genomic regions involved incancers. Proc Natl Acad Sci USA 2004;101:2999–3004.

[6] Michael MZ, O’ Connor SM, van Holst Pellekaan NG, et al. Reducedaccumulation of specific microRNAs in colorectal neoplasia. MolCancer Res 2003;1:882–91.

[7] Yanaihara N, Caplen N, Bowman E, et al. Unique microRNA mo-lecular profiles in lung cancer diagnosis and prognosis. Cancer Cell2006;9:189–98.

[8] Johnson SM, Grosshans H, Shingara J, et al. RAS is regulated by thelet-7 microRNA family. Cell 2005;120:635–47.

[9] Cimmino A, Calin GA, Fabbri M, et al. miR-15 and miR-16 induceapoptosis by targeting BCL2. Proc Natl Acad Sci USA 2005;102:13944–9.

[10] Scott GK, Goga A, Bhaumik D, et al. Coordinate suppression ofERBB2 and ERBB3 by enforced expression of micro-RNA miR-125aor miR-125b. J Biol Chem 2007;282:1479–86.

[11] Eis PS, Tam W, Sun L, et al. Accumulation of miR-155 and BICRNA in human B cell lymphomas. Proc Natl Acad Sci USA 2005;102:3627–32.

[12] Kluiver J, Poppema S, de Jong D, et al. BIC and miR-155 are highlyexpressed in Hodgkin, primary mediastinal and diffuse large B celllymphomas. J Pathol 2005;207:243–9.

[13] Volinia S, Calin GA, Liu C-G, et al. A micro RNA expressionsignature of human solid tumors defines cancer gene targets. ProcNatl Acad Sci USA 2006;103:2257–61.

[14] Sharifi N, Gulley JL, Dahut WL. Androgen deprivation therapy forprostate cancer. JAMA 2005;294:238–44.

[15] Adams BD, Furneaux H, White BA. The micro-ribonucleic acid(miRNA) miR-206 targets the human estrogen receptor-alpha (ERe)and represses ERe messenger RNA and protein expression in breastcancer cell lines. Mol Endocrinol 2007;21:1132–47.

[16] Livak KJ, Schmittgen TD. Analysis of relative gene expression datausing real-time quantitative PCR and the 2(-��CT) method. Methods2001;25:402–8.

[17] Wu W, Sun M, Zou GM, et al. MicroRNA and cancer: Current statusand prospective. Int J Cancer 2007;120:953–60.

[18] Porkka KP, Pfeiffer MJ, Waltering KK, et al. MicroRNA expressionprofiling in prostate cancer. Cancer Res 2007;67:6130–5.

[19] Ozen M, Creighton CJ, Ozdemi M, et al. Widespread deregulation ofmicroRNA expression in human prostate. Cancer Oncogene 2008;27:1788–93.

[20] Nupponen NN, Kakkola L, Koivisto P, et al. Genetic alterations inhormone-refractory recurrent prostate carcinomas. Am J Pathol 1998;153:141–8.

[21] Buttyan R, Sawczuk IS, Benson MC, et al. Enhanced expression ofthe c-myc protooncogene in high-grade, human prostate cancers.Prostate 1987;11:327–37.

[22] Gumerlock PH, Poonamallee UR, Meyers FJ, et al. Activated rasalleles in human carcinoma of the prostate are rare. Cancer Res1991;51:1632–7.

[23] Grady WM. Genomic instability and colon cancer. Cancer MetastasisRev 2004;23:11–27.

[24] Haruta M, Matsumoto Y, Izumi H, et al. Combined BubR1 proteindown-regulation and RASSF1A hypermethylation in Wilms tumorswith diverse cytogenetic changes. Mol Carcinog 2008;47:660–6.

[25] Lerner SE, Blute ML, Bergstralh EJ, et al. Analysis of risk factors forprogression in patients with pathologically confined prostate cancersafter radical retropubic prostatectomy. J Urol 1996;156:137–43.

[26] Akao Y, Nakagawa Y, Kitade Y, et al. Down-regulation of microRNAs-143 and �145 in B-cell malignancies. Cancer Sci 2007;98:1914–20.

[27] Cheng AM, Byrom MW, Shelton J, et al. Antisense inhibition ofhuman miRNAs and indications for an involvement of miRNA in cellgrowth and apoptosis. Nucleic Acids Res 2005;33:1290–7.

[28] Calin GA, Croce CM. MicroRNA signatures in human cancers. NatRev Cancer 2006;6:857–66.

[29] Zhang L, Huang J, Yang N, et al. microRNAs exhibit high frequencygenomic alterations in human cancer. Proc Natl Acad Sci USA2006;103:9136–41.

[30] Martinez I, Gardiner AS, Board KF, et al. Human papillomavirus type16 reduces the expression of microRNA-218 in cervical carcinomacells. Oncogene 2008;27:2575–82.

[31] Neetu D. MicroRNA Expression and Identification of PutativemiRNA Targets in Ovarian Cancer. PLOS ONE 2008;3:e2436.

[32] Bessière D, Lacroix C, Campagne S, et al. Structure-function analysisof the THAP zinc finger of THAP1, a large C2CH DNA-bindingmodule linked to Rb/E2F pathways. J Biol Chem 2008;283:4352–63.

[33] Bozhenok L, Wade PA, Varga-Weisz P. STF-ISWI chromatin remod-eling complex targets heterochromatic replication foci. EMBO J2002;21:2231–41.

[34] Zhou Y, Santoro R, Grummt I. The chromatin remodeling complexNoRC targets HDAC1 to the ribosomal gene promoter and repressesRNA polymerase I transcription. EMBO J 2002;21:4632–40.

269K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 265–269

Original article

miRNA analysis of prostate cancer by quantitative real time PCR:Comparison between formalin-fixed paraffin embedded and

fresh-frozen tissue

Katia R.M. Leite, M.D., Ph.D.a,b,*, Juliana M.S. Canavez, Ph.D.b, Sabrina T. Reis, Ph.D.a,Alberto H. Tomiyama, B.Sc.a, Camila B. Piantino, B.Sc.a, Adriana Sañudo, Ph.D.a,

Luiz Heraldo Camara-Lopes, M.D.b, Miguel Srougi, M.D., Ph.D.a

a Laboratory of Medical Investigation–LIM55, Urology Department, University of Sao Paulo Medical School, Sao Paulo, Brazilb Genoa Biotechnology, Sao Paulo, Brazil

Received 13 April 2009; received in revised form 6 May 2009; accepted 11 May 2009

Abstract

Objective: Micro RNA (miRNA) is a class of small noncoding RNA that plays a major role in the regulation of gene expression, whichhas been related to cancer behavior. The possibility of analyzing miRNA from the archives of pathology laboratories is exciting, as it allowsfor large retrospective studies. Formalin is the most common fixative used in the surgical pathology routine, and its promotion of nucleicacid degradation is well known. Our aim is to compare miRNA profiles from formalin-fixed paraffin embedded (FFPE) tissues withfresh-frozen prostate cancer tissues.

Methods: The expression of 14 miRNAs was determined by quantitative real time polymerase chain reaction (qRT-PCR) in 5 pairedfresh-frozen and FFPE tissues, which were representative of prostate carcinoma.

Results: There was a very good correlation of the miRNA expression of miR-let7c and miR-32 between the fresh-frozen and FFPEtissues, with Pearson’s correlation coefficients of 0.927 (P � 0.023) and 0.960 (P � 0.010), respectively. For the remaining miRNAs, thecorrelation was good with Spearman correlation coefficient of 0.638 (P � 0.001).

Conclusion: Analysis of miRNAs from routinely processed and stored FFPE prostate tissue is feasible for some miRNAs usingqRT-PCR. Further studies should be conducted to confirm the reliability of using stock tissues for miRNA expressiondetermination. © 2011 Elsevier Inc. All rights reserved.

Keywords: Micro RNA; Prostate cancer; Methods; Formalin-fixed; Paraffin-embedded; qRT-PCR

1. Introduction

Identification of the molecular mechanisms related to thedevelopment and progression of neoplasms is a vast area ofresearch and has resulted in enormous transformations inclinical practice, permitting the finding of new diagnostic andprognostic markers and, more interestingly, the developmentof specific target therapy. Ideal clinical specimens for molec-ular biology study are snap fresh-frozen tissues with preservedDNA, RNA, and proteins. However, this practice of snapfreezing requires a special structure and organization that manyhospitals and laboratories do not have [1].

The most common fixative for human samples is aformaldehyde solution at a concentration of 3.7%, whichis referred to as 10% formalin and leads to extensivecrosslinking of all tissue components. As a consequence,nucleic acids suffer different grades of fragmentation de-pending on the conditions of the tissue fixation, processing,and storing [2].

The archives of pathology laboratories are an enormoussource of human samples, allowing for huge retrospectivestudies, which are important in the study of cancer sincemany years of follow-up are imperative to prove the role ofa new molecular marker.

Micro RNA (miRNA) is a class of small noncoding RNAof 19 to 25 nucleotides, whose major role is the regulationof gene expression. Half of them are located in the so-called

* Corresponding author. Tel.: �55-11-30617183; fax: �55-11-32312249.E-mail address: katiaramos@uol.com.br (K.R.M. Leite).

Urologic Oncology: Seminars and Original Investigations 29 (2011) 533–537

1078-1439/$ – see front matter © 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.urolonc.2009.05.008

“fragile sites” of chromosomal DNA, which are frequentlydeleted, amplified, or rearranged in many cases of cancer.miRNAs have been shown to interfere in key cellular func-tions, such as cell proliferation, cell differentiation, andapoptosis, abnormalities that are hallmarks of cancer, whichsuggests that miRNA might be a new class of genes in-volved in human tumorigenesis [3].

Our aim is to study the possibility of using formalinfixed, paraffin embedded (FFPE) material from the archivesto identify miRNA profiles. To achieve this goal, we com-pared the expression of 14 miRNAs in paired fresh-frozenand FFPE tissues representative of prostate carcinoma.

