University of Southern California Dissertations and Theses

Identification of oncogenes cooperating in murine mammary tumorigenesis and transgenic mouse models of breast cancer

(USC Thesis Other)

Identification of oncogenes cooperating in murine mammary tumorigenesis and transgenic mouse models of breast cancer

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content IDENTIFICATION OF ONCOGENES COOPERATING
IN MURINE MAMMARY TUMORIGENESIS
AND TRANSGENIC MOUSE MODELS OF BREAST CANCER
by
Rocio Sagrario Lopez-Diego
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In partial fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR MICROBIOLOGY AND IMMUNOLOGY)
December 2002
Copyright 2002 Rocio Sagrario Lopez-Diego
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3093785
UMI
UMI Microform 3093785
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90089-1695
This dissertation, w ritten by
Rocio Lopez-Diego
under the direction o f h e r dissertation committee, and
approved by a ll its members, has been presented to and
accepted by the D ire cto r o f Graduate and Professional
Programs, in p a rtia l fu lfillm e n t o f the requirements f o r the
degree o f
DOCTOR OF PHILOSOPHY
Date
D ire cto r
Decem ber 1 8 , 2002
Dissertation Committee
C hair
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
“Cancer is so limited, It cannot corrode faith, It cannot shatter hope,
It cannot kill friendships, It cannot cripple love, It cannot destroy peace,
cannot silence courage, It cannot suppress memories,
It cannot conquer the spirit.”
(Anonymous)
This Dissertation is dedicated to the memory of
Sagrario Diego Nunez (Abuelita). You conquered it all.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
My deepest thanks to Dr. Gregory Shackleford, for his trust in my capability
as a scientist and his guidance during the past six years. To Deepa, for
patiently teaching me much of what I know about molecular biology today, for
her constant encouragement and support of my work, and for a friendship
that has enriched my life and that I will cherish forever. To Domingo
Ochavillo, for his unconditional support and sharing with me many hard
times, but also many happy moments. To Sam Wong, a very kind soul, and
always of help to me whenever I needed it. To all my friends, I am a lucky
person because of you; my life in Los Angeles would have never been the
same without the time we have spent together. Finally, to my very dear
family: my parents, aunt, and grandpa. Their believe in me have kept my
dream alive. Mil gracias por todo vuestro cariho, por tantas noches de
desvelos, y por la alegria de compartk escasos pero preciosos dias.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Page
DEDICATION ii
ACKNOWLEDGEMENTS iii
LIST OF FIGURES v
ABSTRACT viii
CHAPTER 1 Introduction 1
CHAPTER 2 Fgfr2 is insertionally activated by MMTV 31
and cooperates with Wnt-1 in murine
mammary tumorigenesis
CHAPTER 3 Generation of Fgfr2DN transgenic mice 53
CHAPTER 4 MMTV insertional mutagenesis in 70
Wnt/Fgfr2DN and Wnt/Fgf bitransgenic
mouse models for mammary oncogene
discovery
CHAPTER 5 A new rat model for mammary oncogene 106
discovery
EPILOGUE 122
REFERENCES 126
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure
number
1 New MMTV provirai insertions in mammary
tumors from MMTV-infected Wnt-1 transgenic
mice
2 MMTV provirai insertion into a common
integration locus
2 Map and orientation of provirai insertions into
the new common insertion locus
The Fgfr2(1 1 1 b)/Kgfr isoform is predominantly
4 expressed in the Wnt-1 mammary tumors
containing MMTV proviruses newly integrated
into the Fgfr2 locus
Relative expression of Fgfr2 in Wnt-1
2 mammary tumors as measured by real-time
RT-PCR
Sketch representing an alternative
2 mechanism for MMTV-mediated Fgfr2
insertional activation in mammary tumors from
MMTV infected Wnt-1 transgenic mice
7 Fgf/Fgfr signal transduction
2 Sketch of the MMTV-Fgfr2DN transgene
Southern blot analysis of Fgfr2DN transgenic
9 mice
MMTV-Fgfr2DN transgene in vitro transfection
into C57MG cells demonstrates its dominant
10 negative effect and the inhibition of FGF
mediated mitogenesis
Page
39
41
42
44
46
51
54
62
65
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES (continued)
Figure
number
11
12
13
14
15
16
17
18
19
20
Northern blot analysis of Fgfr2DN transgene
expression in lactating mammary gland total
RNA
Experimental design of female mouse cohorts
Southern blot screening of Wnt10b/Fgfr2DN
bitransgenic mice
Representative Southern blot screening of
Wnt1/Fgfr2DN and Wnt1/Fgf3 mice
Southern blot analysis of tumor origin in
Wnt10h/Fgfr2DN females infected with XC
cells (MMTV EH-swpF9) versus Mm5MT cells
(MMTV C3H)
New provirai integrations in mammary tumors
from MMTV-infected Wnt10b/Fgfr2DN mice
Histopathology of Wnt10b/Fgfr2DN mammary
tumors
New MMTV provirai insertions in mammary
tumors from MMTV-infected transgenic mice
Incidence of mammary tumors in Wnt1
/Fgfr2DN female mice and control cohorts
New provirai integrations in mammary tumors
from MMTV-infected Wnt1/Fgfr2DN females
Page
67
76
82
83
87
88
90
91
93
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES (continued)
Figure Page
number
21 Macroscopic necropsy findings in MMTV- 95
infected Wnt1/Fgfr2DN transgenic females
22 Mammary tumor histopathology in MMTV- 96
infected Wnt10b/Fgfr2DN and Wnt1/Fgfr2DN
females
Lung metastases (papillary mammary
23 carcinoma) from Wnt1/Fgf3 mouse mammary 97
tumors
24 Incidence of mammary tumors in Wnt1/Fgf3 99
female mice and control cohorts
25 Representative necropsy findings in MMTV- 190
infected Wnt1/Fgf3 females
26 Rat lymph node reactivity to MMTV 116
subcutaneous injection
27 Rats are susceptible to MMTV infection 117
Whole mount analysis of the effect of various
28 treatments on the stimulation of mammary 129
gland proliferation
The rat mammary gland is susceptible to MMTV
29 infection by intraductal injection 121
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
The main goal of this project is to create MMTV-infected animal models of
breast cancer in order to allow the identification of oncogenes that are
implicated in multistep mammary tumorigenesis. Previously, we and others
have established the key role that activation and cooperation of Writ and Fgf
genes plays in this process. This dissertation underscores this idea with the
finding of Fgfr2 as a potential oncogene that is insertionally activated by
MMTV and cooperates with Wnt-1 in mammary tumorigenesis. In order to
identify additional oncogenic events involved in this process, we have
developed three new transgenic mouse models of breast cancer
(Wnt10b/Fgfr2 dominant negative (DN), Wnt1/Fgfr2DN, and Wnt1/Fg3
bitransgenic mice) in which we are using MMTV insertional mutagenesis. In
the first two models, a Wnt oncogenic signal is constitutively active in the
mammary gland, while the activation of cooperative Fgf signaling is blocked
through the overexpression of a dominant negative (tyrosine kinase-
defective) Fgf receptor. In the third model, both Wnt and Fgf oncogenic
signals are activated in the mammary gland. IWe hypothesize that upon
MMTV infection the insertional activation of proto-oncogenes (other than
Wnts and Fgfs) may confer an additional proliferative/growth advantage to
the cells carrying such gene activations. Therefore, those mutated cells
should be naturally selected for in the transgenic mammary glands, and give
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rise eventually to clonal mammary tumors with an accelerated latency
compared to that in uninfected transgenic littermates.
Infected bitransgenic females in all three models developed mammary
adenocarcinomas. In addition, frequent lung metastases were observed in
Wnt1/Fgfr2DN, and Wnt1/Fg3 bitransgenics. Southern blot analysis of
mammary tumor DNAs confirmed the presence of newly integrated MMTV
proviruses in several of these tumors. We are currently using an inverse
polymerase chain reaction (IPCR) approach to analyze viral-cellular junction
fragments isolated from these tumors. In this way, we hope to clone
common gene targets for MMTV insertional activation.
In addition, we have developed a novel animal model based on the infection
of the rat with MMTV. Since the host genetic background is a strong
determinant of which cellular oncogenes are activated by MMTV insertions,
we expect to find a new repertoire of target genes in the rat, in addition to
those already characterized in mice. The rat is a species that has never
been experimentally infected with MMTV. Here we show that rats are
susceptible to MMTV infection and, more specifically, that rat mammary
tissue can be infected by MMTV following direct mammary intraductal
injection.
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1: INTRODUCTION
Breast Cancer impact on Public Health Today.
Breast cancer is the most common cause of cancer among women around
the world, accounting for one third of diagnosed cancers. In the United
States, about 1 in 8 women will suffer this disease during her lifetime, being
more frequent in women over 50 years of age. Breast cancer incidence has
steadily increased at a rate of increase of approximately 0.5% for the past 5
years. This increase is likely due to socio-economic improvement and its
impact on associated changes in lifestyle and reproductive patterns. In 2002,
approximately 192,000 U.S. women will be diagnosed with invasive cancer,
40,600 with in situ cancer, and 40,000 will die from this disease. Currently,
breast cancer is responsible for 18% of female cancer deaths, and ranks as
the second cause of death after lung cancer. (A.C.S. 2002, Greenlee 2001)
Since 1995, however, a steady decrease of 3% in breast cancer mortality
has been documented, possibly related to improvements in early screening
and treatment strategies (Howe 2001). Furthermore, although uncommon
(<1% of all breast carcinomas in the United States), male breast cancer will
affect approximately 1,500 men and approximately 400 will die of the disease
in 2002 (A.C.S. 2002). Numerous risk factors have been associated with the
development of breast cancer, including age, family history, genetic,
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
environmental, hormonal, and nutritional influences. Nevertheless, no risk
factors can be identified in 75% of breast cancer patients (Jardines 2001).
Signals and genes involved in normal mammary gland and
breast cancer development
Mammary gland development begins in the fetus, and a rudimentary
epithelial ductal tree, embedded in a stromal fat pad, is present at birth and
prepuberal stages. However, most of the changes leading to a fully
developed and functional gland take place throughout several postnatal
stages in female life. Steroid and peptide hormones produced during
puberty, pregnancy, lactation, puerpery, and involution, direct the changes
that characterize the development of the gland at each stage (Silberstein
2001, Hansen 2000). Through puberty, ovarian estrogens induce the
extensive branching and differentiation of epithelial ducts and endbuds that
progressively fill the entire fat pad, accompanied with limited alveolar
proliferation. In addition, stromal epidermal growth factor, the progesterone
receptor and Wnt (Wiesen 1999, Gallego 2001, Brisken 2000) signaling
contribute as well to the ductal development at this virgin mammary gland
stage. Many different signals appear to act in concert in the mammary gland
during pregnancy. High levels of progesterone, placental lactogens, the
pituitary hormones prolactin and growth hormone, the osteoclast
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
differentiation factor (RANK), cell-cycie signals (p27, cyclinDI), and fibroblast
growth factor receptor type 2-lllb (Fgfr2-Illb) signals regulate epithelial
alveolar expansion and transformation into the secretory units at this
developmental stage (Liu 1997, Miyoshi 2001, Shillingford 2001,
Hennighausen 2001, Spencer-Dene 2001). During lactation, prolactin and
oxytocin signals induce a characteristic increased density of secretory lobulo-
alveolar units and milk production. Finally, upon cessation of lactation,
lobuio-alveolar apoptosis and collapse ensue, and the gland reverts to a pre
pregnancy stage.
Approximately only 5% to 10% of breast cancer cases result from inherited
mutations in breast cancer susceptibility genes, such as BRCA1 and BRCA2
(Burke 1997), or p53 in the Li-Fraumeni sydrome (Frebourg 2001). These
mutations occur in far less than 1% of the general population (Whittemore
1997). In the remaining 90-95% of breast cancer patients, sporadic genetic
alterations in somatic mammary cells— ranging from single point mutations to
complex chromosome aberrations— underlie breast tumor formation. Many
of such mutagenic events are common to cancer development and
progression in general. It is well established now that cancer is a multistep
process in which the effects of mutations in three main types of genes
— proto-oncogenes, tumor suppressor genes, and DNA repair
genes—cooperate in leading to uncontrolled cell growth, proliferation, and
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tumor formation. In most cases, the non-mutated forms of these genes play
essential roles in the regulation of normal cell growth and tissue/organ
development. Upon mutation, however, signaling mediated by the altered
genes may lead to aberrant cell proliferation and development of malignant
tumors. The mammary gland is no exception to this fact, and various
signaling pathways involved in normal mammary development also play a
key role in mammary tumorigenesis.
The role of genetic factors in breast cancer etiology: mouse
models of breast cancer
The study of cancer genetics in humans presents certain limitations.
Specifically, in the most prevalent forms of cancer, an individual’s tumor
susceptibility is determined not only by key gene players but also by several
weaker tumor modifiers, which are notoriously difficult to identify by
performing population-based genetic studies. The diversity in genetic
background, the difficulty in tumor and tissue collection, and the long latency
of human breast tumors, pose additional obstacles in human cancer
research. By contrast, the mouse offers several distinct advantages and has
proven to be a valuable model system for the controlled study of multistep
tumorigenesis at distinct levels: (a) identification of novel genes and tumor
determinants involved in cancer pathogenesis; (b) establishment of a link
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
between individual gene mutations and specific cellular changes that lead to
tumor development; (c) experimental in vivo assessment of the role of
specific genetic events and their intricate interactions in multistep
tumorigenesis; and (d) identification of previously unsuspected roles of these
genes in normal development and differentiation.
For the past two decades, advances in molecular biology concepts and tools
have made possible the creation of genetically engineered mouse models of
breast cancer. It is now possible to target the transgene effects in a
mammary cell- and time-specific fashion, hence providing a useful approach
to study the genetic and events involved in the molecular basis of tumor
progression at different stages.
No single transgenic model can recapitulate the complete human breast
cancer spectrum, and many transgenic models display tumor characteristics
dissimilar to the human counterparts. Nevertheless, the use of genetically
engineered mouse models of breast cancer can be very instructive and is
justified for the following reasons: first, the mouse and human genomes are
highly similar (>98%); second, genetic mutations involved in human breast
cancer can also induce this disease in transgenic mice; and third, several
breast cancer animal models faithfully resemble the human cases in their
histopathological appearance, kinetics, hormonal dependence, and
metastatic potential, and genetic pathways.
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Full malignant cell transformation in vitro requires the collaboration of two
oncogenes (Rassoulzadegan 1982, Land 1983). In general, in vivo
transgenic expression of an activated oncogene of the inactivation of a tumor
suppressor gene induces the stochastic formation of tumors that are clonal in
origin and that appear only after a variable latency period. These facts
suggest that other mutagenic events, in addition to that of the transgene
expression, are required for tumor onset. Cancer, hence, arises as a result
of the acquisition and expression of multiple genetic lesions over time, in a
step-wise fashion. One of the most effective approaches for the study of
multistep cooperative tumorigenesis in the mouse is the generation of
bitransgenic mice or knockout mice in a transgenic background, to determine
whether these additional mutations can synergistically accelerate tumor
development. This model also allows the study of the parallelism and
connection between signaling pathways. This capability of genetically
engineered mouse models of cancer is particularly valuable, since it allows a
more accurate replication of human breast cancer cases, in which a complex
genetic multistep process appears to be involved.
Both normal mammary gland and tumor development involve the exquisite
orchestration of signaling pathways that regulate cell growth, proliferation,
differentiation, death, and tissue remodeling. Different kinds of oncogenes--
from growth factors and their receptors, to cell cycle regulators, or genes
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
involved in differentiation— and tumor suppressor genes have been
overexpressed or inactivated respectively in various combinations in
transgenic mice. One might therefore expect that transgene-mediated
disruption at any of these levels result in mammary tumor formation. What
important lessons have we learned from these transgenic models thus far?
Transgenic mouse models of gain of function mutations. Transgenes
under the mouse mammary tumor virus (MMTV)-LTR or the whey acidic
protein (WAP) promoter can be used to overexpress various genes
specifically in the mammary epithelium. The MMTV LTR acts as a
promoter/enhancer that is active in the virgin mammary gland, and highly
expressed during pregnancy and lactation, due to the presence of steroid-
responsive elements within the LTR. This hormonal responsiveness is also
responsible for MMTV-LTR expression in other tissues besides the
mammary gland (Pattengale 1989). The WAP, on the other hand, is a milk-
specific promoter with high expression limited to the differentiated mammary
epithelium during late pregnancy and lactation (Andres 1987). Different
oncogenes have been overexpressed in the mammary gland of transgenic
mice under the control of either promoter. The following headings of this
Introduction present the current knowledge derived from these mammary
oncogenesis models.
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C-mvc, Ras, and Tgfa
C-myc is a protooncogene that encodes a nuclear transcription factor
involved in the control of cell proliferation, apoptosis, and DNA synthesis. C-
myc participates in cellular transformation, when apoptosis is suppressed,
through transcriptional up- or downregulation of target genes (Dang 1999).
Altered c-myc expression is a shared feature of many human cancers, and
appears to be particularly important in breast tumorigenesis. Its amplification
can be detected in approximately 16% of breast cancer patients and
correlates with poor prognosis. Furthermore, its rearrangement or
overexpression in breast tumors can de detected in up to 70% of the cases
(Deming 2000). Although MMTV/ or WAP/c-myc transgene overexpression
in the mammary gland predisposes transgenic mice to the development of
mammary adenocarcinomas, these are pregnancy-dependent, random,
solitary tumors of clonal origin, which appear adjacent to normal transgenic
mammary epithelium (Stewart 1984, Leder 1986).
Ras genes encode for GTPases that play a central role in a large variety of
cellular processes. These GTPases act as molecular switches, alternating
between inactive GDP-bound and active GTP-bound states. GTP-bound Ras
recruits and activates a number of effector proteins [Campbell 1998; Katz
and McCormick 1997). The activation of Ras effector signaling pathways
(including those mediated by the epidermal growth factor receptor (EGFR)]
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
has been reported in many breast tumors and cell lines, suggesting that Ras
is activated in primary tumors but, to date, the relevance of Ras activation in
human breast cancer development remains unclear. Similarly to c-myc,
overexpression of MMTV/v-Ha-ras, N-ras or WAPIv-Ha-ras transgenes in the
mammary gland induces breast tumor formation in a stochastic and clonal
fashion, suggesting once more that the Ras oncogene contributes to but is
not the sole cause of breast tumorigenesis in this system (Muller 1991).
Transforming growth factor a (TGFoc) is a secreted glycoprotein growth factor
highly homologous to the epidermal growth factor (EGF). Both factors signal
through the membrane-bound EGF receptor (EGFR/c-ErbB-1) and lead to
the activation of diverse intracellular tyrosine kinase-mediated cascades
(Jamerson 2000). TGfa is expressed at different stages in the normal adult
mammary gland. Deregulation of TGFa expression is involved in partial
transformation of mammary epithelial cells in vitro and in malignant breast
tumorigenesis in vivo. These effects depend upon the establishment of a
positive feedback loop with Egfr, which is overexpressed in approximately
1/2 of breast cancers. Three different Tgfa transgenic mouse models have
been generated to date (Sandgren 1990, Matsui 1990, Sandgren 1995). All
three models display notable hyperplasia and the development of mammary
adenocarcinomas in the multiparous female mammary gland after an
extended latency, once more indicating a requirement for additional
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mutagenic step in leading to mammary tumor formation (Humphreys and
Hennighausen 2000).
Cell cycle genes: cvclin D1
Mammalian cell growth and proliferation is dependent upon tightly controlled
cell cycle progression. Cyclins are critical cell cycle regulators that enable
associated cyclin-dependent kinases (CDKs) to drive cell cycle progression
through the activation/inactivation of key cellular proteins. Cyclin D1 is a G1
phase-specific cyclin whose expression is mostly regulated by extracellular
mitogens. Cyclin D1/CDK complexes promote cell growth by
phosphorylating the retinoblastoma protein (pRB), which is rendered inactive
and releases sequestered transcription factors that promote S-phase entry
(Weinberg 1995). Consistent with its cyclic growth promoting function, Cyclin
D1 plays a key role in normal mammary gland development. Cyclin D1, but
not other cyclins, is specifically induced during pregnancy in the proliferating
mammary epithelium. Furthermore, cyclin D'/-deficient mice display marked
lobulo-alveolar hypoplasia and functional impairment during pregnancy
(Sicinski 1997, Fantl 1999).
Cyclin D1 has an oncogenic effect in a variety of tissues and its upregulation
has been reported for several human malignancies (Sicinski 1997).
Interestingly, its overexpression can be detected in up to 50% of breast
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
carcinomas but not in benign lesions (Barnes 1994, Bartkova 1994,
Weinstat-Saslow 1995). The oncogenic role of cyclin D1 has been further
confirmed in MMTMicydin D1 transgenic mice. As with the c-myc, Ras, and
Tgfa models, transgenic females develop mammary hyperplasia early in life
and focal mammary tumors after a long latency (Wang 1994), reflecting the
need for additional cooperating mutagenic steps. Following along this line,
cooperation between cyclin D1 and other factors such as Src kinases, or
ErbB-2/Neu has been reported during the past years (Lee 1999, Radeva
1997).
Growth factors and their receptors
EGFRs: Activation of the EGFR signaling pathway plays an early and crucial
role in mammary tumorigenesis. This event can be mediated not only
through overexpression of EGFR ligands such as Tgfa, as previously
mentioned, but also through activation of the EGF receptor itself. Four highly
related receptor tyrosine kinases belong to the ECFR family: EGFR/Erb-
B1/HER1, ErbB-2/HER2/Neu), ErbB-3/HER3 and ErbB-4/HER4). EGFR
ligand binding induces formation of active receptor dimers upon
autophosphorylation events within the cytoplasmic domain. This domain, in
turn, recruits and phosphorylates intracellular targets and initiate different
signaling cascades. EGFR receptors-mediated signaling controls critical
developmental as well as normal adult physiological processes (Olayioye
2000).
1 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In the mammary gland, the expression of all EGFR receptors is cell type- and
developmental stage-specific. In general, ErbB1 participates mainly in ductal
growth promotion, while ErbB-2 and ErbB-4 are critical regulators of
lobuloalveolar differentiation and lactation (Olayioye 2000)
ErbB-2/HER2/Neu overexpression is a common feature of 40-50% of human
breast and ovarian tumors. This finding correlates also with poor clinical
prognosis and predicts resistance to hormonal anti-cancer therapy (Slamon
1989, Pegram 1998). The mammary oncogenic potential of ErbB-2 has been
corroborated in MMTV/A/eu transgenic mice, by overexpressing wild-type
ErbB-2 cDNA. Random mammary tumors appear with a high incidence in
these transgenic females and are pregnancy-independent. Moreover, the
tumors only appear in transgene positive regions of the mammary gland,
while the adjacent mammary tissue does not show transgene expression
(Amundadottir 1996, Hutchinson 2000). Strikingly, one study has described
mammary tumors in both females and males (Muller 1988). It thus appears
that ErbB-2 is a potent mammary oncogene, capable of inducing malignant
mammary transformation in single step, unlike other previously discussed
transgenic mouse models. A number of reports suggest that ErbB-2
overexpression results in spontaneous active ErbB-2/ErbB-1 and/or ErbB-
2/ErbB-3 dimer formation. The observation that endogenous ErbB-1 and -2
are frequently overexpressed in ErbB-2 positive breast tumor samples further
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
implicate ErbB receptor dimerization as a crucial event in breast cancer
development.
Fibroblast growth factors (FGF) and FGF receptors (FGFRs): Four Fgf genes
(Fgf3, 4, 7, and 8) were initially identified as common targets for MMTV
insertional activation in MMTV-infected Wnt1 transgenic mice. This aspect
together with their cooperation with Wnt genes in multistep murine mammary
tumorigenesis will be presented in later headings. The mammary oncogenic
potential of Fgf3, 7, and 8 has been confirmed in MMTV/Fgf transgenic
mouse models, in which female mice develop pregnancy-dependent
mammary hyperplasia that eventually progresses to stochastic focal
mammary tumor formation. Although the importance of Fgf-mediated
oncogenesis remains unclear in human breast cancer, overexpression of
several members of the FGF and FGFR families have been detected in some
human breast tumor series (Dickson 2000, Zammit 2001).
