Embed Size (px)
Citation preview
MARTA RIBEIRO JOSÉ CARDOSO
THE ROLE OF HOXB13 IN PROSTATE CANCER
Dissertação de Candidatura ao grau de Mestre em Oncologia
Molecular submetida ao Instituto de Ciências Biomédicas de
Abel Salazar da Universidade do Porto.
Orientador:
Professor Doutor Manuel Rodrigues Teixeira
Professor Catedrático Convidado com “Agregação”
Departamento de Patologia e Imunologia Molecular do
Instituto de Ciências Biomédicas de Abel Salazar da
Universidade do Porto.
Director do Departamento de Genética e do Centro de
Investigação do Instituto Português de Oncologia – Porto.
Co-orientador:
Doutora Paula Cristina Martins dos Santos Paulo
Investigadora de Pós-Doutoramento no Centro de
Investigação do Instituto Português de Oncologia – Porto.
“Learn from yesterday, live for today, hope for tomorrow.
The important thing is to not stop questioning.”
Albert Einstein
Acknowledgements
VII
Em primeiro lugar gostava de expressar a minha profunda gratidão ao Professor
Manuel Teixeira, por me ter recebido no seu grupo de investigação e por me ter orientado
durante esta Tese de Mestrado. Muito obrigada pelo seu apoio, orientação e
disponibilidade.
À Professora Berta Martins, directora do Mestrado em Oncologia, bem como a
todos os professores e pessoas que colaboraram neste programa, por tudo o que me
ensinaram e pela experiência e conhecimento que partilharam.
À Liga Portuguesa Contra o Cancro, pelo apoio financeiro.
Um agradecimento muito especial à minha co-orientadora, Paula Paulo, que
esteve sempre presente durante esta jornada. Obrigada pelo conhecimento que
partilhaste comigo e pelas longas horas que despendeste para que este trabalho fosse
possível. Estarei eternamente grata pelo teu encorajamento, apoio, motivação e acima de
tudo paciência.
Ao Pedro Pinto, por tudo o que me ensinou sobre sequenciação de Sanger e pela
imensa ajuda durante a primeira fase deste projecto. Às meninas da genética, Manela,
Anita, Carla, Catarina, Patrícia e Isabel, por todas as sugestões e conselhos partilhados.
A todo o grupo da Genética do Cancro, Joana, Diogo, Sofia e Diana, pela
excelente recepção e constante apoio. Um agradecimento especial à Joana e à Diana,
pelo conhecimento e ajuda na cultura de células.
Aos meus colegas de mestrado, Maria, Sílvia e Miguel, pelo enorme
companheirismo e partilha durante esta viagem. Muito obrigada por todos os momentos
de riso e descontracção que me proporcionaram. Obrigado ao Tiago, por se ter juntado a
este grupo e estar sempre disponível para ouvir desabafos e frustrações. Obrigada pela
vossa amizade.
A minha maior gratidão vai para a minha família.
Aos meus pais, por me apoiarem sempre a cada passo e a cada escolha, por
estarem presentes nos meus momentos de queda. Obrigado por me ensinarem que tudo
é possível e que eu posso ser quem quiser.
VIII
Ao Zé, pela presença constante e amor incondicional, mesmo nas horas de
silêncio. Obrigado pelo teu apoio diário e por toda a paciência durante este ano. Obrigado
por nunca teres duvidado das minhas capacidades e por me encorajares a seguir em
frente e a lutar pelos meus objectivos. Obrigado por me mostrares como ser uma pessoa
melhor e feliz.
À minha irmã e ao Pedro, por me ensinarem a viver um dia de cada vez e por toda
a ajuda e sábios conselhos nas alturas mais cruciais da minha vida.
À minha avó Fernanda, pelo enorme orgulho que sente por mim e pela força que
me transmite.
À minha avó Júlia e ao meu avô Narciso, que embora já não estejam presentes
serão sempre uma fonte de inspiração. Este trabalho é dedicado a vocês.
Relevant Abbreviations
XI
A Adenine
ACTβ Actin β
AD Androgen-dependent
ADT Androgen-deprivation therapy
AFMS Anterior Fibromuscular Stroma
AI Androgen-independent
AJCC American Joint Comitee on Cancer
Ala Alanine
AR Androgen receptor
Arg Arginine
AS Active surveillance
Asp Aspartic acid
BCA Bicinchoninic acid
BLASTn Nucleotide Basic Local Alignment Search Tool
BPH Benign Prostatic Hyperplasia
BRCA1 Breast cancer 1, early onset
BRCA2 Breast cancer 2, early onset
BSA Bovine serum albumin
C Cytosin
CDK2 Cyclin dependent kinase 2
cDNA Complementary DNA
CGH Comparative Genomic Hybridization
CHK2 Checkpoint kinase 2
COSMIC Catalogue of somatic mutations in cancer
CPs Cystoprostatectomies
Ct Threshold cycle
CYP17A1 Cytochrome P450, family 17, subfamily A, polypeptide 1
Cys Cysteine
dbSNP Single Nucleotide Polymorphism Database
ddH2O Double-distilled water
ddNTPs Dideoxy nucleoside triphosphates
DMSO Dimethylsulfoxide
DNA Deoxyribonucleic acid
dNTPs Deoxynucleoside triphosphates
DRE Digital rectal examination
E2F E2F transcription factor
EGFR Epidermal growth factor receptor
XII
ELAC2 elaC homolog 2
EPHB2 Ephrin receptor B2
ERG v-ets avian erythroblastosis virus E26 oncogene homolog
ETS E-twenty six
ETV1 ETS variant 1
ETV4 ETS variant 4
ETV5 ETS variant 5
FDA Food and Drug Administration
FISH Fluorescence In Situ Hybridization
FLI1 Friend leukemia virus integration 1
FNA Fine-needle aspiration
FOXA1 Forkhead box A1
FSH Follicle-stimulating hormone
G Guanine
G418 Geneticin selective antibiotic
Gly Glycine
Glu Glutamic acid
GnRH Gonadotropin-releasing hormone
GSTP1 Glutathione S-transferase pi 1
GUSB Glucuronidase beta
GWAS Genome-wide association studies
HGPIN High-grade prostatic intraepithelial neoplasia
HGVS Human Genome Variant Society
HOX Homeobox gene
KLF5 Kruppel-like factor 5
KLF6 Kruppel-like factor 6
Leu Leucine
LB Luria-Bertani
LGPIN Low-grade prostatic intraepithelial neoplasia
LH Luteinizing hormone
MEIS Myeloid ecotropic insertion site
MIPOL1 Mirror-image polydactyly 1
mRNA Messenger RNA
MSR1 Macrophage scavenger receptor 1
MTT 3-(4,5-dimethylthiazollyl-2)- 2,5-diphenyltetrazolium bromide
MYC Myc myelocytomatosis viral oncogene homolog
NCBI National Center for Biotechnology Information
XIII
NKX3-1 NK3 homeobox 1
NPTs Normal prostate tissues
Oligo(dT)18 Synthetic single-stranded 18-mer oligonucleotide
ORF Opening reading frame
OS Overall survival
p12 Cyclin-dependent kinase inhibitor 2A protein
p21 Cyclin-dependent kinase inhibitor 1A protein
p25 Cyclin-dependent kinase 5 protein
p27 Cyclin-dependent kinase inhibitor 1B protein
PBS Phosphate buffered saline
PCa Prostate cancer
PCR Polymerase chain reaction
PI3K Phosphatidylinositol-4,5-bisphosphate 3-kinase
PIA Proliferative inflammatory atrophy
PIN Prostatic intraepithelial neoplasia
PLXNB1 Plexin B1
Polyphen-2 Polymorphism Phenotyping v2
Pro Proline
PSA Prostate-specific antigen
PTEN Phosphatase and tensin homolog
qRT-PCR Quantitative reverse-transcription PCR
R1881 Methyltrienolone (synthetic androgen)
RASSF1A Ras association domain family member 1
RB Retinoblastoma
RMA Robust Multi-array Average
RNASEL Ribonuclease L
RP Radical prostatectomy
SDS Sodium Dodecyl Sulfate
SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis
SLC45A3 Solute carrier family 45, member 3
SNPs Single nucleotide polymorphisms
SPOP Speckle-type POZ
SRD5A2 Steroid 5-alpha reductase 2
T Tymine
TALE Three amino acid loop extension
TBS Tris buffered saline
TCF-4 Transcription factor 4
XIV
TMPRSS2 Transmembrane protease, serine 2
TNM Tumor, node and metastasis
TP53 Tumor protein p53
TRUS Transrectal ultrasound imaging
TSG Tumor suppressor gene
Tyr Tyrosine
UICC Union Internationale Contre le Cancer
WT Wild-type
Summary
XVII
BACKGROUND: Worldwide, prostate cancer is the second most frequent cancer in
males, and the sixth cause of cancer related deaths. Many genetic and epigenetic
alterations have already been described, although it is still difficult to differentiate a
clinically insignificant PCa from an aggressive one, meaning that the exact mechanisms
surrounding proliferation, differentiation and tumor progression still need to be explored.
Recently, germline mutations in HOXB13, a homeobox transcription factor gene that is
important in prostate development, were discovered, being HOXB13 p.Gly84Glu (G84E)
the first major genetic mutation associated with familial PCa.
AIMS: The main goal of this Master´s Thesis was to search for mutations in the
HOXB13 gene at the somatic level, and eventually in the germline, and to gain insights
about its role in prostate carcinogenesis by using in vitro models. In addition, we sought to
characterize HOXB13 mRNA expression pattern in normal prostate tissues and in
prostate carcinomas belonging to different molecular subgroups.
METHODOLOGY: To search for mutations in HOXB13, we performed Sanger
sequencing in 178 primary tumor samples obtained from prostatectomies, from a
consecutive series of 200 clinically localized PCa. To investigate the potential oncogenic
role of HOXB13 and of its mutated forms (the previously described G84E and the new
mutation, discovered after Sanger sequencing), we induced de novo overexpression of
these three forms in two prostate cell lines, 22Rv1 and PNT2, by transfecting these cells
with three expression vectors. The wild-type HOXB13 expressing vector was acquired,
and the mutated vectors were created by inducing site-directed mutagenesis in the wild-
type vector. Functional studies were then performed to evaluate cell proliferation and cell
apoptosis in the PNT2-derived cell populations. To evaluate HOXB13 mRNA expression
levels in NPTs and in prostate carcinomas, we used data obtained from exon-level
expression microarrays, which had already been performed by our group.
RESULTS AND DISCUSSION: We found a novel nonsynonymous mutation,
p.Ala128Asp (A128D), in the HOXB13 gene, initially in tumor tissue and then also in a
blood sample of the patient. This novel germline mutation seems to be damaging to
protein function, emphasizing that HOXB13 is involved in prostate carcinogenesis. By
performing phenotypic assays we observed that HOXB13 stimulates cell growth and that
one of its mutated forms, A128D, causes a decrease in apoptotic rates in PNT2-cells, in
the absence of androgens. Both observations suggest a possible oncogenic role for
HOXB13 in prostate carcinogenesis. We also report that HOXB13 is particularly
XVIII
overexpressed in prostate carcinomas with ETV1 rearrangements, and further studies are
needed to study this correlation.
Resumo
XXI
INTRODUÇÃO: O cancro da próstata é o segundo tipo de cancro mais frequente
nos homens em todo o mundo, e a sexta causa de morte por cancro. Já foram descritas
várias alterações genéticas e epigenéticas, no entanto continua a ser difícil distinguir um
carcinoma da próstata clinicamente insignificante de um agressivo, o que significa que os
mecanismos exactos de proliferação, diferenciação e progressão tumoral ainda precisam
de ser explorados. Recentemente foram descritas mutações germinativas no gene
HOXB13, um factor de transcrição homeobox importante no desenvolvimento da próstata,
sendo a variante p.Gly84Glu (G84E) a primeira grande mutação genética a ser associada
a cancro da próstata familiar.
OBJECTIVOS: O principal objectivo desta Tese de Mestrado consistiu em procurar
mutações no gene HOXB13 a nível somático, e eventualmente germinativo, e tentar
perceber o seu papel na carcinogénese prostática através do uso de modelos in vitro.
Adicionalmente, pretendíamos caracterizar o padrão de expressão do mRNA do HOXB13
em tecidos normais prostáticos e em carcinomas prostáticos pertencentes a diferentes
subgrupos moleculares.
METODOLOGIA: Para a pesquisa de mutações no HOXB13, realizámos
sequenciação de Sanger em 178 amostras de tumores primários obtidos através de
prostatectomias, de uma série consecutiva de 200 carcinomas prostáticos clinicamente
localizados. Para investigar um eventual potencial oncogénico do HOXB13, bem como
das suas formas mutadas (a já descrita G84E e a nova mutação, descoberta após
sequenciação de Sanger), induzimos a sobre-expressão de novo destas três formas em
duas linhas celulares prostáticas, 22Rv1 e PNT2, através da transfeção destas células
com três vectores de expressão. O vector de expressão para a forma nativa do HOXB13
foi adquirido, e os vectores mutados foram criados através de mutagénese dirigida no
vector original. Estudos funcionais foram então realizados para avaliar a proliferação
celular e a apoptose nas populações celulares derivadas da PNT2. Para avaliar os níveis
de expressão do mRNA do HOXB13 em tecidos normais prostáticos e em carcinomas
prostáticos, usámos dados obtidos de microarrays de expressão ao nível exónico, que já
tinham sido realizados pelo nosso grupo.
RESULTADOS E DISCUSSÃO: Encontrámos uma nova mutação não sinónima,
p.Ala128Asp (A128D), no gene HOXB13, inicialmente em tecido tumoral e depois
também numa amostra de sangue periférico do paciente. Esta nova mutação germinativa
parece ser prejudicial para a função da proteína, dando ênfase ao facto de que o
XXII
HOXB13 está envolvido na carcinogénese prostática. Através da realização de ensaios
fenotípicos observámos que o HOXB13 estimula o crescimento celular e que uma das
suas formas mutadas, A128D, causa uma diminuição da apoptose nas células PNT2, na
ausência de androgénios. Estas duas observações sugerem um possível papel
oncogénico para o HOXB13 na carcinogénese prostática. Também reportamos que este
gene está particularmente sobre-expresso em carcinomas prostáticos com rearranjos
envolvendo o ETV1, e são necessários mais estudos para perceber esta correlação.
.
Index
Acknowledgements ............................................................................................... V
Relevant Abbreviations ........................................................................................ IX
Summary ............................................................................................................. XV
Resumo .............................................................................................................. XIX
Index ................................................................................................................ XXIII
Figure Index .................................................................................................... XXIX
Table Index ................................................................................................... XXXIII
Introduction ........................................................................................................... 1
1. Global Burden of Cancer .......................................................................... 3
2. The Prostate ............................................................................................ 3
2.1 Anatomy and Histology of the Prostatic Gland .......................................... 3
2.2 Pathology of the prostatic gland ............................................................... 5
3. Prostate Cancer ....................................................................................... 6
3.1 Epidemiology ............................................................................................ 6
3.2 Etiology .................................................................................................... 8
3.3 Screening ................................................................................................. 9
3.4 Diagnosis ................................................................................................11
3.4.1 Histopathological evaluation .............................................................11
3.4.2 Clinical and pathological staging .......................................................12
3.5 Treatment ................................................................................................15
3.5.1 Active surveillance (AS) ....................................................................15
3.5.2 Radical prostatectomy (RP) ..............................................................16
3.5.3 Radiation Therapy.............................................................................16
3.5.4 Hormonal therapy (androgen-deprivation) .........................................16
3.5.5 Chemotherapy ..................................................................................16
3.6 Prostate Cancer Genetic Alterations ........................................................17
3.6.1 Germline mutations and polymorphisms ...........................................17
3.6.2 Somatic gene alterations ..................................................................18
Somatic mutations ..................................................................................18
Copy number alterations ........................................................................20
Chromosomal rearrangements ...............................................................20
3.7 Prostate Cancer Epigenetic Alterations ...................................................21
3.8 The role of androgens .............................................................................21
4. Homeobox Genes ...................................................................................22
4.1 HOXB13 ..................................................................................................24
4.1.1 HOXB13 and PCa .............................................................................25
4.1.2 HOXB13 mutations and PCa risk ......................................................28
Aims .....................................................................................................................31
Materials and Methods .........................................................................................35
1. Search for somatic mutations in HOXB13 ...............................................37
1.1 Samples ..................................................................................................37
1.2 Sanger sequencing .................................................................................37
2. Cell Lines and Cell Culture ......................................................................39
3. Induction of de novo overexpression of HOXB13 in prostate cell lines ....41
3.1 Selection of the study cell model .............................................................41
3.1.1 RNA and protein extraction from prostate cell lines ...........................42
3.1.2 Complementary DNA (cDNA) synthesis ............................................43
3.1.3 Quantitative reverse transcription PCR (qRT-PCR)...........................43
3.1.4 Immunoblotting .................................................................................45
3.2 Construction of the HOXB13 expression vectors .....................................46
3.3 Transfection and selection of the transfected cells ..................................50
4. Phenotypic assays ..................................................................................52
4.1 Cell proliferation assay: MTT colorimetric assay ......................................52
4.2 Cell apoptosis assay ...............................................................................53
5. Exon-array data to compare HOXB13 mRNA expression in PCa samples
from different molecular subtypes .................................................................................55
Results .................................................................................................................57
1. HOXB13 Mutation Analysis .....................................................................59
1.1 Primary prostate carcinomas ...................................................................59
1.2 Prostate cell lines ....................................................................................60
2. Induction of de novo overexpression of HOXB13 in prostate cell lines ....61
2.1 Selection of the study cell model .............................................................61
2.2 Confirmation of site-directed mutagenesis in the HOXB13 expressing
vectors ......................................................................................................................62
3. Analysis of HOXB13 overexpression in transfected cell lines ..................63
4. Biological role of HOXB13 – phenotypic assays ......................................65
4.1 Cell proliferation assay ............................................................................65
4.2 Cell apoptosis assay ...............................................................................66
5. HOXB13 mRNA Expression Pattern by Molecular Subtype .....................67
Discussion ............................................................................................................69
1. HOXB13 mutation analysis in primary prostate carcinomas ....................71
2. Biological role of HOXB13 .......................................................................72
2.1 Selection of the study cell model .............................................................72
2.2 Phenotypic assays ..................................................................................73
3. HOXB13 mRNA Expression Pattern by Molecular Subtype .....................74
Conclusions and Future Perspectives ..................................................................77
References ...........................................................................................................81
Figure Index
Figure 1 - Schematic representation of the sagittal view of the prostate, showing
anatomical subdivisions .................................................................................................... 4
Figure 2 - Evolution from normal epithelium to carcinoma, passing by LGPIN and
HGPIN .............................................................................................................................. 5
Figure 3 - Estimated new cancer cases and deaths in the world, by cancer sites .. 6
Figure 4 - Age-standardized prostate cancer incidence and mortality rates (W) per
100,000 in selected world areas. ...................................................................................... 7
Figure 5 - Estimated cancer incidence and mortality in Portugal in 2008 ............... 8
Figure 6 - Modified Gleason system . ...................................................................11
Figure 7 - Primary treatments for PCa, by age in 2008 . .......................................15
Figure 8 - General structure of the HOX genes and their protein products ...........23
Figure 9 - Genomic organization of the mammalian HOX clusters and their
embryonic expression pattern, which correlates with the chromosomal positioning .........24
Figure 10 - Expression of HOXB13 in adult tissues using real time reverse
transcription-PCR combined with Taqman probe .............................................................25
Figure 11 - HOXB13 mRNA levels in the prostate (last column) and in ten prostate
cell lines according to the status of the activity of the AR ................................................26
Figure 12 - Protein levels of HOXB13 expression in seven prostatic cell lines by
Western Blot (A) and immunocytochemistry (B) ..............................................................27
Figure 13 - Proposed models of HOXB13 dual activity in the presence (A) and in
the absence (B) of androgens .........................................................................................28
Figure 14 - HOXB13 expression in androgen-dependent (AD) and androgen-
independent (AI) tumors .................................................................................................28
Figure 15 - Location of all the missense mutations found in HOXB13 by Ewing and
collaborators ...................................................................................................................30
Figure 16 - TaqMan probe principle ......................................................................44
Figure 17 - Immunoblotting procedure . ................................................................46
Figure 18 - Plasmid map of the Myc-DDK-tagged ORF clone of Homo sapiens
homeobox B13 (HOXB13) as transfection-ready DNA . ...................................................46
Figure 19 - Three-step procedure of the Site-Directed Mutagenesis method .......48
Figure 20 - Transformation of competent cells with the desired plasmids . ...........50
Figure 21 - Schematic representation of the transfection strategy. .......................51
Figure 22 - Schematic representation of the MTT colorimetric assay to quantify cell
viability. ............................................................................................................................53
Figure 23 - Schematic representation of the asymmetric phospholipd composition
of a viable and an apoptotic mammalian cell membrane .................................................54
Figure 24 - DNA sequence chromatogram obtained from PCa tissue from the
patient with the HOXB13 c.383C>A, p.Ala128Asp variant. ...............................................59
Figure 25 - Scores of the predicted effect of HOXB13 c.383C>A, p.Ala128Asp
variant in protein function obtained with the two PolyPhen datasets (HumVar, above and
HumDiv, below) ..............................................................................................................60
Figure 26 - DNA sequence chromatogram obtained from the PCa cell line LNCaP
........................................................................................................................................61
Figure 27 - HOXB13 expression levels in eight prostate cell lines using quantitative
real-time PCR. .................................................................................................................61
Figure 28 - Expression of HOXB13 in eight prostate cell lines, evaluated by
Western blot. ...................................................................................................................62
Figure 29 - DNA sequence chromatogram obtained from the vector with the G84E
mutation. ..........................................................................................................................63
Figure 30 - DNA sequence chromatogram obtained from the vector with the
A128D mutation. ..............................................................................................................63
Figure 31 - HOXB13 expression levels in 22Rv1 transfected cells, normalized for
the correspondent control vector, using quantitative real-time PCR. ................................63
Figure 32 - HOXB13 expression levels in PNT2 transfected cell populations,
normalized for the correspondent control vector, using quantitative real-time PCR. .........64
Figure 33 - HOXB13 expression levels in the selected PNT2 transfected cells,
normalized for the control vector, using quantitative real-time PCR. ................................65
Figure 34 - Quantification of cell viability in PNT2- derived cells (p-Entry, HOXB13-
WT, HOXB13-G84E and HOXB13-A128D) evaluated at two different time-points (72 and
96 hours) .........................................................................................................................66
Figure 35 - Quantification of apoptotic rates in PNT2 derived cells (HOXB13-WT,
HOXB13-G84E, HOXB13-A128D and in the control vector – p-Entry), evaluated after 96h
in culture ..........................................................................................................................67
Figure 36 - Boxplot representing HOXB13 mRNA expression in NPTs and in
prostate carcinomas ........................................................................................................68
Figure 37 - Boxplot representing HOXB13 expression in the different PCa ETS
subgroups and in normal mucosa obtained from CPs. .....................................................68
Table Index
Table 1 - Dietary factors that are thought to reduce PCa risk ............................... 9
Table 2 - Gleason patterns and their main features ..............................................12
Table 3 - AJCC clinical and pathological TNM classification of prostatic tumors .. 13
Table 4 - AJCC stage grouping . ..........................................................................14
Table 5 - PCa predisposition loci discovered by linkage analysis ........................18
Table 6 - PCa genes with somatic mutations ........................................................19
Table 7 – Primers´ sequences used for sequencing the HOXB13 gene ..............37
Table 8 – Principal characteristics of the eight prostate cell lines .........................39
Introduction
3
1. Global Burden of Cancer
Some years ago, heart diseases and pneumonia constituted the leading causes of
death worldwide, however, with the improvement of diet and health conditions of the
populations, these pathologies started to decrease. More recently, cancer, known as the
XXIst century disease, rose to the first and the second positions, in developed and in
developing countries, respectively, concerning causes of death (Jemal et al., 2011). Still,
incidence rates in developed countries are twice those found in developing countries, but
the mortality rates are alike. This is due to a late diagnosis and a lack of access to
appropriate treatment in the latter. Nevertheless, incidence rates in developing world are
also increasing, which is in part explained by the growth and aging of the population,
along with an increase of unhealthy choices and behaviors, like sedentary, smoking and
bad diet habits, that are typical of the western and industrialized countries (Ferlay et al.,
2010; Jemal et al., 2011).