2. Methods

2.1. Patients

Five patients underwent open radical prostatectomy totreat prostate cancer from December, 2007 to January, 2008.The median patient age was 70 years, and ranged from 54 to77 years. The median prostate specific antigen (PSA) was8.7 ng/mL, and ranged from 4.2 to 13.3 ng/mL. The medianGleason score was 8, and varied from 8 to 9, and the mediantumor volume was 8 cm3, and ranged from 2.6 to 10.6 cm3.All patients were staged pT3.

The fresh surgical specimen was sent to the laboratory nolonger than 15 minutes after removal, and was immediatelyexamined by an experienced uropathologist. The prostatewas serially cut and a fragment of 1 cm2 in area that wassuspected of being cancerous was frozen in liquid nitrogenat �170°C. The prostate was than fixed in neutral buffered10% formalin for 6 to 12 hours, and routinely processedin a Shandon Pathcenter automated vacuum processor(Thermo Scientific, Waltham, MA) for a period of 8.5hours. Paraplast (McCormick, Maarn, The Netherlands), alow melting point embedding medium substitute for classicparaffin, was used for embedding. The molecular tests wereperformed 12 months after surgery in December 2008. Par-affin blocks were stocked in an environment with controlledtemperature and humidity.

2.2. miRNA isolation from fresh frozen tissue

Small RNA fractions were isolated and enriched usingthe mirVana miRNA isolation kit (Ambion, Austin, TX)and the cDNA was obtained using the TaqMan miRNAReverse Transcription kit (Applied Biosystems, Foster City,CA). Briefly, 10 ng of miRNA was reverse transcribedusing sequence-specific stem-loop primers for the followingmiRNAs: hsa-miR-let7c; hsa-miR-15a; hsa-miR-16; hsa-miR-21; hsa-miR-25; hsa-miR-32; hsa-miR-100; hsa-miR-143; hsa-miR-145; hsa-miR-146a; hsa-miR-191; hsa-miR-199a; hsa-miR-206; and hsa-miR-218. The miRNAs wereselected based on their predicted target genes. The reactionwas performed in 9600 emulation mode with the following

parameter values: 30 minutes at 16°C, 5 minutes at 42°C, 5minutes at 85°C, and 4°C until use.

The miRNAs were chosen based on their role in cancer.As examples, we cite miR-15a and miR-16, controllers ofBcl2 [4], miR21 a negative controller of PTEN [5,6], as wellas others described in prostate [7,8].

2.3. miRNA isolation from formalin fixed, paraffinembedded tissue

For miRNA isolation from paraffin embedded sampleswe used the RecoverAll total nucleic acid isolation kit(Ambion Austin, TX). Briefly, 10 cuts of 10 �m from theparaffin blocks were incubated in xylene at 50°C for 3minutes to solubilize and remove paraffin from the tissue,and then washed in alcohol solution to remove the xylene.The deparaffinized samples were next subjected to a pro-tease step to digest proteins covalently bound to RNA,DNA, and other proteins. Finally, the miRNA was purifiedby capture on a glass-fiber filter, washing, and elution.High-ethanol washing steps ensure the recovery of smallerRNA fragments (�200 nt), including miRNA. The cDNAwas synthesized as described above.

For all fresh-frozen and FFPE tissue samples, the quan-tification and analysis of the RNA quality of RNA wasassessed using the Nanodrop (Thermo Scientific, Waltham,MA) and Agilent 2100 Bioanalyzer (Agilent Technologies,Waldbronn, Germany), respectively.

2.4. miRNA expression analysis

Quantitative RT-PCR was carried out using the ABI7500 Fast Real-Time PCR System and a Taqman UniversalPCR Master Mix (Applied Biosystems, Foster City, CA).Expression of the individual miRNAs mentioned above wasanalyzed using miRNA sequence-specific primers. ThesemiRNAs were selected from the Sanger miRBase database(http://microrna.sanger.ac.uk/sequences).

miRNA expression levels were assessed by relativequantification and the fold expression changes were deter-mined by the 2���CT method [9]. All RT-PCR were per-formed in duplicate, and small nucleolar RNA RNU43 wasused as an endogenous control.

2.5. Statistical analysis

The correlation among the 14 pairs of samples wasevaluated using the dispersion graphic and the Spearmancorrelation coefficient. As the distribution of the levels ofexpression of miRNAs was skewed, these data were logtransformed for analyses. The results are presented as geo-metric means, the mean value of the log transformed vari-able transformed back into the original units, and the 95%confidence interval (95% CI).

miRNA expression levels from both groups of specimenswere compared using the t-paired test based on log trans-

534 K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 533–537

formed data. For all the statistical analyses, the level ofsignificance was 5%.

3. Results

3.1. Quality of recovered RNA

The integrity and quantity of recovered miRNA wasanalyzed using a NanoDrop 1000 spectrophotometer(Thermo Scientific, Waltham, MA). As shown in Fig. 1,the concentration of miRNA was similar in both the frozentissues and FFPE specimens. Additionally, the 260/280 ab-sorbance ratio of absorbance was around 2.0 for both sam-ples, indicating good quality miRNA. The 260/230 ratiowas 1.26 for fresh tissue and 0.93 for paraffin embeddedtissue. The 230 nm is related to solvent contaminants and, asexpected, it is higher in paraffin embedded tissue, sincethere is a necessity of using xylene. The digital gel from thebioanalyzer showed satisfactory recovery of miRNA fromboth the fresh-frozen and paraffin embedded tissue speci-mens (Fig. 2).

3.2. Comparison of miRNA expression between thefresh-frozen and paraffin embedded prostate tissues

There was a good correlation between the expressionlevels of each miRNA when the fresh-frozen and paraffinembedded tissue results were compared (0.638; P � 0.001)(Fig. 3). Additionally, Fig. 4 shows an example of a goodcorrelation between paired specimens when the individualmiRNA expression is compared. In this particular case,there was concordance in 10 (71.4%) miRNAs when theresults of over- or underexpression of miRNA were consid-ered. For only 4 particular miRNAs, miR-15, miR-25,miR206, and miR-146a, the results showed opposite find-ings. Analyzing individual miRNAs expression using theSpearman correlation test, miR-let7c and miR-32 showedvery good correlation, with a Spearman’s correlation coef-ficient of 0.900 (P � 0.037) for both. This result wasfollowed by miR-100 (0.800; P � 0.104), miR145 (0.600;P � 0.285), and miR-16 and miR206 (0.500; 0.391). Cor-relation coefficients from the others varied from �0.100 to0.400.

There was no pattern of discordance between the FFPEand fresh-frozen tissue specimens, which at some timesshowed underexpression and at other times showed overex-pression (Table 1).

Fig. 1. Graphic from the nanodrop, which shows the good quality of theRNA for both the fresh-frozen (above) and formalin-fixed paraffin embed-ded (below) specimens. The quality determination is based on the 260/280absorbance ratio of 1.98 for both specimens.

Fig. 2. Gel from the Agilent 2100 bioanalyzer illustrating the good recov-ery of miRNA with a band located under 200 bp.

Fig. 3. This dispersion graphic shows a fair correlation between the resultsof miRNA expression in fresh frozen and formalin-fixed paraffin-embed-ded prostate specimens.

Fig. 4. The graphic is an example of the level of expression of 14 miRNAsin paired fresh-frozen and formalin-fixed paraffin embedded prostate can-cer tissue.

535K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 533–537

RNU43 was used as an endogenous control for all thereactions, and the mean of expression of all experimentswas compared between the two groups. There was a faircorrelation between the experiments in regard to the RNU43expression. The Pearson’s correlation test revealed a corre-lation of r � 0.557 (P � 0.119).

4. Discussion

The formalin-fixed, paraffin embedded clinical samplesin the archives of pathology laboratories are a rich source ofinformation regarding neoplasm behavior. The possibilityof analyzing nucleic acid from these specimens makes re-search more flexible and less expensive, since it does notrequires special specimen handling or infrastructure.

Rupp and Locker [10] were the first to extract RNA fromparaffin embedded tissue for northern hybridization, andsince their work a large amount of progress has allowed forthe use of filed material for RNA analyses. Dunn et al. haveextracted RNA from recently processed prostate tissue onan experimental basis and showed 80% concordance be-tween fresh-frozen and formalin-fixed specimens using agenome-wide expression analysis [11].

miRNA is a recently described class of regulatory RNA,which is important in the control of gene expression and isclosely related to the promotion and progression of neo-plasms. There are few reports comparing results of miRNAexpression in paired fresh-frozen and FFPE clinical speci-mens.

By studying paired fresh-frozen and FFPE prostate tissuesthat have been routinely processed and filed, we have shownby qRT-PCR that miRNA expression presented a good cor-relation between specimens (r � 0.638; P � 0.001). Addi-tionally, the better performance of miR-let7c and miR-32demonstrated a good correlation between the two groups ofspecimens (Spearman’s correlation coefficient � 0.900;

P � 0.037). The majority of studies that compare miRNAexpression in fresh-frozen and FFPE tissue samples usuallyshow a good correlation of the results [12,13]. However,most of the studies are based on different methodologiesthan the one herein, mainly microarray [14], or are per-formed without statistical analyses [15,16] and validationby a more specific and sensitive technique as qRT-PCR.Additionally, few studies have analyzed paired specimens,while others are based on cell culture or mouse tissuehandled under experimental conditions, not routinely pro-cessed clinical specimens [17,18].

Different tissues have distinct requirements regardingfixation, with some needing more or less time to be ade-quately fixed. Some authors use a microwave to speedfixation of the prostate since the penetration of formalin iscalculated to be only 1 mm per hour, which is insufficient toachieve deep tissue penetration unless it is maintained insolution for at least 24 hours [19]. Differences in the char-acteristics of each organ justify our study, as this is the firststudy to compare miRNA expression in paired, routinelyprocessed prostate cancer specimens.