Transgenic models of loss of function: tumor suppressor genes.
Certain cell-cycle control and DNA repair proteins, act as tumor supressor
genes that counter the action of oncogenes. Inactivation of these genes on
both chromosomes makes cells much more susceptible to oncogene action,
leads to increased mutability, and are very often inactivated by mutation in
human tumors
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
d53. BRCA1. and PTEN
The p53 tumor suppressor gene encodes for a transcription factor which
regulates a number of celi-cycle control (e.g., p21W A F 1 cyclin-dependent
kinase inhibitor, cyclin G1 and 67, c-kit), DNA repair, pro-apoptotic (hax), and
differentiation-related genes (cytokeratin 19, kappa casein, a smooth muscle
actin) (Cui 2000). The presence of p53 in the cell is a protective mechanism
against various cellular insults, such as DNA breakage or aberrant oncogene
activation (Giaccia 1998). Under these stressful conditions, p53 expression
is activated and leads to either G1 cell-cycle arrest, or to programmed cell
death,
The most common (>50%) genetic lesion in all human cancers is the loss of
heterozygocity (LOH) or mutation of p53, and it its possible that non-
structural p53 loss of function may be involved in an additional number of
cases (Lozano and Elledge 2000, Freedman 1999). In general, loss of p53
activity relates to more agressive tumor phenotypes as well as worse
prognosis and treatment response (Donehower 1996, Wallace-Brodeur
1999). About 40-60% of all human breast cancers display p53 deletions or
mutations, and Li-Fraumeni syndrome patients (who carry an inactive p53
copy) are highly susceptible to many kinds of tumors, especially breast
cancer (Cressman 1999, Malkin 1993).
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Studies in p53+ /" and p53'A knockout (KO) mice are of limited use since,
despite developing an array of other aggressive tumors that frequently lead
to premature death, these mice rarely (<2%) develop mammary tumors
(Donehower 1992, Harvey 1993). This problem was partially circumvented
by the creation of WAP/mutant p53 transgenic mice. While these mice
developed very few tumors, carcinogen-mediated mammary tumorigenesis
was clearly accelerated (Li 1998). A new conditional human p53 knock-in
mouse model has been generated very recently, and it may prove to be
instructive in determining the direct role of p53 in mammary tumorigenesis
(Luo 2001).
BRCA1 is a tumor supressor gene that encodes a transcription factor
involved in the activation of DNA damage repair, G2-M checkpoint control,
centrosome duplication-related genes (Deng 2000). BRCA1 and BRCA2
germline mutations were the first two genetic lesions directly associated with
the genesis of heritable breast cancer. BRCA1 mutations have been
detected in 90% of familial breast/ovarian cancer and in almost 50% of all
patients with familial breast cancer alone (Paterson 1998).
As for p53, null and heterozygous BRCA1 KO mice present problems: while
null KOs are embryonic lethals, the BRCA1+ /~ mice do not develop mammary
tumors (Liu 1996). Conditional mutation of Brcal in the mouse mammary
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
epithelium results however in late onset breast tumor formation. These
tumors displayed a high level of genomic instability, chromosomal alterations
and, interestingly, p53 loss of function mutations, supporting a role for
BRCA1 as a caretaking gene crucial for the maintenance of genomic integrity
(Deng 2000).
PTEN (phosphatase and tensin homolog deleted from chromosome 10)
encodes for a lipid phosphatase that upregulates the activities of the
phosphatidylinositol 3-kinase (PI3K)/AKT pathway. It is currently thought that
PTEN is a tumor suppressor that promotes apoptosis and inhibits cell
spreading, migration, and invasion. Loss of PTEN therefore would promote
cell survival and metastatic potential. (Li 2001)
PTEN loss or mutation is a common finding in primary human brain
glioblastoma, prostate, endometriun, and breast tumors among others, and
other tumor types carry inactivating PTEN somatic mutations as well
(Petrocelli 2001). In general, in approximately 33% of breast cancers the
PTEN protein is abnormally decreased or absent. In addition, two multi
neoplastic, autosomally dominant syndromes (Cowden disease and
Bannayan-Zonana syndrom) are associated with germline PTEN mutation,
and about 20-50% of affected females in these families develops breast
cancer (Eng 1998, Bose 1998). It thus appears that PTEN may offer a
protective role against breast tumorigenesis.
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Three independent mouse lines carrying inactivating Pten mutations have
been made (Di Cristofano 1998, Suzuki 1998, Podsypanina 1999). While
Pten homozygous mutants are embryonic lethals, heterozygous mice
develop various kinds of tumors at approximately 6 months of age, but very
few mammary tumors are actually observed in these animals.
Transgenic mouse models of multistep mammary tumorigenesis.
Bitransaenic mouse models
The generation of c-myd-v-Ha-ras and c-myc/tgfa bitransgenic mice during
the past years has demonstrated that these oncogenes act synergistically in
mammary tumorigenesis. Generalized microscopic hyperplastic and
neoplastic changes can be observed in the mammary gland of these
animals, even at a young age (5 weeks), and they soon develop mammary
tumors in an accelerated and pregnancy-independent fashion. It is currently
thought that, in these models, upregulated EGFR and/or Ras signaling may
abrogate c-myc-driven apoptosis and cooperate with c-myc in promoting cell
cycle progression and proliferation (Amundadottir 1996, Rose-Hellekant
2000). As a whole, these transgenic breast cancer models reflect the
multistep nature of human breast cancer, and have provided strong lines of
evidence in necessary role played by oncogenic cooperation in mammary
tumorigenesis and progression (Muller 1991, Dankort 2000, Hutchinson
2000, Jamerson 2000).
1 7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BRCA1/P53 transgenic mouse models.
The late onset and sporadic nature of mammary tumors observed in
conditional BRCA KO mice suggests that other collaborating events may
participate in tumorigenesis. In order to assess the role played by other
genes in this process, transgenic mice carrying a conditional BRCA1
mutation in a wt p53 or p53+/~ background have been created (Brodie 2001).
Animals with a p53+/- genotype display a striking shortening in mammary
tumor latency, although the tumors still appear in random fashion and display
a variable histologic appearance. While this observation implies a major role
for p53 in Brcal-mediated tumorigenesis, it nevertheless strongly suggests
that additional genetic steps are involved in this process. It is thus not
surprising that other oncogenic mutations frequently associated with sporadic
breast cancer, such as c-myc, cyclin D1 or ErbB-2 overexpression, are also
found in the mammary tumors from this KO model, suggesting a complex
molecular basis for Brea'/-mediated tumorigenesis.
As mentioned before, despite clinical evidence pointing in the direction of a
direct association between inactivation of PTEN--another tumor suppressor
gene--and mammary tumorigenesis, this relationship remained unclear. In
order to better assess the role of PTEN in mammary tumorigenesis, a Pten
+/- in a Wnt1 background mouse model has been created. Interestingly, in
this system, the absence of Pten cooperates with Wnt1 in the development of
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mammary tumors with an accelerated latency, strongly implicating loss of
Pten as an important event in multistep mammary tumorigenesis (Li 2001).
MMTV-induced mammary tumorigenesis and oncogenic cooperation.
The high susceptibility to mammary tumorigenesis of certain inbred mouse
strains is caused by the presence of the milk-transmitted mouse mammary
tumor virus (MMTV) (Tekmal 1997, Hilgers and Bentvelzen 1979, Teich
1982). MMTV is a slowly transforming retrovirus that causes tumorigenesis
via an insertional mutagenesis mechanism (Varmus 1982, Callahan 1996)
(Fig. 1). Proviral DNA integration into the host genome activates the
transcription of or, in other cases, directly mutates adjacent proto-oncogenes
(often named as int genes). This event contributes to the transformation and
clonal outgrowth of the mutated mammary epithelial cells eventually leading
to mammary tumor formation.
MMTV insertional mutagenesis studies in mice present certain advantages.
First, since MMTV-induced mouse mammary tumorigenesis progresses
through three well defined stages: preneoplastic hyperplastic nodules,
malignant tumors, and metastases (mainly to the lung), this model can be
very instructive on the multistep nature of breast tumor formation (Callahan
1996). Second, because MMTV is an insertional mutagen that can be found
physically linked to the activated genes, it can thus be used as a molecular
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tag to identify such gene targets. Finally, this model allows for the
identification of novel or unexpected potential mammary proto-oncogenes, in
contrast to the more limited possibilities offered by the traditional single or
double transgenic mouse models.
Several int loci have been cloned and identified by studying MMTV-induced
mouse mammary tumorigenesis (Tekmal 1997, Nusse and Varmus 1982,
Peters 1989, Roelink 1990, Shackleford 1993, MacArthur and Shankar 1995,
Lee, 1995, van Leeuwen 1995). Most of these proto-oncogenes belong to
either the Writ or the fibroblast growth factor (Fgf) gene family (van Leeuwen
1995). Although structurally different, both Writs and Fgfs play crucial roles,
and often collaborate, in the regulation of embryonic patterning, cell
proliferation, and differentiation in widely divergent species (Dierick 1999,
Nusse and Varmus 1992, Szebenyi 1999, McKeehan 1998).
Wnt and Fgf genes are clearly implicated in murine mammary tumorigenesis.
MMTV-induced transcriptional activation of Writ genes, more frequently
Wnt1, is a key early step in multistep mammary tumorigenesis in mice. Wnt1,
the first protooncogene cloned in mouse mammary tumors, is insertionally
activated in 75% of mammary tumors of MMTV-infected C3H mice (Nusse
and Varmus 1982). Two other Writ genes, Wnt3 and W ntW b, are less
frequently activated by MMTV (Roelink 1990, Lee 1995). In addition to Writ
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
genes, three members of the Fgf family (Fgf3/int2, Fgf4/hst, and Fgf8/AIGF),
the transmembrane receptor Notch-4/int3, and the translation initiation factor
elF3/int6, are targets for MMTV insertional activation (Dickson 1984, Peters
1989, Gallahan and Callahan 1997, MacArthur and Shankar 1995, Lee 1995,
Asano 1997). The generation of Wnt1, Fgf3, and WntlOb transgenic mice
has confirmed the oncogenic potential of these genes. In all cases, the
females develop mammary hyperplasia that eventually gives rise to random
mammary tumors in an accelerated fashion. The stochastic character of
these tumors implies that single proto-oncogene activation contributes to, but
is not sufficient for tumorigenesis (Nusse and Varmus 1982, Tsukamoto
1988, Jonkers 1996).
The use of retroviral insertional mutagenesis in Wnt or Fgf transgenic mice
has allowed the identification of several cooperating mammary oncogenes,
and has confirmed the idea that progressive accumulation of synergistic
gene mutations, whose effects cooperate to favor unrestrained cell
proliferation, underlie tumorigenesis. Analysis of the MMTV integration loci in
mammary tumors from MMTV-infected Wnt1 or Fgf3-transgenics has
revealed preferential activation and expression of Fgf genes (Fgf3, Fgf4,
Fgf8) and the Fgfr2 receptor in Wnt1 transgenics, or activation of Wnt genes
(Wnt1, Wnt3, WntlOb) in Fgf3 transgenics (Shackleford 1993, Lee 1995).
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
This fact, together with the decreased tumor latencies observed in Wnt1/Fgf3
bitransgenics, demonstrate that activation of Wnt and Fgf genes, and strong
oncogenic cooperation between both growth factor families, are crucial
events involved the molecular basis of multistep mammary tumorigenesis.
Oncogenic cooperation between Writs and Tgfa has been demonstrated in
MMTV-infected \NAPI Tgfa transgenics. These mice display accelerated
mammary tumor formation, due to preferential activation of Wnt1 or Wnt3 by
MMTV, thus implicating parallel upregulation of the EGFR and Wnt signaling
pathways as an additional mechanism involved in mammary tumorigenesis
(Schroeder 2000).
The analysis of mammary tumors from MMTV-infected MMTV/neo transgenic
mice has revealed that another member (besides Notch4) of the Notch
family, N otchl is insertionally activated (through N-terminal truncation) and
cooperates with the neu oncogene in the acceleration of mammary
tumorigenesis in these animals. A similarly truncated N o tc h l cDNA
transgene has been shown to elicit malignant cell transformation in vitro, thus
confirming the oncogenic potential of this int locus (Dievart, 1999).
MMTV insertional mutagenesis in mice has revealed the critical role that
early Wnt/Fgf gene activation and cooperation plays in multistep murine
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mammary tumorigenesis. This fact is the key premise of the experimental
mouse models that will be discussed in the body of this Thesis. For this
reason, these two gene families will be presented in more detail below.
Writs and Fgfs
Fgf genes.
The fibroblast growth factor [F gf (murine), FGF (human)] gene family
comprises at least 22 members closely related among species. The
presence of a highly homologous 34 amino-acid core, and a strong affinity for
heparan sulphate proteoglycan (HSPG) characterize this family of
polypeptide factors (Ago 1991, Zhang 1991, Ornitz 2000). Fgfs are
pleiotropic factors, they not only regulate cell proliferation, but also participate
in cell migration and differentiation (Basilico 1992). During vertebrate
development, Fgfs are differentially expressed in a temporo-spatial fashion,
and are frequently involved in signaling across epithelial-mesoderm
boundaries (Hogan 1999). Fgfs are important in embryonic ventral-posterior
fate specification, and face, limb, neural, cartilage, and mammary
development among other processes (Munoz-Sanjuan 2001, Heikinheimo
1994, Martin 2001, Capdevila 2001, Smallwood 1996, Ornitz 2001, Spencer-
Dene 2001). They also play a key role in normal physiological processes
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
such as angiogenesis, tissue repair and injury response, and have recently
been implicated in the inhibition of proliferation and induction of apoptosis in
certain cell types (Cross 2001, Duplan 2002). Furthermore, several Fgfs
were initially cloned as activated oncogenes in a variety of cancers (Dickson
1984, Sakamoto 1986, Delli-Bovi 1987, Zhan 1987, MacArthur 1995). It is
now well established that gene mutations resulting in the deregulation of Fgf
signaling collaborate with other genetic events in the genesis of various
human malignancies.
Most FGFs act extracellularly. They elicit their wide biological effects upon
binding to one of four high affinity fibroblast growth factor receptors. In
general, FGFs are secreted to into the extracellular matrix, were they are
stabilized through their binding to heparin or HSPG. The formation of this
dimeric complex is essential to Fgf biology. It is currently thought that
FGF/heparin or HSPG complexes represent the functionally active FGF form
capable effective binding and activation of the cell’s FGFRs. FGFRs are type
I cell surface proteins that belong to the immunoglobulin (Ig) receptor
superfamily. To date, four receptors, FGFR1-4, sharing a 55-72% identity,
have been identified (Johnson-Williams 1993). The Fgfr comprises two or
three extracellular immunoglobulin (Ig)-like domains, an acidin region
between the first and second Ig-like domains, a transmembrane domain, and
an intracellular catalytic domain with tyrosine kinase (TK) activity. (Figure
signaling). Alternative splicing of sequences corresponding to the third Ig-
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
like domain can generate differentially expressed receptor isoforms lllb (in
cells of epithelial origin) or 3 1 1 c (in cells of mesenchymal lineage) (Yan 1993,
Orr-Urtreger 1993). Each receptor isoform displays different tissue-specificiy
and ligand-binding affinity. In this way, FGF/FGFR signaling may elicit
distinct and biological effects in a tissue- and temporo-specific fashion (Miki
1992). The formation of an HSPG/FGF/FGFR complex induces a
conformational change in this trimer that facilitates the formation of FGFR
homo- or heterodimers on the cell surface. The FGFR dimer is the active
receptor form, in which transphosphorylation events between each
monomer’s TK domains take place. The active phosphorylated TK domains
transduce in turn the FGF signal through recruitment and phosphorylation of
intracellular target proteins. Ras/mitogen-activated protein kinase (MAPK),
PLCy, Src, Crk, and SNT/FRS2 are some of the main FGFR signaling targets
that mediate the pleiotropic effects of FGFs (Powers, 2000).
Deregulation at any step of the FGF/FGFR-mediated signaling pathways
may result in uncontrolled cell proliferation and/or insensitivity to apoptotic
stimuli. Natural selection of mutated cells displaying these phenotypes may
thus lead to tumor formation. Aberrant FGF/FGFR signaling is implicated in
a variety of cancers in animal models as well as in humans.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Three Fgf family members, Fgf3, 4, and 8 , and the Fgf receptor 2(1 M e)
[Fgfr2(lllc)J, are mammary oncogenes that, upon insertional activation by the
mouse mammary tumor virus (MMTV), collaborate with Wnt1 in the
development of mouse mammary tumors (Shackleford 1993, MacArthur
1995, Kapoun 1997, Lopez-Diego (unpublished data)). These genes,
together with FGF1, FGF2, FGF7, and FGFR1, 2, and 4, have also been
found to be overexpressed in some human breast tumors. Altered
intracellular location of FGFR3 is reported in certain human mammary
tumors as well (Theillet 1989, Penault-Lorca 1995, Zammit 2001, Dickson
2000). These facts, together with the essential role played in normal
mammary development, suggest a likely involvement of FGF/FGFR signaling
in human breast cancer genesis and progression. However, the specific role
that inappropriate activation of this pathway may play in the process remains
unclear.
W rits.
The Wnt gene family consists of at least 19 different genes. They encode
highly homologous, and functionally redundant, cysteine-rich secreted
glycoproteins that act as signaling factors involved in a wide variety of
essential processes in different tissues (Cadigan and Nusse 1997). Wnts
regulate embryonic cell fate determination, as well as cell growth,
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
proliferation, death and differentiation in widely divergent species, from
Drosophila to human.
Wnts act through the Frizzled/LRP cell surface-receptor complex (Bhanot
1996, Tamai 2000). Upon Wnt ligand binding, the receptor signals via
Dishevelled and /or Axin, which inactivate the APC/Axin/GSK3p complex. In
the absence of Wnt signal, the complex targets cytoplasmic (3-catemn for
ubiquitination and proteosomal degradation. The Wnt signal inactivates this
complex, resulting in (3-catenin stabilization and its translocation to the
nucleus. Immunohistochemical detection of nuclear p-catenin is considered
a characteristic feature of Wnt pathway activation. Once inside the nucleus,
p-catenin couples with various transcription factors, such as TCF/LEF1, to
direct the expression of several target genes, which vary in each cell type,
but include oncogenes such as c-myc, cyclin D1, and cyclooxygenase-2
(Polakis 2000, Li 2000) (Figure signaling)
Wnt1 was initially identified as the first oncogene responsible for MMTV-
induced mammary tumorigenesis in mice. It is not normally expressed in
adult mammary tissue, nor has its involvement in human breast cancer been
proven. Wnt3 and WntlOb are two mammary proto-oncogenes whose
insertional activation by MMTV induces breast tumor formation in mice as
well (Roelink 1990, Lee 1995, Lane 1997). Unlike Wnt1, these and other
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Wnt genes are expressed in the mammary gland, and are deregulated in
human mammary tumors. The fact that overexpression of Wnt1, Wnt3, and
WntlOb in the mammary gland induces hyperplasia and mammary tumors in
mice, and that several Wnts are differentially expressed during postnatal
mammary development, strongly suggest an important role for Wnt genes
both in mammary tumorigenesis and normal gland morphogenesis (Weber-
Hall 1994, Edwards 1998, Smalley 2001).
Several components and targets of the Wnt signaling pathway, in addition to
Wnt ligands, have been identified as oncogenes or tumor suppressor genes.
Their alteration, leading to activation of this signaling pathway has been
reported in a large group of human cancers including colorectal, melanoma,
hepatocellular carcinoma, hepatoblastoma, medulloblastoma, gastric,
endometrial, prostatic, thyroid, Wilms’, and pilomatricoma tumors (Polakis
2000). Recently, studies in a transgenic mouse model overexpressing active
P-catenin have recapitulated the phenotype observed in Wnt1 mammary
tumors, suggesting that the oncogenic activation of the Wnt pathway induces
indeed murine mammary tumorigenesis via P-catenin and its common gene
targets (Michaelson 2001). Furthermore, up to 60% of human breast
carcinomas show Wnt/p-catenin activation by immunohistochemistry, and
high P-catenin levels relate to poor prognosis in human breast
adenocarcinoma (Lin 2000). Therefore, although the direct role or functional
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
relevance of Wnt overexpression remains to be proven in human tumors, the
facts described here make a strong case in favor of the direct link between
Wnt signaling activation and breast cancer development.
Research Goals
The results presented so far clearly reflect that genetic alterations underlie
cancer development and progression at a fundamental molecular level. At
the same time, we now recognize the multistep nature of cancer and that, as
a general rule, a single gene mutation is not sufficient to induce the formation
of malignant tumors in vivo. Rather, it requires synergism/cooperation
among the complex effects of multiple gene mutations, effects that lead to
uncontrolled cell growth, unresponsiveness to apoptosis, and defective DNA
repair mechanisms.
In murine mammary tumorigenesis, some members of the Wnt and Fgf gene
families play an oncogenic role and cooperation between Wnt and Fgf
signaling is a key early event. Since tumorigenesis is the result of a complex
multistep molecular process, it is only logical then to attempt the creation of
new animal models that allow the identification of additional oncogenic
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
events involved in this process. This has thus been the main focus of my
doctoral research. The specific goals of my research are:
1- To confirm Fgfr2 as an additional target for MMTV insertional activation in
Wnt1 transgenic mice. The completion of this goal will be discussed in
Chapter 2.
2- To create transgenic mouse lines expressing a dominant negative Fgfr2
(Fgfr2DN) in the mammary gland. This part of the project will be
presented in Chapter 3.
3- To conduct MMTV insertional mutagenesis studies in Wnt/Fgfr2DN and
Wnt/Fgf bitransgenic mouse in order to assess tumor formation kinetics
and the role that oncogenes other than Writs and Fgfs may play in
multistep mammary tumorigenesis. The generation of these models and
the information gathered will be discussed in Chapter 4
4- To create a new rat model of breast cancer with the intent of allowing the
identification of other groups genes, different from those identified in
mouse models, involved in mammary tumorigenesis. This model will be
presented in Chapter 5.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2: Fafr2 IS INSERT10NALLY ACTIVATED BY MMTV
AND COOPERATES WITH Wnt-1 IN MURINE MAMMARY
TUMORIGENESIS
INTRODUCTION
The mouse mammary tumor virus (MMTV) is a viral mutagen that induces
the formation of focal mammary tumors in mice upon infection. Random
pro viral integration within cellular loci frequently results in MMTV-mediated
transcriptional activation of nearby cellular proto-oncogenes that are normally
silent in the mammary gland. These molecular events confer an additional
proliferative advantage to the mutated mammary epithelial cells, which can
be naturally selected for and give rise to tumors of clonal nature (Varmus
1982 and 1987, Dickson 1984). In this model, MMTV not only acts as an
insertional activator of cellular genes, but the integrated provirus molecularly
tags these as well, thus facilitating gene identification or cloning.
Wnt-1 was the first gene identified as a common target of insertional
activation in mammary tumors of MMTV-infected mice (Nusse and Varmus
1982). Its in vivo oncogenic potential has been confirmed in transgenic mice,
in which Wnt-1 ectopic expression induces mammary hyperplasia and the
sporadic development of mammary tumors (Tsukamoto 1988). The
characteristics of tumor formation in these animals indicate that mammary
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tumorigenesis is a multi-step process in which, in addition to Wnt-1
activation, other genetic lesions must be present to contribute to tumor
development. It is therefore crucial to establish the hierarchy and functional
implications of such cooperating genetic events in order to understand the
molecular basis of malignant cell proliferation and tumorigenesis
To identify and isolate additional loci involved in multi-step mammary
tumorigenesis, we used MMTV-infected Wnt-1 transgenic mice. The potent
Wnt-1 oncogenic signal is constitutively activated in the mammary gland of
these transgenic animals. Upon MMTV-infection, these mice develop focal
mammary tumors with high incidence and an accelerated latency compared
to those observed in non-infeeted transgenic littermates (Shackleford 1993).