According to the GLOBOCAN statistics for 2008, 12.7 million cancer cases and 7.6
million cancer related deaths are estimated to have occurred in the world, and 56% of
these cases and 63% of these deaths occurred in developing countries. It is to note that a
substantial amount of cancer cases and deaths could be avoided through the adoption of
healthier habits, vaccination, and ultimately, with early detection and appropriate
treatment (Ferlay et al., 2010; Jemal et al., 2011).
2. The Prostate
2.1 Anatomy and Histology of the Prostatic Gland
The prostate gland is a male accessory organ with a size of a walnut that reaches
the weight of approximately 20g by the age of 25 to 30 years. It resembles an inverted
pyramid with an apex and a basis, and resides between the pelvic floor and the urinary
bladder. In the periphery, the prostate is surrounded by a thin layer of connective tissue
that forms the “true” capsule, and in the outside the pelvic fascia forms the “false” capsule
(Dixon, 1999; Hammerich, 2009; McNeal, 1988). The prostatic gland is crucial for the
secretion of fluids needed for semen nutrition (Joshua et al., 2008; Shen and Abate-Shen,
2010; Verma and Rajesh, 2011). Even though the human prostate does not present a true
lobular architecture, John E. McNeal developed a work through many years (McNeal,
1969, 1981, 1988; Shen and Abate-Shen, 2010) in which he described a zonal
architecture model, which is still widely accepted and used in the clinical practice (Dixon,
1999; Selman, 2011; Shen and Abate-Shen, 2010). He described four distinct anatomic
zones, which are represented in Figure 1.
4
Figure 1 - Schematic representation of the sagittal view of the prostate, showing anatomical
subdivisions [adapted from (Verma and Rajesh, 2011)].
There are three glandular zones, which are the peripheral, the central and the
transition zones (Dixon, 1999; Zhai et al., 2010). The peripheral zone makes up 70% of
the volume of the glandular part of the prostate, comprising the prostatic glandular tissue
at the apex and near the capsule, and its ducts, that open into the distal urethra to the
verumontanum. Certain pathologies like chronic prostatitis, postinflammatory atrophy and
carcinoma are more frequently observed in this zone. The central zone has the shape of a
cone that surrounds the ejaculatory ducts and represents approximately 25% of the gland.
Lastly, the transition zone consists in two small lobes of tissue surrounding the lateral and
the anterior parts of the proximal urethra, occupying 5% of the prostate gland (Amin,
2010; Dixon, 1999; Hammerich, 2009; Shen and Abate-Shen, 2010; Verma and Rajesh,
2011; Zhai et al., 2010). Besides the glandular part of the prostate, McNeal also described
a non-glandular zone, called anterior fibromuscular stroma (AFMS) that embraces about
one-third of the entire gland. It is composed of smooth muscle and fibrous tissue and
forms the convexity of the anterior external part of the gland. This structure is important for
both voluntary and involuntary sphincter functions (Amin, 2010; Dixon, 1999; Hammerich,
2009; Verma and Rajesh, 2011; Zhai et al., 2010).
The prostatic gland is mainly constituted by three epithelial cell types, depending
on the anatomical place. The prostatic glands and ducts are covered by secretory cells
that are important for the normal secretory function of the prostate. Near the basal part of
the secretory cells there is a basement membrane that is coated by basal cells. Between
these two, lies an intermediate type of cells called transient amplifying. The last type of
cells is of neuroendocrine kind and is responsible for the regulation of cell growth and
secretory activity (Joshua et al., 2008; Shen and Abate-Shen, 2010).
5
2.2 Pathology of the prostatic gland
Prostate diseases are chronic and progressive and need several years to develop.
Benign prostatic hyperplasia (BPH) is a hyperplastic process, in which there is an
increased number of epithelial and stromal cells and therefore prostatic enlargement,
which results in symptoms in the lower urinary tract. BPH arises more frequently in the
transition zone and it is a very common pathology that increases with aging. Although its
etiology is not fully understood, it is known that androgens are required for its
development and that inflammation might also be involved (Roehrborn, 2008; Sciarra et
al., 2008).
The most likely precursor lesion of prostate cancer (PCa) is high-grade prostatic
intraepithelial neoplasia (HGPIN) and this idea emerged because of some resemblances
between this lesion and PCa. These include co-localization in the peripheral zone,
morphologic transitions, molecular and phenotypic similarities and also epidemiological
data. In this lesion, cells already have neoplastic features and line the pre-existing acini
and ducts. Nuclei are also enlarged and hyperchromatic, although variation is observed,
and there is frequently a prominent nucleoli. There is also cell and nuclear crowding. It
was postulated that HGPIN arises from the low-grade form of prostatic intraepithelial
neoplasia (LGPIN), since its cells have a very similar morphology, with the exception of
the prominent nucleoli (De Marzo et al., 2007; Joshua et al., 2008; Putzi and De Marzo,
2000; Shen and Abate-Shen, 2010). A proposed model for the progression from a normal
epithelium to PCa is represented in Figure 2. It takes several years and is accompanied
by different molecular processes and alterations in important pathways characteristic from
each stage (Sciarra et al., 2008; Shen and Abate-Shen, 2010).
Figure 2 - Evolution from normal epithelium to carcinoma, passing by LGPIN and HGPIN [adapted from
(Sciarra et al., 2008)].
In 1999, proliferative inflammatory atrophy (PIA) was proposed by Angelo De
Marzo and his colleagues as being another possible precursor lesion for prostate cancer,
and it is thought to occur earlier than prostatic intraepithelial neoplasia (PIN) lesions (De
Marzo et al., 1999; De Marzo et al., 2003; Deutsch et al., 2004; Putzi and De Marzo,
6
2000). PIA lesions can be found near PIN lesions and near adenocarcinoma, and include
simple atrophy and postatrophic hyperplasia, in association with variable grades of
inflammation and stromal fibrosis (De Marzo et al., 2007; Wagenlehner et al., 2007). The
cells observed in these lesions are highly proliferative and have low apoptotic rates
(Joshua et al., 2008; Nakai and Nonomura, 2013; Sciarra et al., 2008).
Finally, the last step in the pathway represented in Figure 2 is PCa, which was first
described by the Egyptians (Shen and Abate-Shen, 2010). PCa is very heterogeneous
and usually multifocal, and although there are several types of prostate tumors,
adenocarcinoma is the most frequent one, accounting for 95% of all cases (Boyd et al.,
2012; Eble et al., 2004; Shen and Abate-Shen, 2010). PCa rarely present symptoms,
mostly at the initial stages, and the symptoms may be similar to those referred for BPH
(Eble et al., 2004; Hammerich, 2009). The phenotype of adenocarcinoma is typically
luminal, with the glands lacking a basal membrane and a basal cell layer (Chrisofos et al.,
2007; Shen and Abate-Shen, 2010). PCa frequently metastasizes, mainly to the bones,
but also to pleura, lung and liver (Shen and Abate-Shen, 2010).
3. Prostate Cancer
3.1 Epidemiology
PCa is the most frequently diagnosed malignancy of male genital organs (Eble et
al., 2004; Stasiewicz et al., 2012). Worldwide, it occupies the second position of the most
common cancers in males, with 903,500 new cases, and is the sixth cause of cancer
related deaths in men, with 258,400 deaths in 2008 (Figure 3) (Ferlay et al., 2010; Jemal
et al., 2011).
Figure 3 - Estimated new cancer cases and deaths in the world, by cancer sites [adapted from (Jemal et
al., 2011)].
7
Differences concerning incidence and mortality rates between developed and
developing countries are shown in Figure 4.
Figure 4 - Age-standardized prostate cancer incidence and mortality rates (W) per 100,000 in selected
world areas [adapted from GLOBOCAN 2008 available at http://globocan.iarc.fr/factsheet.asp].
Incidence rates are much higher in economically developed countries of Europe,
North America and Oceania (Jemal et al., 2011) and have been rising since the
introduction of prostate-specific antigen (PSA) testing in the early 1990s (Jemal et al.,
2011; Schroder and Roobol, 2012; Wolf et al., 2010). The relevance of PSA screening for
PCa diagnosis will be discussed further. Mortality rates, on the other hand, have been
decreasing in developed regions, which might be related with an improvement of
treatments. (Ferlay et al., 2010; Schroder and Roobol, 2012).
In Portugal, PCa was the most frequently diagnosed cancer in men in 2008, with
5,140 new cases representing 21.4% of all carcinomas, and the third leading cause of
death, with 2,021 deaths making up 13.8% of all cancer-related deaths (Figure 5). These
proportions are very similar to those seen in the rest of Europe (Ferlay et al., 2010).
8
Figure 5 - Estimated cancer incidence and mortality in Portugal in 2008 [adapted from GLOBOCAN 2008
available at http://globocan.iarc.fr/factsheet.asp].
3.2 Etiology
When compared to other types of cancers, for which some risk factors are known
(smoking in lung cancer, for example), the causes of PCa remain mostly unknown
(Gronberg, 2003). As prostate carcinomas are very heterogeneous, both genetic and
environmental factors are expected to contribute for its etiology (Alvarez-Cubero et al.,
2012).
Over the past years, many epidemiological studies have been trying to associate
some risk factors with an increased risk of PCa, however the results remain controversial.
The differences in incidence rates are thought to be explained, in part, by age, ethnicity,
socioeconomic status, residence area, and environmental factors such as work place,
lifestyle and diet (Alvarez-Cubero et al., 2012; Eble et al., 2004). Concerning diet, while a
high intake of fat, animal proteins, processed meats and a high consumption of dairy
foods have been associated with an increased risk of PCa development (Deutsch et al.,
2004; Gronberg, 2003; Nelson et al., 2003), there are several dietary factors that appear
9
to have a protective effect against PCa (Table 1) (Damber, 2008; Deutsch et al., 2004;
Nelson et al., 2003).
Table 1 - Dietary factors that are thought to reduce PCa risk [adapted from (Deutsch et al., 2004)].
Dietary component Role
Selenium Cell-cycle inhibition and apoptosis
Vitamin E Antiproliferative action
Cruciferous vegetables Cell-cycle inhibition
Lycopene (tomatoes) Free radical scavenger
Resveratrol Free radical scavenger and proapoptotic
Although many risk factors have been suggested, only three are fully established,
which are increasing age, ethnic origin and a positive family history of PCa (American
Cancer Society: Cancer Facts & Figures 2012; Damber, 2008; Jemal, Bray, et al. 2011).
In fact, 60% of the PCa cases are diagnosed in men with more than 65 years old,
and about 97% in men with more than 50 years, with the mean age of diagnosis standing
between 72 and 74 years (American Cancer Society: Cancer Facts & Figures 2012;
Gronberg, 2003). This relationship is not entirely understood, but it is presumed that, with
aging, gene expression alterations may occur in the prostate, mainly in genes associated
with inflammation, senescence and oxidative stress (Shen and Abate-Shen, 2010).
Regarding ethnic origin or race, it is known that the highest incident rates are
observed in men with African ancestry, like Jamaicans and African Americans (American
Cancer Society: Cancer Facts & Figures 2012). After black people, the Caucasians are
the most affected, with the Asians having the lowest risk (Eble et al., 2004).
An hereditary component is estimated to be present in 5-10% of all cases as
observed by a clustering of PCa in some families (American Cancer Society: Cancer
Facts & Figures 2012; Damber, 2008; Gronberg, 2003). Men with affected first-degree
relatives have a two to three-fold increased risk for developing PCa when compared to the
general population (Boyd et al., 2012; Schroeck et al., 2013) and the risk increases with a
decrease in the age at diagnosis of the affected individuals and with the increase in the
number of affected relatives (Gronberg, 2003). The aggregation of cases in families might
be explained by the inheritance of mutations in genes that are associated with PCa
(Damber, 2008). Some have already been identified, and will be discussed later.
3.3 Screening
In many countries PSA testing in combination with digital rectal examination (DRE)
are used as screening tools for the early detection of PCa. PSA is a kallikrein-related
10
serine protease that is essentially produced by the epithelial cells from the ducts and acini
of the prostatic gland. Its levels are typically very low, except when a disruption of the
normal prostatic architecture occurs, causing its release to the circulation (Greene et al.,
2013; Shen and Abate-Shen, 2010).
Men with suspicious DRE in conjunction with a PSA level above 4.0ng/ml are
recommended to proceed to biopsy (Alvarez-Cubero et al., 2012; Heidenreich et al., 2008;
Wolf et al., 2010). However, there is not a true cutoff value allowing the discrimination
between cancer and non-cancer, meaning that the PSA test is not specific for cancer, but
only for prostatic tissue. This is the major disadvantage of PSA testing, however there are
no other biomarkers available (Alvarez-Cubero et al., 2012; Wolf et al., 2010). In fact, PSA
levels can be elevated in other conditions besides PCa, such as BPH, prostatitis, and
trauma, giving false-positive results (Alvarez-Cubero et al., 2012; Greene et al., 2013).
PSA test can also produce false-negative results, so clinically important PCas could be
missed (Alvarez-Cubero et al., 2012; Wolf et al., 2010). Despite being widely use, there is
still a big controversy about the real benefits of PSA screening (Boyle and Levin, 2008)
and it was estimated that 23 to 42% of PSA detected prostate cancers would never be
diagnosed otherwise, indicating that PSA testing may lead to overdiagnosis and
overtreatment. The main reason that favors PSA screening is the fact that it detects the
illness at an early stage, allowing an effective and curative treatment, thus reducing
mortality (Wolf et al., 2010). However, there is not enough proof that the reduced mortality
observed in some countries, like in the USA, for example, is attributable to PSA screening
(Damber, 2008; Wolf et al., 2010).
Before proceeding to biopsy many factors should be taken into account. These
factors include patient´s age and ethnicity, family history, biopsy history, comorbidities, as
well as free and total PSA levels, PSA velocity and density (Greene et al., 2013). Current
recommendations say that older men with a life expectancy lower than 10 years are
unlikely to benefit from screening, but men belonging to high risk groups should pursue
with screening. Concerning the general population, men should be informed about the
pros and cons and make their personal decision (Boyle and Levin, 2008; Wolf et al.,
2010).
Transrectal ultrasound imaging (TRUS) is also used as an aid-tool for patients with
prostate problems. It allows the evaluation of the prostate volume and the delineation and
measurement of potential lesions, but it has limited usefulness in the detection of prostate
cancer (Eble et al., 2004).
Nowadays it is still very difficult to distinguish an indolent cancer from an
aggressive one, which will only be possible with a better understanding of the molecular
mechanisms involved in PCa initiation and progression. This knowledge would allow the
11
identification of new biomarkers and new screening methods that could avoid
overtreatment, and would also allow the development of new treatments with fewer risks
(Shen and Abate-Shen, 2010; Wolf et al., 2010).
3.4 Diagnosis
When something suspicious is found with screening techniques, more exams are
needed to confirm whether or not the patient has PCa and the examination of
histopathological or cytological samples is required to make a definitive diagnosis
(Damber, 2008). The gold standard method to obtain a sample is by transrectal
ultrasound-guided core biopsies (Damber, 2008; Eble et al., 2004; Heidenreich et al.,
2008). It is recommended to take between 10 and 13 biopsy cores, which have a higher
rate of detection than the traditional sextant method. Fine needle aspiration (FNA)
cytology is another available technique, but is no longer widely used because of some
disadvantages, like false-positives and the fact that it does not allow the evaluation of the
Gleason grading (Heidenreich et al., 2008).
3.4.1 Histopathological evaluation
Histopathological grading is essential to access the aggressiveness of the tumor,
to predict the outcome and also to choose the appropriate treatment (Buhmeida et al.,
2006; Damber, 2008; Harnden et al., 2007). In 1966 a grading system was created by
Donald F. Gleason and nowadays it is the most widely used system. The Gleason grading
is based on the glandular architectural pattern of the tumor and comprises five grades of
differentiation (Figure 6 and Table 2), with grade one being the most differentiated and
grade five the most undifferentiated (Buhmeida et al., 2006; Eble et al., 2004).
Figure 6 - Modified Gleason system [adapted from (Epstein, 2010)].
12
Table 2 - Gleason patterns and their main features [adapted from (Eble et al., 2004; Epstein, 2010)].
Gleason patterns
Features
Pattern 1
Very well circumscribed nodule of packed glands with
equal size and shape which do not infiltrate.
Pattern 2
Round or oval glands with smooth ends, less uniform
between each other than in pattern 1, wich can have
minimal invasion.
Pattern 3
Discrete angular units with marked variation in shape
and size, with infiltration in and amongst non-
neoplastic acini.
Pattern 4 Fused or cribiform glands, or defined glands with
poorly formed glandular lumina.
Pattern 5
Almost complete loss of glandular lumina with
formation of solid sheets, solid strands or single cells
that invade the stroma.
A grade must be present in at least 5% of the biopsy specimen in order to be
considered (Ramon, 2007). Nevertheless, prostate cancer presents frequently with more
than one histological pattern, so in order to reflect this heterogeneity, the two most
common patterns have to be taken into account (Buhmeida et al., 2006; Eble et al., 2004).
The final score attributed to the tumor consists on the sum of these two patterns, so the
score is always between two and 10, the less and the more aggressive, respectively
(Damber, 2008; Epstein, 2010). The two patterns and the final score are always reported
(3+4=7, for example). Because grade one is unusual and grade two is frequently mixed
with grade three, resulting in score five, some Gleason scores, like two, three and four are
rarely attributed, making scores six and seven the two most common (Eble et al., 2004).