Formalin adds monomethylol to the amino groups ofbases, which is followed by an electrophilic attack of anamino base by N-methylol, forming a methylene bridgebetween the two amino groups [20]. This phenomenon leadsto extensive crosslinking of all tissue components and re-sults in the fragmentation of nucleic acids. This fragmenta-tion depends on the handling of the specimens and, onaverage, leads to DNA fragments of 300 to 400 bp and RNAfragments of around 200 bp [21]. For miRNAs, this does notseem to be a problem since mature miRNA is only 19 to 25bp. To determine the quality and amount of miRNA ex-tracted from fresh-frozen and FFPE specimens, we used 2different pieces of equipment, the Nanodrop and Bioana-lyzer, both of which indicated good quality miRNA basedon the band patterns and OD ratio that ranged from 1.9 to2.0. Considering the damage that formalin could cause tonucleic acids, it would be sensible to think that the level ofmiRNA expression should be always lower in paraffin em-bedded specimens than in fresh-frozen tissues. However,this was not the case. What we observed was an erraticpattern of miRNA expression. At times, the level of expres-sion of a determined miRNA was higher in the paraffinembedded specimens, while at other times the expressionwas lower.

There are some alternative fixatives to formalin that haverecently been tested [22]. Vollmer et al. [23] have shownthat DNA, RNA, and proteins are protected in HOPE-fixed,paraffin-embedded tissues for at least 8 years. If furtherstudies confirm the superiority of novel fixatives, it mayprovide incentives for pathology laboratories to change theirpractices in order to build a new archive permitting molec-ular research in the future.

Both time and environmental conditions of storing areimportant in nucleic acid fragmentation. Because of thisobservation, we have studied contemporary specimens ar-

Table 1Geometric means [95% CI] of miRNA expression levels by specimens,fresh-frozen, and formalin-fixed, paraffin embedded

Paraffin embedded Fresh-frozen

Mir145 4.778 [0.220; 103.751] 56.699 [12.030; 267.230]Mir16P 0.236 [0.048; 1.173] 0.713 [0.166; 3.056]Mir206 17.391 [0.049; 6145.580] 1.545 [0.076; 31.402]Mir218 12.586 [1.353; 117.108] 24.439 [14.814; 40.316]MirLet7c 5.808 [0.609; 55.376] 86.475 [19.694; 379.698]Mir100 54.476 [8.598; 345.167] 141.441 [32.862; 608.780]Mir143 13.930 [1.469; 132.059] 38.790 [9.932; 151.500]Mir146 0.270 [0.055; 1.329] 0.747 [0.283; 1.973]Mir21 0.658 [0.101; 4.292] 0.087 [0.000; 39.522]Mir15a 0.751 [0.232; 2.428] 1.120 [0.365; 3.438]Mir191 0.373 [0.092; 1.507] 0.455 [0.126; 1.642]Mir199 2.334 [0.321; 16.956] 8.745 [4.429; 17.266]Mir25 1.995 [0.848; 4.691] 3.437 [2.921; 4.045]Mir32 0.801 [0.102; 6.286] 1.454 [0.493; 4.291]

536 K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 533–537

chived for approximately 1 year in a temperature- andhumidity-controlled atmosphere. Different tissues and dis-tinct neoplasms with lower or higher degrees of aggressive-ness should also influence the results. To standardize oursamples, we used only prostate tissue harboring unfavor-able, high Gleason grade, and non-organ confined prostatecarcinoma.

Another possible source of variation could be the heter-ogeneity of the represented tissue. There is a possibility thatdifferent cell types could be present at higher or lowerconcentrations in the fresh-frozen and FFPE tissues. How-ever, this is highly improbable since both samples camefrom the same area of the same specimen and we carefullyverified that tumor was represented in at least 75% of thesamples by examining a control slide stained with hema-toxylin and eosin.

A categorical result with reference to whether miRNA isover- or underexpressed in tissues is also interesting. Tomake that determination, we also analyzed the crude resultof miRNA if over- or underexpressed by both specimens,and in 72.9% of the cases, the results were similar; miR-100, miR-143, and miR-218 were 100% overexpressed inall paired specimens.

The choice of endogenous control (EC) is crucial for thesuccess of the experiment. Use of an unreliable EC may leadto inaccurate, unreliable results, and some authors haveshown that mRNA expression can be made to appear up- ordown-regulated based on the choice of endogenous control[24]. There are some EC candidates, including miR-let7aand miR-16, which were validated by Davoren et al. inbreast tissue [25]. For the prostate, we have shown thatRNU43 was very stable in the experiments and did notdisplay significant differences between fresh-frozen andFFPE specimens.

In conclusion, we have shown that miRNA expressionfrom fresh-frozen and FFPE prostate tissues, which arerepresentative of clinical specimens routinely handled andfiled, could be assessed by qRT-PCR. More studies shouldbe performed in order to confirm the reliability of usingstocked tissue for miRNA expression determination.

References

[1] El-Nagger AK. Methods in molecular surgical pathology. SeminDiagn Pathol 2002;19:56–71.

[2] Godfrey TE, Kim SH, Chavira M, et al. Quantitative mRNA expres-sion analysis from formalin-fixed, paraffin-embedded tissues using 5=nuclease quantitative reverse transcription-polymerase chain reaction.J Mol Diagn 2000;2:84–91.

[3] Calin GA, Sevignani C, Dumitru CD, et al. Human microRNA genesare frequently located at fragile sites and genomic regions involved incancers. Proc Natl Acad Sci USA 2004;101:2999–3004.

[4] Calin GA, Ferracin M, Cimmino A, et al. A MicroRNA signatureassociated with prognosis and progression in chronic lymphocyticleukemia. N Engl J Med 2005;353:1793–801.

[5] Yan LX, Huang XF, Shao Q, et al. MicroRNA miR-21 overexpres-sion in human breast cancer is associated with advanced clinicalstage, lymph node metastasis, and patient poor prognosis. RNA 2008;14:2348–60.

[6] Meng F, Henson R, Wehbe-Janek H, et al. MicroRNA-21 regulatesexpression of the PTEN tumor suppressor gene in human hepatocel-lular cancer. Gastroenterology 2007;133:647–58.

[7] Ambs S, Prueitt RL, Yi M, et al. Genomic profiling of microRNA andmessenger RNA reveals deregulated microRNA expression in pros-tate cancer. Cancer Res 2008;68:6162–70.

[8] Leite KR, Sousa-Canavez JM, Reis ST, et al. Change in expression ofmiR-let7c, miR-100, and miR-218 from high grade localized prostatecancer to metastasis. Urol Oncol 2009 Apr 15 [Epub ahead of print].

[9] Livak KJ, Schmittgen TD. Analysis of relative gene expression datausing real-time quantitative PCR and the 2(���CT) method. Meth-ods 2001;25:402–8.

[10] Rupp GM, Locker J. Purification and analysis of RNA from paraffin-embedded tissues. Biotechniques 1988;6:56–60.

[11] Dunn TA, Fedor H, Isaacs WB, et al. Genome-wide expressionanalysis of recently processed formalin-fixed paraffin embedded hu-man prostate tissues. Prostate 2009;69:214–8.

[12] Hasemeier B, Christgen M, Kreipe HH, et al. Reliable microRNAprofiling in routinely processed formalin-fixed paraffin-embeddedbreast cancer specimens using fluorescence labeled bead technology.BMC Biotechnology 2008;8:90.

[13] Siebolts U, Varnholt H, Drebber D, et al. Tissues from routinepathology archives are suitable for microRNA analyses by quantita-tive PCR. J Clin Patho 2009;62:84–8.

[14] Wang H, Ach RA, Curry B. Direct and sensitive miRNA profilingfrom low-input total RNA. RNA 2007;13:151–9.

[15] Nelson PT, Baldwin DA, Searce LM, et al. Microarray-based, high-throughput gene expression profiling of microRNAs. Nat Methods2004;1:155–61.

[16] Laurie CH, Soneji S, Marafioti T, et al. MicroRNA expression dis-tinguishes between germinal center B cell-like and activated B cell-like subtypes of diffuse large B cell lymphoma. Int J Cancer 2007;121:1156–61.

[17] Li J, Smyth P, Flavin R, et al. Comparison of miRNA expressionpatterns using total RNA extract from matched samples of formalin-fixed paraffin-embedded (FFPE) cells and snap frozen cells. BMCBiotechnol 2007;7:36.

[18] Xi Y, Nakajima G, Gavin E, et al. Systematic analysis of micro RNAexpression of RNA extract from fresh frozen and formalin-fixedparaffin embedded samples. RNA 2007;13:1668–74.

[19] Ruijter ET, Miller GJ, Aalders TW, et al. Rapid microwave-stimu-lated fixation of entire prostatectomy specimens. Biomed-II MPCStudy Group. J Pathol 1997;183:369–75.

[20] Masuda N, Ohnichi T, Kawamoto S, et al. Analysis of chemicalmodification of RNA from formalin-fixed samples and optimizationof molecular biology applications for such samples. Nucleic AcidsRes 1999;27:4436–43.

[21] Lehmann U, Kreipe H. Real-time PCR analysis of DNA and RNAextract from formalin-fixed and paraffin-embedded biopsies. Methods2001;25:409–18.

[22] Vincek V, Nassiri M, Nadji M, et al. A tissue fixative that protectsmacromolecules (DNA, RNA, and protein) and histomorphology inclinical samples. Lab Invest 2003;83:1427–35.

[23] Vollmer E, Galle J, Lang DS, et al. The HOPE technique opens up amultitude of new possibilities in pathology. Rom J Morphol Embryol2006;47:15–9.

[24] McNeill RE, Miller N, Kerin MJ. Evaluation and validation of can-didate endogenous control genes for real-time quantitative PCR stud-ies of breast cancer. BMC Mol Biol 2007;8:107.