Using this experimental approach, we and others previously established that
three members of the fibroblast growth factor gene family (Fgf-3/int-2, Fgf-
4/int-3 and Fgf-8/AIGF) are insertionally activated by MMTV and/or
overexpressed in approximately half of these tumors. (Shackleford 1993,
Dickson 1994, Kwan 1992, MacArthur and Shankar 1995, Peters 1989,
Kapoun 1997). These facts, together with the finding that Wnt genes (Wnt-1
and Wnt-1 Ob) are preferential targets for MMTV insertional activation,
suggest that activation and cooperation of Wnt and Fgf proto-oncogenes, is
an early and key step in the development of murine mammary tumors.
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Although we showed tumor-specific proviral integrations, we could not detect
insertional activation of common Fgf or Wnt integration loci for MMTV in the
remaining group of mammary tumors (Shackleford 1993, MacArthur and
Shankar 1995). In our present study, we have searched for additional known
or uncharacterized genes that may be insertionally activated and cooperate
with Wnt-1 in this group of tumors.
FGFs promote cell growth through mitogenic stimulation, induction of
angiogenesis, or inhibition of apoptosis on diverse target cells. Signaling by
FGFs is mediated through specific transmembrane fibroblast growth factor
receptors (Fgfr 1-4). FGF/FGFR binding leads to the sequential activation of
the receptor’s tyrosine kinase activity and, in turn, of various intracellular
phosphorylation signaling cascades (Powers 2000). Aberrant cell growth
and tumor formation may result from disregulation at any step of these
signaling pathways, including that at the FGF receptor level.
In the present study we report the cloning and identification of a member of
the fibroblast growth factor receptor family, Fgfr2, as a new common genetic
target for MMTV insertional mutagenesis in mammary tumors from Wnt-1
transgenic mice. This report of an additional gene involved in the Fgf
signaling pathway underscores the importance of Wnt/Fgf cooperation in
multi-step mammary tumorigenesis.
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MATERIALS AND METHODS
Southern blot analysis and mapping of newly integrated proviruses in
Wnt-1 mammary tumors. Genomic DNAs (10 fig) from Wnt-1 mammary
tumors, as well as from uninfected non-transgenic mammary gland, were
digested with Xhol or EcoFN, electrophoresed at 30 volts for 30 hours on a
0.6% agarose gel, transferred to a nylon membrane (Hybond-XL, Amersham-
Pharmacia) and analyzed by Southern blot and autoradiography (Kodak X-
OMAT-MS) as previously described (Shackleford 1993, MacArthur and
Shankar 1995).
Preparation and selection of tum or genomic libraries. Proviral-cellular
junction fragments were identified by Southern blot as described above.
Size-selected fragments were then isolated and cloned into appropriate
lambda phage vectors to generate tumor genomic libraries that were
screened as previously described (Shackleford 1993). Clones were isolated
after two to three screening rounds and replating. Inserts were subcloned
into Bluescript (Stratagene) plasmids from which a repeat-free cellular probe
(RH 4.85) was obtained. The latter was used as a probe to screen for MMTV
insertions in the same locus in additional Wnt-1 mammary tumors.
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Exon trapping of Fgfr2 receptor. A 6.6 Kb Bam/Bam cellular genomic
fragment from tumor 14 was subcloned into the “exon trap” vector pML53ln
(gift of M. Reth) in both orientations, and exons 16 and 17 were “trapped”.
The plasmids were transiently transfected into Cos-7 cells. Cells were grown
and harvested, and total RNA was prepared as previously described. The
RNA was used to generate a Fgfr2 cDNA by RT-PCR using an exon trap
protocol as previously described and the cDNA was sequenced (Auch 1990,
Hamaguchi 1992).
RT-PCR analysis of Fgfr2 isoform expression. One-Step RT-PCR
(Qiagen One-Step RT-PCR Kit) was performed on tumor total RNA (50 ng)
after DNAse treatment of each RNA sample. Total RNA was extracted by
the urea-lithium chloride method as previously described (Shackleford 1993
and 1987). RT-PCR conditions: 50°C for 30 min, followed by 95°C for 15
min, and 35 cycles at 94°C for 30 s, 70°C for 1min, 72°C for 1 min, and 72°C
for 10 min. -RT controls as well as Actin RT-PCR positive internal controls
were performed in parallel for each sample. Actin PCR conditions were the
same as before except that the annealing temperature was 57°C. Primers
specific for Fgfr2 were 5’-GGCAGTAAAAACG-GGCGT-GATGGG-3’ (forward
primer), and 5’-CGTGATCT-CCTTCTCTCTCACAGG-31 (reverse primer).
Actin primers were 5’-CCCCATTGAACATGGCATTGTTACC-3’ (forward
primer), and 5’-CT-CTTTGATGTCAC-GCACGATTTCC-3’ (reverse primer).
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Fgfr2-speciftc primers amplified a 215 bp fragment, as visualized after
electrophoresis on a 2% agarose gel. Restriction enzyme digestion with
Hindi or Aval was used to distinguish between the PCR product
corresponding to each potentially expressed isoform
Real-time reverse transcriptase-mediated quantitative PCR. Total RNA
was isolated as described before (Shackleford 1993 and 1987) and 2 fig was
treated with DNase I (Amplification Grade, Gibco-BRL). This sample was
divided in two. Reverse-transcription was carried out with Omniscript
Reverse Transcriptase (Qiagen) or without the enzyme (-RT control), in a
40 jit reaction volume. The reaction mix contained dNTPs (0.5 mM each),
oligo-(dT)is primer (1 pM), RNase inhibitor (10 U/[il), 1x buffer, and enzyme
(4 U). RT-reaction conditions were 37 °C for 90 min, 93 °C for 5 min, 4 °C
cooling, and storage at -20 °C. Real-time PCR was performed with 2 pi of
RT product using the HotStartTaq PCR system (Qiagen) and a GeneAmp
5700 thermocycler. The reaction mix (20 [il total volume) contained dNTPs
(200 pM each), forward and reverse primers (0.25 pM each), 1x buffer, 0.2x
SYBR Green I (Molecular Probes) and 2pM 5.6 ROX dye (Sigma)
fluorogenic probes, and enzyme (2.5 U). The expression of a housekeeping
fi-Actin gene was used as an internal reference for normalization. Primers
and cycling conditions used for Fgfr2 or Actin amplification were the same as
described in b) above, except that 44 cycles were carried out. Controls with
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
no template were run in parallel for each run and always appeared negative.
Each measurement was set up in duplicate and two separate measurements
were performed. The Ct value was defined as the cycle number in which
detected fluorescence reached a threshold value of 0.1.
Statistical analysis. The statistical significance of the real-time quantitative
RT-PCR results was determined by the nested AN OVA model using the
STATISTICA software package.
Probes and sequencing. [y-3 2 PjATP-labe!ed repeat-free DNA probes were
prepared from plasmids as described before (MacArthur and Shankar 1995,
Lee 1995). The following probes were used on Southern and northern blots:
combined 5.4 kb PstI (MMTV gag-pol fragment)+1.2 kb BamH\ (MMTV env
fragment) probe. A unique 1.9 kb Pst\-Xho\ fragment (MMTV gag probe). A
unique 1.2 kb BamH\ fragment (MMTV-e/?v probe). A cellular fragment
(RIH3 4.85) cloned from the genomic Fgfr2 clone in tumor 14. Actin probes
generated from PCR as described above. Pgfr2 (exon 19) PCR fragment
(789 bp), generated under the following cycling conditions: 92°C for 2 min,
and 35 cycles at 92°C for 20 s, 53°C for 30 s, 72°C for 1 min, and 72°C for 10
min; forward primer 5-TCA-CCG-AGC-AGA-GGT-GGG-AAA-ATA-C-3’,
reverse primer 5’-TGA-GTG-GGC-GTA-TCC-AAA-GCA-AAA-C-3’.
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RESULTS
Cloning and mapping of a new common MMTV integration locus in
mammary tumors from MMTV-infected Wnt-1 transgenic mice. Our
previous work demonstrated that MMTV insertional activation of Fgf-3, Fgf-4,
and Fgf-8 accelerates mammary tumor formation in MMTV-infected Wnt-1
transgenic mice (Shackleford 1993, MacArthur and Shankar 1995, Kapoun
1997). This finding indicated that these Fgfs are potent collaborators with
Wnt-1 in mammary tumorigenesis. We were however unable to detect
activation of these or other known proto-oncogene targets for MMTV in a
small proportion of the tumors analyzed (Dickson 1984, Shackleford 1993,
MacArthur and Shankar 1995, Kapoun 1997). Nevertheless, genomic
Southern analysis with an MMTV-gag probe revealed that 72 tumors in this
group contained unique viral-cellular flanking bands, indicating presence of
clonal newly integrated proviruses (Fig. 1). This fact, together with the
kinetics of tumor formation observed in these tumors, suggested that other
insertionally activated proto-oncogenes might be cooperating with the Wnt-1
transgene in the genesis of the tumors (Shackleford 1993). In order to
isolate such genes, we cloned an Xhol-cleaved junction fragment from tumor
14 into X-phage vectors and screened the tumor DNA libraries with an
MMTV-gag probe. A unique 4.85 kb EcoRl-Hindlll cellular DNA fragment
was subcloned into a plasmid and used as a probe to screen for MMTV
insertions in the same locus in additional Wnt-1 mammary tumors with clonal
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 1114 39 42 74 95 1 1 1 C W
29 Kb
19 Kb
15 K b
12 Kb
10 Kb
Probe
Figure 1. New MMTV proviral insertions in mammary tumors from MMTV-
infected W nt-1 transgenic mice. The panel shows the Southern blot analysis of
genomic DNAs isolated from mammary tumors of MMTV-infected Wnt-1 transgenic
mice (designated by tumor number) or from Wnt-1 transgenic normal mammary
gland (W). All DNAs display large MMTV fragments, indicative of endogenous
retroviruses in these mice. Most of the tumor samples have additional rearranged
fragment(s), indicating the presence of clonal, tumor-specific newly integrated MMTV
proviruses. DNAs were digested with X h o I . 32P-labeled MMTV gag-pol+env cDNA
was used as a probe.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
insertions. Southern blot analysis of MMTV-infected Wnt-1 tumor genomic
DNAs detected the presence of rearranged host-viral junction fragments that
hybridized both to MMTV-specific probes and to the cellular probe in four
independent clonal tumors (Fig. 2). This finding suggested that we had
identified a new common MMTV integration site in these tumors. Further
analysis of the tumor DNAs revealed that the new MMTV pro viruses detected
at the common insertion locus were integrated in the same transcriptional
orientation as that of the cellular locus (Fig. 2 and 3).
Detection of expressed genes in the common integration locus.
Identification of expressed genes. We used exon trapping to generate
cDNAs from the isolated tumor DNA clonescorresponding to a
transcriptionally active gene in the new common MMTV insertion locus.
Sequencing of the resulting cDNA revealed that it contained a portion of the
Fgfr2 gene. Restriction enzyme digestion of lambda clone and plasmid
subclones confirmed the location of the MMTV insertion in tumor 14 (data not
shown). Southern blot analyses of tumor DNAs with probes from the locus
and MMTV showed that the new MMTV proviral insertions mapped to the
intron preceding exon 19, the last coding exon (Fig. 2 and 3).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fgfr2 RIM S profe® MMW*g.ag probe
7 14 m 42 3 i 14 39 42 : 95 C
— 22 Kft
i-nm:
i - 1 2 K 8
—10 K B ‘ ik i
^ " . * — 8® ” ■ -
F§fr2 Exon 19 probe MMTV-env probe
Figure 2. MMTV proviral insertion into a common integration locus.
Southern blot showing viral-host rearranged fragments in tumors 7, 14, 39,
and 42, corresponding to new proviral integrations in this locus. Specific
rearrangements of this locus were detected in four Wnt-1 tumors using the
cellular probe Fgfr2-RIH3 (A) or an Fgfr2 (Exon 19)-specific probe (B).
Arrowheads mark fragments that anneal to both Fgfr2 and MMTV probes,
demonstrating the location of the inserted proviruses. MMTV proviral
orientation was determined by probing the blots in (A) or (B) with MMTV-gag
(C) or MMTV-env (D) probes respectively. Tumor DNAs were digested with
EcoRV
of the copyright owner. Further reproduction prohibited without permission.
FSfr2 (mouse,-120 Kb )
New MMTV Integrations in
Wntl Tumors
39 /
/
/
16 17 18 19
R I0 3 Probe (4.85 Kb)
Figure 3. Map and orientation of proviral insertions in the new common
integration locus. The map was established as previously described. A X-phage
clone is indicated at the bottom of the figure. MMTV integrations were determined as
described in the legend to Fig. 2. Arrowheads indicate the position and transcriptional
orientation of the integrated proviruses
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Analysis of the relative expression of Fgfr2 isoforms in Wnt-1 tumors
harboring MMTV integrations in the Fgfr2 locus. The alternative splicing of
exon 8 or exon 9 in the Fgfr2 locus generates two alternatively spliced
isoforms. The Fgfr2(lllb) isoform is mainly expressed by epithelial cells, while
Fgfr2(iilc) is expressed mostly in stromal cells (18). A switch from lllb to life
expression is related with progression of certain prostate and mammary
tumors to a state of steroid hormone-independent growth and to a malignant
phenotype (19-22). For these reasons we sought to determine the isoform
expression pattern in the tumors carrying new MMTV integrations into the
Fgfr2 locus. We also compared this pattern to that in other Wnt-1 mammary
tumors (not containing new MMTV integration in known Wnt, Fgf or Fgfr loci)
as well as in normal mouse mammary gland, where lllb expression
predominates (23). With this purpose, we performed RT-PCR analysis on
total RNA from these tissue samples. PCR primers specific for regions
flanking the alternatively spliced Fgfr2 exon amplified products of the lllb and
lllc isofbrms that were indistinguishable by agarose gel electrophoresis. We
then digested the RT-PCR products with Aval or HincU, restriction enzymes
that cut only within the lllb or lllc exon, respectively, generating fragments of
considerable size difference. The RT-PCR product samples were cut only
completely with Aval, suggesting that Fgfr2(lllb), but not lllc, was
predominantly expressed in all the analyzed tumors, irrespectively of the
status of MMTV integration into a specific locus (Fig. 4B).
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A) B)
M 7 1 4 3 9 4 2 73 6 1 8 26 31 84 127 B, B,
300 bp----
200 bp—
Aval Digest
RT-PCR 215 bp ,
167 bp :
4 M b ft! ifci*
i
■
300 bp— ]
200 bp— j
100 b p _
500 li
400 b
300 b
■
714 39 42 73 6 18 26 31 84 127 B U
215 bp -
Actira
R T-PC R
Hinc II Digest
„ .Aval
f -► a .
- 216 bp
R ~
j | | F: Forward primer t
I I 8: Reverse primer E g M W jwm
213 bp
7 8 1 0 7 9 1 0
Figure 4. The Fgfr2(lllb)/Kgfr isoform is predominantly expressed in the Wnt-1
mammary tumors containing MMTV proviruses newly integrated into the Fgfr2
locus. A) RT-PCR was performed on Wnt-1 mammary tumor (designated by numbers)
or Wnt-1 normal mammary gland (B1, B2) RNAs using primers (F, R) specific for Fgfr2
exons 7 and 10 respectively. While the lllb splice variant, (predominantly expressed in
epithelial cells) contains exon 8, the lllc (predominantly expressed in stromal cells)
contains exon 9. The expected uncut PCR product is -215 bp long (top left panel). -RT
(middle left panel) and Actin +RT-PCR controls (bottom left panel) were included for
each sample. B) Restriction enzyme digestion of the PCR product with Aval (top right
panel) or H indi (bottom right panel) or) was used to distinguish between each potentially
exoressed isoform. U : Uncut RT-PCR Droduct.
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Relative Fgfr2 expression levels in Wnt-1 tumors harboring MMTV
proviral integrations in the Fgfr2 locus. Insertional transcriptional
activation of nearby proto-oncogenes is the molecular mechanism most
commonly implicated in MMTV-mediated mammary tumorigenesis (4, 24-26).
We thus sought to determine the relative level of Fgfr2 expression in the
mammary tumors containing proviral insertions into this genomic locus. We
used real-time reverse transcriptase-mediated quantitative PCR to calculate
the ratio of Fgfr2 mRNA level to fFActin level as an internal positive control.
We analyzed in this way the four mammary tumors (# 7, 14, 39 and 72) with
MMTV-insertions in the Fgfr2 locus, as well as seven control tumors with no
known MMTV integrations near Wnt, Fgf or Fgfr genes. The average
Fgfr2/Actin C t ratio for the group of tumors with integrations in Fgfr2 was
1.498, while the ratio for the control tumor group was 1.428 (Fig. 5). We
were unable to detect statistically significant differences between the average
CT ratios of both groups (nested ANOVA, P = 0.23), indicating that there was
no increase in the levels of Fgfr2 expression in the tumors with integrations in
Fgfr2 versus those in the control tumors. This result suggested that the
presumed MMTV-induced activation of Fgfr2 in these tumors might be
mediated through a mechanism other than transcriptional activation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
’ 43
£
H
U
a
«
o
efc
a
o »
2.2
2.0
1.6
1.4
1.2
1.0
0.6
I
a
I
Box Whisker Plot from Selected Block
Cases 1 through 4
I '
I
I
0
I
T
□
I
I
T1 T39 C26 C44 C115 G129
T14 T42 C30 CIOS C136
~T~ Mean+SD
Mean-SD
□ Mean+SE
. Mean-SE
□ Mean
Fgfr2 (IIIb)+ Fgfr2 (lllb) •
Tumors Tumors
Figure 5. Relative expression of Fgfr2 in W nt-1 mammary tumors as
measured with real-time RT-PCR. Total tumor RNAs reverse-transcribed
and analyzed as described in the Methods section. Fgfr2 expression
levels were normalized to those of the housekeeping gene p-Actin by
calculating the Fgfr2lp-Actin CT ratios. The relative Fgfr2 expression in
four tumors (# 7, 14, 39, 42) with MMTV-insertions in the Fgfr2 locus, as
well as for seven control tumors (26, 30, 44, 106, 115, 129, 138) with no
known MMTV integrations near Writ, Fgf, nor Fgfr genes. There was no
statistically significant difference in the average Fgfr2ip-Actin CT ratios
between both tumor groups (p = 0.23)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DISCUSSION
In previous work we have shown that MMTV infection of Wnt-1 transgenic
mice results in the preferential transcriptional activation of three
protooncogenes in the Fgf family: Fgf-3, Fgf-4, and Fgf-8 (Shackleford 1993,
MacArthur and Shankar 1995, Kapoun 1997). The accelerated formation
and increased incidence of mammary tumors in this model strongly indicates
that cooperation between FGF and WNT signaling pathways is a key event in
murine mammary tumorigenesis. This cooperation is further underscored by
the fact that Writ genes are preferentially activated by MMTV in mammary
tumors in Fgf-3 transgenic mice (Lee 1995, Van Leeuwen 1995). However,
none of the common Writ or Fgf targets for MMTV insertional mutagenesis
can be detected in a small fraction of mammary tumors in MMTV infected
Wnt-1 transgenic mice. We have cloned proviral-host junction fragments
from this group of tumors, and shown that four independent tumors of clonal
origin displayed new MMTV integrations in a common locus that contains the
C-terminus of Fgfr2. Specifically, the proviral insertions mapped to the intron
preceding the last coding exon, with ail proviruses integrated in the same
transcriptional orientation as Fgfr2. Our work underscores once more the
essential role that the collaboration between WNT and FGF oncogenic
signaling plays in multistep mammary tumorigenesis.
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FGFR overexpression and gene rearrangements have been implicated in the
genesis of various human cancer types (Powers 2000, Ishiwata 1998).
Upregulation of FGFR2(lllb) has been reported for pancreatic cancer
(Ishiwata 1998, Kornmann 1997). Moreover, a switch in FGFR2 isoform
differential expression from lllb to lllc in epithelial cells has been related to
progression to steroid hormone-independence and malignancy in prostate
tumors (Yan 1993). This observation and our finding of Fgfr2 insertional
activation by MMTV in Wnt-1 mammary tumors implicate Fgfr2 as an
unexpected putative proto-oncogene. Like the prostate, mammary gland
development and the physiological proliferative changes that take place
cyclically in the mammary epithelium are directly dependent on normal
steroid hormone (estrogen and progesterone) (Silberstein 2001, Hansen
2000). In addition, estrogen signaling through the estrogen receptor-alpha
(ER-a) plays a key role in breast cancer development. Similar to prostate
cancer, breast cancer progression to more aggressive stages is related to
the acquisition of an ER-negative and estrogen-independent phenotype (Pike
1993). Normal organ and tumor development in the prostate and mammary
glands thus share their steroid hormone dependence. In our study we have
not detected a significant change in FGFR2 isoform expression from lllb
(normally expressed in the mammary epithelium) to lllc. Establishing the role
that differential expression of Fgfr2 may play in relation to steroid hormone-
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
independence and malignant progression in mammary tumors remains an
avenue in multistep mammary tumorigenesis research not yet explored.
Insertional transcriptional activation appears as one of the most frequent
mechanisms of MMTV-mediated proto-oncogene activation. However, our
study shows that this does not appear to be the mutagenic event behind
Fgfr2 activation in the mammary tumors from MMTV-infected Wnt-1
transgenic mice (Teich 1982, Nusse 1991). FGFR2(lllb) expression is
normally found in various epithelial cells, including the mammary epithelium,
but not in many carcinoma cell lines (LaRochelle 1995, Carstens 1997) On
the other hand, the amplification of a shorter FGFR2(S!lc) variant that
displays a deleted C-terminal end is characteristic of many gastric cancer cell
lines (Ueda, 1999). The FGFR2 C-terminus contains a negative regulatory
domain required for inhibition of NIH3T3 transformation. It has been
suggested that deletion of this domain, present in different types of proto
oncogenes that encode cellular kinases, leads to kinase activation, and
disregulation of cell proliferation (Ueda 1999, Lodish 2000).
As an alternative mechanism to insertional transcriptional activation, we
believe that, upon proviral insertion within the last intron of the Fgfr2 gene,
use of the polyadenylation signal present in the MMTV LTR results in
premature termination of transcription and the generation of a truncated
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FGFR2 message. This message codes for tyrosine kinase receptor that
lacks the C-terminal negative regulatory domain and is therefore
constitutively active, independently of ligand availability. FGFR2-mediated
constitutive activation of FGF intracellular signaling cascades would
eventually lead to transformation of the insertionally mutated mammary
epithelial cells. Preliminary work by others (O. Tuason) in our lab suggests
that our four tumors of interest may express indeed a chimeric Fgfr2
transcript lacking the last exon and containing intron 18 and MMTV 5’ LTR
sequences.