3.4.2 Clinical and pathological staging
PCa staging is used to describe the extent of the disease, in order to predict the
clinical outcome, define treatment options, and evaluate the response to treatment (Cheng
et al., 2012; Ramon, 2007). Staging of prostate cancer is challenging due to the
complexity of the disease, so an accurate and uniform language is required. The most
commonly used system for PCa staging is the tumor, node and metastasis (TNM) system
of the American Joint Committee on Cancer/Union Internationale Contre le Cancer
(AJCC/UICC) (Cheng et al., 2012; Edge and Compton, 2010). This classification system is
13
based on three parameters, which are the extent of the primary tumor (T), the presence of
involved lymph nodes and its extent (N) and, lastly, the presence of distant metastases
(M) (Edge and Compton, 2010; Ramon, 2007).
TNM staging includes both clinical and pathological staging. The first is based on
information obtained before the first treatment, and the assessment of the tumor extent is
determined by TRUS, DRE and other imaging techniques. On the other side, pathological
staging is accessed through histological examination of the tumor extent in the prostate
gland and its surroundings, which can only be properly achieved after total radical
prostatectomy (Cheng et al., 2012; Hoedemaeker et al., 2000).
The TNM system was revised in 2010 (Table 3) and some changes have been
made in the T3a category, with the incorporation of extraprostatic extension and
microscopic bladder neck invasion. Other alterations include the recognition of the
Gleason score as being the ideal grading system, and the inclusion of prognostic factors
of Gleason score and preoperative PSA into stage grouping. There are four groups,
represented in Table 4, in which a higher number indicates a worst prognosis (Cheng et
al., 2012).
Table 3 - AJCC clinical and pathological TNM classification of prostatic tumors [adapted from (Edge,
2010)].
Primary tumor (T)
CLINICAL
TX – Primary tumor cannot be assessed
T0 – No evidence of primary tumor
T1 – Clinically unapparent tumor neither palpable nor visible by imaging
T1a – Tumor incidental histological finding in 5% or less of tissue resected
T1b – Tumor incidental histological finding in more than 5% of tissue resected
T1c – Tumor identified by needle biopsy (for example, because of elevated PSA)
T2 – Tumor confined within prostate
T2a – Tumor involves one-half of one lobe or less
T2b – Tumor involves more than one-half of one lobe but not both lobes
T2c – Tumor involves both lobes
T3 – Tumor extends through the prostate capsule
T3a – Extracapsular extension (unilateral or bilateral)
T3b – Tumor invades seminal vesicle(s)
T4 – Tumor is fixed or invades adjacent structures other than seminal vesicles, such as
external sphincter, rectum, bladder, levator muscles, and/or pelvic wall
PATHOLOGIC (PT)
14
pT2 – Organ confined
pT2a – Unilateral, one-half of one side or less
pT2b – Unilateral, involving more than one-half of side but not both sides
pT2c – Bilateral disease
pT3 – Extraprostatic extension
pT3a – Extraprostatic extension or microscopic invasion of bladder neck
pT3b – Seminal vesical invasion
pT4 – Invasion of the rectum, levator muscles, and/or pelvic wall
Regional Lymph Nodes (N)
CLINICAL
NX – Regional lymph nodes were not assessed
N0 – No regional lymph node metastasis
N1 – Metastasis in regional lymph node(s)
PATHOLOGIC
pNX – Regional nodes not sampled
pN0 – No positive regional nodes
pN1 – Metastases in regional node(s)
Distant Metastasis (M)
M0 – No distant metastasis
M1 – Distant metastasis
M1a – Nonregional lymph node(s)
M1b – Bone(s)
M1c – Other site(s) with or without bone disease
Table 4 - AJCC stage grouping [adapted from (Edge, 2010)].
Group T N M PSA (ng/ml) Gleason
score
I
T1a-c N0 M0 < 10 ≤ 6
T2a N0 M0 < 10 ≤ 6
T1-2a N0 M0 X X
IIA
T1a-c N0 M0 < 20 7
T1a-c N0 M0 ≥ 10 and ≤ 20 ≤ 6
T2a N0 M0 ≥ 10 and ≤ 20 ≤ 6
T2a N0 M0 < 20 7
T2b N0 M0 < 20 ≤ 7
T2b N0 M0 X X
IIB T2c N0 M0 Any PSA Any Gleason
15
T1-2 N0 M0 ≥ 20 Any Gleason
T1-2 N0 M0 Any PSA ≥ 8
III T3a-b N0 M0 Any PSA Any Gleason
IV
T4 N0 M0 Any PSA Any Gleason
Any T N1 M0 Any PSA Any Gleason
Any T Any N M1 Any PSA Any Gleason
*X: unknown.
3.5 Treatment
There are many treatment options available for PCa management and its choice
relies on the stage and grade of the tumor at the time of diagnosis, patient´s age,
comorbidities and personal believes (Heidenreich et al., 2011; Siegel et al., 2012).
Currently, there is not enough information about what is the best treatment, since few
prospective randomized trials comparing different treatment options have been done
(Heidenreich et al., 2011; Ramon, 2007). The most common used treatments are
represented in Figure 7, in function of age.
*Initial treatment received.
Figure 7 - Primary treatments for PCa, by age in 2008 [adapted from (Siegel et al., 2012)].
3.5.1 Active surveillance (AS)
AS or watchful waiting consists in postponing the treatment until it is necessary
(Heidenreich et al., 2008; Ramon, 2007). It was proposed as a form of reducing
overtreatment in patients with low-risk of PCa progression. It is recommended in men at
low-risk groups, with confined disease, Gleason score < 6, PSA levels < 10ng/ml and with
a life expectancy > 10 years. These recommendations are due to the fact that men
belonging to this risk group have a 20-year PCa-specific survival rate of 80 to 90%
16
(Heidenreich et al., 2011), however they should be monitored periodically with PSA testing
and re-biopsies (Ramon, 2007).
3.5.2 Radical prostatectomy (RP)
RP constitutes the only treatment option with curative intent for localized PCa with
a benefit in cancer-specific survival, when compared to watchful waiting (Damber, 2008;
Heidenreich et al., 2011). In men with a normal erectile function and localized disease,
normally a nerve-sparing approach is made. In men with intermediate and high-risk PCa,
a pelvic lymphadenectomy should be performed, while for men with low-risk, its need and
extent are still in debate. Neoadjuvant therapy with androgen deprivation has no value,
since it does not improve overall survival (OS), and recommendations of adjuvant
androgen deprivation therapy (ADT) after surgery, remain controversial. However,
adjuvant radiation therapy after surgery showed to improve biological survival, OS and 15-
year metastasis-free survival (Heidenreich et al., 2011).
3.5.3 Radiation Therapy
Radiation therapy can be delivered using external-beam radiation or
brachytherapy, or used in combination. They have emerged as alternatives for men with
localized PCa, in whom RP is not recommended (Heidenreich et al., 2008). Both forms
have a disease-free survival identical of RP, for patients with early-stage PCa (Damber,
2008; Jani and Hellman, 2003). Dose-escalation radiotherapy can be used in these
patients or can be combined with hormonal therapy (Damber, 2008).
3.5.4 Hormonal therapy (androgen-deprivation)
Since androgens are essential for PCa development and progression (Shen and
Abate-Shen, 2010), hormonal therapy is the gold standard for patients with more
advanced PCa. Testicular androgens can be eliminated through physical (surgical
removal) or chemical castration. The latter can be accomplished with oestrogens that
reduce the secretion of the gonadotropin-releasing-hormone (GnRH) in the hypothalamus;
with the downregulation of GnRH receptors with agonists or antagonists, thus inhibiting
the pituitary secretion of luteinising hormone (LH) or follicle-stimulating hormone (FSH); or
with the administration of antiandrogens (Damber, 2008; Mottet et al., 2011).
3.5.5 Chemotherapy
Chemotherapy is indicated for the more advanced and metastatic PCa and mainly
when androgen deprivation fails, and should be initiated early in order to improve survival.
There are already some combination regimens with good results. The use of docetaxel in
combination with prednisone seems to be the best treatment for these men when
compared to mitoxantrone, with a survival benefit of approximately three months and a
17
better quality of life, with less morbidities (Mottet et al., 2011). Recently, a vaccine,
Sipuleucel-T was approved by the Food and Drug Administration (FDA) and consists in an
autologous cellular immunotherapy that contains autologous CD54+ cells, stimulating the
immune system. Sipuleucel-T has proven to increase the OS in 4,1 months, however its
concomitant use with chemotherapy and immunosuppressive drugs has not been studied
yet (Mottet et al., 2011).
3.6 Prostate Cancer Genetic Alterations
Cancer is a pathology characterized by genetic alterations that can activate
oncogenes or inactivate tumor suppressor genes (TSGs) (Porkka and Visakorpi, 2004).
These can be inherited, thus predisposing to a higher cancer risk, or acquired during the
progression of the disease. Therefore, PCa can be divided in two types: the familial type,
in which the age at diagnosis is general below 55 years old; and sporadic, where the
individual is exposed, at some point, to an external factor that damages his genetic
material (Alvarez-Cubero et al., 2012; Porkka and Visakorpi, 2004).
The genetic alterations associated with PCa that have been discovered so far are
going to be explored in the following text. It is important to reinforce that although many
genetic and epigenetic alterations are already known, the mechanisms underlying
proliferation, differentiation and tumor progression in the prostatic gland await further
exploitation (Jung et al., 2004a).
3.6.1 Germline mutations and polymorphisms
In order to identify potential genes responsible for familial PCa, many approaches
have been used. The most classic approach is linkage analysis that identifies cancer
susceptibility loci. In Table 5 the loci discovered so far are represented, together with the
chromosomal position and its associated gene (Boyd et al., 2012). The most promising
candidate genes for PCa are RNASEL, MSR1 and ELAC2, which have been described as
hereditary TSG (Alvarez-Cubero et al., 2012). These genes are associated with the
response to infections (Gonzalgo and Isaacs, 2003), and thus linked to the idea that PCa
is somehow associated with inflammation.
After the discover of BRCA1, in the 17q21 region, as a candidate gene for prostate
cancer development, this chromossomal region was studied deeper (Boyd et al., 2012;
Ewing et al., 2012). Consequently, in the beginning of 2012, Ewing et al. screened
approximately 200 genes in this region and found a mutation, p.Gly84Glu (G84E to
simplify), in another gene, HOXB13, which is a homeobox transcription factor implicated in
the development of the normal prostate. More recently, genome wide association studies
(GWAS) allowed the discover of many other susceptibility loci (Boyd et al., 2012). Some
single nucleotide polymorphisms (SNPs) in genes associated with androgen synthesis
18
and metabolism, like the androgen receptor gene (AR), the SRD5A2 gene (encodes part
of 5α-reductase), the CYP17A1 gene (encodes the cytochrome P450), among others,
have also been associated with PCa development (Bosland and Mahmoud, 2011; Boyd et
al., 2012; Damber, 2008; Deutsch et al., 2004).
Altough many inherited genetic alterations have been described, the focus of
research remains in high penetrant genes, such as BRCA1/2 and more recently HOXB13,
as rare genomic variants may be used as biomarkers for the prediction of incidence,
prognosis and response to therapy, or even be incorporated in clinical trials in a near
future (Bambury and Gallagher, 2012). However, because they are rare, it has been
suggested that low-penetrance gene polymorphisms might also contribute for PCa
development. New technologies such as GWAS and SNP arrays are very important to
identify variants that have a moderate effect on the risk of PCa development (Boyd et al.,
2012).
Table 5 - PCa predisposition loci discovered by linkage analysis [adapted from (Boyd et al., 2012)].
3.6.2 Somatic gene alterations
Somatic mutations
Like in other malignancies, somatic mutations are relatively common in PCa, with
the activation of oncogenes or the inactivation of TSGs (Boyd et al., 2012; Jung et al.,
19
2004a). The mutated genes that have already been described are represented in Table 6
(Boyd et al., 2012). Mutations in PTEN, a TSG involved in apoptosis, cell-cycle control
and angiogenesis, are the most frequent. Loss of heterozygosity in 10q23 is also
common, especially in metastatic lesions, which suggests that this TSG might be
important for PCa progression (Boyd et al., 2012; Deutsch et al., 2004). TP53 is also a
TSG, being the most frequently mutated gene in all human cancers, and its mutations are
observed in 20 to 30% of advanced localized PCa tumors (Boyd et al., 2012; Porkka and
Visakorpi, 2004). Other genes that are also mutated in some cases are RNASEL (13% of
cases), EPHB2 (10% of cases), and, less frequently, mutations in RAS, CHK2, SPOP,
EGFR, PLXNB1, FOXA1 and many others (Boyd et al., 2012; Porkka and Visakorpi,
2004).
Another gene that is worth to mention is the AR gene, since it codifies the receptor
that mediates the androgen action in prostate cancer cells (Porkka and Visakorpi, 2004).
Many somatic alterations in this gene have been described, although mutations in
untreated PCa tumors are rare. On the other hand, they are frequently found in androgen-
refractory tumors after treatment with anti-androgens and androgen suppression, which
implicates these mutations in the progression of the disease. Seventy missense mutations
have already been identified in the AR gene (Boyd et al., 2012).
Table 6 - PCa genes with somatic mutations [adapted from (Boyd et al., 2012)].
20
Copy number alterations
The two major techniques that allow the identification of copy number gains and
losses are comparative genomic hybridization (CGH) and fluorescence in situ
hybridization (FISH). The first one revealed that gains at 1q, 2p, 7, 8q and Xq and losses
at 1p, 6q, 8p, 10q, 13q and 16q are the most common alterations in PCa at the
chromosome level. Later on, FISH showed candidate genes for some of the alterations:
gains of MYC (8q24) and AR (Xq12); and deletions of NKX3-1 (8p21), KLF6 (10p15),
PTEN (10q23), P27/Kip1 (12p13) and KLF5 (13q21) (Boyd et al., 2012). NKX3-1 loss of
expression is more common in advanced and refractory tumors (near 80%) and it is
thought that the mechanism involved in the loss of the tumor suppressive activity is
haploinsufficiency, since the other allele is not mutated (Jung et al., 2004a; Porkka and
Visakorpi, 2004).
Chromosomal rearrangements
Genetic alterations such as chromosomal rearrangements are common in both
hematological neoplasias and in solid tumors (Huang and Waknitz, 2009). The discovery
of fusion genes in a high proportion of the prostate tumors had major importance to
understand the molecular mechanisms involved in prostate carcinogenesis (Damber,
2008).
In 2005, Tomlins and his colleagues described chromosomal rearrangements
where the gene TMPRSS2 (a serine protease produced in response to androgens in the
epithelial cells) was fused to ERG or ETV1, two members of the ETS family of
transcription factors. This fusion was reported to be present in 50% of the prostate
cancers, making it the most prevalent fusion gene in any solid tumor. As a result of these
fusions, the ETS gene is under the control of the TMPRSS2 promoter region and
becomes overexpressed (Huang and Waknitz, 2009; Narod et al., 2008; Rubin et al.,
2011; Tomlins et al., 2005).
Besides the TMPRSS2:ERG and TMPRSS2:ETV1 fusions, other ETS genes
(ETV4, ETV5 and FLI1) were reported to be fused at the 5’ end with TMPRSS2 or other 5’
partners, being SLC45A3 (also an androgen-responsive gene) the second most frequent
5´ partner rearranged with all the five ETS genes involved in PCa rearrangements (Boyd
et al., 2012; Paulo et al., 2012a; Tomlins et al., 2009; Tomlins et al., 2007).
Although the TMPRSS2:ERG fusion appears to be a key and early molecular
event in PCa, it does not seem to be enough for the development of the disease. Some
studies have associated this fusion with a more aggressive disease and it is known that
ERG overexpression and PI3K signaling act together in the progression to advanced PCa.
21
There have been some efforts to apply these discoveries to diagnosis, prognosis and
treatment, although more studies are needed (Rubin et al., 2011).
3.7 Prostate Cancer Epigenetic Alterations
Epigenetic alterations, unlike genetic alterations, do not affect the DNA sequence,
but instead affect gene expression by mechanisms such as DNA methylation, histone
modifications, chromatin remodeling and RNA interference. In the prostate, many genes
have already been described to show abnormal methylation levels. Among these,
RASSF1A and GSTP1 are worth to be mentioned (Gonzalgo and Isaacs, 2003).
RASSF1A is found methylated in 60% to 70% of all PCa cases, being more common in
tumors of higher grades (Gonzalgo and Isaacs, 2003). Hypermethylation of GSTP1 is the
most frequent somatic gene alteration in PCa (Henrique and Jeronimo, 2004; Lin et al.,
2001) being considered an initiating event of prostatic carcinogenesis (De Marzo et al.,
2004; Gonzalgo and Isaacs, 2003; Nelson et al., 2001). It is silenced in almost every
prostate carcinomas (90-95%) and in 70% of PIN lesions (Henrique and Jeronimo, 2004;
Porkka and Visakorpi, 2004). GSTP1 is normally expressed in the basal cells, and
encodes the π-class glutathione S-transferase (De Marzo et al., 2004; Nelson et al., 2003)
that belongs to a family of detoxifying enzymes of different xenobiotics (Gonzalgo and
Isaacs, 2003; Porkka and Visakorpi, 2004). GSTP1 is thought to act as a “caretaker”
gene, since it is a scavanger of reactive oxygen species, and its inactivation may cause
genome damage (De Marzo et al., 2004; Deutsch et al., 2004; Nelson et al., 2001). Since
GSTP1 is so important for prostate carcinogenesis, restoring its expression and avoiding
carcinogenic stress may be a good approach for cancer prevention (Nelson et al., 2001).
3.8 The role of androgens
It has long been assumed that androgens are somehow involved in prostate
carcinogenesis, since both the prostate and PCa are androgen-dependent. Despite of a
lack of enough epidemiological evidence, it seems that androgens have a different role in
each step of the prostate carcinogenesis pathway. The oncogenic activity of androgens is
related to their ability to stimulate cell growth, making the cells more prone for the
accumulation of genetic alterations (Bosland and Mahmoud, 2011; Henderson and
Feigelson, 2000).
The binding of androgens to AR causes AR activation and translocation into the
nucleus where it binds to specific DNA response elements on the promoter regions,
interacting with other transcriptional factors and coregulators and activating the
transcription of target genes. The hormone-receptor complex may act as break or
accelerator of transcription. As AR plays a crucial role in the regulation of growth and
differentiation of PCa, androgen deprivation therapy is a good option for patients with
22
metastatic PCa (Jung et al., 2004b; Kim et al., 2010a; Kim et al., 2010b). However, most
of the patients relapse and develop androgen-independent (AI) cancers, where AR
activation and signaling remain intact (Kim et al., 2010b; Shen and Abate-Shen, 2010).
Several mechanisms were proposed for the retained activity of AR signaling in AI
tumors. These include AR amplification that results in overexpression of AR and increases
the sensitivity to low levels of androgens, thus promoting the growth of prostate cancer
cells; and AR mutations that alter the ligand-specificity of the receptor, allowing its binding
and activation by nonandrogens and anti-androgens (De Marzo et al., 2004; Nelson et al.,
2003; Shen and Abate-Shen, 2010).
4. Homeobox Genes
The family of homeobox genes, first described in Drosophila melanogaster,
comprises two subfamilies, class I (clustered homeobox genes – HOX) and class II
(nonclustered or divergent) (Nunes et al., 2003). In terms of evolution, homeobox genes
are highly conserved and arose by duplication from a single ancestral cluster (Daftary and
Taylor, 2006; Lappin et al., 2006). HOX genes codify a family of transcription factors
(Kelly et al., 2011; Kim et al., 2010b) with 39 members described in humans (Chen and
Sukumar, 2003; Daftary and Taylor, 2006). They are organized in four different genomic
clusters: A, B, C and D, each with 9 to 13 genes. The different clusters are located in
7p15, 17q21.2, 12q13 and 2q31, respectively, and comprise 13 paralog groups. The
structure of HOX genes and the proteins encoded by them are represented in Figure 8.
HOX genes are small, being composed of only two exons and one intron (Lappin et al.,
2006). Hox proteins are characterized by their homeobox domain, which is located in the
second exon and is composed of a 183-bp sequence with 61 amino acids. The
homeodomain is a helix-turn-helix-motif with DNA-binding activity, allowing the binding to
specific regions and therefore activating or repressing target genes (Chen and Sukumar,
2003; Daftary and Taylor, 2006; Kelly et al., 2011; Lappin et al., 2006). They harbor an
acidic tail in the COOH extremity and a pentamer near the homeodomain, which is
responsible for the binding of the TALE (three amino acid loop extension) proteins (Lappin
et al., 2006).
23
Figure 8 - General structure of the HOX genes and their protein products [adapted from (Lappin et al.,
2006)].
HOX genes are very important regulators of the positional identity of the organs
and tissues in the anterior-posterior axis during embryonic development, from the level of
the hindbrain to the end of the spine (Daftary and Taylor, 2006; Jung et al., 2004a). They
are known to regulate cell proliferation and differentiation during embryogenesis and
organogenesis (Chen and Sukumar, 2003; Economides and Capecchi, 2003; McMullin et
al., 2009), showing spatial and temporal colinearity. Their expression is tissue-specific and
is related to the stage of development (Kim et al., 2010b; Krumlauf, 1994; McMullin et al.,
2009). Paralogs with lower numbers (3’ genes) are the first to be expressed and mainly in
the anterior and proximal regions of the body axis, while 5’ genes are expressed in a later
stage of the development and in the posterior and distal regions (Figure 9). This
coordinated expression is commonly referred as the HOX code and it determines the
phenotype of a nascent tissue (Daftary and Taylor, 2006; Ewing et al., 2012; Lappin et al.,
2006; McMullin et al., 2009). HOX genes exhibit functional redundancy, which means that
a mutation or loss of expression in one gene can be compensated by the others (Lappin
et al., 2006).