[25] Davoren PA, McNeill RE, Lowery AJ, et al. Identification of suitableendogenous control genes for microRNA gene expression analysis inhuman breast cancer. BMC Mol Biol 2008;9:76.

537K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations 29 (2011) 533–537

MicroRNA-100 Expression is Independently Related toBiochemical Recurrence of Prostate Cancer

Katia R. M. Leite,* Alberto Tomiyama, Sabrina T. Reis, Juliana M. Sousa-Canavez,Adriana Sañudo, Marcos F. Dall’Oglio, Luiz H. Camara-Lopes and Miguel Srougi

From the Laboratory of Medical Investigation, Urology Department, University of São Paulo Medical School (STR, AS, MFDO, MS) andGenoa Biotechnology (JMSC, LHCL), São Paulo (KRML, AT), Brazil

Abbreviations

and Acronyms

miRNA � microRNA

mTOR � mammalian target ofrapamicin

PCR � polymerase chain reaction

PSA � prostate specific antigen

SWI/SNF � switch/sucrosenonfermentable

Submitted for publication May 17, 2010.Study received institutional review board ap-

proval.* Correspondence: Av. Dr. Arnaldo, 455, Room

2145, 01246-903, São Paulo, Brazil (telephone: 5511 30617183; e-mail: katiaramos@uol.com.br).

Purpose: Abnormal miRNA expression has emerged as crucial factors in carci-nogenesis and is important in the comprehension of prostate cancer behavior. Wedetermined the correlation of miRNA expression profiles with prostate cancerprogression.Materials and Methods: We studied frozen specimens from 49 patients treatedfor prostate cancer with radical prostatectomy. We intentionally chose 28 menwithout and 21 with biochemical recurrence, defined as prostate specific antigengreater than 0.2 ng/ml. The expression of 14 miRNAs was determined by quan-titative reverse transcriptase-polymerase chain reaction. All radical prostatec-tomy specimens were studied in toto to determine tumor volume, Gleason scoreand 2002 TNM pathological stage. Benign prostate tissue from benign prostatichyperplasia served as a control.Results: Four miRNAs were related to tumor recurrence. Using the Cox regres-sion test the risk of recurrence was 3.0, 3.3, 2.7 and 3.4 for high levels of miR-100,miR-145, miR-191 and miR-let7c, respectively. When considering statisticallysignificant clinical variables on univariate analysis of biochemical-free survival,prostate specific antigen and tumor volume, results revealed that miR-100 andtumor volume were independently related to tumor recurrence.Conclusions: A high level of miR-100 is related to biochemical recurrence oflocalized prostate cancer in patients treated with radical prostatectomy. The roleof miR-100 during carcinogenesis must be resolved in future studies to betterunderstand the molecular pathways in which miR-100 is involved. This may openthe possibility of using it as a prognostic marker and inspire the development ofa target drug.

Key Words: prostate, carcinoma, microRNAs, prostate specific antigen,neoplasm recurrence

PROSTATE cancer is highly variable andcurrent prognostic indicators have ledto failure to determine its outcome. TheGleason score is frequently underesti-mated1 and PSA can be increased inother conditions, such as benign pros-tatic hyperplasia and prostatitis. Cur-rently there are a number of treat-ments for prostate cancer, ranging from

active surveillance to radical prostatec-tomy to radiotherapy with androgendeprivation for advanced stages of pros-tate cancer, which results in a numberof side effects such as metabolic syn-drome and osteoporosis.2 Due to pros-tate cancer heterogeneity knowledge ofthe molecular mechanisms responsiblefor its outcome is important.

1118 www.jurology.com0022-5347/11/1853-1118/0 Vol. 185, 1118-1122, March 2011THE JOURNAL OF UROLOGY® Printed in U.S.A.© 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. DOI:10.1016/j.juro.2010.10.035

Oncogenes and tumor suppressor genes that areimportant to determine the behavior of other tu-mors, including c-myc, bcl2, p53 and phosphataseand tensin homologue deleted on chromosome 10, areusually altered late in prostate cancer development.Results based on large-scale gene expression profilesare disappointing with few genes confirmed in mul-tiple studies.3

miRNAs are small, noncoding RNAs produced bythe ribonuclease III-enzyme Dicer that act as regu-lators of translation and mRNA stability. More than700 miRNAs are present in Homo sapiens and theyare believed to post-transcriptionally control a thirdof human genes (http://microrna.sanger.ac.uk/cgi-bin/sequences/browse.pl).

Fundamental cellular processes, such as differen-tiation and the developmental timing of an organ-ism, are controlled by miRNAs.4,5 Studies show thatmore than 50% of human miRNAs are located at theso-called fragile sites of chromosomes. These fragilesites are regions that are frequently deleted, ampli-fied or rearranged, as in cases of cancer. miRNAs areinvolved in major cellular processes such as cellproliferation, cell differentiation and apoptosis, ofwhich abnormalities are hallmarks of cancer. Thissuggests that miRNAs may be a new class of genesinvolved in human tumorigenesis.6

The first miRNAs implicated in cancer develop-ment were miR-15 and miR-16. These miRNAs canact as negative regulators of Bcl-27 and each isdown-regulated in up to 60% of B-cell chronic lym-phocytic leukemia cells. Since then, many miRNAshave been described as having altered expressionduring carcinogenesis in different organs, includingthe prostate.8–12 Volinia et al comparing miRNAprofiles in different epithelial tumors and found 15miRNAs commonly involved in carcinogenesis of thepancreas, breast, lung, stomach, colon and pros-tate.9 Porkka et al evaluated only 9 primary pros-tate carcinomas and 6 cell lines, and found underexpression of 37 miRNAs related to the response toandrogen and over expression of 14.10 Ozen et alstudied miRNA expression using a microarray ofonly 16 specimens of prostate adenocarcinoma andnoted general miRNA under expression, especiallymiR-125, miR-145 and miR-let7c.11 Recently we con-firmed miR-let7c under expression in metastaticprostate cancer compared to that in localized highgrade tumors.12

Large studies of expression profiles of miRNAsrelated to the clinical behavior of prostate adenocar-cinoma are still rare in the literature. We investi-gated miRNA expression profiles during prostatecancer progression in patients treated with radicalprostatectomy who were followed a mean of 58.8months.

METHODS

Prostate Tissue SamplesA total of 49 patients underwent radical prostatectomy forlocalized prostate carcinoma. We intentionally chose 21men with biochemical recurrence, defined as PSA 0.2ng/ml or greater, and 28 who did not at a mean 58.5-month followup. Table 1 lists clinical and demographicdetails. The Gleason score was used for grading, tumorvolume is expressed in percent and stage was determinedaccording to the 2002 TNM. As a control, nonneoplastictissue was obtained from surgical specimens in 10 pa-tients who underwent retropubic prostatectomy for benignprostate hyperplasia. Mean age of controls was 68.5 years(median 67, range 61 to 80).

RNA IsolationSmall RNA fractions were isolated and enriched using themirVana™ miRNA isolation kit. cDNA was made usingthe TaqMan® miRNA Reverse Transcription Kit. Briefly,10 ng miRNAs were reverse transcribed using sequencespecific stem loop primers for certain miRNAs, includinghsa-miR-let7c, hsa-miR-15a, hsa-miR-16, hsa-miR-21,hsa-miR-25, hsa-miR-32, hsa-miR-100, hsa-miR-143, hsa-miR-145, hsa-miR-146a, hsa-miR-191, hsa-miR-199a,hsa-miR-206 and hsa-miR-218. These miRNAs were se-lected based on their predicted target genes. The reactionwas performed in 9600 emulation mode with the param-eters 16C for 30 minutes, 42C for 5 minutes, 85C for 5minutes and 4C indefinitely.

AnalysismiRNA expression. Quantitative reverse transcriptase-PCR was done using the ABI® 7500 Fast Real-Time PCRSystem and TaqMan Universal PCR Master Mix. Expres-sion of the individual miRNAs mentioned was analyzedusing miRNA sequence specific primers. These miRNAswere selected for verification based on their predictedtarget genes listed on the Sanger miRBase database(http://microrna.sanger.ac.uk/sequences).

We assessed miRNA expression by relative quantifica-tion using the 2���CT method13 to determine fold changesin expression. All reverse transcriptase-PCR reactionswere done in duplicate. Small nucleolar RNA RNU43served as an endogenous control.

Statistics. Since the distribution of miRNAs levels wasskewed, these data were log transformed for analysis.Results are shown as the geometric mean and the meanvalue of the log transformed variable was transformedback into the original units with the 95% CI. Groups were

Table 1. Demographic characteristics in 49 men whounderwent radical prostatectomy for prostate cancer

Recurrent PC Nonrecurrent PC p Value

Median age (range) 65 (51–76) 61.5 (48–72) 0.269Median ng/ml PSA (range) 11 (3.7–37) 8.5 (2.5–15.5) 0.027Median Gleason score (range) 7 (5–9) 5 (4–9) 0.117Median % tumor vol (range) 20 (4–43) 8.8 (1–26) 0.001No. tumor stage (%):

pT2 10 (47.6) 21 (75.9) 0.107pT3 11 (52.4) 7 (25.0)

MICRORNA-100 EXPRESSION AND BIOCHEMICAL RECURRENCE OF PROSTATE CANCER 1119

compared using the Student t test for the expression levelof each miRNA based on the log transformed data. Todetermine whether miRNA levels were related to time ofrecurrence, defined as PSA 0.2 ng/ml or greater, Kaplan-Meier curves were built based on median miRNA expres-sion. We analyzed the comparison between the curvesusing the log rank test. Also, the recurrence risk wascalculated with the 95% CI. For multivariate analysis aCox regression model was adjusted based on variablessignificant on univariate analysis. For all statistical anal-ysis significance was considered at 5% (p �0.05).