Our report of the putative oncogenic potential of FGFR2(lllb) does not
conflict with previous reports on the tumor suppressive effect of FGFR2(lllb)
expression in rat prostate and human bladder cancer (Feng 1997, Matsubara
1998, Ricol 1999). While deletion of a negative regulatory domain appears
to be the reason for the receptor’s oncogenic potential in our study and the
mentioned reports, other mechanisms, such as expression of dominant
negative FGFR2 variants, or FGFR2-mediated induction of cell cycle
inhibitors (Ricol 1999) could also be responsible for FGFR2-mediated tumor
suppression. Finally, insertional activation of Fgfr2 in MMTV-infected Wnt-1
transgenic mice further is one more piece of evidence pointing at the key role
that activation and cooperation of the Wnt and Fgf signalling pathways plays
in murine mammary tumorigenesis. Similar retroviral insertional mutagenesis
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5’ L T R
U3 RU5
17 18
Truncated Fgfr2 mRNA
Exon 1 / / 17 18
I3 Z Z 3 -
Poly A
19
3’
I
Transcription
Intron 18
- \
U3
iV.V.V.V.V.V.Vr
A
Poly A
Figure 6. Sketch representing an alternative mechanism for MMTV-mediated
F g fr2 insertional activation in mammary tumors from MMTV infected W nt-1
transgenic mice. The recognition of a polyadenylation signal present in the LTR of
new proviruses within the last intron (intron 18) in Fgfr2 locus leads to premature
termination of the Fgfr2 mRNA. While this truncated transcript lacks the exon 18
sequence (exon upstream of the new integrations), it contains a chimeric C-terminus
with partial intron 18 and MMTV LTR sequences of variable length, depending on
the MM TV insertion position within intron 18.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
studies on Wnt/Fgf bitransgenic mice appear therefore as the next logical
approach to identify other oncogenes involved in aberrant cell proliferation
and mammary tumor development.
ACKNOWLEDGEMENTS
I thank H. Varmus for the generous gift of tumor samples; M. Reth for
providing the exon trap vector; S. Twigg for providing plasmid containing the
Fgfr2 intron 18 sequence; I. Tiemann-Boegge and N. Arnheim for providing
hardware, software and technical advice for our real-time quantitative RT-
PCR analysis, and D. Ochavillo for providing technical help with the statistical
analysis of the data.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3: GENERATION OF FGFR2DN TRANSGENIC
MOOSE LINES
EXPERIMENTAL HYPOTHESIS/RATIONALE. SUPPORTING
EVIDENCE
Oncogenic cooperation between the Wnt and Fgf signaling pathways is an
early and crucial event involved in mouse mammary tumorigenesis. Cancer
development requires the accumulation of multiple genetic mutations in a
multistep fashion. One might therefore hypothesize that, in addition to
mutations leading to Wnt and/or Fgf activation, other genetic events may play
a role in mouse mammary tumor formation.
In order to identify other novel or unexpected putative proto-oncogenes
involved in this process, I have taken onto the generation of a transgenic
mouse model overexpressing a truncated Fgf receptor in the mammary
gland. There are four different Fgf receptors: Fgfrl-4. The wild-type Fgfr
comprises two or three extracellular immunoglobulin (Ig)-like domains, an
acidic region between the first and second Ig-like domains, a transmembrane
domain, and an intracellular catalytic domain with tyrosine kinase (TK)
activity (Fig. 7). Alternative splicing of sequences corresponding
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
%
1) Heparin/FGF
Binding
f
4) TK Trans-
Phosphorylation
< ►
#
2) Hep/FGF/FGFR
Complex Formation
3) FGFR
Dimerization
5) Downstream
events ...
Figure 7. Fgf/Fgfr signal transduction. Upon formation of a specific
Fgf/HSPG/Fgfr complex (1,2), Fgfr dimerization (3) is induced, which
allows transphos-phorylation events (4) to activate the receptor dimer.
Once active, the tyrosine kinase receptor dimer mediates the initiation
of an intracellular phosphorylation signaling cascade leading to the
up/down regulation of specific target genes. Truncation of the
cytoplasmic tyrosine kinase domain allows ligand binding and receptor
dimerization, yet it renders non-functional dimmers, incapable of
initiating intracellular phosphorylation events.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to the third Ig-like domain can generate differentially expressed receptor
isoforms lllb (in cells of epithelial origin) or lllc (in cells of mesenchymal
lineage) (Yan 1993, Grr-Urtreger 1993). Each receptor isoform displays
different tissue-specificiy and ligand-binding affinity. In this way, FGF/FGFR
signaling may elicit distinct biological effects in a tissue- and time-specific
fashion (Miki 1992). Fgf ligand binding to its specific receptor induces the
formation of Fgfr homo- or heterodimers on the cell surface. The Fgfr dimer
is the active receptor form, in which transphosphorylation events between
each monomer’s TK domains take place. The active phosphorylated TK
domains transduce in turn the FGF signal through the recruitement and
phosphorylation of intracellular target proteins.
I hypothesize that signaling by multiple endogenous Fgfrs can be abolished
by a single type of truncated Fgfr, lacking the tyrosine kinase domain. This
receptor should be able to bind specific Fgf ligands as well as to dimerize
with endogenous Fgf receptors expressed on the cell surface of mammary
epithelial cells. However, since the transgenic receptor lacks its catalytic
domain, it will form nonfunctional heterodimers upon its interaction with wild-
type Fgfrs. Since Fgfr dimerization can lead to the formation of homo as well
as heterodimers, and each receptor monomer binds more specifically
different Fgfs, overexpression of the truncated Fgfr transgene could thus act
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
as a dominant negative receptor blocking, partially or completely, Fgf/Fgfr-
mediated endogenous intracellular signaling in the mammary gland.
In fact, similar approaches have been successfully used to study the effects
of Fgf signaling and its impairment at different levels. NIH 3T3 cell
transformation can be inhibited by the expression of truncated (kinase-
negative) FGFRs (Li 1994). Expression of truncated Fgfrl transgenes,
lacking most of their cytoplasmic domain, blocks wild-type Fgfr-mediated
signal transduction in Xenopus oocytes, and results in disrupted eye lens
development in mice (Ueno 1992, Robinson 1995). Additional supporting
evidence arises from experiments in which targeted expression of kinase-
negative Fgfr2 transgenes to the mouse lung or to the mammary gland
results in a block of epithelial differentiation and branching morphogenesis,
or in impaired lobuloalveolar development in each organ respectively (Peters
1994, Jackson 1997).
MATERIALS and METHODS
Construction of the Fgfr2DN transgene. A 2 Kb truncated mouse Fgfr2
lltb(Bek) cDNA was released by EcoRl digestion of a pBS-KS7Fgfr2 cDNA
plasmid. The fragment was purified (Qiagen DNA Gel Extraction Kit)
following agarose gel electrophoresis [1% agarose gel (Gibco-BRL) in 1X
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Tris-Acetate-EDTA running buffer]. The fragment was ligated into a linearized
pBS-KS' vector with EcoRi ends and correct fragment orientation was
determined by restriction enzyme digest with EcoRV, Sac I, J3g/I,and Apa\.
The MMTV-LTR (1.7 Kb) was excised (Bg/ll/H/ndlll) from a MTV-Catfi LTR
plasmid and cloned into (BamH\/Hind\\\) the pBK-CMV vector (Stratagene),
which was renamed as MTV(ACIa)LTR/CMV 5.9. This vector was then cut
with Notl/Clal and the truncated Fgfr2 cDNA fragment (Bek TK', 2.0 Kb) with
the same compatible ends was excised from pBS-KS' and cloned into it. The
correct construction of the MTV(ACIa)LTR-Bek TK' transgene was checked
by restriction enzyme digest with Sail, Sspl, Hindlll, Sall/Sspl, Sall/Hindlll,
Notl/Clal, Nsi/Sall, and subsequent Southern blot with MMTV-LTR- or Fgfr2
cDNA-specific probes. Finally, restriction enzyme digest (Nsi/Sall) was
performed to release a 4.9 Kb construct. This construct was purified (Qiagen
DNA Gel Extraction Kit) following agarose gel electrophoresis [1% agarose
gel (Gibco-BRL) in 1X Tris-Acetate-EDTA running buffer], and sent for
mouse blastocyst injection.
Restriction enzyme digests. Plasmid DNA or genomic mouse DNA was
digested (37°C, 2-16 hours) in a reaction mixture containing Bovine Serum
Albumine solution (BSA, 0.1 mg/ml), 1x supplied enzyme buffer, restriction
enzyme (1-5 U/pg DNA) [New England Biolabs (NEB)], and double distilled
sterile water, up to a 20-50 p! volume.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Restriction enzyme fragment end-blunting. Fill-in blunting of the fragment
ends was performed (room temperature, 15 min) in a mix with Klenow
polymerase (NEB, 1 U/pg DNA) and all four dNTPs (0.5 mM each), and was
followed by Klenow inactivation (75°C, 10 min).
Nucleic acid extractions. Bacterial plasmid DNA was extracted using an
alkaline lysis protocol or the Qiagen Midi/Maxiprep DNA extraction kit.
Genomic DNA from mouse tail clippings was obtained by a
phenol:chloroform extraction method as previously described, except that
serum separation tubes were used in the extractions (Shackleford 1993).
Briefly, the tissue samples were incubated (55°C, 16 h) in digestion buffer
(10 mM Tris-CI pH 8.0, 10 mM, EDTA pH 8.0, 0.5%SDS, Proteinase K 50
pg/ml). The digestion mix was then subjected to two phenol:chloroform (1:1)
extractions, followed by DNA precipitation from the aqueous phase in 2.5 vol,
100% ethanol+0.1 vol 5 M sodium acetate. The DNA pellets were air dried
and resuspended in buffer (10 mM Tris-CI 1, EDTA 1mM pH 8.0) and stored
at 4°C. Tumor RNAs were isolated following the urea-lithium chloride
method (Auffray 1980) and stored as isopropanol precipitates at-20°C.
Cloning into plasmid vectors. DNA fragments were cloned into pBS-KS' or
pBK-CMV vectors (Stratagene) upon ligation (16°C, 16 h) with T4 DNA ligase
following manufacturer’s instructions (New England Biolabs).
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In vitro DNA transfection of mammaiian ceils. C57MG mammary
epithelial cells (2 X 105 cells/well) were grown in 6-well-plates under the
following conditions: DMEM medium, 10 % Fetal Bovine Serum (FBS), 10X
penicillin-streptomycin, insuline (10 fig/ml), 5% C 02 ) 37°C. Upon 80%
confluence, the cells were stably transfected with either a linearized (Sail)
pCMV-MTV(ACIa)LTR-Bek TK' DNA construct, or linearized pBK-CMV vector
alone. The transfection was performed following manufacturer’s (Gibco-
BRL) instructions. Briefly, the trasfection mix contained: DNA (2 |ig) in
OptiMem serum-free medium (100 pi, Gibco-BRL)+Lipofectamine (15 ja I,
Gibco-BRL) in OptiMem serum-free medium (100 jjiI, Gibco-BRL). The cells
were incubated with the transfection mix in serum-free medium (0.8 ml) for
7.5 h, and then washed gently in PBS and incubated in normal growth
medium overnight. The stably transfected cell clones were cells were
subjected to G418-selection (600pg/ml, 7days).
Independent stably transfected clones were grown in 6-well plates up to 40%
confluence. At this point, in order to induce MMTV-driven transgene
expression, half of the clones received dexamethasone (Dex,10-7 M)
treatment for the remaining duration of the experiment. As cells reached
75% confluence, they were grown in the presence or absence of heparin (1
(xg/ml) plus human recombinant FGF2 or FGF4 protein (2 or 0.4 ng/mi,
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PreproTech inc.). The levels of cell proliferation were assesed at days 2, 3, 4,
7, and 11 after initial FGF treatment.
Southern blot analysis. Genomic DNAs (10 pg) from mouse tail clippings
were digested with HindlU, electrophoresed at 40 volts for 12 hours on a 1%
agarose gel and transferred to nylon membranes (Hybond-XL, Amersham-
Pharmacia) as previously described (Shackleford 1993, MacArthur and
Shankar 1995). The blots were UV-cross linked, prehybridized for 45 min to
1 h at 65°C, and hybridized overnight (65°C) with radiolabeled DNA probes.
The hybridization buffer contained 0.5 M sodium phosphate (pH 7.2), 1mM
EDTA (pH 8.0), 1% (w/v) BSA (fraction V), 7% (w/v) sodium dodecyl sulfate
(SDS), and 15% (v/v) formamide. Blots were washed (3 X 30 min, 65°C) in
40 mM sodium phosphate (pH 7.2)-1mM EDTA (pH 8.0)-1% SDS, and
analyzed by autoradiography with intensifying screens (Kodak X-OMAT-MS)
as previously described (Shackleford 1993, MacArthur and Shankar 1995).
Northern blot analysis. RNAs were resuspended in a solution containing
50 mM N-2-hydroxyethylpiperazine-N’-2-ethasulfonic acid (HEPES, pH 7.0),
10 mM sodium acetate, 1 mM EDTA (pH 8.0), 0.25 jig of ethidium
bromide/ml, 0.66 M formaldehyde, and 50 % (v/v) formamide, denatured at
65°C for 5 min, and electrophoresed (30-45 V, 6-16 h) in 0.8% agarose gels
(with the same running buffer minus formamide). Following photography the
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
gel was blotted overnight onto nylon membranes. Cross-linking,
prehybridization, hybridization, washing, and exposure were as for DNA
except that the hybridization buffer contained 30% (v/v) formamide.
Probes. A [v-3 2 P]ATP-labeled repeat-free mouse Fgfr2 TK' cDNA (2.0 kb,
EcoRX) or rat glyceraldehyde-3-phosphate dehydrogenase (Gapdh) cDNA
probes were prepared from plasmids as previously described (MacArthur and
Shankar 1995, Lee 1995).
RESULTS
Construction of the Fgfr2DN transgene and generation of Fgfr2DN
transgenic mouse lines. I have constructed an MMTV-Fgfr2D/V transgene
(4.9 Kb) that consists of an MMTV LTR promoter driving the expression of a
truncated murine Fgfr2 cDNA (lacking the tyrosine kinase domain). The 3’
end of the construct also includes the splice and poly-adenylation signals of
SV40 DNA (Fig. 8). The purified construct was sent to the (JSC/Norris
transgenic facility, for the generation of C57BL/6SJL Fgfr2DN transgenic
mice. The use of the MMTV-LTR promoter targets transgene expression to
the mammary epithelium. In addition, the presence of steroid-responsive
elements within the LTR regulates overexpression of the transgene under
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S al I
N si I
H III
H III
, . 1TV ( 1 /
D N -F g J r 2
DNA
SV40
,.A
1.4 Kb 2.0 Kb J).5 Kb.
>
0 j
4.9 Kb
Figure 8. Sketch of the MMTV -F g fr2D N transgene. The 4.9 K b
(Sall/Nsil) transgene contains an MMTV-LTR that drives the
expression of a kinase negative Fgfr2DN cDNA. SV40 poly(A) and
intron sequences are also included. H IM : HindlW
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
conditions of steroid hormone stimulation, such as during pregnancy and
lactation (Cato 1989). The positive transgenic founders were screened by
Southern blot analysis of tail genomic DNA, using an Fgfr2DN radiolabeled
probe (Fig. 9). Five transgenic lines were established by breeding the
founder mice to BALB/CByJ mice to preclude homozygosity of C57BL/6
alleles that could inhibit MMTV infection (Pucillo 1993). A total of 357 mice
were generated and screened by Southern blot, as previously described,
during the process.
Demonstration of the dominant negative effect of Fgfr2DN transgene
expression in vitro. C57MG mammary epithelial cells were stably
transfected with either the MMTV- Fgfr2DN construct cloned into a linearized
pBK-CMV vector, or with linear empty vector alone. G418-selected stably
transfected clones were then grown in the presence or absence of
dexamethasone (Dex, 10"7 M)—to activate MMTV-driven transgene
expression— and recombinant mouse FGF2 (2 ng/ml). Those cells grown in
the presence of FGF2 were also treated with heparin (1 jig/ml), since HSPG
is required for FGF/FGFR complex formation. FGF2 untreated, -/+Dex, cells
displayed the normal flat and cuboidal monolayer morphology that
characterizes epithelial cells. Upon FGF2 treatment mock transfected cells
responded displayed a spindle-shaped, bi-refringent appearance, with most
cells in a high proliferative state. These features are characteristic of
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mitogen-mediated cell transformation such as that elicited by FGF2. In
contrast, cells transfected with the MMTV-Fgfr2 construct, were more
resistant to FGF-induced mitogenesis. This observation suggested that in
vitro overexpression of the truncated Fgfr2 construct in these mammary
epithelial cells was indeed blocking FGF signaling as expected, very possibly
through a dominant negative effect.
Analysis of Fgfr2DN transgene expression in the mammary gland of
five MMTV-Fgfr2DN transgenic mouse lines. The expression level of the
truncated Fgfr2DN transgene was determined by northern blot analysis of
lactating mammary gland total RNA (20 pig) from two offspring females from
each transgenic founder (numbers 9, 52, 59, 66, 67). Offspring from three
transgenic lines (number 52, 66, 67) displayed marked overexpression of the
transgene in the mammary gland (Fig. 11). Transgenic mice from the
transgene-expressing lines were thus selected to generate Wnt10b/Fgfr2DN
and Wnt1/Fgfr2DN bitransgenic mice. The creation of these mice will be
discussed in the following chapter.
DISCUSSION and SUMMARY
An MMTV-Fgfr2DN transgene, lacking its intracytoplasmic tyrosine kinase
domain, has been created. In vitro expression of this transgene in C57MG
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
B 52 59 66 62 69 67
m il H i* . ,
W m .......................... l ^ g g l
gipilii*.-,.
H it — 4 Kb
3 Kb
m m (B m —
Fig. 9. Southern blot analysis of Fgfr2-D N transgenic
m ice. Fgfr2 cDNA was used as a probe. Positive
transgenics display a 2 Kb band. Higher molecular weight
bands correspond to restriction enzyme fragments generated
within the endogenous Fgfr2 locus. 52, 59, 66, 67: positive
transgenics. 62, 69: non-transgenic littermates. B: BALC/ByJ
non-transgenic control. The diagram below indicates the
Fgfr2DN fragment used as a probe (in red). Hill: HindlU
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
UNTREATED____________ FGF2 (2 na/mtl
Figure 10. MMTSf-FgfrlDN transgene in vitro transfection into C57MG cells demonstrates its dominant negative
effect and the inhibition of Fgf-mediated cell mitogenesis. Top half: cells transfected with mock vector control. Bottom
half: Cells transfected with truncated MMTV-Fgfr2DN construct. Cells were grown in the absence (-DEX) or presence
(+DEX) of dexamethasone (to stimulate transgene expression) and/or FGF2. Bottom right panel shows inhibition of cell
proliferation and regression to a flat cell (non-transformed) phenotype in MMTV-Fgfr2DA/-transfected cells despite FGF2
treatment. ■
O n
ON
F Is from _JL _5 2 _ _52 66 67
f o s# r " " . ........................— ~
6 Kb
3.3 Kb
mm
* fr* F gfr2 B N
CA PD Ff
Figure 11. Northern blot analysis of Fgfr2D N transgene
expression in lactating mammary gland total RNA. Top
panel: Fgfr2 cDNA probe. Bottom panel: Gapdh cDNA
probe. 20 pg of total RNA was analyzed. B: control non-
transgenic lactating mammary gland total RNA. Markers on
the right indicate the size of the endogenous (6 Kb) and
truncated transgenic (3.3 Kb) message. Numbers on top of
each lane indicate the progenitor transgenic mouse
founder.
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mammary epithelial cells can inhibit FGF-mediated mitogenesis. Despite the
evidence, the reversion to a non-transformed phenotype observed in these
cells was not complete. It has been reported that a block in FGF-induced
transformation is precluded by the addition of excess growth factor (Li 1994).
It is thus possible that besides eliciting a dominant negative effect, truncated
receptors may act in part through competition with the endogenous receptor
for direct FGF binding. In our hands, complete reversion to a non
transformed phenotype was observed when a lower (0.4 ng/ml)
concentration of FGF2 was used (not shown), corroborating the previous
report.
Five independent MMYsf-Fgfr2DN transgenic mouse lines have been
created, as a first step leading to the generation of MMTV-infected
Wnt/Fgfr2DN bitransgenic mice. The MMTV-LTR present in the transgene
targets expression specifically to the mammary gland. Three of these
transgenic lines displayed strong transgene expression levels in the
mammary gland during lactation, and were thus chosen as ideal candidates
for crossbreeding to Wnt transgenic mice. Previous reports suggest a key
role for Fgfr2 in normal mammary gland development. Specifically,
expression of a kinase defective Fgfr2 transgene in the mammary gland led
to inhibition of lobuloalveolar development during pregnancy and lactation,
and reduced lactogenesis, resulting in an inability to sustain pups (Jackson
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1997). We have not observed lactation-related problems in our transgenic
mice. One may thus argue that normal mammary development and
lactogenesis in our mice may reflect a partial rather than a complete
dominant negative effect of our MMTsf-Fgfr2DN transgene. This scenario
could be due to different reasons, ranging from preferential dimerization with
some but not all endogenous Fgfr types, or low transgene/endogenous Fgfr
expression ratio, to the existence of very high affinity endogenous receptors,
increased circulating Fgf levels, or a number of other unknown factors in our
animals. Nevertheless, our observation does not categorically imply a lack of
effect of our transgene in the mammary gland. Since our mouse strain is
different from that used in the reported experiments, it is also possible that
genetic modifiers characteristic of each strain may account for the observed
differences in mammary gland physiology.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4. MMTV INSERT80NAL MUTAGENESIS IN
WNT/FGFR2DN AND WNT/FGF BITRANSGENtC MOUSE
MODELS FOR MAMMARY ONCOGENE DISCOVERY
EXPERIMENTAL RATIONALE
The use of mouse mammary tumor virus (MMTV) insertional mutagenesis in
transgenic mice is a very useful approach to identify novel or unexpected
proto-oncogenes implicated the development of mammary tumors. The
mouse mammary tumor virus (MMTV) is a biological carcinogen that induces
murine mammary tumorigenesis by insertional mutagenesis (Tekmal 1997,
Teich 1982, Callahan 1996). MMTV proviral integration in the proximity of
cellular proto-oncogenes may result in activation of their expression. This
event may confer a selective growth advantage to the mutated mammary
epithelial cells and facilitates their clonal outgrowth, eventually leading to
tumor formation. In tumor DMAs containing newly integrated proviruses,
MMTV is physically linked to adjacent insertionally activated proto
oncogenes, thus it can be used as a molecular tag that facilitates the cloning
and identification of the activated gene.
MMTV-induced transcriptional activation of Wnt genes, more frequently
Wnt1, is a key early step in multistep mammary tumorigenesis in mice. The
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
stochastic development of mammary tumors observed in Wnt1 and in Fgf3-
transgenic mice suggests that single proto-oncogene activation contributes to
but is not sufficient for tumorigenesis (Peters 1989, Tsukamoto 1988).
Analysis of the MMTV integration loci in mammary tumors from MMTV-
infected Wnt1 or Fgf3-transgenics has revealed preferential activation and
expression of Fgf genes (Fgf3, Fgf4, Fgf8) and the Fgfr2 receptor in Wnt1
transgenics (see chapter 2), or activation of Wnt genes (Wnt1, WntlOb) in
Fgf3 transgenics (Shackleford 1993, MacArthur and Shankar 1995, Lee
1995, Peters 1989). In addition to Wnt1, the mammary oncogenic potential of
WntlOb and Fgf3 has been confirmed by generation of single transgenic
mouse models.
The preferential activation of Fgf genes in Wnt transgenics and vice versa, in
addition to the decreased tumor latencies observed in Wnt1/Fgf3
bitransgenics, demonstrate that activation of Wnt and Fgf genes, and strong
oncogenic cooperation between both growth factor families, are crucial
events underlying the genetic basis of multistep mammary tumorigenesis
(Kwan 1992). Nonetheless, not much is known about any additional
oncogenic cooperation events leading to mammary tumorigenesis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hypothesis/Rationale. The overall aim of this project is to isolate and
characterize novel or unexpected genes that, in addition to Fgfs, collaborate
with early activation of Wnt genes in multistep mammary tumorigenesis.