24
Figure 9 - Genomic organization of the mammalian HOX clusters and their embryonic expression
pattern, which correlates with the chromosomal positioning [adapted from (Daftary and Taylor, 2006)].
HOX genes are very important for limb and pulmonary development in vertebrates,
and many limb malformations and congenital lung abnormalities have already been
associated with specific genes of this family (Lappin et al., 2006). They are also known for
their involvement in the development of secondary sexual structures, like male accessory
organs and female mammary glands (Jung et al., 2004a).
Besides their part in embryogenesis and organogenesis, some HOX genes
continue to be expressed in adult tissues that maintain developmental plasticity, mediating
differentiation processes, homeostasis and organ functioning (Daftary and Taylor, 2006;
Kelly et al., 2011). More recently deregulation of HOX expression patterns have been
associated with many types of cancers (Cross and Burmester, 2008), such as prostate,
breast, ovarian, endometrial, lung, renal, neuroblastoma, colorectal, leukemias and
pancreatic, among others (Chen and Sukumar, 2003; Gray et al., 2011; Habel et al., 2013;
Jung et al., 2005; Kelly et al., 2011; Zhao et al., 2005). HOX genes also exhibit tissue-
specificity in tumors, meaning that different tumors show a specific expression profile. The
exact role of HOX genes is different depending on the type of malignancy, but it is known
that they might interfere in different hallmarks of cancer (Chen and Sukumar, 2003;
McMullin et al., 2009; Nunes et al., 2003). It has also been suggested that HOX genes
that are expressed during embryonic development but show low or absent expression in
adulthood can be overexpressed or re-expressed in neoplasias (Chen and Sukumar,
2003; McMullin et al., 2009).
4.1 HOXB13
HOXB13 was first described by Zeltser and her colleagues in 1996, being the last
discovered homeobox gene, and encodes the transcription factor HOXB13 (Zeltser et al.,
25
1996). It belongs to the paralog group 13 and is located 70kb upstream of HOXB9. Its
expression maintains the temporal and spatial colinearity typical of other HOX genes and
is identical to other HOX-13 genes, being expressed in the urogenital sinus, posterior
extent of the spinal cord and tailbud (Jung et al., 2004a; Jung et al., 2004b; Kim et al.,
2010b; Zeltser et al., 1996) and not being expressed in the genital tubercule or the limbs
(Zeltser et al., 1996).
HOX13 paralogs are particularly important for the prostate development, with
HOXC13 being the only member of the group that is not expressed during embryogenesis
(Economides and Capecchi, 2003; Jung et al., 2004b). Of these, HOXB13 paralog is the
only that maintains a high expression level in adults, and its expression is confined to the
prostate, distal colon and rectum (Figure 10) in an androgen-independent fashion (Ewing
et al., 2012; Jung et al., 2004b; Sreenath et al., 1999; Zeltser et al., 1996). Interestingly,
this contrasts with the expression pattern of NKX3.1, another androgen-dependent
homeobox gene essential for prostate development (Bieberich et al., 1996; Sreenath et
al., 1999).
Figure 10 - Expression of HOXB13 in adult tissues using real time reverse transcription-PCR combined
with Taqman probe [adapted from (Jung et al., 2004b)].
Economides and his co-workers demonstrated that mice with homozygous
mutations leading to loss of function of HOXB13 have defects in the ventral prostate
epithelium, malformations of the duct tips and an absence of secretory proteins such as
p12 and p25. An overgrowth in all the major structures derived from the tail bud is also
observed (Economides and Capecchi, 2003; Economides et al., 2003). These studies
strengthen the importance of HOXB13 for the normal development of the prostate.
4.1.1 HOXB13 and PCa
Since HOXB13 is highly expressed in the prostate and its expression seems to be
maintained in PCa (Edwards et al., 2005), multiple studies have tried to evaluate its
26
function in both the normal prostate and in prostate carcinogenesis (Ewing et al., 2012;
Jung et al., 2004a; Jung et al., 2004b; Kim et al., 2010b). Despite its relevance for the
development of the normal prostate already described (Economides and Capecchi, 2003;
Huang et al., 2007; Norris et al., 2009), its exact function is not yet fully understood,
specially what concerns prostate carcinogenesis. In fact, the role of HOXB13 in prostate
cancer development remains highly controversial, since HOXB13 has been suggested to
act both as an oncogene and as a TSG (Ewing et al., 2012; Jung et al., 2004a).
However, it seems to be consistent that HOXB13 is only expressed in AR-
expressing cells and that its role is dependent on the cell type (Figure 11 and 12) and on
the cellular environment (AR status and androgen stimulation) (Jung et al., 2004a; Jung et
al., 2004b; Kim et al., 2010a; Kim et al., 2010b). Nevertheless, HOXB13 expression is not
directly regulated by androgens (Jung et al., 2004a; Jung et al., 2004b; Sreenath et al.,
1999; Zeltser et al., 1996).
Figure 11 - HOXB13 mRNA levels in the prostate (last column) and in ten prostate cell lines according
to the status of the activity of the AR. P69, PrEc and Pz-HPV7 are benign prostate cell lines and the others
are tumor-derived cell lines [adapted from (Jung et al., 2004b)].
27
Figure 12 - Protein levels of HOXB13 expression in seven prostatic cell lines by Western Blot (A) and
immunocytochemistry (B) [adapted from (Kim et al., 2010b)].
In 2004, Jung et al. induced de novo expression of HOXB13 in the HOXB13-
negative and AR-negative PC-3 PCa cell line, and observed growth suppression along
with alterations in cell morphology, typical of terminal differentiated cells. This effect was
also evident in vivo, with a decrease in tumor volume five weeks after cell implantation.
This growth inhibition was proved to result from cell cycle arrest at G1 phase caused by
the down-regulation of cyclin D1, which is essential for accelerating this phase. The down-
regulation of cyclin D1 was shown to be caused by HOXB13 suppression of the β-
catenin/TCF-4 activity, since TCF-4 is an essential transcription factor of cyclin D1
expression. This observation lead to the assumption that HOXB13 loss might represent a
growth advantage for prostatic tumors (Jung et al., 2004a).
The same group also studied the role of HOXB13 in the HOXB13-positive and AR-
positive LNCaP PCa cell line, demonstrating both by its overexpression and by its
suppression, that HOXB13 plays different roles depending on the androgen stimuli (Jung
et al., 2004b; Kim et al., 2010b). Briefly, when HOXB13 is overexpressed in the presence
of androgens, it causes repression of androgen-activated AR transcriptional activity, by
physically interacting with AR (Figure 13), modulating its positive growth signal and
leading to an inhibition in cell growth (Kim et al., 2010a). On the other hand, in an
androgen-free environment HOXB13 promotes cell proliferation through the inhibition of
the TSG p21, allowing the activation of E2F mediated signaling (Figure 13) (Kim et al.,
2010b). The same effect regarding the influence of androgens was observed when
HOXB13 was suppressed in LNCaP cells, with the presence of R1881 conferring a stimuli
for growth, and its absence inhibiting cell growth (Jung et al., 2004b; Kim et al., 2010a).
28
Regarding HOXB13 expression in PCa, there are still no certainties. A microarray
study demonstrated that HOXB13 expression levels in PCa were higher than those found
in normal adjacent tissue (Edwards et al., 2005), while others suggest that no such
changes could be observed due to the heterogeneity and multifocal nature of PCa, or
report a mixed pattern with HOXB13-positive and negative PCa cells (Jung et al., 2004a;
Kim et al., 2010a; Kim et al., 2010b). Nevertheless, it was demonstrated that hormone-
refractory tumors overexpress HOXB13 when compared to androgen-dependent
carcinomas (Figure 14) (Kim et al., 2010b).
Figure 13 - Proposed models of HOXB13 dual activity in the presence (A) and in the absence (B) of
androgens. In the presence of androgens HOXB13 can either sequester AR coactivators or recruit more
corepressors, thus inhibiting the AR-mediating growth signals. In the absence of androgens HOXB13 inhibits
the TSG p21 leading to the activation of the cyclin dependent kinase 2 (CDK2), which together with the
hyperphosphorylated RB releases the transcription factor E2F that activates the transcription of genes
promoting cell proliferation [adapted from (Kim et al., 2010a; Kim et al., 2010b)].
Figure 14 - HOXB13 expression in androgen-dependent (AD) and androgen-independent (AI) tumors
[adapted from (Kim et al., 2010b)].
4.1.2 HOXB13 mutations and PCa risk
As mentioned before, at the beginning of 2012, Ewing and his colleagues (Ewing
et al., 2012) screened more than 200 genes in the 17q21-22 region, a region that had
29
previously been suggested as a possible location for a PCa susceptibility gene, by linkage
studies (Cropp et al., 2011; Lange et al., 2007; Lynch and Shaw, 2013). By sequencing
germline DNA from patients and controls, they found a recurrent, albeit rare, mutation in
HOXB13, G84E, now identified as rs138213197, in men of European descent, which co-
segregated with the disease in families. The G84E mutation leads to a change of an
adenosine for a guanine in the second position of codon 84, causing a nonconservative
substitution of glutamic acid for glycine. This mutation is localized in the first MEIS
(myeloid ecotropic insertion site) interacting domain (Figure 15), which is a highly
conserved domain that is thought to fine-tune HOX function by mediating the binding to
members of the MEIS homeobox protein family, which act as HOX cofactors (Ewing et al.,
2012; Witte et al., 2013).
The G84E mutation is found in 0.1% of the general population while in PCa
patients it reaches 1.4% and it increases to 3.1% in patients with early-onset of the
disease (≤55 years old) and with a positive family history (Ewing et al., 2012). This
increase in strength of association was also corroborated by other studies, but no
differences were found in Gleason grade between carriers and noncarriers (Akbari et al.,
2012; Karlsson et al., 2012; Kluzniak et al., 2013; Witte et al., 2013). Accordingly to
Ewing, the location of the mutation might be the explanation for how prostate
carcinogenesis is promoted and its nature indicates a role of HOXB13 more compatible
with an oncogene than a TSG. It is important to note that G84E carriers do not lose the
mutated allele, neither the expression of HOXB13, and that the mutated allele (A) is more
frequently transmitted from parents to the affected offspring than the normal allele (G)
(Ewing et al., 2012; Xu et al., 2013).
Other groups have confirmed the association of the G84E germline mutation with a
significantly increased risk of familial prostate cancer, although with lower frequency rates
(Akbari et al., 2012; Breyer et al., 2012; Chen and Sukumar, 2003; Karlsson et al., 2012;
Kluzniak et al., 2013; Laitinen et al., 2013; MacInnis et al., 2013; Schroeck et al., 2013;
Stott-Miller et al., 2013; Witte et al., 2013; Xu et al., 2013). Haplotype studies have
revealed that it is a founder mutation, more common in individuals of European descent,
being more frequent in the Nordic countries than in North America and Australia (Karlsson
et al., 2012; Xu et al., 2013). These studies also suggest that this mutation probably has
occurred in Northern Europe (Chen et al., 2013).
Although G84E variant only accounts for a small fraction of PCa cases, it confers a
significantly increased risk of hereditary prostate cancer (Ewing et al., 2012; Lynch and
Shaw, 2013), being the first major genetic mutation associated with familial PCa (Bessede
and Patard, 2012; Karlsson et al., 2012).
30
Ewing et al., also found four additional missense variants, although much rarer
than G84E, and these are shown in Figure 15. The p.Arg229Gly (R229G) was found in
one patient from an African-American family and in his two brothers with PCa, and
p.Gly216Cys (G216C) was found in two half-brothers with PCa, from an African-
Caribbean family. The p.Tyr88Asp (Y88D) and p.Leu144Pro (L144P) were found in
LAPC4 and LNCaP cell lines, respectively, both PCa-derived and androgen-dependent.
All of the mutations were predicted to affect protein structure and function (Ewing et al.,
2012).
Figure 15 - Location of all the missense mutations found in HOXB13 by Ewing and collaborators
[adapted from (Ewing et al., 2012)].
With the intent of addressing the association of HOXB13 mutations with PCa in the
Chinese population, Lin et al. found a novel nonsynonymous mutation, p.Gly135Glu
(G135E), resulting in the substitution of a glycine to a glutamic acid. It also appears to be
a founder mutation that is associated with an increased risk of PCa in Chinese men. In
contrast with the G84E mutation, this variant is located in the second MEIS interacting
domain that is also a highly conserved domain (Lin et al., 2013).
The fact that other variants besides G84E had also been found, reinforces the idea
of the involvement of HOXB13 in prostate carcinogenesis, which is now established as a
PCa susceptibility gene, with a relative risk of approximately 20 for carriers of the G84E
mutation, and with a moderate penetrance in the average population (Ewing et al., 2012;
Karlsson et al., 2012; Kluzniak et al., 2013). However, the utility of these discoveries need
to be translated into the clinics, in order to identify men at higher risk which could benefit
from early screening (Xu et al., 2013). It is also important to further explore the biological
consequences of this variant and to evaluate how confined it is to prostate carcinogenesis
(Kluzniak et al., 2013; Schroeck et al., 2013).
Aims
33
Given the recent discovery that germline mutations in the HOXB13 gene confer an
increased risk for familial PCa, the main goal of this Master Thesis´s project was to search
for somatic mutations and expression alterations of the HOXB13 gene in prostate
cancer. Therefore, the specific goals of this project were:
To search for mutations in the entire coding sequence of the HOXB13
gene in a series of primary tumor samples from patients with clinically
localized PCa;
To evaluate the oncogenic potential of specific mutations in prostate
carcinogenesis using in vitro models;
To compare HOXB13 mRNA expression levels in normal prostate tissues
and in different molecular subtypes of prostate carcinomas.
Materials and Methods
37
1. Search for somatic mutations in HOXB13
Since some germline mutations in HOXB13 gene have already been found in PCa
patients and demonstrated to confer a higher risk for the disease (Chen et al., 2013;
Ewing et al., 2012; Kluzniak et al., 2013; Lin et al., 2013; Lynch and Shaw, 2013; Witte et
al., 2013; Xu et al., 2013), we have decided to start by searching somatic mutations in this
gene. In order to accomplish this goal, the chosen strategy was DNA sequencing of the
entire coding region of the gene.
1.1 Samples
To search for somatic mutations in HOXB13, a retrospective analysis was
performed in primary tumor samples that had been collected from 200 patients with
clinically localized PCa, diagnosed and treated with radical prostatectomy at the
Portuguese Oncology Institute (Porto, Portugal) between 2001 and 2006. The patients
were diagnosed at a mean age of 62 years (lower age of 49 years and highest age of 75
years), with mean PSA levels and Gleason scores of 9.07 and 6.77, respectively. Of the
initial series, only 178 cases were further analyzed, due to problems of quality of some of
the samples.
The LNCaP, PNT2, 22Rv1 and NCI-H660 prostate cell lines were also sequenced.
DNA from the patient samples had previously been extracted by our group using
the phenol-chlorophorm method using phase-lock gel (5 PRIME, Hamburg, Germany) and
DNA from the prostate cell lines was extracted using the illustra TriplePrep kit (GE
Healthcare, Life Sciences, Cleveland, USA), accordingly to the manufacturer´s
instructions.
1.2 Sanger sequencing
Since HOXB13 gene has 855bp, it was necessary to perform the sequencing in
separate, for the two exons. The primers for amplification of coding regions of the two
exons were designed using the Primer-BLAST design tool and their specificity was
evaluated with the Nucleotide Basic Local Alignment Search Tool (BLASTn) from the
National Center for Biotechnology Information (NCBI), and were acquired from Metabion
(Martinsried, Germany). The targeted region was amplified in two amplicons and the
respective primer sequences are represented in Table 7.
Table 7 – Primers´ sequences used for sequencing the HOXB13 gene (NM_006361.5).
Primer forward sequence Primer reverse sequence
Exon 1 5´-CGAGCTGGGAGCGATTTA-3´ 5´-AGCTCCAAGTCTCCCTCCTC-3´
38
Exon 2 5´-TTGCACGTGCGCCTGTAGGG-3´ 5´-GTCTCCCCAGGACACCCCCA-3´
For the amplification reaction, a total volume of 30μl was used, with 1xTaq Buffer
with (NH4)2SO4 (Thermo Fisher Scientific, Rockford, USA), 2.5mM MgCl2 (Thermo Fisher
Scientific), 250mM dNTP mix (Thermo Fisher Scientific), 0.5M of each primer, 0.05U of
Taq DNA Polymerase (Thermo Fisher Scientific) and 100 to 500ng of DNA. The PCR
reaction was performed in the thermocycler with an initial step of denaturation at 94ºC for
10 minutes; followed by five cycles with denaturation at 94ºC for 1 minute, annealing at
64ºC for 1 minute, and extension at 72ºC for 2 minutes; and five identical cycles with a
lower annealing temperature (62ºC). A third round of amplification included 25 cycles with
another annealing temperature (60ºC) and the reaction ended with a final extension step
at 72ºC for 7 minutes. The amplification was confirmed by electrophoresis in a 2% (w/v)
agarose gel stained with ethidium bromide.
Next, the PCR product was purified using a mix of two enzymes from Thermo
Fisher Scientific: exonuclease I (20μg/μl) and Fast Thermosensitive Alkaline Phosphatase
(1μg/μl), in a proportion of 1:2. The purification was carried out in the thermocycler at 37ºC
for 30 minutes for proper enzyme activity, followed by 15 minutes at 85ºC, to inactivate
enzymes. For this reaction, 1μl of the mix (ExoSap) is added to 10μl of PCR product. This
allows the removal of unincorporated primers and also degrades the nucleotides that have
not been incorporated, thus the resulting PCR product is completely cleaned-up for
sequencing (Werle et al., 1994).
For the sequencing reaction, we only used the forward primers and the DNA
amplicons were sequenced using the the BigDye® Terminator v3.1 Cycle Sequencing Kit
from Applied Biosystems® (Life Technologies). The reaction consisted on mixing 1.9μl of
sequencing buffer, 1μl of Terminator Ready Reaction Mix, containing dNTPs, ddNTPs-
fluorocromes, MgCl2 and Tris-HCl buffer, 3.5M of the already mentioned primers and
lastly, 100 to 500ng of DNA, with a final reaction volume of 10μl. The PCR program at the
thermocycler consisted on an initial denaturation step at 96ºC for 10 minutes, followed by
35 cycles of denaturation at 96ºC for 10 seconds, annealing at 52ºC for 5 seconds and
extension at 60ºC for 6 minutes. In order to reduce possible contaminants, the sequencing
product was purified with IIlustra Sephadex® G-50 fine columns (GE Healthcare, Life
Sciences, Cleveland, USA), according to the manufacturer´s instructions. Finally, to
stabilize the DNA, the samples were eluted in 12μl of deionized formamide (Applied
Biosystems). The sequencing was performed on a 3500 Genetic Analyzer (Applied
Biosystems) and all the sequences were compared with the HOXB13 RefSeq sequence
39
(NM_006361.5) for variant detection, using the Mutation Surveyor V4.0.7 software
(Softgenetics, State College, PA, USA).
2. Cell Lines and Cell Culture
Seven PCa cell lines, namely VCaP, NCI-H660, PC-3, 22Rv1, LNCaP, MDA-PCa-
2b and DU145 and one non-malignant prostatic cell line, PNT2, were grown in a 37ºC
environment with 5% CO2 and were frequently tested for contaminations with Mycoplasma
spp., using the PCR Mycoplasma Detection Set, from Clontech Laboratories (Mountain
View, CA, USA).
All the cell lines except MDA-PCa-2b were grown in media from GIBCO®
(Invitrogen, Carlsbad, CA, USA). 22Rv1, LNCaP, PNT2 and DU145 were grown in RPMI-
1640 medium; VCaP in DMEM medium and PC-3 in F-12. All the media were
supplemented with 10% Fetal Bovine Serum – FBS (GIBCO®) – with 1%
Penicillin/Streptomycin (GIBCO®) at a final concentration of 1%. NCI-H660 was grown in
RPMI-1640 medium supplemented with 1x Insulin-transferrin-selenium, 10nM of β-
estradiol, 10nM of hydrocortisone, 1x L-glutamine, and also FBS and
Penicillin/Streptomycin at a final concentration of 5% and 1%, respectively. MDA-PCa-2b
was grown in BRFF medium (Gentaur, Kampenhout, Belgium) supplemented with 20%
FBS and 1% Penicillin/Streptomycin.
VCaP and PNT2 were acquired from ECACC (European Collection of Cell
Cultures), PC-3, DU145 and LNCaP from the Leibniz Institute DSMZ (German Collection
of Microorganisms and Cell Cultures), NCI-H660 and MDA-PCa-2b from ATCC (American
Type Culture Collection) and 22Rv1 were kindly provided by Dr. David Sidransky from
Johns Hopkins University.