RESULTS

Figure 1 shows the expression of the 14 miRNAs inall patients. There was no statistically significantdifference between mean expression levels of themiRNAs and prostate cancer recurrence (table 2).However, relatively low miR-100, miR-145, miR-191and miR-191 levels correlated with a better out-come, as shown by Kaplan-Meier curves of biochem-ical-free survival (fig. 2). The Cox regression testshowed that the risk of recurrence was 3.0 timesmore likely when miR-100 was higher than 39.4(p � 0.019), 3.3 times more likely when miR-145 washigher than 3.8 (p � 0.011), 2.7 times more likelywhen miR-191 was higher than 1.3 (p � 0.043) and3.4 times more likely when miR-let7c was higher

than 5.1 (p � 0.012). Univariate analysis revealedthat serum PSA and tumor volume were signifi-cantly different in patients with vs without recur-rent prostate cancer (mean 15.2 vs 8.3 ng/ml,p � 0.027). Mean tumor volume in patients with vswithout recurrence was 22.5% vs 10.0% (p � 0.001).To determine statistically significant clinical vari-ables related to biochemical-free survival we usedmultivariate analysis to identify the importance of the

A

0,0001

0,001

0,01

0,1

1

10

100

1000

10000

100000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

B

0,01

0,1

1

10

100

1000

10000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

miR-100 miR-143 miR-145 miR-146a miR-15a miR-16 miR-191 miR-199a miR-206 miR-21 miR-218 miR-25 miR-32 miR-Let7c

miR-100 miR-143 miR-145 miR-146a miR-15a miR-16 miR-191 miR-199a miR-206 miR-21 miR-218 miR-25 miR-32 miR-Let7c

Figure 1. miRNA profiles in individuals without (A) and with (B) biochemical recurrence at mean 58.5-month followup

Table 2. miRNA expression in patients with and withoutrecurrent prostate cancer after radical prostatectomy

miRNAMean Recurrent PC

(95% CI)Mean Nonrecurrent

PC (95% CI)p

Value

100 50.03 (21.43–116.81) 33.65 (13.71–82.60) 0.52143 10.11 (5.79–17.64) 9.27 (4.67–18.44) 0.85145 5.05 (2.65–9.63) 3.08 (1.37–6.93) 0.35146a 1.49 (0.66–3.35) 0.77 (0.38–1.54) 0.2115a 0.87 (0.49–1.54) 0.67 (0.36–1.25) 0.5516 1.15 (0.57–2.32) 0.70 (0.37–1.32) 0.28191 1.71 (0.90–3.25) 0.89 (0.42–1.90) 0.21199a 0.75 (0.45–1.27) 0.69 (0.39–1.21) 0.81206 1.11 (0.49–2.51) 1.70 (0.71–4.06) 0.4821 0.43 (0.26–0.72) 0.25 (0.12–0.53) 0.26218 45.58 (19.63–105.86) 26.92 (10.85–66.80) 0.4025 0.81 (0.46–1.45) 0.78 (0.52–1.16) 0.8932 2.12 (0.90–5.02) 2.46 (1.12–5.41) 0.80Let7c 5.37 (2.91–9.92) 4.95 (3.33–7.36) 0.81

MICRORNA-100 EXPRESSION AND BIOCHEMICAL RECURRENCE OF PROSTATE CANCER1120

isolated factors. Results showed that percent tumorvolume (1.052, 95% CI 1.017–1.087, p � 0.003) andmiR-100 greater than 39.40 (vs 39.40 or less) (3.648,95% CI 1.384–9.616, p � 0.009) were independentlyrelated to the risk of tumor recurrence.

DISCUSSION

Our results show that miR-100 expression greaterthan 39.4 together with tumor volume are indepen-dently related to biochemical recurrence in patientstreated with radical prostatectomy for prostate can-cer. miR-100 targets the mRNA of SWI/SNF related,matrix associated, actin dependent regulator ofchromatin, subfamily a, member 5, which is part ofthe SWI/SNF chromatin remodeling complex thatregulates the transcription of a number of genes byremodeling nucleosomes situated at promoter re-gions. SWI/SNF becomes incorporated into nascentpre-ribonucleoprotein and acts post-transcription-ally to regulate the amount of mRNA synthesizedfrom a given promoter and the type of alternativetranscript produced.14 Functional diversificationamong SWI/SNF complexes allows the eukaryoticcell to fine tune and integrate the execution of di-verse biological programs involved in the expres-sion, maintenance and duplication of its genome.

Suppression of this essential mechanism by miR-100 would lead to alterations in gene expression andcause genomic instability and success of the neoplas-tic process.

DNA replication and repair also use SWI/SNFcomplexes, which promote chromatin remodeling,increasing DNA accessibility, and facilitating repli-cation and chromatin assembly. After DNA damagechromatin remodeling factors related to SWI/SNFcomplexes have a key role in nucleotide excisionrepair and double strand break repair.15 In the ab-sence of functional SWI/SNF complexes tumor pro-gression may occur since DNA methylation, repair,recombination and replication are affected. A num-ber of cancer cell lines and human tumors lack SWI/SNF. Clinical studies have shown that loss of ex-pression of the SNF5/INI1/BAF47 subunit is relatedto aggressive pediatric rhabdoid tumors, rhabdo-myosarcoma, medulloblastoma and choroid plexuscarcinoma.16 There is evidence that the SWI/SNFcomplex also interacts with the product of importantproto-oncogenes and tumor suppressor genes,17 andis implicated in androgen and estrogen dependenttumors.18

Shen et al reported that the Brm adenosinetriphosphatase, a component of certain SWI/SNFcomplexes, has a significant antiproliferative func-

Bio

chem

ical

-free

sur

viva

l

Bio

chem

ical

-free

sur

viva

l

months

Risk IC 95% p-valueMir100

>39.4 x ≤39.4 3.045 [1.200 ; 7.737] 0.019

Risk IC 95%

months

p-value Mir145

>3.8 x ≤3.8 3.347 [1.319 ; 8.496] 0.011

Bio

chem

ical

-free

sur

viva

l

Bio

chem

ical

-free

sur

viva

l

months

Risk IC 95% p-valueMir191

>1.28 x ≤1.28 2.642 [1.030 ; 6.780] 0.043

Risk IC 95% months

p-value MirLet7c

>5.14 x ≤5.14 3.359 [1.311 ; 8.604] 0.012

Figure 2. Kaplan-Meier curves show miR (Mir)-100, miR-145, miR-191 and miR-let7c expression biochemical survival rates

MICRORNA-100 EXPRESSION AND BIOCHEMICAL RECURRENCE OF PROSTATE CANCER 1121

tion in the prostate, protecting against the transi-tion between androgen dependent and independentstages.19 Expression of these genes is controlled bynutrients, such as resveratrol in wine and grapes,related to protection of the aging process and cancerdevelopment. These nutrients are also involved inprotecting against prostate cancer.20,21

The expression of miR-100 in tumors is controver-sial, and miR-100 is over and under expressed indifferent types of ovarian and prostate cancer. Da-hiya et al noted that miR-100 is over expressed inovarian cancer cell lines and tissues, mostly in se-rous high grade adenocarcinoma,22 while Nagarajaet al analyzed ovarian cancer cell lines and observedthat miR-100 is under expressed in 8 of 10 tumors.23

Using a lentivirus system they also noted thatmTOR was decreased in OVSAYO cells when miR-100 was over expressed, confirming mTOR as a tar-

get of miR100. mTOR is a component of an impor-tant pathway involved in cell growth, proliferationand survival. Phosphatase and tensin homologuedeleted on chromosome 10, a negative regulator ofthis pathway, is involved in prostate carcinogenesis.

In agreement with our results Jiang et al found a4-fold increase in miR-100 in the LNCaP cell line.24

Lin et al reported that miR-100 is over expressed inandrogen independent prostate cancer comparedwith that in androgen dependent prostate cancer.25

CONCLUSIONS

Few groups have examined the role of miRNA inprostate cancer. Other studies must be done to clar-ify the importance of miRNA in prostate carcinogen-esis. Also, future experimental studies may shedlight on the role of miR-100 in tumorigenesis.

REFERENCES

1. Moreira Leite KR, Camara-Lopes LH, Dall’OglioMF et al: Upgrading the Gleason score in ex-tended prostate biopsy: implications for treat-ment choice. Int J Radiat Oncol Biol Phys 2009; 3:353.

2. Freedland SJ, Eastham J and Shore N: Androgendeprivation therapy and estrogen deficiency in-duced adverse effects in the treatment of pros-tate cancer. Prostate Cancer Prostatic Dis 2009;12: 333.

3. Cooper CS, Campbell C and Jhavar S: Mecha-nisms of disease: biomarkers and molecular tar-gets from microarray gene expression studies inprostate. Cancer 2009; 4: 677.

4. Lee RC and Ambros V: An extensive class ofsmall RNAs in Caenorhabditis elegans. Science2001; 294: 862.

5. Reinhart BJ, Slack FJ and Basson M: The 21-nucleotide let-7 RNA regulates developmentaltiming in Caenorhabditis elegans. Nature 2001;403: 901.

6. Calin GA, Sevignani C, Dumitru CD et al: HumanmicroRNA genes are frequently located at fragilesites and genomic regions involved in cancers.Proc Natl Acad Sci USA 2004; 101: 2999.

7. Calin GA, Dumitru M, Shimizu R et al: Frequentdeletions and down regulation of micro-RNAgenes miR15 e miR16 and 13q14 in chronic lym-phocytic leukemia. Proc Natl Acad Sci USA 2002;99: 15524.

8. Michael MZ, O’ Connor SM, van Holstpelle-kaanqq NG et al: Reduced accumulation of spe-cific microRNAs in colorectal neoplasia. Mol Can-cer Res 2003; 1: 882.

9. Volinia S, Calin GA, Liu C-G et al: A micro RNAexpression signature of human solid tumors de-fines cancer gene targets. PNAS 2006; 2257: 61.

10. Porkka KP, Pfeiffer MJ, Waltering KK et al: Mi-croRNA expression profiling in prostate cancer.Cancer Res 2007; 67: 6130.