With this purpose, I propose to use MMTV-insertional mutagenesis in
transgenic mice carrying two transgenes expressed in the mammary gland:
WntlOb and a dominant negative form of the Fgfr2 lllc receptor isoform
(Fgfr2DN). While Wnt oncogenic signaling is constitutively activated in the
mammary gland of these mice, expression of the Fgfr2DN transgene (a
truncated, tyrosine kinase-deficient, non-functional Fgfr2 lllc receptor
isoform) will abolish cooperative Fgf signaling. A transgene for the Fgfr2 lllc
splice variant has been chosen over other receptor variants because it has
been shown to interact with a higher number of Fgf types (Omitz 1996). This
fact, together with the demonstrated dominant negative effect of tyrosine
kinase-truncated Fgfrs on endogenous Fgf signaling (Li 1994, Ueno 1992,
Robinson 1995, 1994, Jackson 1997) lead us to believe that Fgfr2DN
expression should abolish signaling by many, if not all, Fgfs in the mammary
gland.
Upon MMTV-infection in our bitransgenic mouse models, Wnts, Fgfs, and
other proto-oncogenes will be activated through random proviral insertional
mutagenesis. Since the Wnt and Fgf signals will be overexpressed and
blocked respectively in the mammary epithelial cells of WntlOb/ Fgfr2DN
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transgenics, I hypothesize that the insertional activation of proto-oncogenes
other than Writs and Fgfs may confer an additional proliferative/growth
advantage to the cells carrying such gene activations. Therefore, those
mutated cells should be naturally selected for in the transgenic mammary
glands. Eventually, the clonal outgrowth of these cells should give rise to
mammary tumors with an accelerated latency compared to that in infected
non-transgenic littermates. Tumors arising due to proviral insertions contain
proviruses physically linked to the activated proto-oncogenes. The newly
integrated proviruses thus serve as a molecular tag that facilitates
identification or cloning of the activated genes within the new proviral
integration loci in the mammary tumors analyzed.
In order to ensure that we have access to sufficient number of potential
candidate tumors for our insertional mutagenesis and oncogenic cooperation
studies, I also proposed to create two additional Wnt1/Fgfr2DN and
Wnt1/Fgf3 bitransgenic mouse models. Moreover, it is also possible that
variation in the constitutive Wnt and Fgf overexpression patterns of each
experimental group may lead to identification of a wider range of novel
oncogenes. The same working hypothesis and rationale, as originally
postulated for our MMTV-infected Wnt10b/DNFgfr2 bitransgenics, apply for
the Wnt1/DNFgfr2 experimental group. In a similar fashion, in MMTV-
infected Wnt1/Fgf3 bitransgenics, we expect that constitutive overexpression
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of Wnt1 and Fgf3 oncogenic signals will lead to the genesis of clonal
mammary tumors displaying the insertional activation of oncogenes that
cooperate with both Writs and Fgfs in multistep mammary tumorigenesis.
In order to carry out our insertional mutagenesis studies, it was initially
decided to use a modified hybrid MMTV provirus that is composed of the 5’-
half from Mtv1 and a 3’-half from MMTV(C3H). This provirus also carries a
bacterial supF gene within both U3-LTR regions ( Shackleford 1989, Jiang
1999). The supF gene encodes a bacterial suppressor tRNA gene, and its
expression in tumors arising from MMTV-infected mammary epithelial cells
should enable rapid isolation of proviral integration sites as previously
reported. Briefly, we hope to clone proviral-cellular tumor DNA fragments
into specific X-phage vectors (Xgft/VES.XB or XZap, depending on the size of
the fragment to be cloned) to create subgenomic tumor DNA libraries. These
X-phage vectors contain amber (STOP) mutations in lytic growth genes and
are thus unable to grow in wild type E. Coli (supF*3 ). The presence of supF, a
bacterial suppressor tRNA gene, will preclude lambda screening for proviral
insertion sites. Only those X-phage vectors containing viral-cellular junction
fragments (which carry supF) will form plaques that can then be selected.
The junction fragments obtained in this manner can be screened for the
putative activated oncogenes using an inverse polymerase chain reaction
(IPCR) approach (Lee 1995, Li 1999, Akasaka 2000, Raponi 2000).
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Due to experimental problems, that will be discussed in detail within some of
the following sections, we were forced to change our MMTV-infection
approach. Instead of injecting the hybrid MMTV-supF provirus, we resorted
to the injection of wild type MMTV-C3H viral stock.
The basic experimental design that was followed in this project is described
in figure 12. Briefly, single transgenic mice were mated to generate sufficient
numbers of positive bitransgenic or single transgenic females. For each
bitransgenic mouse model, we created 3 cohorts (n=40-50 mice/cohort) of
female mice: a bitransgenic cohort, and one cohort for each single transgene
type, as controls. Half of the females (n=20-25) in each cohort were infected
with MMTV at 3-4 weeks of age. The other half remained uninfected as
negative controls. In order to stimulate ongoing mammary epithelial cell
division, and hence MMTV infection spreading and expression, all females
were uninterruptedly bred, and allowed to lactate during 7-10 days after each
pregnancy, until mammary tumors were detected. The tumor DNAs were
then analyzed for the presence of new MMTV integrations. Tumors DNAs
carrying such integrations were used to generate viral-cellular junction
fragments that will serve to clone insertionally activated genes using IPCR.
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
^ Wnt10b/Fgfr2 DN
W nt1/Fgfr2DN
W nt1/Fgf3 '
A
WntlOb
W nt1
A
Un-infected
N= 20
Un-infected
N- 20
MMTV
Infected
N- 20
MMTV
Infected
N= 20
Fgfr2 DN
Fgf3
A
Un-infected
N= 20
MMTV
Infected
N= 20
Figure 12. Experimental Design of Female Mouse Cohorts
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MATERIALS and METHODS
Nucleic acid extractions. Bacterial plasmid DNA was extracted using an
alkaline lysis protocol or the Qiagen Midi/Maxiprep DNA extraction kit.
Genomic DNA from mouse tail clippings and mammary tumors was obtained
by a phenol:chloroform extraction method as previously described, except
that serum separation tubes were used in the extractions (Shackleford
1993). Briefly, tissue samples were digested (55°C, 16 h) in digestion buffer
(10 mM Tris-CI pH 8.0, 10 m i, EDTA pH 8.0, 0.5%SDS, Proteinase K 50
pg/ml). The digestion mix was then subject to two phenol:chloroform (1:1)
extractions, followed by supernatant DNA precipitation in 2.5 vol, 100%
ethanol+0.1 vol 5 M sodium acetate. The DNA pellets were air dried and
resuspended in buffer (10 mM Tris-CI 1, EDTA 1mM pH 8.0) and stored at
4°C. Tumor RNAs were isolated following the urea-lithium chloride method
(Auffray 1980) and stored as isopropanol precipitates at -20°C.
Restriction enzyme digests. Plasmid DNA or genomic mouse DNA was
digested (37°C, 2-16 hours) in a mix containing 1X Bovine Serum Albumin
solution (BSA, 10 mg/ml), 1x supplied enzyme buffer, restriction enzyme (1-5
U/pg DNA) [New England Biolabs (NEB)], and double distilled sterile water,
up to a 20-50 pi volume.
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Southern blot analysis. Genomic DNAs (10 p.g) from mouse tail clippings
or from mammary tumors where digested with restriction enzymes (New
England Biolabs), electrophoresed (tail DNAs: 70 volts for 7 hours, 1%
agarose gel; tumor DNAs: 40 volts for 12 hours on a 0.6% agarose gel) and
transferred to nylon membranes (Hybond-XL, Amersham-Pharmacia) as
previously described (Shackleford 1993, MacArthur and Shankar 1995). The
blots were UV-cross linked, prehybridized 45 min to 1 h at 65°C, and
hybridized overnight (65°C) with radiolabeled DNA probes. The hybridization
buffer contained 0.5 M sodium phosphate (pH 7.2), 1mM EDTA (pH 8.0), 1%
BSA (fraction V), 7% sodium dodecyl sulfate (SDS), and 15% (vol/vol)
formamide. Blots were washed (3 X 30 min, 65°C) in 40 mM sodium
phosphate (pH 7.2)-1mM EDTA (pH 8.0)-1% SDS, and analyzed by
autoradiography with intensifying screens (Kodak X-OMAT-MS) as
previously described (Shackleford 1993, MacArthur and Shankar 1995).
Northern blot analysis. RNAs were resuspended in a solution containing
50 mM N-2-hydroxyethylpiperazine-N’-2-ethasulfonic acid (HEPES, pH 7.0),
10 mM sodium acetate, 1 mM EDTA (pH 8.0), 0.25 fig of ethidium
bromide/ml, 0.66 M formaldehyde, and 50 % ( vol/vol) formamide, denatu
red at 65°C for 5 min, and electrophoresed (30-45 V, 6-16 h) in 0.8%
agarose gels (with the same running buffer minus formamide). Following
photography the gel was blotted overnight onto nylon membranes. Cross-
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
linking, prehybridization, hybridization, washing, and exposure were as for
DNA except that the hybridization buffer contained 30% (vol/vol) formamide.
Probes. [y-3 2 P]ATP-labeled repeat-free mouse Wnt1 cDNA (3.3 Kb),
WntlOb cDNA (2.0 Kb), Fgf3 cDNA (2.1 Kb), Fgfr2 TK cDNA (2.0 kb,
EcoRl), a unique 1.9 kb Pstl-Xhol fragment (MMTV gag probe), the MMTV
LTR fragment (1.4 Kb, Pstl-Pstl, from the pLIVEH plasmid), and rat
glyceraldehyde-3-phosphate dehydrogenase (Gapdh) cDNA probes were
prepared from plasmids as previously described (MacArthur and Shankar
1995, Lee 1995).
MMTV infection. Wnt10b/Fgfr2DN virgin female mice and control cohorts
(3-4 weeks old) were injected i.p. with 107 live XC/EH-supF9 cells and/or
Mm5MT live cells (ATCC, CRL-1637, lot number 204115). The rat XC/EH-
supFQ cell line produces a pathogenic hybrid MMTV consisting primarily of
MMTV(C3H) sequences. The Mm5MT mouse cell line produces wild-type
pathogenic MMTV(C3H). Wnt1/Fgfr2DN and Wnt1/Fgf3 virgin female mice
received one i.p. and ten s.c. injections of C3H-MMTV viral stock (1.57X 10s
viral particles/[xl). All females in each cohort were continuously bred
following MMTV-injection to enhance viral infection.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Stimulation of pregnancy/lactation cycles and collection of tum or
specimens. Female mice were continuously bred to males in order to
stimulate mammary proliferation and spreading of MMTV infection. After
delivery, females were allowed to nurse for 7-10 days and pups were
sacrificed (CO2 euthanasia) at that time. The females were inspected
weekly, visually and by palpation, for development of mammary tumors.
Tumors, when found, were allowed to reach 3 cm in diameter and then
surgically resected under anesthesia (Isoflurane). Samples from tumors and
other tissues were quickly frozen (liquid nitrogen) and stored at -80°C until
further use. Tumor samples were also fixed and preserved in 10% formaline
for hystological analysis, as well as partially minced and cryopreserved
(liquid nitrogen) in DMEM+20%FBS+10%DMSO medium (1.5 ml) in gas-tight
vials.
RESULTS and DISCUSSION
a) Generation and screening of MMTV-infected Wnt10b/Fgfr2DN,
Wnt1/Fgfr2DN and Wnt1/Fgf3 transgenic mice. Previously generated
Fgfr2DN transgenic mice (see Chapter 3) were bred to WntlOb transgenic
mice (C57BI6xSJL/J) that had been created in our laboratory or to a
purchased Wnt1 male mouse (The Jackson Lab, B6SJL[t4/foff]HeV). The
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
purchased Wnt1 transgenic was also mated to two Fgf3 transgenic females
mice (inbred FVB/N; TG.NR line) kindly provided by Dr. Philip Leder. The
WntlOb mice generated in our lab express a transgene that contains the
mouse WntlOb gene driven by the Wnt-1 promoter and an MMTV LTR-
enhancer. The Wnt1 mice express a transgene containing the mouse Wnt1
gene driven by the the Wnt-1 promoter and an MMTV LTR-enhancer. The
Fgf3 mice carry a transgene with the murine wild-type Fgf3 cDNA under the
control of a truncated MMTV LTR promoter (lacking sequence 5’ to the C/al
site) and with SV40 transcriptional processing signals (Muller 1990)
Offspring from these matings were screened by Southern blot analysis of
genomic tail DNA (Fig. 13, 14). The presence of the WntlOb transgene was
detemined following restriction enzyme digest with BamHI restriction enzyme
digest and blot probing with a WntlOb cDNA fragment. Similarly, Wnt1
transgenics were screened with BamHI and a Wnt1 cDNA probe. Fgf3
transgenics were detected upon Hindlll digest and probing with Fgf3 cDNA.
Fgfr2DN transgenic were screened as previously described (Chapter 3). The
expected sizes of transgene-specific bands were: WntlOb, 2.5 Kb+4.5 Kb
bands; Wnt1, 4.0 Kb+2.3 Kb bands; Fgf3, 1.8 Kb band; Fgfr2DN, 2Kb band.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13 14 15 16 17 18 19 20 B B+P T G # 13 14 15 16 17 18 19 20 B B+P
mm mmm
W w w IP
BamHI Digest
WntlOb probe
Figure 13. Southern blot screening of W nt10b/Fgfr2DN bitransgenic mice.
Left panel: WntlOb transgenic mice display 2.5 and 4.5 Kb bands. Right panel:
Fgfr2D N transgenic mice display a 2.0 Kb band. Numbers on top indicate
transgenic mouse identification number. Molecular weight markers are displayed
between the panels
-
_ 5 Kb _
_ 4 Kb _
#
. J
: '”: v ...I* ' ■ . • ■ ■ ■ ■ ■ : ■ t
_ 3 Kb _
.... IN * ,. *
_ 2 Kb _
mmmmm m
Hindlll Digest
Fgjfr2 DN probe
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A)
4.0 K b,
m m m w ® * ■
Wnt1 probe
« - - v- BamHI digest
2.3 K b ►
177 178 179 180 181 182 183 184 185 188 187
Fgfr2 probe
Hindlll digest
2.0 Kb~— ►
B)
4.0 K b — ► I M I l M l W nt1 probe
iiiM i m m H M t # 4WI BamHI digest
2.3 Kb— ►
135 136 137 138 139 140 141 142 143 144 145
Fgf3 probe
Hindill digest
1.8 Kb— -*■>
Figure 14. Representative Southern blot screening of W nt1/Fgfr2DN and
Wnt1/Fgf3 mice. Tail DNAs were digested with BamHI or Hind\\\ and probed with
Wnt1, Fgfr2 or Fgf3 cDNA respectively. Wnt1 transgenics display 2.3 Kb and 4.0 K b
bands. Fgfr2DN transgenics display a 2.0 K b band. Fgf3 transgenics display a 1.8
Kb band. W nt1/Fgfr2D N transgenics (#’s 183, 184, 186, 187) and Wnt1/Fgf3
bitransgenics (#’s 143, 145) display all characteristic bands.
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
For each mouse model study, 3 female cohorts (n=40-50 each) were
generated: 1) Wnt/Fgf or Wnt/Fgfr2DN, 2) Wnt, and 3) Fgf or Fgfr2DN, (Fig.
12). A total of 313 offspring from Wnt10b/Fgfr2DN matings, 308 offspring
from Wnt1/Fgfr2DN, and 260 offspring from Wnt1/Fgf3 matings were
screened in this manner.
MMTV-insertional mutagenesis in Wnt10b/Fgfr2DN bitransgenic mice.
We infected half of the females from each experimental cohort with MMTV at
3-4 weeks of age, while the other half remained uninfected. XC rat sarcoma
cells, producing a hybrid MMTV provirus, composed of the 5’-half from Mtv1
and a 3’-half from C3H MMTV, and carrying a bacterial supF gene (Jiang
1999), were initially used for the infection of the Wnt10/Fgfr2DN females and
corresponding single transgenic controls. Stably transfected rat XC cell
clones, that produce different hybrid MMTV variants, were previously
generated in Dr. Shackleford’s lab. Prior to injection, the levels of MMTV
expression in these clones were characterized by northern blot analysis of
their total RNA (20 fxg) after growth in the presence of dexamethasone, a
glucocorticoid that stimulates MMTV expression. The expression level of
MMTV in the EH-supFQ clone was considered suitable for our experiment,
when compared to the expression level of wild-type MMTV(C3H)-expressing
XC clones. The EH-supF9 clone was further expanded and used for MMTV
infection {via intraperitoneal injection) of our mice.
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
During a seven month-long period following infection with MMTV-EH supFB,
only two Wnt10/Fgfr2DN females developed mammary tumors, and only one
(# 136) had new MMTV integrations. This represents a much lower incidence
than we initially expected. Unrelated experiments done by others in my
mentor's lab suggest that our WntlOb mice may be losing transgene
expression over time. This problem may be contributing to the low tumor
incidence and prolonged latency, that we have observed in our bitransgenic
animals. In addition, we have also considered the possibility that our mice
may not be efficiently infected with the retrovirus, due perhaps to some
unknown immune-compatibility problems between host and virus that are
associated with IP injection and the dependence of MMTV infection on B and
T-cell activation and expansion. In order to circumvent this unforeseen
potential problem, our bitransgenic mouse cohort was re-injected with the
wild-type MMTV(C3H)-producing MmSMT mouse cell line. This wild-type
virus has been reported to strongly infect and induce mammary tumor
formation in mouse strains similar to ours.
Approximately between one to two months after the new infection, 50% of
the females developed very large mammary tumors always around the
injection area. Some of these animals also developed massive intra
abdominal tumor masses, and two-thirds of all the injected animals
eventually succumbed to them. The very short latency and the locations of
the new tu-
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mors suggested that the carrier Mm5MT cells may have not been rejected by
the infected animals and that the tumors may be comprised of MmSMT cells.
Southern blot analysis of tumor DNAs from these mice confirmed that some
of them were of MmSMT origin as shown by the similar numbers and band
pattern of proviruses and MmSMT cells (Fig. 15 and data not shown). We
therefore considered that all the tumors which quickly arose after the second
round of infection were not good candidates for our retroviral insertional
mutagenesis studies.
Out of the remaining surviving Wnt10b/Fgfr2DN females (n=9), five of them
eventually developed a total of ten independent mammary tumors with
varying postinfection latencies. Southern blot analysis demonstrated the
presence of new proviral integrations in eight of these tumors (Fig. 16). Half
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49 258 292 136 241* 126. 206*206. 129>k
i
Figure 15. S o u th e rn blot analysis of tumor origin in Wnt10h/Fgfr2DN
females infected with XC cells (MMTV EH-supF9) versus MmSMT cells
(MMTV C3H). DNAs from tumors arising from Mm5MT cells display a band
smear corresponding to hundreds of proviruses present in these cells. A :
abdominal tumor. M : mammary tumor. L : liver. MG: mammary gland. EcoRI
digest, MMTV-Env probe. Molecular weight markers are displayed on the left.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
136 266iaii 64l 5 64 Lt 74 ai 74 as
4.4 KB
2.0 KB
Figure 16. New proviral integrations in mammary tumors from
MMTV-infected W nt10b/Fgfr2D N m ice. Tumor DNAs (num ber
indicated on top of each lane) were digested with BglW and blots
were probed with an MM TV Bglll-Bglll 4.2 Kb probe. A 2.0 Kb band
corresponds to endogenous proviruses, while a 4.4 Kb band
indicated the presence of newly integrated proviruses. 266ta n :
negative control (tail DNA from uninfected Wnt10b/Fgfr2DN
female).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of these tumors (#’s 64|_i, 64ls, 74ri, 136, 72) contained new hybrid fVSMTV
proviruses, while the other half (#’s 74L 5 , 207L 1 + L 2 , 207R 1 , 207r2, 207R 3 )
contained both hybrid and C3H proviruses. Macroscopically, the tumors
were solid, well encapsulated masses. The histopathology of these tumors
was predominantly of the papillary and lobular mammary adenocarcinoma
type (Fig. 17, 22). No distant metastases were observed in the affected
animals.
In order to identify rearranged viral/cellular junction DNA fragments that could
be used for cloning and further study, we analyzed the 10 tumors DNAs by
Southern blot, after Xho\ restriction enzyme digestion, using an MMTV-Gag
probe. As shown in Figure 18, multiple rearranged fragments could be
detected in all Wnt10/Fgfr2DN mammary tumor DNAs.
MMTV-insertional mutagenesis in Wnt1/Fgfr2DN and Wnt1/Fgf3
bitransgenic mice. Half of the females in the Wnt1/Fgfr2DN and Wnt1/Fgf3
mouse cohorts, as well as the corresponding single transgenic control
females, were injected with wild-type MMTV(C3H). This viral strain was
chosen for injection, since it is highly pathogenic in our mouse strain
(Andervont 1964) with the hope to avoid any immune incompatibility-related
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 17. Histopathology of W nt10b/F gfr2D N mammary tumors. The
figure shows abnormal proliferation of ductal, lobular, and connective stroma
invading the normal mammary adipose parenchyme (top panels: 100X
magnification, bottom panels: 200X magnification; Hematoxylin-Eosine stain)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Tumor I 64u 74l5 74l5 748 1 72 135 207 , 2072 2073 2074 187, 1872 219188, 188, 1863 8 2 F U
12 K B _
8 KB—
7 KB —
6 KB—
5 KB—
EcoRI
digest
T u m o r I §4, 74 7 4 l 5 ? 4, 1 3 5 2 0 7 , S ? 2 2 0 7 3 2 0 7 4 1 8 7 , 1 8 7 a 219185,184184 8 2 R *
12 KB —
8 KB
6 KB _
4 KB —
2 KB—
1
1 KB —
0.5 KB —
Xhol
digest
4 - - -
Figure 18. New MMTV proviral insertions in mammary tumors from MMTV-
infected transgenic mice. The panel shows the Southern blot analysis of genomic
DNAs isolated from mammary tumors of MMTV-infected Wnt10h/Fgfr2DN (# 64L i to
2074 J, Wnt1/Fgfr2DN (# 187, to 1863 j, and Wnt1/Fgf3 transgenic mice (# 82 R4). A ll
DNAs display large MMTV fragments, indicative of endogenous retroviruses in these
mice. Most of the tumor samples have additional rearranged fragment(s), indicating
the presence of clonal, tumor-specific newly integrated MMTV proviruses. DNAs were
digested with EcoRI (top) or Xhol (bottom),
as a probe.
32
P-labeled MMTV gag cDNA was used
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
problems, as well as to increase the number of mammary tumors produced
in our animals.
Currently, 16 of 19 MMTV-infected Wnt1/Fgfr2DN females have developed
mammary tumors, with a median latency of 5.4 months, and tumor load
values between one and four tumors per animal (Fig. 19). Median tumor
latencies for the control groups were as follows: uninfected Wnt1/Fgfr2DN,
3.6 months; infected Wnt1, 4.8 months; uninfected Wnt1, 3.9 months;
infected Fgfr2DN, 9.8 months; uninfected Fgfr2DN, no tumors developed.
Southern blot analysis of mammary tumor DNAs revealed the presence of
new pro viruses in four of the tumors (Fig. 20). Rearranged cellular-viral
junction fragments could be detected as well in these tumors by Southern
blotting (Fig. 18).
Upon necropsy examination, mammary tumors varied from a solid, well
encapsulated mass to well defined cystic and very necrotic multilobulated
tumors. Lung metastases (Fig. 21, 23) were macroscopically manifest in 5
animals, and hepato-splenomegaly was observed in 8 animals (Fig. 21).
Histologically, mammary tumors were invasive papillary carcinomas (Fig. 22).
In the MMTV-infected Wnt1/Fgf3 group, hyperplastic mammary glands could
be observed in all animals approximately one month after MMTV infection.
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
< D
a )
L L
0
E
3
®
U
1
£
100
90
80
70
60
50
40
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12
Age (months)
-MMTV-Wnt1/Fgfr2DN
TG (n=19)
-Wnt1/Fgfr2DN TG (n7)
MMTV-Wnt1 TG (n20)
Wnt1 TG (n20)
■MMTV-Fgfr2DN TG
(n20)
■Fgfr2DN TG (n20)
Figure 19. Incidence of mammary tumors in W nt1/Fgfr2DN female mice
and control cohorts. The percentage of animals in each cohort remaining
free of palpable tumors was plotted at monthly intervals as a function of age.