The main features of the prostatic cell lines are summarized in Table 8.
Table 8 – Principal characteristics of the eight prostate cell lines [adapted from the ATCC website
available at http://www.atcc.org/].
Cell line Cell type
(ATCC) Origin (ATCC)
Androgen
response
PSA
expression
ETS
rearrangements
PNT2 Epithelial;
adherent
Established by
immortalization of
normal adult
prostatic epithelial
cells. The primary
culture was
obtained from a
-
40
prostate of a 33
year-old male post
mortem (Berthon et
al., 1995).
22Rv1
Carcinoma
; epithelial;
adherent
Derived from a
xenograft that was
serially propagated
in mice after
castration-induced
regression and
relapse of the
parental, androgen-
dependent CWR22
xenograft (Knouf et
al., 2009;
Sramkoski et al.,
1999).
+ + -
DU145
Carcinoma
; epithelial;
adherent
Derived from a
brain metastasis of
a 69 year- old
Caucasian male.
- - -
PC3
Grade IV,
adenocarci
noma;
epithelial;
adherent
Initiated from a
bone metastasis of
a grade IV prostatic
adenocarcinoma
from a 62 year-old
Caucasian male.
- - -
MDA-
PCa-2b
Adenocarci
noma
epithelial;
adherent
Established from a
bone metastasis of
a 63 year-old black
male with
androgen-
independent
adenocarcinoma of
the prostate.
+ +
MIPOL1-ETV1
(Tomlins et al.,
2007)
LNCaP
Carcinoma
epithelial;
adherent,
single cells
and loosely
attached
clusters
Derived from a left
supraclavicular
lymph node
metastasis of a 50
year-old Caucasian
male.
+ +
Cryptic insertion of
ETV1 into an
intronic sequence
from the MIPOL1
locus at 14q13.3–
14q21.1 (Tomlins et
al., 2007).
41
VCaP Epithelial;
adherent
Established in from
a vertebral bone
metastasis from a
59 year-old
Caucasian male
with hormone
refractory prostate
cancer.
+ + TMPRSS2:ERG
(Narod et al., 2008).
NCI-H660
Epithelial
neuroendo
crine;
floating
and
attached
clusters
Derived from a
lymph node
metastasis of a
small cell
carcinoma, stage E.
- - TMPRSS2:ERG
(Mertz et al., 2007)
3. Induction of de novo overexpression of HOXB13 in prostate
cell lines
In order to evaluate the oncogenic potential of the G84E mutation first described
by Ewing (Ewing et al., 2012) and of the p.Ala128Asp (A128D, to simplify) mutation
discovered along this work, we aimed to establish a cell model of prostate tumors
harboring these mutations by transfecting prostate-derived cell lines with different
expression vectors: HOXB13 wild-type, HOXB13 with the two mutations and a control
vector. With these in vitro models, we expect to clarify if the overexpression of HOXB13
and its mutated forms results in an oncogenic phenotype.
3.1 Selection of the study cell model
In order to choose the prostate-derived cell lines to induce HOXB13
overexpression, some criteria had to meet. Since we have decided to study the role of
HOXB13 both in the benign and in the malign contexts, we selected one of each kind.
The first criterion was the absence of HOXB13 mutations and, accordingly to the
literature, of the eight prostate-cancer cell lines already sequenced (LNCaP, PC3, DU145,
22Rv1, E006AA, VCaP, MDA-PCa-2b, and LAPC4), only LNCaP and LAPC4 were found
to harbor nonsynonymous mutations in this gene (Ewing et al., 2012), so they were
excluded. Other prostate-derived cell lines were also sequenced, and the results will be
presented later.
42
For the malignant context, we also intended to choose a cell line that is androgen-
responsive, which could then be an interesting model to study the influence of androgens
in HOXB13´s biological role. We also decided to choose a PCa-derived cell line without
known ETS rearrangements, since we are not certain that these rearrangements would
not interfere with HOXB13 expression.
Another condition was that the chosen cell lines would have low or basal level
expression of HOXB13 and, according to the literature, we could choose DU145, PC3,
22Rv1, P69, PrEc and Pz-HPV7 (Jung et al., 2004b; Kim et al., 2010a). However, we also
evaluated HOXB13 expression both at the RNA and at the protein level, by quantitative
real-time reverse transcription PCR (qRT-PCR) and immunoblotting, respectively.
To meet all these criteria, and after obtaining our own results (shown later), the
chosen cell lines were PNT2 and 22Rv1.
3.1.1 RNA and protein extraction from prostate cell lines
RNA from all the prostate cell lines available at the Cancer Genetics Department
of IPO-Porto, which included PNT2, 22Rv1, DU145, PC-3, MDA-PCa-2b, LNCaP, VCaP
and NCI-H660, had already been extracted by our group, using the TRIzol® Reagent
(Ambion®, Life Technologies, Carlsbad, CA, USA).
RNA from sub-confluent 22Rv1- and PNT2-derived cell populations (transfected
cell lines) was extracted using the TRIzol® Reagent. First the samples were homogenized
with 500µl of TRIzol, which lyses cells and dissolves its components. Then, 100µl of
chloroform were added, to separate phases, and RNA was precipitated with 250µl of
100% isopropanol. Finally, the RNA was washed with 250µl of 75% ethanol and
ressuspended in 25µl of RNase-free water. RNA concentrations and purity were
measured in a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies,
Wilmington, DE, USA).
Protein extracts from the eight prostatic cell lines were also available and had
been obtained using RIPA lysis buffer (Santa Cruz Biotechnology, CA, USA). The same
extraction method was used for protein extraction of 22Rv1- and PNT2-derived cell
populations. Concisely, sub-confluent cells were washed twice with PBS and 150µl of
RIPA were added. The flasks were scraped and the cell lysate was set on ice for 30
minutes, followed by centrifugation at 180rcf for 30 minutes at 4ºC in order to obtain the
protein extract. For the measurement of protein concentrations, the Thermo
Scientific™Pierce™BCA Protein Assay (Rockford, USA) was used. The assay consists on
a colorimetric detection, based on the reduction of Cu+2 to Cu+1 by protein in an alkaline
medium, which results in a purple-colored product due to the chelation of two molecules of
bicinchoninic acid (BCA) with the Cu+1. The reaction consisted on mixing 200µl of the BCA
43
solution with 25µl of each sample of unknown concentration, followed by a period of
incubation at 37ºC for 30 minutes. The absorbance was then read at a wavelength of
562nm in the FLUOstar Omega microplate reader (BMG Labtech). Protein quantification
was achieved using the standard curve method, in which a curve is built using a series of
known consecutive dilutions of BSA (bovine serum albumin). For these controls, the
reaction was exactly the same as described for the samples.
3.1.2 Complementary DNA (cDNA) synthesis
RNA from the prostatic cell lines had already been converted into cDNA using the
TransPlex Whole Transcriptome Amplification Kit (Sigma-Aldrich, Steinheim, Germany).
For 22Rv1- and PNT2-derived cell populations, cDNA was synthetized using the
RevertAid H Minus First Strand cDNA Synthesis Kit from Thermo Scientific. Briefly, 1µg of
template RNA was mixed with 1µl of oligo (dT)18 primer in a total reaction volume of 12µl,
incubated at 65ºC for 5 minutes and chilled on ice for 2 minutes. Then, 8µl of a mix
containing 1xReaction Buffer, 20U of RiboLock RNase Inhibitor, 1mM of dNTP Mix and
200U of RevertAid H Minus M-MuLV Reverse Transcriptase in a final volume of 20µl,
were added. The reaction took place at 42ºC for 60 minutes, followed by enzyme
inactivation at 70ºC for 5 minutes.
3.1.3 Quantitative reverse transcription PCR (qRT-PCR)
In order to evaluate HOXB13 expression at the RNA level, quantitative real-time
reverse transcription PCR was performed. HOXB13 relative expression levels were
evaluated in the eight available prostatic cell lines, as well as in the 22Rv1- and PNT2-
derived cell populations.
HOXB13 primers and probe were designed using the Primer3 software (v.0.4.0)
(Untergrasser A, 2012 ) using the criteria of the Primer Express® Software for Real-Time
PCR version 3.0 (Applied Biosystems), and its specificity was evaluated with the BLASTn.
Both primers and probe were purchased from Metabion. The primers’ sequences were 5´-
GTTGCCAGGGAGAACAGAAC-3´ (forward) and 5´-TGTACGGAATGCGTTTCTTG-3´
(reverse). The probe used was a TaqMan® probe, that is a dual labeled fluorogenic probe
and its sequence was 5´-AAGGCAGCATTTGCAGACTCCAGC-3´. The probe has a FAM
fluorophore linked to the 5´ end and a quencher dye (TAMRA) at the 3´ end. The principle
of TaqMan® probes is represented in Figure 16.
44
Figure 16 - TaqMan probe principle. A – During the annealing step, both the primers and the probe anneal
with the template and due to the proximity of the quencher and the fluorophore, fluorescence is strongly
reduced. B- At the extension step, Taq DNA polymerase extends the primer and when it reaches the probe, it
cleaves the fluorophore (5´ 3´ exonuclease activity), releasing it from the quencher inhibition, so the
fluorescence can be measured. The emitted signal is then measured and is proportional to the amount of
accumulated PCR product [Adapted from Critical factors for successful Real-time PCR (2010) available at the
Qiagen´s website].
The reactions were run in 96-well plates on the Applied Biosystems 7500 Real-
time PCR system (Applied Biosystems). Reactions were performed in triplicate and for
each reaction 2µl of cDNA were mixed with 0.373µM of each primer, 0.25µM of the dual
labeled probe, 1xSensiFASTTMProbe Lo-ROX mastermix (BIOLINE, London, UK) and
nuclease-free water for a final reaction volume of 20µl. The reactions were performed
using the following conditions: 50ºC for 2 minutes and 95ºC for 10 min, followed by 45
cycles at 95ºC for 15 seconds and 60ºC for 1 minute.
For the analysis, HOXB13 expression levels were normalized to the expression of
a housekeeping gene, Beta glucuronidase (GUSB), in order to normalize differences in
RNA input. The primers for this housekeeping gene were acquired from Applied
Biosystems as a predesigned TaqMan® Endogenous Control. A standard curve was built
using five consecutive ten-fold dilutions of a positive control template (DNA from the
LNCaP cell line) to determine amplification efficiencies of both the target and the
endogenous reference gene.
For the quantification of HOXB13 relative expression levels in the eight prostatic
cell lines, the comparative threshold cycle (Ct) method was used, in which a ∆Ct value is
calculated and describes the difference between the Ct of the target gene (HOXB13) and
the Ct of the housekeeping gene (GUSB). The Ct (from cycle threshold) is inversely
45
proportional to the initial copy number and it represents the cycle at which the
amplification plot crosses the threshold value, where there is a significantly increase in
fluorescence.
For the 22Rv1- and PNT2-derived cell populations, the ∆∆Ct method was used, in
which the relative expression levels of all samples were normalized for the levels of the
cells transfected with the control vector (p-Entry). This value was calculated by subtracting
the average ∆Ct of the reference sample to the sample of interest.
3.1.4 Immunoblotting
Immunoblotting is a technique that uses antibodies to identify specific proteins
among a number of unrelated protein species, in a given sample. Briefly, proteins are
separated by electrophoresis and transferred to membranes that are incubated with a
specific primary antibody, followed by incubation with a labeled secondary antibody
(Figure 17) (Magi and Liberatori, 2005).
To separate protein extracts, a 10% SDS-polyacrylamide gel electrophoresis
(SDS-PAGE) was performed. First, 30µg of protein extract from each prostate cell line
were mixed with 5x loading buffer (0.5M Tris-HCl - pH 6.8, 2% SDS, 10% glycerol and
0.05% bromophenol blue) and incubated at 95ºC for 5 minutes to denature proteins.
Then, 20µl of each sample were loaded in the 10% gel and were run at 120V/cm. After the
electrophoresis, the proteins were transferred to a 0.1µm nitrocellulose membrane
(Sigma-Aldrich) and for that, a “sandwich” consisting on the nitrocellulose membrane, the
gel, sponges and filter paper (all pre-wet in 1x Tris-Glycine buffer with 20% methanol) was
assembled. The transfer was run at 50V/cm for 1 hour. The membrane was blocked in a
blocking solution (5% non-fat-dry milk in TBS-0.1% Tween 20) for 1 hour, and incubated
overnight (4ºC, under agitation) with the Anti-HOXB13 antibody (Abcam, Cambridge, MA,
USA) diluted in blocking solution (1:1000). In the following day, the membrane was
washed four times (10 minutes each) with TBS-0.1% Tween 20, and incubated for 1 hour
at room temperature with the secondary antibody, goat anti-rabbit IgG (1:100000) from
Santa Cruz Biotechnology. This antibody is linked to a reporter enzyme, the horseradish
peroxidase (HRP), allowing the detection of the HOXB13 protein. The membrane was
washed again before proceeding to chemiluminescence detection. For
chemiluminescence detection, the Immun-Star™ WesternC™ kit (Bio-Rad, Munich,
Germany) was used and the specific bands were visualized by autoradiography.
46
Figure 17 - Immunoblotting procedure. Proteins are separated by gel electrophoresis according to their
molecular weight and transferred to a nitrocellulose membrane by electroblotting. The membrane is then
incubated with specific antibodies, and detection is performed by autoradiography [adapted from
http://www.bio.davidson.edu/genomics/method/Westernblot.html].
3.2 Construction of the HOXB13 expression vectors
An HOXB13 expression vector was purchased from Origene (Rockville, MD, USA)
and is represented in Figure 18. This vector has the particularity of having a Myc-DDK tag
fused at the C-terminal of the ORF, which allows easy purification and antibody detection
using the 4C5-AntiDDK monoclonal antibody (Origene). The selection antibiotic for this
vector is kanamycin, at a final concentration of 25µg/ml.
Figure 18 - Plasmid map of the Myc-DDK-tagged ORF clone of Homo sapiens homeobox B13
(HOXB13) as transfection-ready DNA [adapted from the Origene website, available at
www.origene.com/orf_clone/trueclone/NM_006361/RC209991/HOXB13.aspx].
The HOXB13 plasmid was expanded in Stellar™ Competent Cells from Clontech
Laboratories. To do such, 25µl of competent cells were mixed with 0.05ng of the plasmid
47
in a 14-ml round-bottom tube and then placed on ice for 30 minutes. The cells were
exposed to a heat shock at 42ºC for 45 seconds and again placed on ice for 2 minutes.
SOC medium (Clontech Laboratories), previously warmed, was added and the mix was
incubated by shaking for 1 hour at 37ºC. 2.5ng of a control vector, pUC18 (Origene), were
transformed in parallel to evaluate the efficiency of the transformation. 100µl of the
transformed culture were plated in selective medium: Luria-Bertani (LB) agar with 25µg/ml
of kanamycin (Sigma-Aldrich) for the HOXB13 plasmid, and LB agar with 50µg/ml of
ampicillin (GIBCO®) for the pUC18 control, and incubated overnight at 37ºC. The solid
medium was made by mixing 1% (w/v) Bacto™ Agar with 1% (w/v) Bacto™ Tryptone,
0.5% (w/v) Bacto™ Yeast Extract (all from Becton, Dickinson and Company, Sparks, MD,
USA) and 0.25% (w/v) sodium chloride.
To create the mutated vectors (G84E and A128D), mutagenesis was induced in
the wild-type HOXB13 vector, using the QuikChange II Site-Directed Mutagenesis Kit
(Figure 19) from Agilent Technologies (Santa Clara, CA, USA). This technique allows site-
specific mutations in double-stranded plasmids and is expected to generate mutants with
an efficiency >80%. First, two sets of oligonucleotide primers were designed using the
Quick-Change Primer Design Program (Agilent Technologies) to create the desired
mutations, and were acquired from Sigma-Aldrich. The primers sequences were the
following:
A128D Forward: 5’-CGCCCCACTGAGTTTGACTTCTATCCGGG-3’
A128D Reverse: 5’-CCCGGATAGAAGTCAAACTCAGTGGGGCG-3’
G84E Forward: 5’-GGTTACTTTGAAGGCGGGTACTACTCCTGCC-3’
G84E Reverse: 5’-GGCAGGAGTAGTACCCGCCTTCAAAGTAACC-3’
48
Figure 19 - Three-step procedure of the Site-Directed Mutagenesis method [adapted from the Instruction
manual of the QuikChange II Site-Directed Mutagenesis Kit, available at
http://www.chem.agilent.com/Library/usermanuals/Public/200523.pdf].
The first step consisted on the mutant strand synthesis, using the HOXB13 vector
and the two sets of oligonucleotide primers for the desired mutations. The primers were
extended in the thermocycler by the PfuUltra High fidelity DNA polymerase, thus
generating a mutated plasmid, with staggered nicks. This enzyme has the particular
feature of having an 18-fold higher fidelity in DNA synthesis than the regular Taq DNA
polymerase, thus avoiding unwanted errors during the process. For the mutant strand
synthesis reactions, 50ng of dsDNA template were used, with 125ng of each primer,
1xreaction buffer, 1µl of dNTP mix, and 0.05U of PfuUltra High fidelity DNA polymerase in
a final reaction volume of 50µl in double-distilled water (ddH2O). Two different reactions
were performed, one for each mutation. Besides the two mutant strand synthesis
reactions, an additional reaction was done, to evaluate the mutagenesis efficiency
according to the manufacturer´s instructions. Briefly, a pWhitescript 4.5-kb control plasmid
(Agilent Technologies) was used, which contains a stop codon in the β-galactosidase
gene, instead of a glutamine. Using the oligonucleotides provided by the manufacturer,
the mutagenesis reaction will reverse the stop codon into the original glutamine codon,
creating a point mutation in the plasmid that will allow the expression of full-length β-
galactosidase. Therefore it allows distinguishing the colonies with and without the
insertion, as it will be explained later. The amplification program in the thermocycler
consisted in an initial denaturation step at 95°C for 30 seconds, followed by 12 cycles of
denaturation at 95°C for 30 seconds, annealing at 55°C for 1 minute and extension at
49
68°C for 6 minutes. After the thermal cycling reaction the products were set on ice for 2
minutes.
After the extension, the amplified product was treated with an endonuclease
provided in the kit, Dpn I, which digests the parental methylated and hemimethylated DNA
template, thus allowing the selection of the newly synthetized mutated DNA (step 2-
Figure 19). The reaction consisted on adding 10U of the endonuclease to each
amplification reactions, followed by incubation on the thermocycler for 1 hour at 37°C.
Lastly, XL1-Blue supercompetent cells were transformed with the vectors
containing the desired mutations by mixing 4µl of the DNA product treated with Dpn I with
50µl of supercompetent cells. pUC18 control plasmid was also transformed in parallel to
evaluate the efficiency of the transformation. The transformation procedure was identical
of that described before using Stellar™ Competent Cells. Bacteria transformed with the
mutated plasmids were plated in LB agar containing 25µg/ml of kanamycin, and
pWhitescript and pUC18 controls, plated in LB agar containing 50µg/ml of ampicillin. For
the pWhitescript and pUC18 transformations, 40µl of blue-white select screening reagent
(40mg/ml of X-Gal and 40mg/ml of IPTG- Sigma-Aldrich) were added to the selective LB
medium. After transformation, pWhitescript and pUC18 colonies expressing full-length β-
galactosidase appeared blue.
Single colonies were picked from each LB agar plate transformed with the different
plasmids, and cultured in liquid LB (without agar), at 37ºC for 15.5 hours in the
Thermolyne ROSI 1000™ shaking incubator. Cells were pelleted by centrifugation in the
centrifuge 5810 R (Eppendorf, Hamburg, Germany) for 30 minutes at 3220rcf. Then, the
plasmid DNA was extracted using the NucleoSpin® Plasmid kit (Macherey-Nagel, Düren,
Germany) according to the manufacturer’s instructions, and the identity of each plasmid
was confirmed by Sanger sequencing.
Before proceeding to the transfection of prostate cell lines, an empty vector to
serve as control was created using the original HOXB13 expression vector. The ORF was
excised with two restriction enzymes, FastDigest Sal I and FastDigest Xho I (Thermo
Scientific) and then the plasmid was re-ligated with 5U/µl of T4 DNA Ligase (Thermo
Scientific).
For the digestion reaction, 2µg of plasmid DNA (HOXB13 wild-type), 1xFastDigest
Green Buffer, 2µl of FastDigest Sal I and 2µl of FastDigest Xho I (all from Thermo
Scientific) were used, in a final volume of 20µl in ddH2O. The reaction was incubated in a
thermocycler at 37ºC for 1 hour followed by enzyme inactivation at 65ºC for 10 minutes.
The mix was loaded in a 1.5% (w/v) agarose gel stained with ethidium bromide, to
separate the bands, with the heaviest band being the empty, linearized plasmid, and the
lighter band the HOXB13 insert. After electrophoresis, the heaviest DNA band was cut
50
under UV light and purified with the illustra™ GFX™ PCR DNA and Gel Band Purification
Kit, from GE Healthcare. The basic procedure consists on dissolving the agarose with a
chaotropic agent, then binding the DNA to a glass fiber matrix, followed by the wash of
proteins and salt contaminants. Lastly, the DNA was eluted in a low ionic strength buffer.