11. Ozen M, Creighton CJ, Ozdemir M et al: Wide-spread deregulation of microRNA expression inhuman prostate. Cancer Oncogene 2008; 27:1788.

12. Leite KR, Sousa-Canavez JM, Reis ST et al:Change in expression of miR-let7c, miR-100, andmiR-218 from high grade localized prostate can-cer to metastasis. Urol Oncol 2009; Epub aheadof print.

13. Livak KJ and Schmittgen TD: Analysis of relativegene expression data using real-time quantitativePCR and the 2(���CT) method. Methods 2001;25: 402.

14. Tyagi A, Ryme J, Brodin D et al: SWI/SNF asso-ciates with nascent pre-mRNPs and regulatesalternative pre-mRNA processing. PLoS Genet2009; 5: e1000470.

15. Osley MA, Tsukuda T, Nickoloff JA et al: ATP-dependent chromatin remodeling factors andDNA damage repair. Mutat Res 2007; 618: 65.

16. Glaros S, Cirrincione GM, Muchardt C et al: Thereversible epigenetic silencing of BRM: implica-tions for clinical targeted therapy. Oncogene2007; 26: 7058.

17. Halliday GM, Bock VL, Moloney FJ et al: SWI/SNF: a chromatin-remodelling complex with a

role in carcinogenesis. Int J Biochem Cell Biol2009; 41: 725.

18. Garcia-Pedrero JM, Kiskinis E, Parker MG et al:The SWI/SNF chromatin remodeling subunitBAF57 is a critical regulator of estrogen receptorfunction in breast cancer cells. J Biol Chem 2006;281: 22656.

19. Shen H, Powers N, Saini N et al: The SWI/SNFATPase Brm is a gatekeeper of proliferative con-trol in prostate cancer. Cancer Res 2008; 68:10154.

20. Lekli I, Ray D and Das DK: Longevity nutrientsresveratrol, wines and grapes. Genes Nutr 2010;5: 55.

21. Slusarz A, Shenouda NS, Sakla MS et al: Com-mon botanical compounds inhibit the hedgehogsignaling pathway in prostate cancer. Cancer Res2010; 70: 3382.

22. Dahiya N, Sherman-Baust CA, Wang TL et al:Micro RNA expression and identification of puta-tive miRNA target in ovarian cancer. PLOS 2008;3: e2436.

23. Nagajara AKI, Creighton CJ, Yu Z et al: A linkbetween mir-100 and FRAP/mTOR in clear cellovarian cancer. Mol Endocrinol 2010; 24: 447.

24. Jiang J, Lee EJ, Gusev Y et al: Real-time expres-sion profiling of microRNA precursors in humancancer cell lines. Nucleic Acids Res 2005; 33:5394.

25. Lin SL, Chiang A, Chang D et al: Loss of mir-146afunction in hormone-refractory prostate cancer.RNA 2008; 14: 417.

MICRORNA-100 EXPRESSION AND BIOCHEMICAL RECURRENCE OF PROSTATE CANCER1122

Original article

MicroRNA expression profiles in the progression of prostate cancer—from high-grade prostate intraepithelial neoplasia to metastasis�

Katia R. M. Leite, M.D., Ph.D.a,b,*, Alberto Tomiyama, B.Sc.a,b, Sabrina T. Reis, B.Sc.a,Juliana M. Sousa-Canavez, Ph.D.b, Adriana Sañudo, B.Sc.a, Luiz H. Camara-Lopes, M.D.b,

Miguel Srougi, M.D., Ph.D.a

a Laboratory of Medical Investigation, Urology Department, University of Sao Paulo Medical School, Sao Paulo, Brazilb Genoa Biotechnology, Sao Paulo, Brazil

Received 3 May 2011; received in revised form 2 July 2011; accepted 5 July 2011

Abstract

Introduction: Models of the multistep process related to cancer progression have been designed for many cancers including prostate. The aimof this study is to propose a new model including a possible role for recently described micro RNAs in prostate cancer (CaP) progression.

Methods: Sixty-three patients underwent radical prostatectomy to treat localized prostate carcinoma. The specimens of 15 patients wererepresentative of high grade prostate intraepithelial neoplasia (HGPIN). Fourteen specimens represented localized favorable CaP, and 34unfavorable, mostly non-organ-confined disease. Representing the advanced disease we studied 4 metastatic androgen-independent CaP and2 cell lines. Micro RNAs were isolated using the mirVana miRNA Isolation kit and cDNA was obtained using the TaqMan miRNA ReverseTranscription kit to the miRNAs: hsa-miR-let7c, hsa-miR-15a, hsa-miR-16, hsa-miR-21, hsa-miR-25, hsa-miR-32, hsa-miR-100, hsa-miR-143, hsa-miR-145, hsa-miR-146a, hsa-miR-191, hsa-miR-199a, hsa-miR-206, and hsa-miR-218. Quantitative RT-PCR was carried out usingthe ABI 7500 Fast Real-Time PCR System and the TaqMan Universal PCR Master Mix. miRNA expression levels were measured byrelative quantification, and fold expression changes were determined by the 2–��CT method. The small nucleolar RNA RNU43 was usedas an endogenous control.

Results: Except for miR-21 and miR-206, the expression levels of all miRNAs significantly changed during the progression of CaP.Interestingly, there was a significant global loss of miRNA expression between HGPIN and metastasis at 2 important steps. The first wasrelated to the transition from HGPIN to invasive adenocarcinoma, and the second was related to the transition from localized to metastaticadenocarcinomas.

Conclusion: Through the analysis of 14 miRNAs in 4 groups of prostate lesions, which reproduced the progression of CaP, we showedthat there is a global loss of miRNA expression at 2 distinct steps. The first related to the transition between HGPIN and localized invasivecarcinoma, and the second associated with the transition from localized to metastatic CaP. The importance of our study is in theidentification of possible miRNAs and miRNA-targeted genes involved in the progression of prostate carcinogenesis that may helpthe development of potential diagnostic or prognostic markers as well as the design of new target therapies. © 2011 Elsevier Inc.All rights reserved.

Keywords: Micro RNA; Prostate cancer (CaP); Carcinogenesis; Prognosis; Diagnosis

1. Introduction

Since the description of the progression of carcinogene-sis in colorectal cancer by Vogelstein et al. in 1988 [1], anequivalent model of the multistep progression of prostate

carcinogenesis has been developed with the purpose offinding specific genetic alterations that can be used in thediagnosis and prognosis of this disease [2]. The TMPRSS2-ETS gene fusion, the loss of 8p, 10q, 16q, and 18q and thegain of 8q, as well as the recovery of telomerase activity, areobserved early in the transition from high-grade prostateintraepithelial neoplasia (HGPIN) to invasive carcinoma[3,4]. The accumulation of additional genetic and epige-netic abnormalities is necessary for the progression of the

� This work was Funded by FAPESP – no. 2008/58276–3.* Corresponding author. Tel.: �55 11 30617183; fax: 55 11 32312249.E-mail address: katiaramos@uol.com.br (K.R.M. Leite).

Urologic Oncology: Seminars and Original Investigations xx (2011) xxx

1078-1439/$ – see front matter © 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.urolonc.2011.07.002

disease, but most of these abnormalities are associa-ted with the development of metastasis and androgen-independent growth.

MicroRNAs (miRNAs) are small non-coding RNAs, pro-duced by the RNase III-transcribed enzyme Dicer, that regulatetranslation and the stability of messenger RNA (mRNA). Thereare more than 1424 miRNAs described in Homo sapiens(http://www.mirbase.org/cgi-bin/browse.pl?org�hsa), which arethought to be responsible for controlling the expression ofone-third of human genes [5]. Many are located in so-calledfragile sites of DNA, and the same alterations that areassociated with genes implicated in the process of carcino-genesis, such as translocations, deletions, and amplifica-tions, have been found in genes encoding miRNAs [6].Recently, profiling of miRNAs has been described in tu-mors, including prostate tumors, but only clinical specimenswill represent the multistep process of prostate carcinogen-esis and should be used for miRNA profiling [7–9].

The purpose of this study was to compare miRNA ex-pression in prostate lesions that represent various steps ofprostate carcinogenesis, namely, HGPIN, localized carci-noma with favorable characteristics, localized carcinomawith unfavorable characteristics and metastatic prostate can-cer (CaP), to elucidate the genetic steps involved in theprogression of carcinogenesis in prostate tumors.

2. Methods

2.1. Prostate tissue samples and cell lines

Sixty-three patients underwent radical prostatectomyto treat localized prostate carcinoma. Samples from 15patients selected for molecular studies were representa-tive of HGPIN without invasive carcinoma. Fourteen pa-tients exhibited favorable clinical and pathological charac-teristics, and 34 displayed unfavorable prognostic factors.The clinical and demographic details are outlined Table 1. AGleason score was employed for grading, tumor volume wasexpressed as a percentage, and stage was determined followingTNM 2010. A second group of patients consisted of 4 menwith a mean age of 63.3 years, median of 63, variable from 59to 68, from whom we studied the metastatic, androgen-independent CaP specimens. Also, LNCaP (FGC clone), awell-known CaP cell line, along with another lineage devel-oped in our laboratory, which was derived from a high-gradelocal recurrent adenocarcinoma, were subjected to analysis.LNCaP is a cell line derived from a metastatic, androgen-dependent prostate adenocarcinoma in the lymph node of a50-year-old man. The PcBra1 cell line, developed by ourgroup, originates from a local, advanced, obstructive, Gleason-score-9 (4 � 5), castrate refractory CaP from a 62-year-oldman subjected to transurethral resection.

As a control, non-neoplastic tissue was obtained fromsurgical specimens from 10 patients who underwent retro-pubic prostatectomy for the treatment of benign prostate

hyperplasia. The mean age of patients was 68.5 years, witha median of 67, variable from 61 to 80 years.