The number of animals in each group are indicated in parentheses.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
187 187 166 193 193b 204b 219 186 186 186 186 186
L2+3 L4 L3 R3 L2 L2 R2+3 L4 R4 R5 R2+3 L2
4,4 KB -
2.o kb - - - - - * + - m mm
Figure 20. New proviral integrations in mammary tumors from MMTV-
infected W nt1/Fgfr2DN females. Tumor DNAs (number indicated on top of
each lane) were digested with BglI I and blots were probed with an MMTV Bglll-
Bglll 4.2 Kb probe. A 2.0 Kb band corresponds to endogenous proviruses, while
a 4.4 Kb band indicated the presence of newly integrated proviruses.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
B)
Figure 21. Macroscopic necropsy findings in MMTV-infected Wnt1/Fgfr2DN transgenic females. A) Mammary tumor (L1+ L .2
location). B) Top: normal lungs. Bottom: massive bilateral lung metastases. C) Splenomegaly.
v o
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
•» ** '* £ r .%
4 ? ''& \ * > r/e- r-r* . -V • *’ - ■ • $*%-* ~
'B m $ m B pss®
§ m
= _ = __
Figure 22. Mammary tumor histopathology in MMTV-infected W nt10b/Fgfr2DN and W nt1/Fgfr2DN females. Hematoxylin-
Eosine stain. A) Papillary mammary carcinoma, probably invasive. B) Solid type papillary mammary carcinoma, invasive. C)
Ductal mammary carcinoma, invasive. D) Tubular mammary carcinoma. E) Lobular carcinoma, alveolar type, invasive. F)
Metaplastic carcinoma, spindle type. A, C, E, F : Wnt10b/Fgfr2DN tumors. B, D : Wnt1/Fgfr2DN tumors.
v o
C T \
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Figure 23. Lung metastases (papillary mammary carcinoma) from W nt1/Fgf3 mouse mammary tumors. Hematoxylin-Eosine
stain. A, B, and C each correspond to a different bitransgenic animal. T : Mammary tumor metastasis. P: Pulmonary parenchyme.
Solid arrow: Tumor/lung parenchyme boundary.
'O
The hyperplasia was pregnancy-dependent in the majority of cases and was
also observed in uninfected Wnt1/Fgf3 female controls. Seventeen out of
nineteen MMTV-infected bitransgenic females have developed mammary
tumors to date. The median tumor latency is 3 months, and tumor load
values range from 1 to 6 independent tumors per animal (Fig. 24). Data for
uninfected bitransgenic controls is not available at present. Median tumor
latency values for other control groups were as follows: infected Wnt1, 4.8
months; uninfected Wnt1, 3.9 months; infected Fgf3, 6.7 months; uninfected
Fgf3, 9.2 months.
In comparison to the mammary tumors observed in MMTV-infected
W nt1/Fgfr2DN females, necropsy examination of Wnt1/Fgf3 females showed
all glands to be affected to different degrees, varying from focal or diffuse
hyperplastic nodules to generalized mammary tumors. Common necropsy
findings were lung metastases and hepato-splenomegaly (Fig. 25).
Southern blot analysis of Wnt1/Fgf3 mammary tumor DNAs from MMTV
infected females was negative for the presence of newly integrated
proviruses as well as for rearranged viral-cellular junction fragments. This
finding suggests that the tumors developed so far in these animals are the
result of the cooperative oncogenic effect of the Wnt1 and Fgf3 rather than a
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
< D
O
E
3
H
£
00
90
80
70
60
50
40
30
20
10
0
1 2 3 4 5 6 7 8 9 1011 12 13 14
MMTV-Wnt1 /Fgf3
TG (20)
-®—MMTV-Wntl TG
(20)
-a—Wn tl TG (20)
■ MMTV-Fgf3 TG
(21)
■Fgf3 TG(20)
Age (mo)
Figure 24. Incidence of mammary tumors in W nt1/Fgf3 female mice and
control cohorts. The percentage of animals in each cohort remaining free of
palpable tumors was plotted at monthly intervals as a function of age. The number
of animals in each group are indicated in parentheses.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A )
Figure 25. Representative necropsy findings in MMTV-infected W nt1/Fgf3
females. A) Generalized mammary tumors & magnification (right). B) Lung
metastases (solid arrowhead). C) Splenomegaly (cm scale)
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
consequence of mammary tumor acceleration due to MMTV-induced
insertional activation of additional oncogenic collaborators.
DISCUSSION and SUMMARY
The discovery, more than twenty years ago, of Wnt1-th e first proto
oncogene directly implicated in mouse mammary tumorigenenesis-as a
preferential target for MMTV proviral tagging insertions, opened the door to
studies in the field of retroviral insertional mutagenesis and oncogenic
cooperation in murine breast cancer. Since then, it has become clear that
the collaborative/synergistic effects of specific genetic mutations play a key
role in mammary tumorigenesis and progression. Among these genetic
events, it is now well established that mutations leading to the activation and
cooperation of Fgf and W nt signaling are key early events involved in the
process.
In order to identify oncogenes other than Fgfs that cooperate with Wnt genes
in multistep mammary tumorigenesis, we have created three new
bitransgenic mouse models of breast cancer: Wnt10h/Fgfr2DN,
Wnt1/Fgfr2DN, and Wnt1/Fgf3 transgenic mice. In all three, either through
concom ittant overexpression of transgenic Wnt and Fgf signals (Wnt1/Fgf3
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
model), or through activation of the former and blocking of the latter
(Wnt10b/Fgfr2DN and Wnt1/Fgfr2DN models), MMTV-mediated insertional
mutagenesis of genes other than Writs and Fgfs should give rise to the
formation of clonal tumors with an accelerated latency.
Our Wnt10b/Fgfr2DN were initially infected with a hybrid MMTV provirus,
which, despite its proven mammary tumorigenicity, did not lead to the
appearance of hybrid MMTV-infected mammary tumors but in 1/25 of the
females infected. Efficient MMTV infection and spreading to the mammary
gland depends on the immune compatibility between the MMTV strain and
the host strain (specific lymphoid MHC II receptor type). In fact, certain
inbred mouse strains (e.g., C57BL/6, l/LnJ) are highly resistant to MMTV
infection due to MHC II incompatibility. The Wnt10b/Fgfr2DN mice have a
mixed C57BI6, SJL/J, and BALB/cByJ background. It is thus possible that
the expression of an unknown genetic modifier in this mixed strain may lead
to resistance to infection by the hybrid MMTV strain. In order to circumvent
this potential caveat, the bitransgenic females were reinfected with mouse
wild-type MMTV(C3H)-producing Mm5MT cells. This viral strain is highly
pathogenic in a wide variety of mouse strains. To our surprise, some of the
mice developed very aggressive and disseminated tumors with a very short
latency after infection (< 2 mo). The latency and location of these tumors
suggested another potential problem, that they were arising from non-
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rejected Mm5MT themselves and not from the MMTV-infected mammary
epithelium. Our tumor analysis confirmed that indeed this was the case, for
some tumors rendering those tumor samples invalid for further study.
However, some mice in the cohort did not develop such rapidly growing
tumors, and in these mice 10 independent mammary tumors arose several
months later, all containing new MMTV insertions (five containing hybrid
provirus, and five with new MMTV(C3H)+hybrid proviral integrations).
Southern blot analysis of the tumor DNAs revealed the existence of multiple
rearranged viral-cellular junction fragments in all of the tumors. These
fragments are candidates harboring potential insertionally activated proto
oncogenes, and are currently being screened using various long inverse
PGR approaches.
Due to the dual time of infection and other mentioned problems, a realistic
estimation of tumor incidence and latency was not possible in this
bitransgenic model. The pathology report on the mammary tumors produced
in these animals revealed a variety of tumor types: papillary lobular, ductal,
and metaplastic invasive carcinomas. The implications of the tumor type
tumor heterogeneity in terms of its relation to activation of specific proto
oncogene types remain unclear at this moment and should require further
study.
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in order to optimize MMTV-infection efficiency while avoiding the problems
that we encountered in the Wnt10b/Fgfr2DN project, our Wnt1/Fgfr2DN and
Wnt1/Fgf3 mice were infected with wild-type MMTV(C3H) virus directly. All
MMTV-infected Wnt1/Fgfr2DN females developed mammary tumors, most of
them corresponding to papillary carcinomas. Despite the high tumor
incidence, only four tumors contained newly integrated proviruses. The
observed tumor latency in these animals was 5.4 months. Surprisingly, this
constitutes a higher value than that observed in the uninfected Wnt1/Fgfr2DN
control groups (3.6 months). The difference in sample size between both
groups (19 vs. 7, respectively) is significant at the moment, and the latency
value for the controls may not be truly representative. Another possibility that
may explain the unexpected delay in tumor formation in the infected
bitransgenics is that the MMTV-infection status may modify the action of
certain effectors involved in the control of mammary proliferation. With this
respect, for example, it has been reported that MMTV-infected mice are more
susceptible to mammary proliferation stimulation by progesterone, whereas
those without the virus respond better to beta-estradiol (Lee 1983). The
proliferative response of the mouse mammary gland to ovarian hormones,
and perhaps other factors, can thus be modified by mammary tumor virus
infection, and may affect the incidence of mammary tumor formation in our
model.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MMTV-infected Wnt1/Fgf3 females developed generalized metastatic
mammary papillary carcinomas with high incidence and an accelerated
median latency of 3 months compared to the control groups (infected Wnt1,
4.8 months; uninfected Wnt1, 3.9 months; infected Fgf3, 6.7 months;
uninfected Fgf3, 9.2 months). Data for uninfected bitransgenic controls are
not available at present. It is hence not possible to determine if MMTV-
infection results in significant tumor latency shortening, which may reflect
additional oncogenic activations in these tumors. We have not been able to
detect any new proviral integrations in any of the tumors developed in the
infected bitransgenics so far. Thus we suggest that strong expression and
oncogenic cooperation between the Wnt1 and Fgf3 transgenes is leading to
the generalized mammary neoplasia well before the effects of MMTV-
induced oncogenic activations may be apparent. It is thus possible that this
model may not be useful for our study purpose, but this conclusion must wait
until tumor kinetics data from the uninfected bitransgenic females are
available.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTERS. A HEW RAT MODEL FOR SlAliWARY ONCOGENE
DISCOVERY
EXPERIMENTAL RATIONALE and GOALS
V\fe have developed a novel model for mammary oncogene discovery based on infection
of tie rat with MMTV, in which we will attempt to unveil new genes that, in addition to
those identified in MMTV-infected mice, are implicated in the genesis of breast cartoer.
Since tie strain of mouse is known to be a strong determinant of which cellular
oncogenes are activated by MMTV insertions, we expert to find a new repertoire of target
genes involved in mammary tumorigenesis in the rat V M th this purpose, rats will be
infected with MMTV, in the hope hope that it will induce mammary tumors, which will be
analyzed for new clonal MMTV insertions and for insertionally activated oncogenes.
The rat is a species that has never been experimentally infected with MMTV. The
question arises then of this projects feasibility. Nevertheless, there is strong indirect
evidence in the literature in support of the rat susceptibility to MMTV infection in vivo.
First, rat cells are susceptible to MMTV infection in vitro, and the MMTV promoter is
functional in this system, as well as in transgenic rate (Bueffi 1981, Salmons 1985,
Shackleford 1988, Ringold 1983, Stallcup 1983, Davies 1999). Second, MMTV
infection of and expression in T and B lymphoid cells is required for its infectious cyde and
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
its spreading to tie mammary gland. In this respect, rat T-cels responding to MMTV
superantigens, indicating that this Symphoicktependent infection is intact in this species.
Third, MMTV is an ecotropic vims that can infect various species, and other murine
ecotropic viruses (i.e., MLV) can in fa t rats in vivo and are tumorigenic via insertional
mutagenesis (Tsichlis 1983, Vijaya 1987, Bear 1989, Lazo 1990, Shin 1993, Villeneuve
1993). Fourth, rate do not have endogenous MMTV prom ises nor express MMTV
envelope proteins, which could otherwise interfere with the binding between MMTV and
its cellular receptor (Ringold 1977). Finally, there is no data to date contradicting the
hypothesis that rate are not susceptible to in vivo MMTV infection.
The development of this new rat model for retroviral insertional mutagenesis offers
therefore much potential for mammary oncogene discovery. Our initial specific goals are:
1- Demonstration of rat mammary gland susceptibility to MMTV-infection.
- Three breast cancer-susoeptible rat strains will be infected with wiki-type
MMTV-C3H. In order to overcome any potential tropism features of this virus
we will deliver high vims stock dosed both intraperitoneally, and via
mammary intraductal injection.
2- Demonstration o f ra t mammary tumorigenesis by MMTV.
The presence of new donal MMTV proviral insertions in tumor DNAs
tom the infated animals will be tested by Southern blot analysis.
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MATERIALS and METHODS
A- D em oretation of rat susceptibility to SMMT¥ infection
a) MMTV footpad injection and determination of lymph node readMtv. Ten week-
old naive female rats (n=3/strain) from tie Sprague-Dawfey, F344 Lewis, and \Afetar~
Furth inbred strains were injected in the right and left hind footpad with MMTV-C3H viral
stock (1|xl, 5X108 viral particles). Three additional animals from each strain remained non-
injecfced as negative controls. Six days post-injection, the popliteal and inguinal draining
lymph nodes were removed bilaterally, under surgical anesthesia. Lymph node images
were captured and measured digitally under stereoscopic magnification (2.5X). Lymph
nodes were subsequently snap frozen in liquid nitrogen until further manipulation.
b) PCR analysis of lymph node MMTV infection. Lymph node DNAs from each of
the MMTV-injected and control rats were extracted as previously described. PCR was
then performed to amplify an MMTV LTR-spedfic fragment (307 bp) under the following
conditions: 95°C, 15min; 94°C, 1 min, 50°C, 1 min, 72°C, 1 min, for 35 cycles; 72°C, 1 0
min; cooling at 4°C. The MMTV LTR-spedfic primers used were: forward: 5’ -
ATAGGAGACAGGTGGTGGCAAC-3’; reverse: S ’-CAC-TGTCCCCTCCTTGGTAT-
GG-3’. PCR products were visualized after gel electrophoresis (80 volts, 40 min) in a
1.7% Agarose-TBE gel. Gels were blotted overnight onto a nylon membrane (Hybond-
XL, Amersham-Pharrnada) in 10XSSC buffer. The blots were UV cross-linked, and
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hybridized (65°C, overnight) to an MMTV LTR (PP 14) radiolabeled probe. The blots
were then washed (3X, 30 min), and exposed to film (Kodak X-Omat MR) for auto
radiography (-80 °C overnight).
B- MitffTV insertional mutagenesis In the raft
a) Rat MMTV infection and stimulation o f mammary gland prdiferation. Intradudal
mammary injections were performed as follows: unde" Isoiurane® anesthesia, the ventral
side of each rat was shaved to expose the 12 mammary nipples. Nipples were
individually clipped and 1% J of treatment solution were slowly infused into the central
mammary lacteal, using a 27-gauge blunt needle (Hamrnion)
Eleven Sprague Dawiey female rats were subjected to intraductal injection of one of tie
following treatments at day 0:
Rat #1) R1 and R3 mammary glands (MG): PBS-BSA (0.5 mg/ml)+ Polybrene
(hexadimethrine bromide, 80 fig/ml, Sigma-Aldrich)+ Fast Green (FG, 0.5 mg/ml, Sigma-
Aldrich). R4 and R6 MG: PBS-BSA (0.5 mg/mi)+Polybrene +Fast Green + MMTV-C3H
viral stock (tot: 34I-R-51/7-334, Meby Laboratories Inc.).
Rat #2) 1 1 and 15 MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green. R 1 and R4
MG: PBS-BSA (0.5 mg/ml)+Polybrene +Fast Green + MMTV-C3H viral stock.
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rat #3) L2 and L3 MG: PBS-BSA (0.5 mg/m!)+ Polybrene + Fast Green. R2 and R3
MG: PBS-BSA (0.5 mg/mi)+ Polybrene + Fast Green +MMTV-C3H viral stock+ human
recombinant keratinocyte growth factor (KGF, 5ng/|jl, PreproTech Inc.).
Rat #4) 1 1 and L6 MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green. R 1 and R6
MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green +MM7V-C3H viral stock + human
recombinant KGF (5ng/pl, PreproTech Inc.).
Rat#5i L 1 and L6 MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green. R 1 and R6 MG:
PBS-BSA (0.5 mg/ml)+ Pdybr@ne+ Fast Green +1/10 MM7V-C3H viral stock+ human
recombinant KGF (5ng/pl, PreproTech Inc.).
Rat #6) L 1 and 15 MG: PBS-BSA (0.5 mg/mQ+ Polybrene + Fast Green. R 1 and R5
MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green+MMTV-C3H viral stock+ human
recombinant KGF (5ng/pl, PreproTech fix:.).
Rat #7) L4 and L5 MG: PBS-BSA (0.5 mg/ml)+ Polybrene. R4 and R5 MG: PBS-BSA
(0.5 mg/mi)+ Polybrene + Fast Green +MMTV-C3H viral stock+ human recombinant
KGF (Sng/pl, PreproTech Inc.).
Rat #8) 1 1 and L4 MG: PBS-BSA (0.5 mg/ml)+ Polybrene. R 1 and R4 MG: R 1 and R6
MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green +MMTV-C3H viral stock.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rat #9) L 1 and L5 MG: PBS-BSA (0.5 mg/ml)+ Polybrene+ Fast Green. R 1 a id R5 MG:
PBS-BSA (0.5 mgfml)+ Polybrene + Fast Green +MMTV-C3H viral stock+ human
recombinant KGF (5ng/|jl, PreproTech Inc.).
Rat #10) L1 and L6 MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Gneen. R 1 and R6
MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green +MMTV-C3H viral stock.
Rat #11) 12 and 14 MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green. R2 and R4
MG: PBS-BSA (0.5 mg/ml)+ Polybrene + Fast Green +MMTV-C3H viral stock+ human
recombinant KGF (5ng/|J, PreproTech Inc.).
In addition, rats # 2, 3, 4 and 5 received subcutaneous injections of perphenazine (3
mg/Kg)+0.95 NaCI+0.02N HC! at days -2, -1, and day 0. Rat #1 received NaCi+HCI only
as a control. Rats # 3, 8, and 9 received subcutaneous injections of p-estradid (100 ng,
Sigma)+progesterone ( 1 mg, Sigma) on the same days plus at days 1 through 5. Rats #
10 and 1 1 received p-esiradioi (100 ng)+ hydrocortisone acetate ( 1 mg, Sigma) on the
same days.
b) Mammatv aland whole mount preparation. The intraductaly injected mammary
glands and their non-injected contralateral counterparts were resected at day 3 or day 6
after treatment. The tissue was spread on a glass slide, fixed in Camay's fixative (2-4 h at
room temperature), dehydrated in a reverse ethanol wash series, and stained in carmine
111
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
aluminum overnight. On the following day, the tissue was dehydrated in an ethanol
series, cleared in Xylene, and mounted on Pemnourrt lor photographic documentation.
c) PCR analysis of MMTV mammary aland infection. Mammary gland DNAs from
each of the treated rats were extracted as previously described (Shackleford 1993). PCR
was then performed to amplify an MMTV Enwspedfic fragment (499 bp) under the
following cycling conditions: 95°C, 15min; 94°C, 1 min, 52°C, 1 min, 72°C, 1 min, for 35
cycles; 72°C, 10 min; cooling at 4°C. The MMTV Env-specific primers used were:
forward: 5’-TGAATCTAGCCCCCATCAAAGAG-3’; reverse: 5’-TGAGTTCCCCAAAG-
TAGTCAACCAG-3’. As a positive PCR control, a rat pActiin gene fragment (484 bp)
was also amplified from the same DNAs in parallel reactions under the following
conditions: 95°C, 15 min; 94 °C, 1 min, 54°C, 1 min, 72°C, 1 min, for 25 cycles; 72°C, 10
min; cooling at 4°C. The rat p-Actin specific primers used were: forward: 5-AATGGAG-
CCCCTGTCCTGATA-CTC-3’; reverse: 5XTCTTTGATGTCACGCACGATTTCC-3’.
PCR products were eledrophoresed (80 volts, 40 min) in a 1.7% agarose gel, and the gel
was capillary blotted onto a nylon membrane (Hybond-XL, Amersham-Phairnada). The
blots were UV cross-linked and hybridized (65°C, overnight) to an MMTV-Env probe (BB
1.2) (as described in Chapter 3). The blots were then washed (3X, 30 min), and exposed
to film (Kodak, X-Ornat MR) with intensifying screens at -80°C overnight
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RESULTS
A- Demonstration of rat susceptibility to MMTV infection
a) Determination o f h/mph node m acM v and PCR analysis o f h/moh node MMTV
infection. It is generally accepted that, open infection of the host, MMTV spreads to the
mammary gland through the infection of EFIymphocytes, which mediate the presentation
of an MMTV-encoded super-antigen to T-ceils. These activated T-oels, in turn, stimulate
the expansbn of the MMTV-infected R-cell population and its subsequent migration to
other locations in the body. In the rats subjected to MMTV footpad injection, lymph node
reactive inflammation indirectly reflects MMTV infection and super-antigen function. V\fe
show that the draining popliteal and inguinal lymph nodes tom our three strains of
MMTV-injected rats displayed marked reactive inflammation compared to toe non
injected controls. The Wistar and Sprague Dawtey strains were the strongest responders
(Fig. 26). This finding suggests that these rat strains are indeed susceptible to MMTV
infection. Concordant with this finding, combined PCR and Southern blot analysis on the
lymph node DMA’s, detected the presence of a MMTV-spedfic PCR product only in
those DNAs corresponding to the reactive interned lymph nodes but not in the non
injected controls (Fig. 27). This result further confirmed the susceptibility to MMTV
infection of all three rat strains.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
B- M M W Insertional nntytaaeniesis in the rat m am m ary qiamri
a) Stimulation o f mammary aland odifemtion. Efficient cell infection by MMTV
depends on actively ongoing cel division. V\fe therefore reasoned fia t continuous
exogenous stimulation of the mammary gland proliferation should induce optimal
spreading of MMTV infection in the gland. V\fe have tested and compared various
exogenous treatments that induce mammary gland proliferation comparable to that
observed normally during pregnancy and lactation. Eleven Sprague Dawiey female rate
were subjected to intraductal injection of estradiol, progesterone, hydrocortisone acetate,
perphenazine, and/or KGF in various combinations, all of them known mammary
mitogens. The analysis of mammary whole mounts revealed that all treatments induced
a certain level of mammary proliferation in comparison compared to untreated animals
(Fig. 28). The levels of proliferation observed were nevertheless variable upon type of
treatment, with estradfol+hydrooortisone bang the most potent mammogenie
combination, and administration of KGF alone bang the least effective. In addition, we
observed that add- on of KGF to estradid+hydrocortisone or esfradd+progesterene
combinations resulted in a decrease of the mitogenic levels induced by the steroidal
combination alone.
b) . Rat mammary susceptibility to MMTV infection. The DNAs from fie treated
mammary glands were analyzed by PCR in Oder to assess the presence of new MMTV
proviruses, thus reflecting successful MMTV infection of the mammary gland. Figure 29
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
shows that MMW-speciftc fragments could only be detected in those glands that had
previously displayed higher lewis of proliferation upon exogenous treatment
(perphenazine or estrogen^ydrooortisone). This finding indicated that the mammary
gland is indeed susceptible to MMTV infection. However, infection depends on the
existence of adequate levels of mammary cell proliferation, which favors MMTV
spreading in the gland.