After the digestion, the linear vector was re-ligated. For this reaction 50ng of the linear
DNA were combined with 1xT4 Buffer (Thermo Scientific), 0.1U of T4 DNA ligase (Thermo
Scientific) and nuclease-free water in a final volume of 50µl. This vector was also
expanded in Stellar™ Competent Cells and purified as previously described, and its
identity was confirmed by Sanger sequencing.
The transformation procedure is schematically summarized in Figure 20.
Figure 20 - Transformation of competent cells with the desired plasmids [adapted from
http://www.acgtinc.com/dna_rna_extraction.htm].
To simplify nomenclature, the empty vector will be referred as p-Entry and the
mutated vectors as HOXB13-G84E and HOXB13-A128D.
3.3 Transfection and selection of the transfected cells
After expansion and isolation of the four vectors (p-Entry, HOXB13-WT, HOXB13-
G84E and HOXB13-A128D), these were transfected into the chosen cell lines, using the
TurboFectin 8.0™ transfection reagent (Origene). TurboFectin is a lipid/histone based
transfection reagent recommended by Origene for the transfection of TrueORF clones, for
gene overexpression. Having in mind the criteria mentioned before, the cell line availability
at our lab, as well as the results regarding HOXB13 sequencing and expression levels
(shown later), the chosen prostate-derived cell lines were 22Rv1 and PNT2.
51
One day before transfection, 22Rv1 and PNT2 cells were plated in 60mm Petri
dishes at a density of 5x105 cells and 1,5x105, respectively, in complete RPMI-1640
medium (GIBCO®), in order to reach a confluence of approximately 60% at the day of
transfection. In the day of transfection, 250µl of Opti-MEM medium (GIBCO®) were mixed
with TurboFectin 8.0™ (one for each vector and cell line) and incubated at room
temperature for 5 minutes. The DNA from each vector was added to the complex formed
before, and incubated at room temperature for 30 minutes. Three different ratios of
TurboFectin-DNA were used for each vector, in order to achieve the best transfection
efficiency:
3:1 – 7.5µl of TurboFectin for 2.5µg of DNA;
3:2 – 7.5µl of TurboFectin for 5µg of DNA;
6:1 – 15µl of TurboFectin for 2.5µg of DNA.
The TurboFectin-DNA containing mixtures were then added drop-wise to the cells
and incubated for 48 hours at 37ºC with 5% CO2. The transfection scheme is illustrated in
Figure 21.
Figure 21 - Schematic representation of the transfection strategy. PNT2 and 22Rv1 cell lines were
simultaneously transfected with the four vectors, each with three different transfection conditions (TurboFectin-
DNA ratios).
52
After 48 hours, the cells were cultured in selective medium - complete RPMI
medium supplemented with G418 (GIBCO®). The concentrations of G418 that induced
more than 50% of cell death in 24-48 hours - 900µg/ml for 22Rv1 and 450µg/ml and for
PNT2 – were selected to establish stable cell populations, which was achieved after 4 to 5
weeks.
4. Phenotypic assays
In order to have an insight into the biological role of HOXB13 and the phenotypic
impact of the G84E and A128D mutations, two different phenotypic assays were
performed. These assays were only performed in the PNT2-derived cell populations, as it
will be explained later.
4.1 Cell proliferation assay: MTT colorimetric assay
The MTT (3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazoliumbromide) colorimetric
assay was first described by Mosmann in 1983, and allows reliable and accurate
quantification of the number of viable growing cells in proliferation and cytotoxicity studies,
using a spectrophotometer. The basic principle consists on the cleavage of the yellow
MTT to formazan (blue/purple) by the mitochondrial reductases, with the number of living
cells being directly proportional to the amount of formazan formed (Denizot and Lang,
1986; Gerlier and Thomasset, 1986; Mosmann, 1983; Sylvester, 2011).
PNT2-derived cell populations were plated in 96-well plates at a density of 1x104
cells per well. After plating, cells were incubated at 37ºC with 5% CO2 and evaluated in
five time points (0, 24, 48, 72 and 96 hours) after adherence. At each time point, culture
medium was removed and 100µl of MTT solution (Sigma-Aldrich) at 0.5mg/ml in complete
growth medium were added. After an incubation period of 2 hours, which is the time
required for the formation of formazan crystals, the MTT solution was removed from the
wells and 50µl of dimethyl sulfoxide (DMSO; Merck, Darmstadt, Germany), were added to
dissolve the crystals (Figure 22). The 96-well plate was then placed in the FLUOstar
Omega microplate reader (BMG Labtech, Ortenberg, Germany), and the optical density
was measured at a wavelength of 540nm with background correction at 630nm. The
results were analyzed with the Omega Data analysis software (BMG Labtech).
53
Figure 22 - Schematic representation of the MTT colorimetric assay to quantify cell viability [adapted
from (Ali-Boucetta et al., 2011)].
Two independent experiments were done, each including nine wells per condition.
Data was expressed as a ratio of the results obtained with each transfected cell line and
the control vector. The statistical analysis was performed using the Student’s t test and
the data was considered statistically significant when the p values were below 0.05.
4.2 Cell apoptosis assay
The term “apoptosis” was first proposed in 1972 by Kerr, Wyllie and Currie (Kerr et
al., 1972), as being a different form of cell death with some particular morphologic
characteristics, like condensation of the nucleus and the cytoplasm, accompanied with the
breaking up of the cell into fragments. Nowadays, apoptosis is considered a programmed
cell death, occurring as a homeostatic process during development processes and aging,
and also as a protective mechanism against noxious agents and diseases (Elmore, 2007).
Aberrations in this genetically determined process have been implicated in autoimmune
and neurodegenerative diseases, heart conditions and cancer (Saddick, 2013).
To evaluate and quantify apoptosis levels the APOPercentage™ Apoptosis Assay
kit (Biocolor, County Antrim, UK) was used. During apoptosis, phosphatidylserine that is
usually in the inner part of the mammalian cell membrane is translocated to the outer part.
This movement allows the binding of the dye used that is incorporated into the committed
cell (Figure 23). As the cell shrinks, the dye becomes more concentrated, since this
accumulation is unidirectional but it can be chemically released. The amount of released
intracellular dye can then be measured using a microplate colorimeter.
54
Figure 23 - Schematic representation of the asymmetric phospholipd composition of a viable and an
apoptotic mammalian cell membrane. In a viable cell, the membrane is asymmetric and is maintained by
enzymes called flippases. In opposition, in apoptotic cells, the activity of these enzymes is compromised and
is inactivated by other enzymes, termed floppases [adapted from the Biocolor’s protocol, available at
www.biocolor.co.uk].
In this assay, PNT2-derived cell populations were seeded in 96-well plates at a
density of 1x104 cells per well, and incubated at normal growth conditions with RPMI
medium for 96 hours. Besides the three replicates per condition, two additional wells were
used as positive apoptotic controls, in which the medium was substituted by an apoptotic
inducer (hydrogen peroxide at 10mM) 2 hours before the assay. For apoptosis evaluation,
100µlof APOPercentage dye dissolved in medium (1:20) were added to each well. After 2
hours of incubation, the dye was removed and cells were washed twice with 200µl of PBS
to remove the excess of unbound dye. The dye was released from the apoptotic cells
using 100µl of APOPercentage dye release reagent and the absorbance was measured in
the FLUOstar Omega microplate reader (BMG Labtech) at 550nm with background
correction at 630nm. The results were analyzed with the Omega Data analysis software
(BMG Labtech).
Three independent experiments were performed, and results are expressed as a
normalized ratio of each transfected cell line and the empty vector. The Student’s t test
was performed and a p value less than 0.05 was considered statistically significant.
55
5. Exon-array data to compare HOXB13 mRNA expression in
PCa samples from different molecular subtypes
For the analysis of HOXB13 mRNA expression levels in tumor samples from
different ETS subgroups, we used the data obtained from exon-level expression
microarrays that had previously been performed by our group (Paulo et al., 2012b). As
described, we used a series of 48 tumor samples selected from the consecutive series of
200 clinically localized PCa, already typed for ETS rearrangements (Paulo et al., 2012a).
They represented different molecular subtypes, namely 21 samples with ERG
rearrangement, 11 samples with ETV1 rearrangement, two samples with other ETS
rearrangements (one with ETV4 and one with ETV5 rearrangements), and 14 samples
without known ETS. Nine normal prostate tissues (NPTs) obtained from
cystoprostatectomies (CPs) were also screened for control purposes.
Briefly, RNA was extracted and 1µg was processed into cDNA and hybridized to
GeneChip Human Exon 1.0 ST arrays. To obtain gene-level Robust Multi-array Average
(RMA)-normalized expression values, the Affymetrix Expression Console v1.1 software
was used (Paulo et al., 2012b).
Results
59
1. HOXB13 Mutation Analysis
1.1 Primary prostate carcinomas
Sequences were analyzed using the Sequencing Analysis Software v5.4 (Applied
Biosystems) and the Mutation Surveyor V4.0.7 (Softgenetics). Of the 178 prostate
carcinomas sequenced, one heterozygous alteration was found in one case (P308T), at
the position 383 of exon 1 (Figure 24). The mutation consists of a nonsynonymous
substitution of an adenosine for a cytosine (transition, c.383C>A) in the second position of
codon 128 (GCC→GAC), resulting in a nonconservative substitution of aspartic acid for
alanine. According to the Human Genome Variant Society (HGVS) nomenclature, this
alteration should be referred to as p.Ala128Asp. This mutation has not been described in
any published article or in any of the databases consulted, namely, Exome Variant Server,
dbSNP (the database of known DNA sequence variants from NCBI), COSMIC (Catalogue
of somatic mutations in cancer), or Ensembl, the last two from the Wellcome Trust Sanger
Institute.
In order to determine the somatic or germline nature of the mutation found in the
tumor tissue, a DNA sample was extracted from peripheral leukocytes and sequenced,
and it was demonstrated that the mutation was present in the germline.
Figure 24 - DNA sequence chromatogram obtained from PCa tissue from the patient with the HOXB13
c.383C>A, p.Ala128Asp variant.
The possible effect of this missense mutation on the structure and function of the
HOXB13 protein was evaluated using the Polymorphism Phenotyping v2 (PolyPhen-2)
software that uses sequence and structure-based information for prediction (Adzhubei et
al., 2013). Two pairs of datasets from the software were used to evaluate potentially
deleterious effects (HumanVar and HumanDiv). The first contains 13,032 human disease-
causing mutations from the UniProt database, as well as 8,946 human nonsynonymous
60
SNPs without annotated involvement in disease. HumDiv was compilled from 3,155
damaging alleles annotated in the UniProt as causing Mendelian diseases and affecting
protein function or stability, with 6,321 differences between human proteins and their
related mammalian homologs (Adzhubei et al., 2010). According to the two datasets, the
A128D alteration is predicted to be probably damaging to protein function, with a
probability of 0.997 by the HumVar and 1 by the HumDiv (Figure 25).
Figure 25 - Scores of the predicted effect of HOXB13 c.383C>A, p.Ala128Asp variant in protein
function obtained with the two PolyPhen datasets (HumVar, above and HumDiv, below). The score
increases from green to red, with a higher score meaning an increased damage to protein function [adapted
from PolyPhen software available at http://genetics.bwh.harvard.edu/pph2/].
Besides the A128D variant, we have also found the two synonymous variants in
exon 1, c.366C>T and c.513T>C, which do not change the encoded aminoacid. These
variants have already been described as being SNPs, with the dbSNP IDs being rs8556
and rs9900627, respectively (Ewing et al., 2012; Lin et al., 2013). Both SNPs were found
in 36 subjects in our study, with a frequency of approximately 20%.
1.2 Prostate cell lines
The LNCaP, PNT2, 22Rv1 and NCI-H660 prostate cell lines were sequenced to
select our study cell model. Only LNCaP presented an alteration in HOXB13, c.431T>C
(Figure 26), resulting in a nonsynonymous heterozygous substitution of a proline for a
leucine at codon 144, p.Leu144Pro (L144P), which is located in the second MEIS
interacting domain (Ewing et al., 2012; Williams et al., 2005).
61
Figure 26 - DNA sequence chromatogram obtained from the PCa cell line LNCaP.
2. Induction of de novo overexpression of HOXB13 in prostate
cell lines
2.1 Selection of the study cell model
HOXB13 mRNA expression levels in the cell lines was evaluated using
quantitative real-time PCR (Figure 27). NCI-H660, DU145 and PNT2 do not express
HOXB13, in opposition to the other prostate cancer cell lines, with MDA-PCa-2b being the
cell line with higher expression levels, followed by LNCaP, PC-3, 22Rv1, and VCaP.
Figure 27 - HOXB13 expression levels in eight prostate cell lines using quantitative real-time PCR.
HOXB13 expression at the protein level was also evaluated, using Western Blot
(Figure 28). As demonstrated by qRT-PCR, MDA-PCa-2b is the cell line with higher
expression levels of HOXB13, and LNCaP, VCaP, 22Rv1 and PC-3 also show expression
of this protein. NCI-H660, DU145 and PNT2 do not express HOXB13.
62
Figure 28 - Expression of HOXB13 in eight prostate cell lines, evaluated by Western blot.
After evaluation of HOXB13 mutation status and HOXB13 expression levels in the
available prostatic cell lines, the choice to establish HOXB13 models of prostatic
carcinogenesis could be made. As it was mentioned before, two prostatic cell lines with no
HOXB13 mutations were chosen.
To explore if the de novo overexpression of HOXB13 and its mutated forms are
somehow involved in PCa initiation, PNT2, a non-malignant prostate-derived cell line was
chosen. PNT2 was chosen since it was the only non-malignant cell line available at our
lab, and it has minor expression and no mutations of the HOXB13 gene.
For the malignant context, 22Rv1 was the PCa-derived cell line that best fited in
the defined criteria, altough its HOXB13 expression is not as low as we would prefer. The
other cell lines did not fulfilled the criteria for different reasons: LNCaP was excluded
because it has a nonsynonymous mutation in the HOXB13 gene; MDA-PCa-2b was
shown to have high HOXB13 expression levels and it has the MIPOL1:ETV1
rearrangement; VCaP and NCI-H660 also harbor an ETS rearrangement
(TMPRSS2:ERG) and the latter is androgen-independent; and PC-3 and DU145 were
also excluded because they are androgen-independent. We excluded all the cell lines with
ETS rearrangements because the tumor sample with the A128D mutation does not harbor
any rearrangement, (Paulo et al., 2012a) and these could eventually interfere with
HOXB13 expression in our in vitro models.
2.2 Confirmation of site-directed mutagenesis in the HOXB13 expressing
vectors
As mentioned before, Sanger sequencing was performed to confirm the success of
the mutagenesis. The mutagenesis was indeed successful, since we could induce the
substitution of an adenosine for a guanine to construct the vector with the G84E mutation
(Figure 29), and the substitution of an adenosine for a cytosine to construct the vector
harboring the A128D mutation (Figure 30).
63
Figure 29 - DNA sequence chromatogram obtained from the vector with the G84E mutation.
Figure 30 - DNA sequence chromatogram obtained from the vector with the A128D mutation.
3. Analysis of HOXB13 overexpression in transfected cell lines
After the transfection of 22Rv1 and PNT2 cell lines with the four different vectors
(p-Entry, HOXB13-WT, HOXB13-G84E and HOXB13-A128D), the expression of HOXB13
in the altered cells was evaluated by quantitative real-time PCR (Figures 31 and 32).
Figure 31 - HOXB13 expression levels in 22Rv1 transfected cells, normalized for the correspondent
control vector, using quantitative real-time PCR.
64
As it can be observed in Figure 31, there are virtually no differences in the levels of
HOXB13 expression between the cells transfected with different HOXB13 expression
vectors and the control vectors (p-Entry). It remains to be clarified whether the observed
HOXB13 expression is due to “silencing” of the exogenous or the endogenous HOXB13,
as these cell populations are resistant to the selective antibiotic. For the purpose of this
Thesis, these 22Rv1 established cell populations were not useful and no further studies
were done with them.
Figure 32 - HOXB13 expression levels in PNT2 transfected cell populations, normalized for the
correspondent control vector, using quantitative real-time PCR.
Regarding PNT2 transfected cell populations (Figure 32), the expression levels
vary not only depending on the vector that was inserted, but also between different
transfection conditions (TurboFectin-DNA ratio). In fact, there is not a common
transfection condition that is more efficient than the others. Due to this fact, for each
HOXB13 expressing vector, we decided to choose two populations to pursue our studies
taking the expression levels into account. Thus, the two transfection conditions of each
vector with higher HOXB13 expression were selected. Regarding the control vector (p-
Entry), as there were no significant differences between the three established populations,
one was randomly chosen. The selected cell populations and their HOXB13 expression
levels normalized for the control vector are represented in Figure 33.
65
Figure 33 - HOXB13 expression levels in the selected PNT2 transfected cells, normalized for the
control vector, using quantitative real-time PCR.
4. Biological role of HOXB13 – phenotypic assays
In order to investigate a potential oncogenic role of HOXB13 in prostate
carcinogenesis, two phenotypic assays, namely MTT proliferation assay and apoptosis
assay, were performed as previously described.
4.1 Cell proliferation assay
The role of HOXB13 and its mutated forms, G84E and A128D, in cell proliferation
was evaluated in vitro using the MTT colorimetric assay. Proliferation results were only
considered for three different time-points, because the growth rates obtained for the 24
and 48 hours’ time-points were not consistent between experiments.
The overexpression of the wild-type form of HOXB13 has shown to have a growth
stimulatory effect on PNT2 cell line at 72 hours. However, this effect was only statistically
significant for one of the overexpressing populations (P1; p=0.03), being borderline for the
other cell population (P2; p=0.054). This tendency is maintained at 96 hours, although the
values do not reach statistical significance. Because these data were obtained from only
two independent experiments, additional replicates will strengthen the significance of the
results. None of the mutations seem to alter PNT2 growth rate when compared to the
control vector (Figure 34).
66
Figure 34 - Quantification of cell viability in PNT2- derived cells (p-Entry, HOXB13-WT, HOXB13-G84E
and HOXB13-A128D) evaluated at two different time-points (72 and 96 hours). Bars represent standard
deviation.
4.2 Cell apoptosis assay
We evaluated if the overexpression of HOXB13 and of its two mutated forms could
influence apoptosis in PNT2-derived cell populations, and quantitative analysis was
performed through the measurement of the concentration of the dye used, in a microplate
reader.
We observed that cells with overexpression of HOXB13-A128D variant have
decreased apoptotic rates when compared to the control (Figure 35). This result was
statistically significant (p=0.02) for the P2 population and borderline for the P1 population.
The wild-type form of HOXB13 and the mutated form G84E do not seem to influence cell
death through apoptosis in vitro. As for the MTT assay, and although these data were
obtained from three independent experiments, additional replicates may strengthen these
observations.
67
Figure 35 - Quantification of apoptotic rates in PNT2 derived cells (HOXB13-WT, HOXB13-G84E,
HOXB13-A128D and in the control vector – p-Entry), evaluated after 96h in culture. Bars represent
standard deviation.
5. HOXB13 mRNA Expression Pattern by Molecular Subtype
HOXB13 mRNA expression levels were evaluated in NPTs and in prostate tumors
from different molecular subtypes, using the exon-array data available from our group.
When we compared HOXB13 expression between all PCa and NPTs, no statistically
significant differences were found (Figure 36). Regarding HOXB13 mRNA expression
levels in prostate tumors, we have observed differences between different molecular
subtypes, although these were not statistically significant (Kruskal-Wallis test, p value =
0.053). For the several two-group comparisons, Mann-Whitney tests were performed, and
a significantly higher expression of HOXB13 was observed in tumors with ETS
rearrangements involving the PEA3 family members (ETV1, ETV4 and ETV5) when
compared to ERG-positive tumors (p value = 0.004), as illustrated in Figure 37.
68
Figure 36 - Boxplot representing HOXB13 mRNA expression in NPTs and in prostate carcinomas. RMA
- Robust Multi-array Average.
Figure 37 - Boxplot representing HOXB13 expression in the different PCa ETS subgroups and in
normal mucosa obtained from CPs. RMA - Robust Multi-array Average.
Discussion
71
The HOX genes are highly conserved genes that belong to the homeobox
superfamily of transcription factors, and their coordinated expression is responsible for the
correct pattern formation of the body, during embryonic development (Daftary and Taylor,
2006; Ewing et al., 2012; Jung et al., 2004b). Deregulation of their expression have been
associated with many types of neoplasias, such as PCa; however, the mechanisms by
which they promote carcinogenesis are not entirely understood and could be different
depending on the type of malignancy (Chen and Sukumar, 2003; Gray et al., 2011; Habel
et al., 2013; Jung et al., 2005; Kelly et al., 2011; Zhao et al., 2005).