2.2. RNA isolation

Small RNA fractions were isolated from fresh-frozenprostate tissues that were previously examined to guaranteethat more than 75% of specimen was representative ofcancer or HGPIN. An enrichment was performed using themirVana miRNA Isolation kit (Ambion, Austin, TX), andcDNA was obtained using the TaqMan miRNA ReverseTranscription kit (Applied Biosystems, Foster City, CA).Briefly, 10 ng of miRNA was reverse-transcribed usingsequence-specific stem-loop primers to the followingmiRNAs: hsa-miR-let7c, hsa-miR-15a, hsa-miR-16, hsa-miR-21, hsa-miR-25, hsa-miR-32, hsa-miR-100, hsa-miR-143, hsa-miR-145, hsa-miR-146a, hsa-miR-191, hsa-miR-199a, hsa-miR-206, and hsa-miR-218, which were selectedbased on their predicted target genes. The reaction wasperformed in 9600 Emulation mode with the followingparameters: 30 min at 16°C, 5 min at 42°C, 5 min at 85°C,and 4°C until samples were removed.

2.3. miRNA expression analysis

Quantitative RT-PCR was carried out using the ABI7500 Fast Real-Time PCR System and the Taqman Univer-sal PCR Master Mix (Applied Biosystems, Foster City, CA).The expression of the individual miRNAs mentioned abovewas analyzed using miRNA sequence-specific primers. ThesemiRNAs were selected for verification based on their predi-cated target genes, as listed in the Sanger miRBase database(http://microrna.sanger.ac.uk/sequences).

miRNA expression levels were measured by relativequantification between prostate lesions, HGPIN, localizedand metastatic adenocarcinoma, and BPH representativetissue that was considered as normal, and fold expression

Table 1Demographic, clinical and pathologic characteristics of the patients

Favorable CaP Unfavorable CaP P

Age (rears-old)Median 59.5 65 0.062Range 47–72 50–79

PSA (ng/mL)Median 7.6 10.2 0.016Range 3.2–25 4.2–40

Gleason scoreMedian 6 8 �0.001Range 6–7 8–10

Tumor volume %Median 12 23.5 0.001Range 1.3–30 3–88

Stage TNMpT2 14 (100%) 11 (32.4%) �0.001pT3 0 23 (67.6%)

2 K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations xx (2011) xxx

changes were determined by the 2–��CT method [10]. AllRT-PCR experiments were performed in duplicate, and thesmall nucleolar RNA RNU43 was used as an endogenouscontrol.

2.4. Statistical analysis

Because the distribution of the expression levels of themiRNAs were skewed, these data were log-transformed foranalyses; the results are presented as geometric means, withthe mean value of the log-transformed variable transformedback into its original unit, and 95% confidence interval(95% CI) are presented. The comparison of miRNA expres-sion levels between the groups was performed using oneway ANOVA based on transformed-log data. For all statis-tical analyses, we considered the level of significance to be5% (P � 0.05).

3. Results

The results are presented in Table 2. Except formiR-21 and miR-206, the expression levels of all miR-NAs significantly changed during the progression of CaP.Interestingly, there was a significant global loss ofmiRNA expression between HGPIN and metastasis at 2important steps. The first is related to the transition fromHGPIN to invasive adenocarcinoma, and the second isrelated to the transition from localized to metastatic ad-enocarcinomas (Fig. 1). There is no important alterationin miRNA expression between favorable and unfavorablelocal adenocarcinomas.

All miRNAs were overexpressed in all cases of HG-PIN compared with benign prostate tissue. In fact, miR-100 was the only miRNA that remained overexpressedduring all steps of the carcinogenesis process, but a

significant reduction of this miRNA was also observedduring cancer progression.

In localized favorable and unfavorable carcinomas, 57%and 43% of miRNAs, respectively, were underexpressed. Inmetastases, 86% of miRNAs were underexpressed com-pared with benign tissue.

4. Discussion

Through the analysis of 14 miRNAs in 4 groups ofprostate lesions, which reproduced the progression ofCaP, we showed that there is a global loss of miRNAexpression at 2 distinct steps. The first is related to thetransition between HGPIN and localized invasive carci-noma, and the second is associated with the transitionfrom localized to metastatic CaP.

Our data confirm prior reports that indicate a predomi-nant down-regulation of miRNA in neoplastic cells, result-ing in a loss of control of fundamental cellular processesrelated to tumor progression, such as cell proliferation anddifferentiation [11].

MicroRNAs 16, 25, 146a, and 191 exhibited a shift inexpression between HGPIN and localized, invasive adeno-carcinoma, together with miR-143 and 218. Previously,miR-16 was described as underexpressed in CaP by othergroups, and the reconstitution of miR-16 expression levelsled to growth arrest and apoptosis in prostate-cancer celllines [12]. Bcl2 is a well-characterized target for miR-16;it was one of the first to be associated with the develop-ment of neoplasias, namely, B-cell lymphomas and leu-kemias. In prostate tumors, Esquela-Kerscher and Slack[6] found that miRNA-16 is underexpressed in 60% ofcases, a result that has been confirmed by other research-ers [13].

Lin et al., using CaP cell lines, have found that miR-146ais underexpressed in hormone-independent tumors, indicat-

Table 2Mean expression of 14 miRNA in HGPIN, favorable, unfavorable, and metastatic CaP

Variable HGPIN Favorable Unfavorable Metastasis/cell lines

n Mean [95% CI] n Mean [95% CI] n Mean [95% CI] n Mean [95% CI] P value

Mir100 15 108.2 66.3 176.7 14 14.7 7.6 28.3 34 18.0 9.5 34.2 6 8.3 1.1 61.8 0.002Mir143 15 71.1 40.9 123.7 14 1.6 0.6 3.9 34 1.8 0.8 3.9 5 0.2 0.01 2.9 �0.001Mir145 15 5.0 2.5 9.8 14 1.6 0.6 4.3 34 3.6 1.4 9.2 5 0.04 0.003 0.9 0.001Mir146a 15 2.3 1.3 4.2 14 0.1 0.05 0.1 34 0.4 0.2 0.8 6 0.04 0.002 0.9 0.003Mir15a 15 1.9 1.2 3.1 14 0.1 0.05 0.3 34 0.3 0.1 0.7 6 1.3 0.4 3.8 �0.001Mir16 15 2.0 0.6 6.0 14 0.1 0.04 0.1 34 0.3 0.1 0.6 6 0.5 0.1 2.3 �0.001Mir191 15 4.5 2.9 6.9 14 0.4 0.28 0.5 34 0.5 0.3 0.7 6 0.2 0.04 0.9 �0.001Mir199a 15 1.5 0.8 2.8 14 0.2 0.09 0.6 31 1.1 0.4 2.9 6 0.6 0.02 20.3 0.224Mir206 15 2.9 1.3 6.1 14 2.8 0.7 10.2 33 1.2 0.7 2.3 6 0.7 0.05 9.8 0.244Mir21 15 1.9 1.1 3.3 14 0.04 0.02 0.09 34 0.1 0.05 0.3 6 0.3 0.08 0.9 �0.001Mir218 15 112.3 62.2 202.8 14 1.8 0.6 5.6 34 10.9 5.6 21.4 6 0.9 0.04 20.3 �0.001Mir25 15 1.9 1.2 3.1 14 0.2 0.1 0.6 34 0.6 0.3 1.5 6 0.9 0.3 3.8 0.007Mir32 15 6.0 3.5 10.3 14 0.3 0.2 0.4 34 1.1 0.6 2.0 6 6.2 0.4 101.7 0.004MIrLet7c 15 31.8 17.6 57.4 14 3.1 1.7 5.6 34 7.5 3.6 15.7 6 0.9 0.2 4.7 �0.001

3K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations xx (2011) xxx

ing that this underexpression is an important alteration re-lated to advanced prostate carcinoma [14]. Furthermore, Linand colleagues demonstrated that ROCK1 mRNA is a targetof miR-146a. ROCK1 is activated when hyaluronan bindsto its receptor CD168, enhancing cell migration and metas-tasis; therefore, miR-146a should be important in inhibitingthe invasiveness of tumor cells. Our results suggest that theloss of miR-146a expression is an early event related tothe invasive capability of cells characterized in situ asneoplastic.

We observed an up-regulated expression of miR-21 andmiR-191 in HGPIN but found a reduction in expression inlocalized, invasive carcinomas, which persisted until metas-tasis. The miR-21 and miR-191 microRNAs have beenreported as overexpressed in a number of tumor types,including prostate tumors [7] and hepatocarcinomas. Priorreports indicate that miR-21 targets PTEN, an importanttumor suppressor gene related to the negative control ofphosphatidylinositol-(3,4,5)P3 (PI3K). In fact, miR-191 in-hibition was recently demonstrated to decrease tumorgrowth in in vitro and in vivo models of hepatocellularcarcinomas, indicating that miR-191 may be a potentialtherapeutic target [15]. We propose that both miRNAs arerelated to the first steps of the progression of carcinogenesisin the prostate and are important to the invasiveness oftumor cells.

The microRNAs let7c, 143, 145 and 218 are suggested tobe involved in the transition from localized prostate adeno-carcinoma to metastasis. These miRNAs have been consis-tently associated with CaP. Indeed, miR-145 is one of themain miRNAs that has been shown to be down-regulated inCaP [8,9]; it has hundreds of target mRNAs, but it wasexperimentally proven to target the insulin receptor sub-strate-1 (IRS-1) [16]. IRS-1 is the major substrate for both

the insulin receptor (IR) and the insulin-like growth factorreceptor (IGF-IR), both of which are involved in cell pro-liferation, apoptosis inhibition, and cell differentiation. TheIGF axis has been associated with the initiation and pro-gression of CaP. Liao et al. also linked up-regulated IGF-Iand IGF-II expression to high-grade tumors [17]. IGF-IRhas an important role in the metastatic process and is in-volved in cell adhesion, migration, invasion, angiogenesis,and metastatic growth at distant organ sites. After overex-pressing miR-145 in a colorectal cell line, Shi et al. dem-onstrated that growth arrest occurred together with a de-crease in the levels of IRS-1 and IGF-IR protein [18].In colorectal cancer, previous studies have found that tumorcells expressing a dominant-negative form of IGF-IR fail toproduce liver metastases.