DI SCUSSI ON and SUMMARY
V\fe have developed a novel animal model for mammary oncogene discovery based on
the infection of the rat with MMTV. The rat is a species that has never before been
experimentally infected with MMTV. In our experiments, we first demonstrated the
susceptibility of three rat strains to MMTV infection. Following MMTV injection in the hind
footpad of the animals, the draining popliteal and inguinal lymph nodes displayed marked
reactive inflammation. Concordant with this finding, PCR analysis detected the presence
of an MMTV specific product only in those DMAs corresponding to the inflamed lymph
nodes.
More importantly, we have shown for the first time (by PCR analysis) that the rat
mammary gland is susceptible to MMTV infection upon in w o intraductal injection of the
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
M M T V
IN F E C T E D
UNINFECTED
CONTROLS
S P R A G U E DA W LEY
□□□□
FISCHER 344 W ISTAR
m *
□ □ □ D Q D O O D
!l)i|l!II IIIT|III
(ttTSHC j METRIC
Figure 26. Rat lymph node reactivity to MMTV subcutaneous injection. Three female rats from each Sprague Dawley,
F344, and Wistar (outbred) strain, were injected in the right and left hind footpads with MMTV (C3H) virus (top row). Three
additional animals from each strain were kept uninfected as negative controls (bottom row). Popliteal and inguinal lymph
nodes were resected six days after the injection. All lymph nodes from the infected rats displayed an evident inflammatory
reactivity, compared to the uninfected controls, with the Wistar and Sprague Dawley strains showing the strongest
response. The lymph node inflammatory reaction indirectly reflects MMTV infection and superantigen function
M M T V ip s ila te ra i fo o tp a d Injection
4 * » 4 ® 4 » J S B = 4 ® *st " I "
PCR
Nift,
Conlret
S prague
D a w tiy
m T • • • •
F344 (L e w s )
W is ta r
ropiM Ilfp M I rXtpSBBm ip ra r e p M S i
Figure 27. Rats are susceptible to MMTV infection. Three female rats from
each rat strain were injected in the right and left hind footpads with MM TV virus.
Three uninfected animals from each strain were used as negative controls.
Popliteal and inguinal lymph nodes were resected six days after injection.
Lymph node DNAs were analyzed by PCR with MMTV LTR-specific primers,
and shown with an MMTV LTR probe. The presence of an MMTV-specific band
in the lymph node DNAs corresponding to the injected side confirms the
susceptibility of all three rat strains to MMTV infection.
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
virus. H ie use of this retroviral delivery approach targets more specifically the organ of
interest, the mammary gland. In addition, it avoids the time delay and unforeseen
immune-oompatibility problems between host and virus that are associated with IP
injection due to the dependence of MMTV infection on B and T -ceH activation and
expansion. We have also oonfirmed that the proliferating mammary gland is amenable
to more efficient infection by MMTV, as expected due to the characteristics or the
retroviral life cycle. A high state of mammary proliferation is observed physiologically
during pregnancy and lactation, due to the existenoe of high estrogen, progesterone,
prolacfine, and placental lactogen among a myriad of hormones. Other substances can
act as direct or indirect mammary epithelium mitogens as well. Among these, it has
previously teen shown that keratinocyte growth factor (KGF/FGF7) is a potent direct
mitogen for tie mammary epithelium in wo. (Ulich 1994). Perphenazine, on the other
hand, indirectly stimulates mamogenesis and lactogenesis through the elevation of
endogenous serum prolactin. In our animal system, a combination of
estrogen+hydrocortisone acetate appeared to induoe the highest level of mammary
proliferation, comparable to that observed during pregnancy and lactation. Since the
MMTV LTR contains regulatory elements that respond to steroid hormone and growth
factor stimulation (Cato 1989), we expect that using such combination should induce
higher levels of MMTV expression, thereby facilitating further spreading of MMTV
infection as an additional advantage.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V\fe have demonstrated that the rat mammary gland can be infected by MMTV in mo,
and that intraductal injection is a viable method for viral delivery. Interestingly, higher
levels of MMTV proviruses can be detected (by PCR) in infected mammary glands from
animals treated with estradid+hydnocortisone, compared to untreated animals or to those
treated with other substance cxxmbinafoins. Combining exogenous mammary hormonal
stimulation together v w fo direct MMTV intraductal induction seams to be thus an ideal
approach that optimizes viral delivery to a gland in a proliferative state (required for MMTV
infection spreading) while precluding the need to subject the animals to uninterrupted
cycles of pregnancy/lactation, (with the time delay which cycling through these phases
implies).
V S fe suggest that this approach could be routinely used to effectively tavor MMTV
replicafon, to expend the pool of infected cells, and to increase the potential target
population for MMTV insertional mutagenesis in this new animal breast cancer model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
> -
<
a
KGF+
MMTV+Pb+D MMTV+Pb+D
K O F +
MMTV+Pb
f
» • * - , if. *
/ ■ . , 1
Perphen Perphen
♦ +KGF
MMTV+Pb+D MMTV+Pb+D
Perphen+ Perphen*
E2+Pr+ E2+Pr+KQF* E2+Pr+KGF+ E2+Pr+KGF+ E2*HC+ E2*HC+KGF+
MMTV+Pb+D MMTV+Pb+D MMTV+Pb+D 1f10 MMTV+Pb+D MMTV*Pb*D MMTV+Pb+D
Figure 28. Whole mount analysis of the effect of various treatments on the stimulation of mammary gland
proliferation. Eleven Sprague Dawley female rats were subjected to different treatments (middle panel), delivered via
subcutaneous or intraductal injection (bold). For those glands injected intraductally, the contralateral glands remained
non-injected, to serve as negative controls (top and bottom pannel: bottom picture rows). Whole mammary glands were
resected at day 3 (top pannel) or day 6 (bottom pannel) after intraductal injection. While subcutaneous injection of
perphenazine stimulated the mammary gland noticeably, treatment with estradiol+hydrocortisone acetate induced the
highest level of mammary proliferation, similar to that found physiologically during rat pregnancy and lactation. PBS:
Phosphate buffered saline. Pb: Polybrene. KGF: recombinant human keratinocyte growth factor. M M TV: C3H mouse
mammary tumor virus. D: Fast Green dye. Perphen: perphenazine. E2: p-estradiol. Pr: Progesterone. HC:
hydrocortisone acetate.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
w e tl O TiRM W NJICTW f 6»l«IR*JV«JeCTO
RW* 1 I 3 1 S 4 ? 8 8 ' IS 1 1 t i 3 4 S 8 » I i M 1t
f^W ITV # + 4 « + « # s + a + a + „ + H + ., + a + * , + ., + » + ® + s 4 = ® + b 4 =>4*4** 4 * 4 ® 4 ® ^ |||
KGF + +* + ♦+ + + * + ++ H 444.4+ + 4 4 4 4 4 * | | 1
Ptrphtnailrw * + + ♦+ + 4 * + + + + + + + # S i *f S i
Estradiol + * + + + + * + + + + + + + * + h h h h e g *
Progesterone + + + + * R f
Hydrocortisone n** h h
• a
s
^««^»*N.*5PS!!2?IP?i5:S58SSS8aa8l5Sa858858JSfe8S!t5m«?^?
MMTV LTft probe 1 • •
ACTIN probe
Figure 29. The rat mammary gland is susceptible to MMTV infection by intraductal injection. Eleven Sprague Dawley
female rats received intraductal MMTV injections in one of their mammary glands (odd numbers 1-47). The contralateral gland
remained uninjected as an internal negative control for each animal (even numbers 2-48). In addition, one rat (# 1) also received
PBS-only injection in separate glands (numbers 3 and 27), with the contralateral gland remaining uninjected (4), and both glands
to serve as negative controls for the experiment. The remaining ten rats received different combinations of agents (on top left, in
blue) to exogenously stimulate the proliferation of their mammary epithelial cells before and/or after MMTV injection. Whole
mammary glands were resected at days 3 and 6 post-MMTV injection. Mammary gland DNAs were analyzed by PCR with
M MTV LTR- specific primers and by Southern blot with MMTV LTR and Actin probes. MMTV-infected lymph node DNA was also
used as a positive control. MMTV LTR-specific products could be detected (both at day 3 and 6 post-injection) in those glands
that had received exogenous stimulation with perphenazine or with an estradiol+hydrocorisone acetate combination.
EPILOGUE
In order to identify activated proto-oncogenes that cooperate with Wnt genes
in mammary tumorigenesis, we have used MMTV as an insertional mutagen
in the mammary gland of various transgenic mouse models and a rat model.
The analysis of mammary tumors in MMTV-infected Wnt-1 mice revealed
that the Fgfr2 gene is a new common locus for MMTV integration.
Interestingly, this gene does not appear to be activated by MMTV through
classic enhancer or promoter insertion mechanisms but through a rare
polyadenylation insertion mechanism instead. We suggest that MMTV
integration in the C-terminus may result in the deletion of a C-terminal
negative regulatory domain and the generation of an activated truncated
receptor, thereby facilitating oncogenic Fgf signaling. Our finding adds onto
previous reports on the oncogenic role of Fgfr2. Furthermore, the activation
of this component in the Fgf signaling pathway underscores the importance
of Wnt/Fgf oncogenic cooperation in murine mammary tumorigenesis.
Three new bitransgenic mouse models have been developed
(Wn110b/Fgfr2DN, W nt1/Fgfr2D N and Wnt1/Fgf3 mice) and MMTV
insertional mutagenesis has been used in these animals. Actively breeding
MMTV-infected bitransgenic Wnt10b/Fgfr2DN and Wnt1/Fgfr2DN females
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
have developed mammary tumors containing new MMTV proviral
integrations. New proviral integrations couldn’t be detected in mammary
tumors from MMTV-infected Wnt1/Fgf3 mice. The animals developed
generalized early mammary neoplasia, likely due to the potent combined
oncogenic effect of the Wnt and Fgf signals overexpressed in the mammary
gland in these mice. Mammary tumors observed in the Wnt10b/Fgfr2DN
bitransgenics were pleomorphic, including mammary papillary, ductal,
lobuloalveolar, and metaplastic carcinomas. Wnt1/Fgfr2DN and Wnt1/Fgf3
animals predominantly developed mammary papillary carcinomas, and lung
metastases were frequently observed. The different histological and invasive
features displayed by these tumors, compared to those from
Wnt10h/Fgfr2DN mice, may reflect differences in the insertional activation of
specific proto-oncogenes in each bitransgenic mouse model. DNA’s from
these tumors are currently being analyzed, using various PCR-based
approaches, in order to isolate new activated oncogenes involved in mouse
mammary tumorigenesis.
In addition to the new transgenic mouse models, we have created a novel
animal model based in the infection of the rat with MMTV. The rat is a
species that is never been experimentally infected with this mammary
retrovirus. For the first time, we show that rats are susceptible to MMTV
infection. Furthermore, following exogenous stimulation of mammary
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
proliferation with a combination of steroid hormones, rat mammary tissue
was successfully infected by MMTV using direct intraductal injection. We
therefore suggest that the use of these optimized conditions improve the
efficiency of targeted mammary retroviral insertional mutagenesis studies,
and should further facilitate the identification of new mammary oncogenes
using this approach.
It is of crucial importance to achieve a deeper understanding of the
mechanisms that are involved in the genesis and progression of cancer. The
study of cancer genetics in humans presents serious limitations, thus
laboratory animal models that resemble specific aspects of human cancer
have been used. The development of animal models for oncogene discovery
can be a very instructive tool, but is often a lenghty and cumbersome
process. The work presented here has paved the way for the generation of
high number of mammary tumors which, in the near future, can be studied to
identify and clone new oncogenes involved in breast cancer development.
Several genetic mutations displayed in human breast cancer are used as
helpful markers for preventive, diagnostic, and prognostic purposes.
Knowing what genes are involved in breast cancer development and
progression may thus be directly translated into the design of more integral
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and effective therapies that will improve the survival and quality of life of
those affected by this and perhaps other malignancies.
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
[1-188]
1. ACS (2002). Cancer Facts and Figures.
2. Ago, H., Kitagawa, Y., Fujishima, A., Matsuura, Y., and Katsube, Y.
(1991). Crystal structure of basic fibroblast growth factor at 1.6 A
resolution. J Biochem (Tokyo) 110: 360-363.
3. Akasaka, H., et al. (2000). Molecular anatomy of BCL6 translocations
revealed by long-distance polymerase chain reaction-based assays.
Cancer Res 60: 2335-2341.
4. Altmann, C. R., and Brivanlou, A. H. (2001). Neural patterning in the
vertebrate embryo. Int Rev Cytol 203: 447-482.
5. Amundadottir, L. T., Merlino, G., and Dickson, R. B. (1996).
Transgenic mouse models of breast cancer. Breast Cancer Res Treat
39: 119-135.
6. Andres, A. C., et al. (1987). Ha-ras oncogene expression directed by a
milk protein gene promoter: tissue specificity, hormonal regulation,
and tumor induction in transgenic mice. Proc Natl Acad Sci U S A 84:
1299-1303.
7. Arman, E., Haffner-Krausz, R., Chen, Y., Heath, J. K., and Lonai, P.
(1998). Targeted disruption of fibroblast growth factor (FGF) receptor
2 suggests a role for FGF signaling in pregastrulation mammalian
development. Proc Natl Acad Sci U S A 95: 5082-5087.
8. Asa no, K., Merrick, W. C., and Hershey, J. W. (1997). The translation
initiation factor elF3-p48 subunit is encoded by int-6, a site of frequent
integration by the mouse mammary tumor virus genome. J Biol Chem
272: 23477-23480.
9. Auch, D., and Reth, M. (1990). Exon trap cloning: using PCR to
rapidly detect and clone exons from genomic DNA fragments. Nucleic
Acids Res 18; 6743-6744.
10. Auffray, C., and Rougeon, F. (1980). Purification of mouse
immunoglobulin heavy-chain messenger RNAs from total myeloma
tumor RNA. Eur J Biochem 107: 303-314.
11. Barnes, S., Peterson, G., Grubbs, C., and Setchell, K. (1994).
Potential role of dietary isoflavones in the prevention of cancer. Adv
Exp Med Biol 354: 135-147.
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12. Bartkova, J., et al. (1994). Cyclin D1 protein expression and function
in human breast cancer. Int J Cancer 57: 353-361.
13. Basilico, C., and Moscatelli, D. (1992). The FGF family of growth
factors and oncogenes. Adv Cancer Res 59: 115-165.
14. Bear, S. E., et al. (1989). Provirus insertion in Tpl-1, an Ets-1-related
oncogene, is associated with tumor progression in Moloney murine
leukemia virus-induced rat thymic lymphomas. Proc Natl Acad Sci U S
A 86: 7495-7499.
15. Bhanot, P., et al. (1996). A new member of the frizzled family from
Drosophila functions as a Wingless receptor. Nature 382: 225-230.
16. Bose, S., Wang, S. I., Terry, M. B., Hibshoosh, H., and Parsons, R.
(1998). Allelic loss of chromosome 10q23 is associated with tumor
progression in breast carcinomas. Oncogene 17: 123-127.
17. Brisken, C., et al. (2000). Essential function of Wnt-4 in mammary
gland development downstream of progesterone signaling. Genes
Dev 14: 650-654.
18. Brisken, C., et al. (1999). Prolactin controls mammary gland
development via direct and indirect mechanisms. Dev Biol 210: 96-
106.
19. Brisken, C., et al. (1998). A paracrine role for the epithelial
progesterone receptor in mammary gland development. Proc Natl
Acad Sci USA 95: 5076-5081.
20. Brodie, S. G., et al. (2001). Inactivation of p53 tumor suppressor gene
acts synergistically with c- neu oncogene in salivary gland
tumorigenesis. Oncogene 20: 1445-1454.
21. Brown, A. M. (2001). Wnt signaling in breast cancer: have we come
full circle? Breast Cancer Res 3: 351-355.
22. Buetti, E., and Diggelmann, H. (1981). Cloned mouse mammary tumor
virus DNA is biologically active in transfected mouse cells and its
expression is stimulated by glucocorticoid hormones. Cell 23: 335-
345.
23. Burke, W., et al. (1997). Recommendations for follow-up care of
individuals with an inherited predisposition to cancer. II. BRCA1 and
BRCA2. Cancer Genetics Studies Consortium. Jama 277: 997-1003.
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24. Cadigan, K. M., and Nusse, R. (1997). Wnt signaling: a common
theme in animal development. Genes Dev 11: 3286-3305.
25. Callahan, R. (1996). MMTV-induced mutations in mouse mammary
tumors: their potential relevance to human breast cancer. Breast
Cancer Res Treat 39: 33-44.
26. Campbell, S. L , Khosravi-Far, R., Rossman, K. L , Clark, G. J., and
Der, C. J. (1998). Increasing complexity of Ras signaling. Oncogene
17: 1395-1413.
27. Capdevila, J., and Izpisua Belmonte, J. C. (2001). Patterning
mechanisms controlling vertebrate limb development. Annu Rev Cell
Dev Biol 17: 87-132.
28. Carstens, R. P., Eaton, J. V., Krigman, H. R., Walther, P. J., and
Garcia-Bianco, M. A. (1997). Alternative splicing of fibroblast growth
factor receptor 2 (FGF-R2) in human prostate cancer. Oncogene 15:
3059-3065.
29. Cato, A. C., et al. (1989). The regulation of expression of mouse
mammary tumor virus DNA by steroid hormones and growth factors. J
Steroid Biochem 34: 139-143.
30. Celli, G., LaRochelle, W. J., Mackem, S., Sharp, R., and Merlino, G.
(1998). Soluble dominant-negative receptor uncovers essential roles
for fibroblast growth factors in multi-organ induction and patterning.
Embo J 17: 1642-1655.
31. Cressman, V. L., et al. (1999). Growth retardation, DNA repair defects,
and lack of spermatogenesis in BRCA1-deficient mice. Mol Cell Biol
19: 7061-7075.
32. Cross, M. J., and Claesson-Welsh, L. (2001). FGF and VEGF function
in angiogenesis: signalling pathways, biological responses and
therapeutic inhibition. Trends Pharmacol Sci 22: 201-207.
33. Cui, X. S., and Donehower, L. A. (2000). Differential gene expression
in mouse mammary adenocarcinomas in the presence and absence of
wild type p53. Oncogene 19: 5988-5996.
34. Dang, C. V. (1999). c-Myc target genes involved in cell growth,
apoptosis, and metabolism. Mol Cell Biol 19: 1-11.
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35. Dankort, D. L , and Muller, W. J. (2000). Signal transduction in
m a m m a ry tu m o rig en esis: a tran sg en ic p ers p e c tiv e . Oncogene 1 9 :
1038-1044.
36. Daphna-lken, D., et al. (1998). MMTV-Fgf8 transgenic mice develop
mammary and salivary gland neoplasia and ovarian stromal
hyperplasia. Oncogene 17: 2711-2717.
37. Davies, B. R., Platt-Higgins, A. M., Schmidt, G., and Rudland, P. S.
(1999). Development of hyperplasias, preneoplasias, and mammary
tumors in MMTV- c-erbB-2 and MMTV-TGFalpha transgenic rats. Am
J Pathol 155: 303-314.
38. Delli Bovi, P., et al. (1987). An oncogene isolated by transfection of
Kaposi's sarcoma DNA encodes a growth factor that is a member of
the FGF family. Cell 50: 729-737.
39. Deming, S. L, Nass, S. J., Dickson, R. B., and Trock, B. J. (2000). C-
myc amplification in breast cancer: a meta-analysis of its occurrence
and prognostic relevance. Br J Cancer 83: 1688-1695.
40. Deng, C. X., and Brodie, S. G. (2000). Roles of BRCA1 and its
interacting proteins. Bioessays 22: 728-737.
41. Deng, C. X., and Scott, F. (2000). Role of the tumor suppressor gene
Brcal in genetic stability and mammary gland tumor formation.
Oncogene 19: 1059-1064.
42. Di Cristofano, A., Pesce, B., Cordon-Cardo, C., and Pandolfi, P. P.
(1998). Pten is essential for embryonic development and tumour
suppression. Nat Genet 19: 348-355.
43. Dickson, C., and Fantl, V. (1994). Fgf-3, an oncogene in murine breast
cancer. Cancer Treat Res 71: 331-343.
44. Dickson, C., Smith, R., Brookes, S., and Peters, G. (1984).
Tumorigenesis by mouse mammary tumor virus: proviral activation of
a cellular gene in the common integration region int-2. Cell 37: 529-
536.
45. Dickson, C., Spencer-Dene, B., Dillon, C., and Fantl, V. (2000).
Tyrosine kinase signalling in breast cancer: fibroblast growth factors
and their receptors. Breast Cancer Res 2: 191-196.
46. Dierick, H., and Bejsovec, A. (1999). Cellular mechanisms of
wingless/Wnt signal transduction. Curr Top Dev Biol 43: 153-190.
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47. Dievart, A ., Beaulieu, N., and Jolicoeur, P. (1999). Involvement of
Notch 1 in the development of mouse mammary tumors. Oncogene
18: 5973-5981.
48. Donehower, L. A. (1996). Effects of p53 mutation on tumor
progression: recent insights from mouse tumor models. Biochim
Biophys Acta 1242: 171 -176.
49. Donehower, L. A., et al. (1995). Deficiency of p53 accelerates
mammary tumorigenesis in Wnt-1 transgenic mice and promotes
chromosomal instability. Genes Dev 9: 882-895.
50. Donehower, L. A., et al. (1992). Mice deficient for p53 are
developmentally normal but susceptible to spontaneous tumours.
Nature 356: 215-221.
51. Duplan, S. M., Theoret, Y., and Kenigsberg, R. L. (2002). Antitumor
Activity of Fibroblast Growth Factors (FGFs) for Medulloblastoma May
Correlate with FGF Receptor Expression and Tumor Variant. Clin
Cancer Res 8: 246-257.
52. Edwards, P. A. (1998). Control of the three-dimensional growth
pattern of mammary epithelium: role of genes of the Wnt and erbB
families studied using reconstituted epithelium. Biochem Soc Symp
63: 21-34.
53. Eng, C., and Peacocke, M. (1998). PTEN and inherited hamartoma-
cancer syndromes. Nat Genet 19: 223.
54. Fantl, V., Edwards, P. A., Steel, J. H., Vonderhaar, B. K., and Dickson,
C. (1999). Impaired mammary gland development in Cyl-1 (-/-) mice
during pregnancy and lactation is epithelial cell autonomous. Dev Biol
212: 1- 11 .
55. Feng, S., Wang, F., Matsubara, A., Kan, M., and McKeehan, W. L.
(1997). Fibroblast growth factor receptor 2 limits and receptor 1
accelerates tumorigenicity of prostate epithelial cells. Cancer Res 57:
5369-5378.
56. Frebourg, T., et al. (2001). [Li-Fraumeni syndrome: update, new data
and guidelines for clinical management]. Bull Cancer 88: 581-587.
57. Freedman, D. A., and Levine, A. J. (1999). Regulation of the p53
protein by the MDM2 oncoprotein— thirty-eighth G.H.A. Clowes
Memorial Award Lecture. Cancer Res 59: 1-7.
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58. Gallahan, D., and Callahan, R. (1997). The mouse mammary tumor
associated gene INT3 is a unique member of the NOTCH gene family
(NOTCH4). Oncogene 14: 1883-1890.
59. Gallego, M. I., et al. (2001). Prolactin, growth hormone, and epidermal
growth factor activate Stat5 in different compartments of mammary
tissue and exert different and overlapping developmental effects. Dev
Biol 229: 163-175.
60. Ghosh, A. K., et al. (1996). Molecular cloning and characterization of
human FGF8 alternative messenger RNA forms. Cell Growth Differ 7:
1425-1434.
61. Giaccia, A. J., and Kastan, M. B. (1998). The complexity of p53
modulation: emerging patterns from divergent signals. Genes Dev 12:
2973-2983.
62. Greenlee, R. T., Hill-Harmon, M. B., Murray, T., and Thun, M. (2001).
Cancer statistics, 2001. CA Cancer J Clin 51: 15-36.