HOXB13, such as other HOX13 paralogs, is very important for normal prostate
development (Economides and Capecchi, 2003; Huang et al., 2007), but it is the only that
continues to be expressed through adulthood and its expression levels are also
maintained in PCa (Ewing et al., 2012; Jung et al., 2004a; Jung et al., 2004b; Kim et al.,
2010b). The role of HOXB13 in prostate carcinogenesis remains elusive so far, with some
studies implicating HOXB13 as a TSG and others as an oncogene, and its mode of action
seems to depend on the cellular environment and on the cell type (Jung et al., 2004a;
Jung et al., 2004b; Kim et al., 2010b).
More recently, some germline mutations were described in the HOXB13 gene,
namely G84E in populations of European descent (Ewing et al., 2012) and G135E in
Chinese men (Lin et al., 2013), which seem to confer a higher risk for PCa development.
However, the exact mechanisms by which they promote carcinogenesis are not
understood. Furthermore, it is unknown if HOXB13 somatic mutations play a role in
prostate cancer in addition to inherited predisposition. We therefore started by searching
for HOXB13 variants in prostate carcinomas, with the idea to test any mutation found for
its somatic or germline nature.
1. HOXB13 mutation analysis in primary prostate carcinomas
The recurrent mutations G84E and G135E respectively in Western and Chinese
populations (Ewing et al., 2012; Lin et al., 2013) were not detected in this series. While it
was not expected that the Chinese mutation would be found, we cannot completely
exclude that the G84E mutation exists in our population because of the limited number of
cases analyzed. However, this variant is known to be more frequent in the Nordic
countries than in South Europe.
Interestingly, we here describe for the first time the presence of the
nonsynonymous mutation p.Ala128Asp in the HOXB13 gene, which, like the other
mutations so far detected, is a heterozygous germline mutation. It is important to mention
that, like in G84E carriers, HOXB13 expression is maintained. This variant was predicted
72
to be probably damaging to protein function by two different in silico analyses (Adzhubei
et al., 2013). The patient carrying this germline mutation had prostate cancer, bladder
cancer and gastric cancer diagnosed at 65, 58 and 74 years of age. Since his family
history is otherwise unremarkable besides a sister with gastric cancer, it would be
important to test additional relatives to find out who else is at risk and if this is a de novo
germline mutation.
It is conceivable that we would find additional cases with HOXB13 germline
mutations if we increased our sample size and if we had selected patients with a positive
family history or early-onset of the disease, instead of sequencing DNA from a
consecutive series of prostate carcinomas. However, this strategy was appropriate for our
initial purpose of finding somatic HOXB13 mutations, while at the same time having the
possibility to test if any detected variant was somatic or germline, which was exactly what
happened in this case. To assess the importance of this novel variant in PCa
susceptibility, it will be essential to evaluate not only more cases (subjects with PCa) but
also controls, in order to estimate the relative risk of PCa development. Nevertheless, the
discovery of this new genetic variant further implicates HOXB13 in prostate
carcinogenesis, and different populations may have their own founder mutations
predisposing to prostate cancer.
Regarding the two SNPs found, rs8556 and rs9900627, they were found with
similar frequencies as those previously reported (Ewing et al., 2012). Although many
SNPs associated with PCa risk have already been described (Karlsson et al., 2012; Xu et
al., 2013), the odd ratios (ORs) are so low that it prevents their use in clinical practice
(Akbari et al., 2012).
2. Biological role of HOXB13
Given the importance of HOXB13 in prostate biology and since there is still a
debate surrounding its importance in prostate carcinogenesis, we tried to clarify if it
behaves like a TSG or like an oncogene. We also intended to gain insight on the
phenotypic consequences of the G84E mutation and of the new A128D variant discovered
in this work, so we investigated for the first time the potential oncogenic role of these
mutations in prostate carcinogenesis.
2.1 Selection of the study cell model
To choose our study models, HOXB13 was first sequenced in three prostate cell
lines, NCI-H660, PNT2 and LNCaP, to verify if they harbored any HOXB13 mutations. The
first two have not been sequenced yet by any other group (or at least it has not been
published), but we found no mutations. Regarding LNCaP, we have confirmed the
73
presence of a nonsynonymous mutation, p.Leu144Pro, already described by Ewing and
collaborators (Ewing et al., 2012). Besides performing Sanger sequencing, we have also
evaluated HOXB13 expression both at the mRNA and protein level. The pattern of
expression of HOXB13 in our cell lines was similar to those described in the literature
(Jung et al., 2004b; Kim et al., 2010b), except for the PC-3 cell line. It has been described
that only AR-expressing prostate cells express HOXB13 (Jung et al., 2004b; Kim et al.,
2010b) and PC-3, being an AR-negative PCa cell line, should not express HOXB13.
Regarding NCI-H660 and VCaP, the obtained results are consistent with their AR status:
VCaP expresses HOXB13, while NCI-H660 (AR-negative) only presents minor
expression.
22Rv1 and PNT2 were the chosen cell lines where we induced de novo
overexpression of the HOXB13-WT form and HOXB13 mutated forms. However, we could
not increase HOXB13 expression in the 22Rv1 PCa-derived cell line. It was confirmed by
qRT-PCR and by immunoblotting that the 22Rv1 wild-type already expresses HOXB13,
and we do not know for sure what the transfected cells are expressing. As mentioned
before, these cell populations are resistant to the selective antibiotic, so the transfection
was not the problem. It is possible that the transfected cells might be “silencing” the
exogenous or the endogenous HOXB13, by an unknown mechanism. Using a specific
antibody for the exogenous HOXB13 (4C5-AntiDDK monoclonal antibody) would allow
clarifying this phenomenon. Thus, only PNT2 was used as a cell model to study cell
proliferation and apoptosis.
2.2 Phenotypic assays
Functional evaluation revealed that overexpression of HOXB13-WT in PNT2-
derived cells influences their proliferative phenotype, with an increase in cell proliferation.
These results were only statistically significant for one time-point (72 hours), but this
tendency was also present for the other evaluated time-point. This might be due to the
fact that only two independent experiments were performed, and increasing the number of
experiments would increase the consistency of the results. The results only reached
statistical significance for the P1 population, and one possible explanation might be the
fact that this population expresses more HOXB13 than P2. This increase in cell
proliferation was consistent with another study, although in a different cell line (Kim et al.,
2010b), where the overexpression of HOXB13 in the LNCaP PCa-derived cell line
markedly promoted cell proliferation in the absence of androgens, by inhibiting the
expression of the p21 TSG that activates the RB-E2F signaling, as explained before. The
results obtained by Kim and collaborators indicate that HOXB13 overexpression confers a
growth advantage for PCa cells in an androgen-deprived environment, and could explain
74
why some patients relapse after androgen ablation (Kim et al., 2010a; Kim et al., 2010b).
In our model, more studies are needed to understand the mechanism by which HOXB13
overexpression promotes cell proliferation. We also would like to perform the same
experiment in the presence of androgens, to verify if, in contrast, HOXB13 would have an
inhibitory effect in cell growth, as described by other groups (Jung et al., 2004a; Jung et
al., 2004b). Another observation was that apoptosis tended to decrease in PNT2-derived
cells harboring the A128D mutation. As verified for the cell proliferation assay, the results
only reach statistical significance for the population with higher HOXB13 expression (P2),
which might indicate a dose effect. Once more it is possible that by increasing the number
of independent experiments the significance of the results could be strengthen.
The results obtained with the two phenotypic assays suggest a carcinogenic
mechanism consistent with a gain of function, typical of oncogenes. However, more
studies are needed to confirm this hypothesis. One possible way to confirm these results
could be by using an opposite approach, such as silencing HOXB13 in a cell line with high
expression levels, like MDA-PCa-2b, for example, and confirm if the absence of HOXB13
would have a suppressive effect on cell growth. Although we did not reach any
conclusions regarding the G84E mutation, in the first description of this mutation (Ewing et
al., 2012) there was a suspicion that it might also act as an oncogene due to its recurrent
nature. Since there is a lack of truncating mutations in HOXB13 in patients with PCa
(Ewing et al., 2012), it is more plausible that both mutations have a gain of function
mechanism.
One limitation of our study was the fact that we only evaluated two functional
capabilities associated to cancer, which were increased cell proliferation and evasion of
apoptosis. We also should evaluate if HOXB13 and its mutated forms are somehow
involved in other hallmarks of cancer, such as tissue invasion and metastasis, since
sooner or later almost every cancers acquire these abilities, which are the primary causes
for a fatal outcome (Behrens, 1993; Hanahan and Weinberg, 2000). Regarding these two
features, we could perform two phenotypic assays: a colony formation assay that
evaluates anchorage-independent growth of cancer cells and a cell invasion assay.
3. HOXB13 mRNA Expression Pattern by Molecular Subtype
We have also proposed to verify if HOXB13 mRNA expression was altered in
tumors, when compared to NPTs, but we have not found such global difference. Our
results are in agreement with other studies (Jung et al., 2004a; Jung et al., 2004b; Kim et
al., 2010a; Kim et al., 2010b) that state that it is difficult to observe such differences.
Additionally, those studies reported that different populations of HOXB13-negative and
75
HOXB13-positive cells could be observed in the same tumor, which could be a reasonable
explanation for the absence of such differences. On the other hand, other studies reported
a higher expression of HOXB13 in tumors, when compared to the normal prostate
(Edwards et al., 2005; Grossmann et al., 2001). However, all of these studies used
matching benign tissue adjacent to tumor site, which has not been obtained by
microdissection techniques and which raises doubts about the truly benign nature of these
samples, since PCa is very heterogeneous and usually multifocal. To avoid these
problems, here we used NPTs collected from CPs specimens of bladder cancer patients.
The last goal of this project was to find out if HOXB13 mRNA expression was
somehow related to the molecular subtype of the tumor, more specifically with different
ETS subgroups, and we found that HOXB13 was overexpressed in the tumors belonging
to the PEA3 family. Unfortunately, we have no other studies to compare with our results.
Since the majority of samples from the PEA3 family were samples with an ETV1
rearrangement (11 of 13), it will be interesting to study the role of HOXB13 in the ETV1
context. For that we could use the LNCaP PCa-cell line with ETV1 silencing (shETV1),
which was already established in our lab, and induce HOXB13 overexpression in the
presence and absence of androgen stimuli and evaluate the phenotypic impacts. It has
been previously described that overexpression of HOXB13 in LNCaP has different
consequences depending on the presence or absence of androgens (Jung et al., 2004b;
Kim et al., 2010b) and this model will clarified if this opposite effect is in fact only
dependent on androgens or if, otherwise, is related with ETV1. Additionally, it would be
interesting to evaluate the phenotypic impact of silencing HOXB13 in the MDA-PCa-2b
cell line, which also has an ETV1 rearrangement and shows high expression but no point
mutations in HOXB13.
Conclusions and Future
Perspectives
79
A new nonsynonymous mutation, p.Ala128DAsp, was identified in the
HOXB13 gene and was predicted to be damaging to protein structure. This
reinforces the fact that HOXB13 is involved in prostate cancer
development. To find more cases harboring this genetic alteration, in the
future we could screen more cases and select subjects with a family history
of PCa and early-onset of the disease.
By performing functional assays in PNT2-derived cells in the absence of
androgens, we observed alterations in two hallmarks of cancer, both
suggesting a potential oncogenic mechanism for HOXB13 in prostate
carcinogenesis. Specifically, we saw an increase in cell proliferation in cells
overexpressing the wild-type form of the gene, and a decrease in apoptosis
in cells overexpressing the novel mutated form A128D. However, more
replicates are required to strengthen the significance of the results, mainly
for the cell proliferation assay.
To validate the observations of the functional assays performed in the
PNT2 cell line, we could use an opposite approach, using a PCa-derived
cell line with HOXB13 expression and perform in vitro silencing.
We also intend to verify whether 22Rv1-derived cell populations are
expressing the exogenous HOXB13, so we will use a specific antibody for
the exogenous HOXB13. If this expression is confirmed, 22Rv1 PCa-cell
line could be used, as initially planned, as a cell model to study further the
role of HOXB13 in prostate carcinogenesis.
We also would like to perform more functional assays, like colony formation
and invasion assays, to see if HOXB13 and its mutated forms are involved
in other stages of carcinogenesis (tissue invasion and metastasis). All of
these functional studies should also be replicated in an androgen rich
environment, to verify if PNT2-derived cells would behave differently.
In this study, HOXB13 mRNA levels were compared for the first time
between different PCa molecular subtypes, and we here report that
HOXB13 is mainly overexpressed in tumors with ETV1 rearrangements.
Thus, we aim to study the role of HOXB13 in an ETV1 context, using the
LNCaP-shETV1 model and the MDA-PCa-2b cell line.
References
83
Adzhubei, I., Jordan, D.M., and Sunyaev, S.R. (2013). Predicting functional effect of
human missense mutations using PolyPhen-2. Current protocols in human
genetics / editorial board, Jonathan L Haines [et al] Chapter 7, Unit7 20.
Adzhubei, I.A., Schmidt, S., Peshkin, L., Ramensky, V.E., Gerasimova, A., Bork, P.,
Kondrashov, A.S., and Sunyaev, S.R. (2010). A method and server for
predicting damaging missense mutations. Nature methods 7, 248-249.
Akbari, M.R., Trachtenberg, J., Lee, J., Tam, S., Bristow, R., Loblaw, A., Narod, S.A.,
and Nam, R.K. (2012). Association between germline HOXB13 G84E mutation
and risk of prostate cancer. Journal of the National Cancer Institute 104, 1260-
1262.
Ali-Boucetta, H., Al-Jamal, K., and Kostarelos, K. (2011). Cytotoxic Assessment of
Carbon Nanotube Interaction with Cell Cultures. In Biomedical Nanotechnology,
S.J. Hurst, ed. (Humana Press), pp. 299-312.
Alvarez-Cubero, M.J., Saiz, M., Martinez-Gonzalez, L.J., Alvarez, J.C., Lorente, J.A.,
and Cozar, J.M. (2012). Genetic analysis of the principal genes related to
prostate cancer: A review. Urologic oncology.
American Cancer Society. (2012). Cancer facts and figures 2012
http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/document
s/document/acspc-031941.pdf
Amin, M.K., A.; Tazeen, N.; Yasoob, M. (2010). Zonal Anatomy of Prostate. Annals of
KEMU 16, 138-142.
Bambury, R.M., and Gallagher, D.J. (2012). Prostate cancer: germline prediction for a
commonly variable malignancy. BJU international 110, E809-818.
Behrens, J. (1993). The role of cell adhesion molecules in cancer invasion and
metastasis. Breast cancer research and treatment 24, 175-184.
Berthon, P., Cussenot, O., Hopwood, L., Leduc, A., and Maitland, N. (1995). Functional
expression of sv40 in normal human prostatic epithelial and fibroblastic cells -
differentiation pattern of nontumorigenic cell-lines. International journal of
oncology 6, 333-343.
84
Bessede, T., and Patard, J.J. (2012). Words of wisdom. Re: Germline mutations in
HOXB13 and prostate-cancer risk. European urology 61, 1062.
Bieberich, C.J., Fujita, K., He, W.W., and Jay, G. (1996). Prostate-specific and
androgen-dependent expression of a novel homeobox gene. The Journal of
biological chemistry 271, 31779-31782.
Bosland, M.C., and Mahmoud, A.M. (2011). Hormones and prostate carcinogenesis:
Androgens and estrogens. Journal of carcinogenesis 10, 33.
Boyd, L.K., Mao, X., and Lu, Y.J. (2012). The complexity of prostate cancer: genomic
alterations and heterogeneity. Nature reviews Urology 9, 652-664.
Boyle, P., B. Levin, et al. (2008). World cancer report 2008. World Health Organization,
Lyon, Distributed by WHO Press.
Breyer, J.P., Avritt, T.G., McReynolds, K.M., Dupont, W.D., and Smith, J.R. (2012).
Confirmation of the HOXB13 G84E germline mutation in familial prostate
cancer. Cancer epidemiology, biomarkers & prevention : a publication of the
American Association for Cancer Research, cosponsored by the American
Society of Preventive Oncology 21, 1348-1353.
Buhmeida, A., Pyrhonen, S., Laato, M., and Collan, Y. (2006). Prognostic factors in
prostate cancer. Diagnostic pathology 1, 4.
Chen, H., and Sukumar, S. (2003). HOX genes: emerging stars in cancer. Cancer
biology & therapy 2, 524-525.
Chen, Z., Greenwood, C., Isaacs, W.B., Foulkes, W.D., Sun, J., Zheng, S.L.,
Condreay, L.D., and Xu, J. (2013). The G84E mutation of HOXB13 is
associated with increased risk for prostate cancer: results from the REDUCE
trial. Carcinogenesis 34, 1260-1264.
Cheng, L., Montironi, R., Bostwick, D.G., Lopez-Beltran, A., and Berney, D.M. (2012).
Staging of prostate cancer. Histopathology 60, 87-117.
Chrisofos, M., Papatsoris, A.G., Lazaris, A., and Deliveliotis, C. (2007). Precursor
lesions of prostate cancer. Critical reviews in clinical laboratory sciences 44,
243-270.
Cropp, C.D., Simpson, C.L., Wahlfors, T., Ha, N., George, A., Jones, M.S., Harper, U.,
Ponciano-Jackson, D., Green, T.A., Tammela, T.L., et al. (2011). Genome-wide
85
linkage scan for prostate cancer susceptibility in Finland: evidence for a novel
locus on 2q37.3 and confirmation of signal on 17q21-q22. International journal
of cancer Journal international du cancer 129, 2400-2407.
Cross, D.S., and Burmester, J.K. (2008). Functional characterization of the HOXB13
promoter region. Med Oncol 25, 287-293.
Daftary, G.S., and Taylor, H.S. (2006). Endocrine regulation of HOX genes. Endocrine
reviews 27, 331-355.
Damber, J.E.A., G. (2008). Prostate cancer. Lancet 371, 1710-1721.
De Marzo, A.M., DeWeese, T.L., Platz, E.A., Meeker, A.K., Nakayama, M., Epstein,
J.I., Isaacs, W.B., and Nelson, W.G. (2004). Pathological and molecular
mechanisms of prostate carcinogenesis: implications for diagnosis, detection,
prevention, and treatment. J Cell Biochem 91, 459-477.
De Marzo, A.M., Marchi, V.L., Epstein, J.I., and Nelson, W.G. (1999). Proliferative
inflammatory atrophy of the prostate: implications for prostatic carcinogenesis.
The American journal of pathology 155, 1985-1992.
De Marzo, A.M., Meeker, A.K., Zha, S., Luo, J., Nakayama, M., Platz, E.A., Isaacs,
W.B., and Nelson, W.G. (2003). Human prostate cancer precursors and
pathobiology. Urology 62, 55-62.
De Marzo, A.M., Platz, E.A., Sutcliffe, S., Xu, J., Gronberg, H., Drake, C.G., Nakai, Y.,
Isaacs, W.B., and Nelson, W.G. (2007). Inflammation in prostate
carcinogenesis. Nature reviews Cancer 7, 256-269.
Denizot, F., and Lang, R. (1986). Rapid colorimetric assay for cell growth and survival.
Modifications to the tetrazolium dye procedure giving improved sensitivity and
reliability. Journal of immunological methods 89, 271-277.
Deutsch, E., Maggiorella, L., Eschwege, P., Bourhis, J., Soria, J.C., and Abdulkarim, B.
(2004). Environmental, genetic, and molecular features of prostate cancer. The
lancet oncology 5, 303-313.
Dixon, J.S.C., P. H.; Gosling, J. A. (1999). Anatomy and function of the prostate gland.
In Textbook of prostatitis, J.C. Nickel, ed. (Oxford, UK: Isis Medical Media), pp.
33-39.
86
Eble, J.N., Sauter, G., Epstein, J.I., Sesterhenn, I.A., and Research, W.H.O.a.I.A.f.C.
(2004). Pathology and Genetics of Tumours of the Urinary System and Male
Genital Organs (Lyon: IARC Press).
Economides, K.D., and Capecchi, M.R. (2003). Hoxb13 is required for normal
differentiation and secretory function of the ventral prostate. Development 130,
2061-2069.
Economides, K.D., Zeltser, L., and Capecchi, M.R. (2003). Hoxb13 mutations cause
overgrowth of caudal spinal cord and tail vertebrae. Developmental biology 256,
317-330.
Edge, S.B., and Compton, C.C. (2010). The American Joint Committee on Cancer: the
7th edition of the AJCC cancer staging manual and the future of TNM. Annals of
surgical oncology 17, 1471-1474.
Edge, S.B.B., D.R.; Compton, C.C.; Fritz, A.G.; Greene, F.L.; Trotti, A. (Eds.) (2010).
AJCC Cancer Staging Manual, 7th edition edn (Berlin: Springer).
Edwards, S., Campbell, C., Flohr, P., Shipley, J., Giddings, I., Te-Poele, R., Dodson,
A., Foster, C., Clark, J., Jhavar, S., et al. (2005). Expression analysis onto
microarrays of randomly selected cDNA clones highlights HOXB13 as a marker
of human prostate cancer. British journal of cancer 92, 376-381.