The microRNA let-7c was described as underexpressedin a small number of clinical specimens and CaP cell linesby several authors [8,13]. Let-7 is one of the most thor-oughly characterized miRNAs and is part of a family com-posed of multiple members with overlapping functions,which generally involve regulating genes related to cell-cycle control and proliferation [19]. Overexpression of let-7can suppress cancer-cell growth in both in vitro and in vivotumor models [20–23]. Furthermore, let-7 has been shownto negatively regulate RAS [24], HMGA2 [25], c-Myc [26],CDC25A CDK6 and cyclin D2 [27]. Additionally, let-7 hasbeen reported to be underexpressed in lung tumors [28,29]that are associated with shortened survival [30].

MicroRNA-143 is involved in adipocyte differentiationand is reported to be down-regulated in colorectal cancer,B-cell malignancies and bladder cancer. Furthermore, themiR-143 target, RAS, and the levels of the RAS proteinwere significantly decreased after the introduction of miR-143 in experiments with bladder-cancer cells, without alter-

Fig. 1. Graphic representation of microRNAs involved in the progression of prostate carcinogenesis from HGPIN to metastasis. (Color version of figure isavailable online.)

4 K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations xx (2011) xxx

ations in the mRNA levels of H-RAS and K-RAS; thisresult indicates a posttranscriptional level of gene expres-sion control [31].

E6, a protein expressed by high-risk human papilloma-virus 16, has been shown to specifically induce underex-pression of miR-218, and down-regulation of this miRNAwas identified in the progression of cervical cancer fromintraepithelial neoplasia to invasive cancer. Importantly,miR-218 targets laminin 5 �3 (LAMB3). LAMB3 is over-expressed in some cancers and promotes cell migration andtumorigenesis in human keratinocytes [32].

Interestingly, miR-100 was not only expressed by alllesions representative of the progression of prostate car-cinogenesis but also indicated a progressive down-regulation from HGPIN to metastasis. We observed pre-viously that miR-100 could act as an oncomir becauselower levels of its expression are associated with lowerrates of biochemical recurrence [33]. In contrast, down-regulation of miR-100 was previously described in hor-mone-refractory carcinomas [8]. These discrepanciesshould be examined in future studies.

Interestingly, we did not find any significant changein the expression of miRNAs between low-grade, pT2,and high-grade, pT3, carcinomas. It is possible that othermiRNAs are involved in this transition. Recently,Schaefer et al. have shown a correlation between miR-31,miR-96, and miR-205 and the Gleason score as well asbetween miR-125b, miR-205 and miR-222 and tumorstage [34]. Future studies could clarify miRNAs involvedin the transition from pT2 low grade to pT3 high gradeprostate carcinomas.

It is important to address that we have explored fewmiRNAs that could be involved in prostate carcinogenesisand a wider search using microarrays platforms should beinteresting to confirm our findings.

The practical importance of our study is in the iden-tification of possible miRNAs and miRNA-targeted genesinvolved in the progression of prostate carcinogenesis.Our data, together with complementary miRNA thatcould be involved in the different steps of prostate car-cinogenesis, may help the development of diagnostic orprognostic tumor markers as well as the design of newtarget therapies.

References

[1] Vogelstein B, Fearon ER, Hamilton SR, et al. Genetic alterationsduring colorectal-tumor development. N Engl J Med 1988;319:525–32.

[2] Bostwick DG, Pacelli A, Lopez-Beltran A. Molecular biology ofprostatic intraepithelial neoplasia. Prostate 1996;29:117–34.

[3] Tomlins AS, Laxman B, Varambally S, et al. Role of the TM-PRSS2-ERG gene fusion in prostate cancer. Neoplasia 2008;10:177– 88.

[4] Leite KR, Srougi M, Darini E, et al. Telomerase activity in localizedprostate cancer: Correlation with histological parameters. Int BrazJ Urol 2001;27:341–7.

[5] He L, Hannon GJ. Micro-RNAs: Small RNAs with a big role in generegulation. Nat Rev Genetic 2004;5:522–32.

[6] Esquela-Kerscher A, Slack FJ. Oncomirs-microRNAs with role incancer. Nat Rer Cancer 2006;6:259–69.

[7] Volinia S, Calin GA, Liu CG, et al. A miRNA expression signature ofhuman solid tumors defines gene targets. Proc Natl Acad Sci U S A2006;103:2257–61.

[8] Porkka KP, Pfeiffer MJ, Waltering KK, et al. MiRNA expressionprofiling in prostate cancer. Cancer Res 2007;67:6130–5.

[9] Ambs S, Prueitt RL, Yi M, et al. Genomic profiling of microRNA andmRNA reveals deregulated microRNA expression in prostate cancer.Cancer Res 2008;68:6162–70.

[10] Livak KJ, Schmittgen TD. Analysus IF relative gene expression datausing real-time quantitative PCR and the 2(–��CT) method. Meth-ods 2001;25:402–8.

[11] Lu J, Getz G, Miska EA, et al. MicroRNA expression profiles classifyhuman cancers. Nature 2005;435:834–8.

[12] Bonci D, Coppola V, Musumeci M, et al. The miR-15a-miR16-1cluster controls prostate cancer by targeting multiple oncogenic ac-tivities. Nat Med 2008;14:1271–7.

[13] Ozen M, Creighton CJ, Ozdemir M, et al. Widespread deregulation ofmicroRNA expression in human prostate cancer. Oncogene 2008;27:1788–93.

[14] Lin S-L, Chiang A, Chang D, et al. Loss of mir-146a function inhormone-refractory prostate cancer. RNA 2008;14:417–24.

[15] Elyakim E, Sitbon E, Faerman A, et al. Hsa-miR-191 is a candidateoncogene target for hepatocellular carcinoma therapy. Cancer Res2010;70:8077–87.

[16] La Rocca G, Badin M, Shi B, et al. Mechanism of growth inhibitionby MicroRNA 145: The role of the IGF-I receptor signaling pathway.J Cell Physiol 2009;220:485–91.

[17] Liao Y, Abel U, Grobholz R, et al. Up-regulation of insulin-likegrowth factor axis components in human primary prostate cancercorrelates with tumor grade. Hum Pathol 2005;36:1186–96.

[18] Shi B, Sepp-Lorenzino L, Prisco M, et al. Micro RNA 145 targets theinsulin receptor substrate-1 and inhibits the growth of colon cancercells. J Biol Chem 2007;282:32582–90.

[19] Reinhart BJ, Slack FJ, Basson M, et al. The 21-nucleotide let-7 RNAregulates developmental timing in Caenorhabditis elegans. Nature2000;403:901–6.

[20] Kumar MS, Erkeland SJ, Pester RE, et al. Suppression of non-smallcell lung tumor development by the let-7 microRNA family. ProcNatl Acad Sci U S A 2008;105:3903–8.

[21] Esquela-Kerscher A, Trang P, Wiggins JF, et al. The let-7 microRNAreduces tumor growth in mouse models of lung cancer. Cell Cycle2008;7:759–64.

[22] Slack FJ, Basson M, Liu Z, et al. The lin-41 RBCC gene acts in theC. elegans heterochronic pathway between the let-7 regulatoryRNA and the LIN-29 transcription factor. Mol Cell 2000;5:659 – 69.

[23] Lin SY, Johnson SM, Abraham M, et al. The C. elegans hunchbackhomolog, hbl-1, controls temporal patterning and is a probablemicroRNA target. Dev Cell 2003;4:639–50.

[24] Johnson SM, Grosshans H, Shingara J, et al. RAS is regulated by thelet-7 microRNA family. Cell 2005;120:635–47.

[25] Lee YS, Dutta A. The tumor suppressor microRNA let-7 represses theHMGA2 oncogene. Genes Dev 2007;21:1025–30.

[26] Sampson VB, Rong NH, Han J, et al. MicroRNA let-7a down-regulates MYC and reverts MYC-induced growth in Burkitt lym-phoma cells. Cancer Res 2007;67:9762–70.

[27] Johnson CD, Esquela-Kerscher A, Stefani G, et al. The let-7 mi-croRNA represses cell proliferation pathways in human cells. CancerRes 2007;67:7713–22.

[28] Johnson SM, Grosshans H, Shingara J, et al. RAS is regulated by thelet-7 microRNA family. Cell 2005;120:635–47.

5K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations xx (2011) xxx

[29] Takamizawa J, Konishi H, Yanagisawa K, et al. Reduced expression ofthe let-7 microRNAs in human lung cancers in association with short-ened postoperative survival. Cancer Res 2004;64:3753–6.

[30] Takamizawa J, Konishi H, Yanagisawa K, et al. Reduced expres-sion of the let-7 microRNAs in human lung cancers in associationwith shortened postoperative survival. Cancer Res 2004;64:3753– 6.

[31] Lin T, Dong W, Huang J, et al. MicroRNA-143 as a tumor suppressorfor bladder cancer. J Urol 2009;181:1372–80.

[32] Martinez I, Gardiner AS, Board KF, et al. Human papilloma virustype 16 reduces the expression of microRNA-218 in cervical carci-noma cells. Oncogene 2008;27:2575–82.

[33] Leite KRM, Tomiyama A, Reis ST, et al. Micro RNA 100 expressionis independently related to biochemical recurrence in prostate cancer.J Urol 2011;185:1118–22.

[34] Schaefer A, Jung M, Mollenkopf H-J, et al. Diagnostic and prognosticimplication of microRNA profiling in prostate carcinoma. Int J Can-cer 2010;126:1166–76.

6 K.R.M. Leite et al. / Urologic Oncology: Seminars and Original Investigations xx (2011) xxx