Elmore, S. (2007). Apoptosis: a review of programmed cell death. Toxicologic
pathology 35, 495-516.
Epstein, J.I. (2010). An update of the Gleason grading system. The Journal of urology
183, 433-440.
Ewing, C.M., Ray, A.M., Lange, E.M., Zuhlke, K.A., Robbins, C.M., Tembe, W.D.,
Wiley, K.E., Isaacs, S.D., Johng, D., Wang, Y., et al. (2012). Germline
mutations in HOXB13 and prostate-cancer risk. N Engl J Med 366, 141-149.
Ferlay, J., Shin, H.R., Bray, F., Forman, D., Mathers, C., and Parkin, D.M. (2010).
Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008.
International journal of cancer Journal international du cancer 127, 2893-2917.
Gerlier, D., and Thomasset, N. (1986). Use of MTT colorimetric assay to measure cell
activation. Journal of immunological methods 94, 57-63.
87
Gonzalgo, M.L., and Isaacs, W.B. (2003). Molecular pathways to prostate cancer. The
Journal of urology 170, 2444-2452.
Gray, S., Pandha, H.S., Michael, A., Middleton, G., and Morgan, R. (2011). HOX genes
in pancreatic development and cancer. JOP : Journal of the pancreas 12, 216-
219.
Greene, K.L., Albertsen, P.C., Babaian, R.J., Carter, H.B., Gann, P.H., Han, M., Kuban,
D.A., Sartor, A.O., Stanford, J.L., Zietman, A., et al. (2013). Prostate specific
antigen best practice statement: 2009 update. The Journal of urology 189, S2-
S11.
Gronberg, H. (2003). Prostate cancer epidemiology. Lancet 361, 859-864.
Grossmann, M.E., Wood, M., and Celis, E. (2001). Expression, specificity and
immunotherapy potential of prostate-associated genes in murine cell lines.
World journal of urology 19, 365-370.
Habel, L.A., Sakoda, L.C., Achacoso, N., Ma, X.J., Erlander, M.G., Sgroi, D.C.,
Fehrenbacher, L., Greenberg, D., and Quesenberry, C.P., Jr. (2013).
HOXB13:IL17BR and molecular grade index and risk of breast cancer death
among patients with lymph node-negative invasive disease. Breast cancer
research : BCR 15, R24.
Hammerich, K.H.A., G. E.; Wheeler, T. M. (2009). Anatomy of the prostate gland
and surgical pathology of prostate cancer. In Prostate Cancer, H.S. Hricak, P.
T., ed. (Cambridge: Cambridge University Press.), pp. 1-10.
Hanahan, D., and Weinberg, R.A. (2000). The hallmarks of cancer. Cell 100, 57-70.
Harnden, P., Shelley, M.D., Coles, B., Staffurth, J., and Mason, M.D. (2007). Should
the Gleason grading system for prostate cancer be modified to account for high-
grade tertiary components? A systematic review and meta-analysis. The lancet
oncology 8, 411-419.
Heidenreich, A., Aus, G., Bolla, M., Joniau, S., Matveev, V.B., Schmid, H.P., and
Zattoni, F. (2008). EAU guidelines on prostate cancer. European urology 53,
68-80.
Heidenreich, A., Bellmunt, J., Bolla, M., Joniau, S., Mason, M., Matveev, V., Mottet, N.,
Schmid, H.P., van der Kwast, T., Wiegel, T., et al. (2011). EAU guidelines on
88
prostate cancer. Part 1: screening, diagnosis, and treatment of clinically
localised disease. European urology 59, 61-71.
Henderson, B.E., and Feigelson, H.S. (2000). Hormonal carcinogenesis.
Carcinogenesis 21, 427-433.
Henrique, R., and Jeronimo, C. (2004). Molecular detection of prostate cancer: a role
for GSTP1 hypermethylation. European urology 46, 660-669; discussion 669.
Hoedemaeker, R.F., Vis, A.N., and Van Der Kwast, T.H. (2000). Staging prostate
cancer. Microscopy research and technique 51, 423-429.
Huang, L., Pu, Y., Hepps, D., Danielpour, D., and Prins, G.S. (2007). Posterior Hox
gene expression and differential androgen regulation in the developing and
adult rat prostate lobes. Endocrinology 148, 1235-1245.
Huang, W., and Waknitz, M. (2009). ETS gene fusions and prostate cancer. American
journal of translational research 1, 341-351.
Jani, A.B., and Hellman, S. (2003). Early prostate cancer: clinical decision-making.
Lancet 361, 1045-1053.
Jemal, A., Bray, F., Center, M.M., Ferlay, J., Ward, E., and Forman, D. (2011). Global
cancer statistics. CA: a cancer journal for clinicians 61, 69-90.
Joshua, A.M., Evans, A., Van der Kwast, T., Zielenska, M., Meeker, A.K., Chinnaiyan,
A., and Squire, J.A. (2008). Prostatic preneoplasia and beyond. Biochimica et
biophysica acta 1785, 156-181.
Jung, C., Kim, R.S., Lee, S.J., Wang, C., and Jeng, M.H. (2004a). HOXB13
homeodomain protein suppresses the growth of prostate cancer cells by the
negative regulation of T-cell factor 4. Cancer research 64, 3046-3051.
Jung, C., Kim, R.S., Zhang, H., Lee, S.J., Sheng, H., Loehrer, P.J., Gardner, T.A.,
Jeng, M.H., and Kao, C. (2005). HOXB13 is downregulated in colorectal cancer
to confer TCF4-mediated transactivation. British journal of cancer 92, 2233-
2239.
Jung, C., Kim, R.S., Zhang, H.J., Lee, S.J., and Jeng, M.H. (2004b). HOXB13 induces
growth suppression of prostate cancer cells as a repressor of hormone-
activated androgen receptor signaling. Cancer research 64, 9185-9192.
89
Karlsson, R., Aly, M., Clements, M., Zheng, L., Adolfsson, J., Xu, J., Gronberg, H., and
Wiklund, F. (2012). A Population-based Assessment of Germline HOXB13
G84E Mutation and Prostate Cancer Risk. European urology.
Kelly, Z.L., Michael, A., Butler-Manuel, S., Pandha, H.S., and Morgan, R.G. (2011).
HOX genes in ovarian cancer. Journal of ovarian research 4, 16.
Kerr, J.F., Wyllie, A.H., and Currie, A.R. (1972). Apoptosis: a basic biological
phenomenon with wide-ranging implications in tissue kinetics. British journal of
cancer 26, 239-257.
Kim, S.D., Park, R.Y., Kim, Y.R., Kim, I.J., Kang, T.W., Nam, K.I., Ahn, K.Y., Bae, C.S.,
Kim, B.Y., Park, S.S., et al. (2010a). HOXB13 is co-localized with androgen
receptor to suppress androgen-stimulated prostate-specific antigen expression.
Anatomy & cell biology 43, 284-293.
Kim, Y.R., Oh, K.J., Park, R.Y., Xuan, N.T., Kang, T.W., Kwon, D.D., Choi, C., Kim,
M.S., Nam, K.I., Ahn, K.Y., et al. (2010b). HOXB13 promotes androgen
independent growth of LNCaP prostate cancer cells by the activation of E2F
signaling. Molecular cancer 9, 124.
Kluzniak, W., Wokolorczyk, D., Kashyap, A., Jakubowska, A., Gronwald, J., Huzarski,
T., Byrski, T., Debniak, T., Golab, A., Gliniewicz, B., et al. (2013). The G84E
mutation in the HOXB13 gene is associated with an increased risk of prostate
cancer in Poland. The Prostate 73, 542-548.
Knouf, E.C., Metzger, M.J., Mitchell, P.S., Arroyo, J.D., Chevillet, J.R., Tewari, M., and
Miller, A.D. (2009). Multiple integrated copies and high-level production of the
human retrovirus XMRV (xenotropic murine leukemia virus-related virus) from
22Rv1 prostate carcinoma cells. Journal of virology 83, 7353-7356.
Krumlauf, R. (1994). Hox genes in vertebrate development. Cell 78, 191-201.
Laitinen, V.H., Wahlfors, T., Saaristo, L., Rantapero, T., Pelttari, L.M., Kilpivaara, O.,
Laasanen, S.L., Kallioniemi, A., Nevanlinna, H., Aaltonen, L., et al. (2013).
HOXB13 G84E mutation in Finland: population-based analysis of prostate,
breast, and colorectal cancer risk. Cancer epidemiology, biomarkers &
prevention : a publication of the American Association for Cancer Research,
cosponsored by the American Society of Preventive Oncology 22, 452-460.
90
Lange, E.M., Robbins, C.M., Gillanders, E.M., Zheng, S.L., Xu, J., Wang, Y., White,
K.A., Chang, B.L., Ho, L.A., Trent, J.M., et al. (2007). Fine-mapping the putative
chromosome 17q21-22 prostate cancer susceptibility gene to a 10 cM region
based on linkage analysis. Human genetics 121, 49-55.
Lappin, T.R., Grier, D.G., Thompson, A., and Halliday, H.L. (2006). HOX genes:
seductive science, mysterious mechanisms. The Ulster medical journal 75, 23-
31.
Lin, X., Qu, L., Chen, Z., Xu, C., Ye, D., Shao, Q., Wang, X., Qi, J., Zhou, F., Wang, M.,
et al. (2013). A novel germline mutation in HOXB13 is associated with prostate
cancer risk in Chinese men. The Prostate 73, 169-175.
Lin, X., Tascilar, M., Lee, W.H., Vles, W.J., Lee, B.H., Veeraswamy, R., Asgari, K.,
Freije, D., van Rees, B., Gage, W.R., et al. (2001). GSTP1 CpG island
hypermethylation is responsible for the absence of GSTP1 expression in human
prostate cancer cells. The American journal of pathology 159, 1815-1826.
Lynch, H.T., and Shaw, T.G. (2013). Familial prostate cancer and HOXB13 founder
mutations: geographic and racial/ethnic variations. Human genetics 132, 1-4.
MacInnis, R.J., Severi, G., Baglietto, L., Dowty, J.G., Jenkins, M.A., Southey, M.C.,
Hopper, J.L., and Giles, G.G. (2013). Population-based estimate of prostate
cancer risk for carriers of the HOXB13 missense mutation G84E. PloS one 8,
e54727.
Magi, B., and Liberatori, S. (2005). Immunoblotting techniques. Methods Mol Biol 295,
227-254.
McMullin, R.P., Mutton, L.N., and Bieberich, C.J. (2009). Hoxb13 regulatory elements
mediate transgene expression during prostate organogenesis and
carcinogenesis. Developmental dynamics : an official publication of the
American Association of Anatomists 238, 664-672.
McNeal, J.E. (1969). Origin and development of carcinoma in the prostate. Cancer 23,
24-34.
McNeal, J.E. (1981). The zonal anatomy of the prostate. The Prostate 2, 35-49.
McNeal, J.E. (1988). Normal histology of the prostate. The American journal of surgical
pathology 12, 619-633.
91
Mertz, K.D., Setlur, S.R., Dhanasekaran, S.M., Demichelis, F., Perner, S., Tomlins, S.,
Tchinda, J., Laxman, B., Vessella, R.L., Beroukhim, R., et al. (2007). Molecular
characterization of TMPRSS2-ERG gene fusion in the NCI-H660 prostate
cancer cell line: a new perspective for an old model. Neoplasia 9, 200-206.
Mosmann, T. (1983). Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays. Journal of immunological
methods 65, 55-63.
Mottet, N., Bellmunt, J., Bolla, M., Joniau, S., Mason, M., Matveev, V., Schmid, H.P.,
Van der Kwast, T., Wiegel, T., Zattoni, F., et al. (2011). EAU guidelines on
prostate cancer. Part II: Treatment of advanced, relapsing, and castration-
resistant prostate cancer. European urology 59, 572-583.
Nakai, Y., and Nonomura, N. (2013). Inflammation and prostate carcinogenesis.
International journal of urology : official journal of the Japanese Urological
Association 20, 150-160.
Narod, S.A., Seth, A., and Nam, R. (2008). Fusion in the ETS gene family and prostate
cancer. British journal of cancer 99, 847-851.
Nelson, W.G., De Marzo, A.M., Deweese, T.L., Lin, X., Brooks, J.D., Putzi, M.J.,
Nelson, C.P., Groopman, J.D., and Kensler, T.W. (2001). Preneoplastic
prostate lesions: an opportunity for prostate cancer prevention. Annals of the
New York Academy of Sciences 952, 135-144.
Nelson, W.G., De Marzo, A.M., and Isaacs, W.B. (2003). Prostate cancer. N Engl J
Med 349, 366-381.
Norris, J.D., Chang, C.Y., Wittmann, B.M., Kunder, R.S., Cui, H., Fan, D., Joseph, J.D.,
and McDonnell, D.P. (2009). The homeodomain protein HOXB13 regulates the
cellular response to androgens. Molecular cell 36, 405-416.
Nunes, F.D., de Almeida, F.C., Tucci, R., and de Sousa, S.C. (2003). Homeobox
genes: a molecular link between development and cancer. Pesquisa
odontologica brasileira = Brazilian oral research 17, 94-98.
Paulo, P., Barros-Silva, J.D., Ribeiro, F.R., Ramalho-Carvalho, J., Jeronimo, C.,
Henrique, R., Lind, G.E., Skotheim, R.I., Lothe, R.A., and Teixeira, M.R.
(2012a). FLI1 is a novel ETS transcription factor involved in gene fusions in
prostate cancer. Genes, chromosomes & cancer 51, 240-249.
92
Paulo, P., Ribeiro, F.R., Santos, J., Mesquita, D., Almeida, M., Barros-Silva, J.D.,
Itkonen, H., Henrique, R., Jeronimo, C., Sveen, A., et al. (2012b). Molecular
subtyping of primary prostate cancer reveals specific and shared target genes
of different ETS rearrangements. Neoplasia 14, 600-611.
Porkka, K.P., and Visakorpi, T. (2004). Molecular mechanisms of prostate cancer.
European urology 45, 683-691.
Putzi, M.J., and De Marzo, A.M. (2000). Morphologic transitions between proliferative
inflammatory atrophy and high-grade prostatic intraepithelial neoplasia. Urology
56, 828-832.
Ramon, J.D., L. J. (Eds.) (2007). Prostate Cancer (Berlin: Springer).
Roehrborn, C.G. (2008). Pathology of benign prostatic hyperplasia. International
journal of impotence research 20 Suppl 3, S11-18.
Rubin, M.A., Maher, C.A., and Chinnaiyan, A.M. (2011). Common gene
rearrangements in prostate cancer. Journal of clinical oncology : official journal
of the American Society of Clinical Oncology 29, 3659-3668.
Saddick, S.Y. (2013). In vivo and in vitro studies on apoptosis in OSE cells and
inclusion cysts of pregnant heifers. Saudi Journal of Biological Sciences 20,
281-289.
Schroder, F.H., and Roobol, M.J. (2012). Prostate cancer epidemic in sight? European
urology 61, 1093-1095.
Schroeck, F.R., Zuhlke, K.A., Siddiqui, J., Siddiqui, R., Cooney, K.A., and Wei, J.T.
(2013). Testing for the recurrent HOXB13 G84E germline mutation in men with
clinical indications for prostate biopsy. The Journal of urology 189, 849-853.
Sciarra, A., Mariotti, G., Salciccia, S., Autran Gomez, A., Monti, S., Toscano, V., and Di
Silverio, F. (2008). Prostate growth and inflammation. The Journal of steroid
biochemistry and molecular biology 108, 254-260.
Selman, S.H. (2011). The McNeal prostate: a review. Urology 78, 1224-1228.
Shen, M.M., and Abate-Shen, C. (2010). Molecular genetics of prostate cancer: new
prospects for old challenges. Genes & development 24, 1967-2000.
93
Siegel, R., DeSantis, C., Virgo, K., Stein, K., Mariotto, A., Smith, T., Cooper, D.,
Gansler, T., Lerro, C., Fedewa, S., et al. (2012). Cancer treatment and
survivorship statistics, 2012. CA: a cancer journal for clinicians 62, 220-241.
Sramkoski, R.M., Pretlow, T.G., 2nd, Giaconia, J.M., Pretlow, T.P., Schwartz, S., Sy,
M.S., Marengo, S.R., Rhim, J.S., Zhang, D., and Jacobberger, J.W. (1999). A
new human prostate carcinoma cell line, 22Rv1. In vitro cellular &
developmental biology Animal 35, 403-409.
Sreenath, T., Orosz, A., Fujita, K., and Bieberich, C.J. (1999). Androgen-independent
expression of hoxb-13 in the mouse prostate. The Prostate 41, 203-207.
Stasiewicz, D., Staroslawska, E., Brzozowska, A., Mocarska, A., Losicki, M., Szumilo,
J., and Burdan, F. (2012). [Epidemiology and risk factors of the prostate
cancer]. Pol Merkur Lekarski 33, 163-167.
Stott-Miller, M., Karyadi, D.M., Smith, T., Kwon, E.M., Kolb, S., Stanford, J.L., and
Ostrander, E.A. (2013). HOXB13 mutations in a population-based, case-control
study of prostate cancer. The Prostate 73, 634-641.
Sylvester, P.W. (2011). Optimization of the tetrazolium dye (MTT) colorimetric assay
for cellular growth and viability. Methods Mol Biol 716, 157-168.
Tomlins, S.A., Bjartell, A., Chinnaiyan, A.M., Jenster, G., Nam, R.K., Rubin, M.A., and
Schalken, J.A. (2009). ETS gene fusions in prostate cancer: from discovery to
daily clinical practice. European urology 56, 275-286.
Tomlins, S.A., Laxman, B., Dhanasekaran, S.M., Helgeson, B.E., Cao, X., Morris, D.S.,
Menon, A., Jing, X., Cao, Q., Han, B., et al. (2007). Distinct classes of
chromosomal rearrangements create oncogenic ETS gene fusions in prostate
cancer. Nature 448, 595-599.
Tomlins, S.A., Rhodes, D.R., Perner, S., Dhanasekaran, S.M., Mehra, R., Sun, X.W.,
Varambally, S., Cao, X., Tchinda, J., Kuefer, R., et al. (2005). Recurrent fusion
of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science
310, 644-648.
Untergrasser A, C.I., Koressaar T, Ye J, Faircloth BC, Remm M, Rozen SG (2012 ).
Primer3 - new capabilities and interfaces. Nucleic acids research 40
94
Verma, S., and Rajesh, A. (2011). A clinically relevant approach to imaging prostate
cancer: review. AJR American journal of roentgenology 196, S1-10 Quiz S11-
14.
Wagenlehner, F.M., Elkahwaji, J.E., Algaba, F., Bjerklund-Johansen, T., Naber, K.G.,
Hartung, R., and Weidner, W. (2007). The role of inflammation and infection in
the pathogenesis of prostate carcinoma. BJU international 100, 733-737.
Werle, E., Schneider, C., Renner, M., Volker, M., and Fiehn, W. (1994). Convenient
single-step, one tube purification of PCR products for direct sequencing. Nucleic
acids research 22, 4354-4355.
Williams, T.M., Williams, M.E., and Innis, J.W. (2005). Range of HOX/TALE superclass
associations and protein domain requirements for HOXA13:MEIS interaction.
Developmental biology 277, 457-471.
Witte, J.S., Mefford, J., Plummer, S.J., Liu, J., Cheng, I., Klein, E.A., Rybicki, B.A., and
Casey, G. (2013). HOXB13 mutation and prostate cancer: studies of siblings
and aggressive disease. Cancer epidemiology, biomarkers & prevention : a
publication of the American Association for Cancer Research, cosponsored by
the American Society of Preventive Oncology 22, 675-680.
Wolf, A.M., Wender, R.C., Etzioni, R.B., Thompson, I.M., D'Amico, A.V., Volk, R.J.,
Brooks, D.D., Dash, C., Guessous, I., Andrews, K., et al. (2010). American
Cancer Society guideline for the early detection of prostate cancer: update
2010. CA: a cancer journal for clinicians 60, 70-98.
Xu, J., Lange, E.M., Lu, L., Zheng, S.L., Wang, Z., Thibodeau, S.N., Cannon-Albright,
L.A., Teerlink, C.C., Camp, N.J., Johnson, A.M., et al. (2013). HOXB13 is a
susceptibility gene for prostate cancer: results from the International Consortium
for Prostate Cancer Genetics (ICPCG). Human genetics 132, 5-14.
Zeltser, L., Desplan, C., and Heintz, N. (1996). Hoxb-13: a new Hox gene in a distant
region of the HOXB cluster maintains colinearity. Development 122, 2475-2484.
Zhai, L., Madden, J., Foo, W.C., Palmeri, M.L., Mouraviev, V., Polascik, T.J., and
Nightingale, K.R. (2010). Acoustic radiation force impulse imaging of human
prostates ex vivo. Ultrasound in medicine & biology 36, 576-588.
95
Zhao, Y., Yamashita, T., and Ishikawa, M. (2005). Regulation of tumor invasion by
HOXB13 gene overexpressed in human endometrial cancer. Oncology reports
13, 721-726.