63. Hamaguchi, M., et al. (1992). Establishment of a highly sensitive and
specific exon-trapping system. Proc Natl Acad Sci U S A 89: 9779-
9783.
64. Hansen, R. K., and Bissell, M. J. (2000). Tissue architecture and
breast cancer: the role of extracellular matrix and steroid hormones.
Endocr Relat Cancer 7: 95-113.
65. Harvey, M., et al. (1993). Spontaneous and carcinogen-induced
tumorigenesis in p53-deficient mice. Nat Genet 5: 225-229.
66. Heikinheimo, M., Lawshe, A., Shackleford, G. M., Wilson, D. B., and
MacArthur, C. A. (1994). Fgf-8 expression in the post-gastrulation
mouse suggests roles in the development of the face, limbs and
central nervous system. Mech Dev 48: 129-138.
67. Hennighausen, L , and Robinson, G. W. (2001). Signaling pathways in
mammary gland development. Dev Cell 1: 467-475.
68. Hilgers, J. (1979). Mammalian oncornaviruses: an introductory review.
Acta Microbiol Acad Sci Hung 26: 149-152.
69. Hogan, B. L. (1999). Morphogenesis. Cell 96: 225-233.
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70. Howe, H. L, et al. (2001). Annual report to the nation on the status of
cancer (1973 through 1998), featuring cancers with recent increasing
trends. J Natl Cancer Inst 93: 824-842.
71. Humphreys, R. C., and Hennighausen, L. (2000). Transforming growth
factor alpha and mouse models of human breast cancer. Oncogene
19: 1085-1091.
72. Hutchinson, J. N., and Muller, W. J. (2000). Transgenic mouse models
of human breast cancer. Oncogene 19: 6130-6137.
73. Ishiwata, T., Friess, H., Buchler, M. W., Lopez, M. E., and Korc, M.
(1998). Characterization of keratinocyte growth factor and receptor
expression in human pancreatic cancer. Am J Pathol 153: 213-222.
74. Jackson, D., Bresnick, J., and Dickson, C. (1997). A role for fibroblast
growth factor signaling in the lobuloalveolar development of the
mammary gland. J Mammary Gland Biol Neoplasia 2: 385-392.
75. Jackson, D., et al. (1997). Fibroblast growth factor receptor signalling
has a role in lobuloalveolar development of the mammary gland. J Cell
Sci 110: 1261-1268.
76. Jamerson, M. H., Johnson, M. D., and Dickson, R. B. (2000). Dual
regulation of proliferation and apoptosis: c-myc in bitransgenic murine
mammary tumor models. Oncogene 19: 1065-1071.
77. Jardines L, H. B., Doroshow JH, Fisher P, Weitzel J, and. Theriault RL
(2001). "Breast Cancer Overview: Risk Factors, Screening, Genetic
Testing, and Prevention.,".
78. Jiang, Z., and Shackleford, G. M. (1999). Mouse mammary tumor
virus carrying a bacterial supF gene has wild-type pathogenicity and
enables rapid isolation of proviral integration sites. J Virol 73: 9810-
9815.
79. Johnson, D. E., and Williams, L. T. (1993). Structural and functional
diversity in the FGF receptor multigene family. Adv Cancer Res 60: 1-
41.
80. Jonkers, J., and Berns, A. (1996). Retroviral insertional mutagenesis
as a strategy to identify cancer genes. Biochim Biophys Acta 1287:
29-57.
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81. Kapoun, A. M., and Shackleford, G. M. (1997). Preferential activation
of Fgf8 by proviral insertion in mammary tumors of Wnt1 transgenic
mice. Oncogene 14: 2985-2989.
82. Katz, M. E., and McCormick, F. (1997). Signal transduction from
multiple Ras effectors. Curr Opin Genet Dev 7: 75-79.
83. Korach, K. S., et al. (1996). Estrogen receptor gene disruption:
molecular characterization and experimental and clinical phenotypes.
Recent Prog Horm Res 51: 159-186.
84. Kornmann, M., Ishiwata, T., Beger, H. G., and Korc, M. (1997).
Fibroblast growth factor-5 stimulates mitogenic signaling and is
overexpressed in human pancreatic cancer: evidence for autocrine
and paracrine actions. Oncogene 15: 1417-1424.
85. Kwan, H., et al. (1992). Transgenes expressing the Wnt-1 and int-2
proto-oncogenes cooperate during mammary carcinogenesis in
doubly transgenic mice. Mol Cell Biol 12: 147-154.
86. Land, H., Parada, L. F., and Weinberg, R. A. (1983). Tumorigenic
conversion of primary embryo fibroblasts requires at least two
cooperating oncogenes. Nature 304: 596-602.
87. Lane, T. F., and Leder, P. (1997). Wnt-1 Ob directs hypermorphic
development and transformation in mammary glands of male and
female mice. Oncogene 15: 2133-2144.
88. LaRochelle, W. J., et al. (1995). Specific receptor detection by a
functional keratinocyte growth factor- immunoglobulin chimera. J Cell
Biol 129: 357-366.
89. Lazo, P. A., Klein-Szanto, A. J., and Tsichlis, P. N. (1990). T-cell
lymphoma lines derived from rat thymomas induced by Moloney
murine leukemia virus: phenotypic diversity and its implications. J Virol
64: 3948-3959.
90. Leder, A., Pattengale, P. K., Kuo, A., Stewart, T. A., and Leder, P.
(1986). Consequences of widespread deregulation of the c-myc gene
in transgenic mice: multiple neoplasms and normal development. Cell
45: 485-495.
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91. Lee, F. S., Lane, T. F., Kuo, A., Shackleford, G. M L , and Leder, P.
(1995). Insertional mutagenesis identifies a member of the Wnt gene
family as a candidate oncogene in the mammary epithelium of int-
2/Fgf-3 transgenic mice. Proc Natl Acad Sci L J S A 92: 2268-2272.
92. Lee, R. J., et al. (1999). pp60(v-src) induction of cyclin D1 requires
collaborative interactions between the extracellular signal-regulated
kinase, p38, and Jun kinase pathways. A role for cAMP response
element-binding protein and activating transcription factor-2 in pp60(v-
src) signaling in breast cancer cells. J Biol Chem 274: 7341-7350.
93. Li, B., et al. (1998). A transgenic mouse model for mammary
carcinogenesis. Oncogene 16: 997-1007.
94. Li, Y., Basilico, C., and Mansukhani, A. (1994). Cell transformation by
fibroblast growth factors can be suppressed by truncated fibroblast
growth factor receptors. Mol Cell Biol 14: 7660-7669.
95. Li, Y „ Hively, W. P., and Varmus, H. E. (2000). Use of MMTV-Wnt-1
transgenic mice for studying the genetic basis of breast cancer.
Oncogene 19: 1002-1009.
96. Li, Y., et al. (2001). Deficiency of Pten accelerates mammary
. oncogenesis in MMTV-Wnt-1 transgenic mice. BMC Mol Biol 2: 2.
97. Lin, S. Y., et al. (2000). Beta-catenin, a novel prognostic marker for
breast cancer: its roles in cyclin D1 expression and cancer
progression. Proc Natl Acad Sci U S A 97: 4262-4266.
98. Liu, X., et al. (1997). Stat5a is mandatory for adult mammary gland
development and lactogenesis. Genes Dev 11: 179-186.
99. Lodish, H., Berk, A., Zipursky, S.L., Matsudaira, P., Baltimore, D., and
Darnell, J. (2000). "Molecular Cell Biology," W.H. Freeman & Co.
100. Lozano, G., and Elledge, S. J. (2000). p53 sends nucleotides to repair
DNA. Nature 404: 24-25.
101. Luo, J. 1., et al. (2001). Knock-in mice with a chimeric human/murine
p53 gene develop normally and show wild-type p53 responses to DNA
damaging agents: a new biomedical research tool. Oncogene 20: 320-
328.
102. MacArthur, C. A., Shankar, D. B., and Shackleford, G. M. (1995). Fgf-
8, activated by proviral insertion, cooperates with the Wnt-1 transgene
in murine mammary tumorigenesis. J Virol 69: 2501-2507.
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103. Malkin, D. (1993). p53 and the Li-Fraumeni syndrome. Cancer Genet
Cytogenet 66: 83-92.
104. Marchetti, A., et al. (1991). Host genetic background effect on the
frequency of mouse mammary tumor virus-induced rearrangements of
the int-1 and int-2 loci in mouse mammary tumors. J Virol 65: 4550-
4554.
105. Martin, G. (2001). Making a vertebrate limb: new players enter from
the wings. Bioessays 23: 865-868.
106. Matsubara, A., Kan, M., Feng, S., and McKeehan, W. L. (1998).
Inhibition of growth of malignant rat prostate tumor cells by restoration
of fibroblast growth factor receptor 2. Cancer Res 58: 1509-1514.
107. Matsui, Y., Halter, S. A., Holt, J. T., Hogan, B. L , and Coffey, R. J.
(1990). Development of mammary hyperplasia and neoplasia in
MMTV-TGF alpha transgenic mice. Cell 61: 1147-1155.
108. McKeehan, W. L , Wang, F., and Kan, M. (1998). The heparan sulfate-
fibroblast growth factor family: diversity of structure and function. Prog
Nucleic Acid Res Mol Biol 59: 135-176.
109. Michaelson, J. S., and Leder, P. (2001). beta-catenin is a downstream
effector of Wnt-mediated tumorigenesis in the mammary gland.
Oncogene 20: 5093-5099.
110. Miki, T., et al. (1992). Determination of ligand-binding specificity by
alternative splicing: two distinct growth factor receptors encoded by a
single gene. Proc Natl Acad Sci U S A 89: 246-250.
111. Miyoshi, K., et al. (2001). Signal transducer and activator of
transcription (Stat) 5 controls the proliferation and differentiation of
mammary alveolar epithelium. J Cell Biol 155: 531-542.
112. Muller, W. J. (1991). Expression of activated oncogenes in the murine
mammary gland: transgenic models for human breast cancer. Cancer
Metastasis Rev 10: 217-227.
113. Muller, W. J., Sinn, E., Pattengale, P. K., Wallace, R., and Leder, P.
(1988). Single-step induction of mammary adenocarcinoma in
transgenic mice bearing the activated c-neu oncogene. Cell 54: 105-
115.
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114. Munoz-Sanjuan, I., and A, H. B. (2001). Early posterior/ventral fate
specification in the vertebrate embryo. Dev Biol 237: 1-17.
115. Nusse, R. (1991). Insertional mutagenesis in mouse mammary
tumorigenesis. In "Retroviral insertion and oncogene activation" (H. J.
Kung, and Vogt, P.K., Ed.), pp. 43-45, Springer-Verlag, New York.
116. Nusse, R. (1992). The Wnt gene family in tumorigenesis and in normal
development. J Steroid Biochem Mol Biol 43: 9-12.
117. Nusse, R., and Varmus, H. E. (1982). Many tumors induced by the
mouse mammary tumor virus contain a provirus integrated in the
same region of the host genome. Cell 31: 99-109.
118. Nusse, R., and Varmus, H. E. (1992). Wnt genes. Cell 69: 1073-1087.
119. Olayioye, M. A., Neve, R. M., Lane, H. A., and Hynes, N. E. (2000).
The ErbB signaling network: receptor heterodimerization in
development and cancer. Embo J 19: 3159-3167.
120. Ornitz, D. M. (2000). FGFs, heparan sulfate and FGFRs: complex
interactions essential for development. Bioessays 22: 108-112.
121. Ornitz, D. M. (2001). Regulation of chondrocyte growth and
differentiation by fibroblast growth factor receptor 3. Novartis Found
Symp 232: 63-76.
122. Ornitz, D. M., and Itoh, N. (2001). Fibroblast growth factors. Genome
Biol 2.
123. Orr-Urtreger, A., et al. (1993). Developmental localization of the
splicing alternatives of fibroblast growth factor receptor-2 (FGFR2).
Dev BioH 58: 475-486.
124. Paterson, J. W. (1998). BRCA1: a review of structure and putative
functions. Dis Markers 13: 261-274.
125. Pattengale, P. K., et al. (1989). Animal models of human disease.
Pathology and molecular biology of spontaneous neoplasms occurring
in transgenic mice carrying and expressing activated cellular
oncogenes. Am J Pathol 135: 39-61.
126. Pedchenko, V. K., and Imagawa, W. (2000). Pattern of expression of
the KGF receptor and its ligands KGF and FGF- 10 during postnatal
mouse mammary gland development. Mol Reprod Dev 56: 441-447.
136
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
127. Peg ram, M. D., Pauletti, G., and Slamon, D. J. (1998). HER-2/neu as
a predictive marker of response to breast cancer therapy. Breast
Cancer Res Treat 52: 65-77.
128. Peters, G., Brookes, S., Smith, R., Placzek, M., and Dickson, C.
(1989). The mouse homolog of the hst/k-FGF gene is adjacent to int-2
and is activated by proviral insertion in some virally induced mammary
tumors. Proc Natl Acad Sci U S A 86: 5678-5682.
129. Peters, K., et al. (1994). Targeted expression of a dominant negative
FGF receptor blocks branching morphogenesis and epithelial
differentiation of the mouse lung. Embo J 13: 3296-3301.
130. Petrocelli, T., and Slingerland, J. M. (2001). PTEN deficiency: a role in
mammary carcinogenesis. Breast Cancer Res 3: 356-360.
131. Pike, M. C., Spicer, D. V., Dahmoush, L , and Press, M. F. (1993).
Estrogens, progestogens, normal breast cell proliferation, and breast
cancer risk. Epidemiol Rev 15: 17-35.
132. Podsypanina, K., et al. (1999). Mutation of Pten/Mmac1 in mice
causes neoplasia in multiple organ systems. Proc Natl Acad Sci U S A
96: 1563-1568.
133. Polakis, P. (2000). Wnt signaling and cancer. Genes Dev 14: 1837-
1851.
134. Powers, C. J., McLeskey, S. W., and Wellstein, A. (2000). Fibroblast
growth factors, their receptors and signaling. Endocr Relat Cancer 7:
165-197.
135. Pucillo, C., Cepeda, R., and Hodes, R. J. (1993). Expression of a
MHC class II transgene determines both superantigenicity and
susceptibility to mammary tumor virus infection. J Exp Med 178: 1441-
1445.
136. Radeva, G., et al. (1997). Overexpression of the integrin-linked kinase
promotes anchorage- independent cell cycle progression. J Biol Chem
272: 13937-13944.
137. Raponi, M., Dawes, I. W., and Arndt, G. M. (2000). Characterization of
flanking sequences using long inverse PCR. Biotechniques 28: 838-
840, 842, 844.
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
138. Rassoulzadegan, M., et al. (1982). The roles of individual polyoma
virus early proteins in oncogenic transformation. Nature 300: 713-718.
139. Ricol, D., et al. (1999). Tumour suppressive properties of fibroblast
growth factor receptor 2- illb in human bladder cancer. Oncogene 18:
7234-7243.
140. Ringold, G. M. (1983). Regulation of mouse mammary tumor virus
gene expression by glucocorticoid hormones. Curr Top Microbiol
Immunol 106: 79-103.
141. Ringold, G. M., Cardiff, R. D., Varmus, H. E., and Yamamoto, K. R.
(1977). Infection of cultured rat hepatoma cells by mouse mammary
tumor virus. Cell 10: 11-18.
142. Robinson, M. L , MacMillan-Crow, L. A., Thompson, J. A., and
Overbeek, P. A. (1995). Expression of a truncated FGF receptor
results in defective lens development in transgenic mice. Development
121: 3959-3967.
143. Roelink, H., Wagenaar, E., Lopes da Silva, S., and Nusse, R. (1990).
Wnt-3, a gene activated by proviral insertion in mouse mammary
tumors, is homologous to int-1/Wnt-1 and is normally expressed in
mouse embryos and adult brain. Proc Natl Acad Sci U S A 87: 4519-
4523.
144. Rose-Hellekant, T. A., and Sandgren, E. P. (2000). Transforming
growth factor alpha- and c-myc-induced mammary carcinogenesis in
transgenic mice. Oncogene 19: 1092-1096.
145. Salmons, B., Sailer, R. M., Baumann, J. G., and Gunzburg, W. H.
(1995). Construction of retroviral vectors for targeted delivery and
expression of therapeutic genes. Leukemia 9 Suppl 1: S53-60.
146. Sandgren, E. P., Luetteke, N. C., Palmiter, R. D., Brinster, R. L., and
Lee, D. C. (1990). Overexpression of TGF alpha in transgenic mice:
induction of epithelial hyperplasia, pancreatic metaplasia, and
carcinoma of the breast. Cell 61: 1121-1135.
147. Sandgren, E. P., et al. (1995). Inhibition of mammary gland involution
is associated with transforming growth factor alpha but not c-myc-
induced tumorigenesis in transgenic mice. Cancer Res 55: 3915-
3927.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
148. Schroeder, J. A., Troyer, K. L , and Lee, D. C. (2000). Cooperative
induction of mammary tumorigenesis by TGFalpha and Wnts.
Oncogene 19: 3193-3199.
149. Shackleford, G. M., MacArthur, C. A., Kwan, H. C., and Varmus, H. E.
(1993). Mouse mammary tumor virus infection accelerates mammary
carcinogenesis in Wnt-1 transgenic mice by insertional activation of
int-2/Fgf-3 and hst/Fgf-4. Proc Natl Acad Sci U SA90: 740-744.
150. Shackleford, G. M., and Varmus, H. E. (1987). Expression of the
proto-oncogene int-1 is restricted to postmeiotic male germ cells and
the neural tube of mid-gestational embryos. Cell 50: 89-95.
151. Shackleford, G. M., and Varmus, H. E. (1988). Construction of a
clonable, infectious, and tumorigenic mouse mammary tumor virus
provirus and a derivative genetic vector. Proc Natl Acad Sci U S A 85;
9655-9659.
152. Shackleford, G. M., Willert, K., Wang, J., and Varmus, H. E. (1993).
The Wnt-1 proto-oncogene induces changes in morphology, gene
expression, and growth factor responsiveness in PC12 cells. Neuron
11: 865-875.
153. Shillingford, J. M., and Hennighausen, L. (2001). Experimental mouse
genetics-- answering fundamental questions about mammary gland
biology. Trends Endocrinol Metab 12: 402-408.
154. Shin, S., and Steffen, D. L. (1993). Frequent activation of the Ick gene
by promoter insertion and aberrant splicing in murine leukemia virus-
induced rat lymphomas. Oncogene 8: 141-149.
155. Sicinski, P., and Weinberg, R. A. (1997). A specific role for cyclin D1
in mammary gland development. J Mammary Gland Biol Neoplasia 2:
335-342.
156. Silberstein, G. B. (2001). Postnatal mammary gland morphogenesis.
Microsc Res Tech 52: 155-162.
157. Slamon, D. J., et al. (1989). Studies of the HER-2/neu proto-oncogene
in human breast and ovarian cancer. Science 244: 707-712.
158. Smalley, M. J., and Dale, T. C. (2001). Wnt signaling and mammary
tumorigenesis. J Mammary Gland Biol Neoplasia 6: 37-52.
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
159. Spencer-Dene, B., et al. (2001). Fibroblast growth factor signalling in
mouse mammary gland development. Endocr Retat Cancer 8: 211-
217.
160. Stallcup, M. R., and Washington, L. D. (1983). Region-specific
initiation of mouse mammary tumor virus RNA synthesis by
endogenous RNA polymerase II in preparations of cell nuclei. J Biol
Chem 258: 2802-2807.
161. Stewart, T. A., Pattengale, P. K., and Leder, P. (1984). Spontaneous
mammary adenocarcinomas in transgenic mice that carry and express
MTV/myc fusion genes. Cell 38: 627-637.
162. Suzuki, A., et al. (1998). High cancer susceptibility and embryonic
lethality associated with mutation of the PTEN tumor suppressor gene
in mice. Curr Biol 8: 1169-1178.
163. Szebenyi, G., and Fallon, J. F. (1999). Fibroblast growth factors as
multifunctional signaling factors. Int Rev CytoliBB: 45-106.
164. Tamai, K., et al. (2000). LDL-receptor-related proteins in Wnt signal
transduction. Nature 407: 530-535.
165. Teich, N., Wyke, J.A., Mak, T, Bernstein, A. and Hardy, W (1982).
Pathogenesis of retrovirus-induced disease. In "The Molecular Biology
of Tumor Viruses, Part III, RNA Tumor Viruses, Chapter 10, R.A." (R.
A. Weiss, Teich, N., Varmus, H.E., and Coffin, J.M., Ed.), Cold Spring
Harbor Laboratory, Cold Spring Harbor, New York.
166. Tekmal, R. R., and Keshava, N. (1997). Role of MMTV integration
locus cellular genes in breast cancer. Front Biosci 2: d519-526.
167. Theillet, C., et al. (1989). Amplification of FGF-related genes in human
tumors: possible involvement of HST in breast carcinomas. Oncogene
4: 915-922.
168. Tsichlis, P. N., Strauss, P. G., and Hu, L. F. (1983). A common region
for proviral DNA integration in MoMuLV-induced rat thymic
lymphomas. Nature 302: 445-449.
169. Tsukamoto, A. S., Grosschedl, R., Guzman, R. C., Parslow, T., and
Varmus, H. E. (1988). Expression of the int-1 gene in transgenic mice
is associated with mammary gland hyperplasia and adenocarcinomas
in male and female mice. Cell 55: 619-625.
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
170. Ueda, T., et al. (1999). Deletion of the carboxyl-terminal exons of K-
sam/FGFR2 by short homology-mediated recombination, generating
preferential expression of specific messenger RNAs. Cancer Res 59:
6080-6086.
171. Ueno, H., Gunn, M., Dell, K., Tseng, A., Jr., and Williams, L. (1992). A
truncated form of fibroblast growth factor receptor 1 inhibits signal
transduction by multiple types of fibroblast growth factor receptor. J
Biol Chem 267: 1470-1476.
172. Ulich, T. R., et al. (1994). Keratinocyte growth factor is a growth factor
for mammary epithelium in vivo. The mammary epithelium of lactating
rats is resistant to the proliferative action of keratinocyte growth factor.
Am J Pathol 144: 862-868.
173. van Leeuwen, F., and Nusse, R. (1995). Oncogene activation and
oncogene cooperation in MMTV-induced mouse mammary cancer.
Semin Cancer Biol 6: 127-133.
174. Varmus, H. E. (1982). Recent evidence for oncogenesis by insertion
mutagenesis and gene activation. Cancer Surveys 2: 301.
175. Varmus, H. E. (1987). Oncogenes and transcriptional control. Science
238: 1337-1339.
176. Vijaya, S., Steffen, D. L , Kozak, C., and Robinson, H. L. (1987). Dsi-
1, a region with frequent proviral insertions in Moloney murine
leukemia virus-induced rat thymomas. J Virol 61: 1164-1170.
177. Villeneuve, L , Jiang, X., Turmel, C., Kozak, C. A., and Jolicoeur, P.
(1993). Long-range mapping of Mis-2, a common provirus integration
site identified in murine leukemia virus-induced thymomas and located
160 kilobase pairs downstream of Myb. J Virol 67: 5733-5739.
178. Wallace-Brodeur, R. R., and Lowe, S. W. (1999). Clinical implications
of p53 mutations. Cell Mol Life Sci 55: 64-75.
179. Wang, T. C., et al. (1994). Mammary hyperplasia and carcinoma in
MMTV-cyclin D1 transgenic mice. Nature 369: 669-671.
180. Weber-Hall, S. J., Phippard, D. J., Niemeyer, C. C., and Dale, T. C.
(1994). Developmental and hormonal regulation of Wnt gene
expression in the mouse mammary gland. Differentiation 57: 205-214.
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
181. Weinberg, R. A. (1995). The retinoblastoma protein and cell cycle
control. Cell 81: 323-330.
182. Weinstat-Saslow, D., et al. (1995). Overexpression of cyclin D mRNA
distinguishes invasive and in situ breast carcinomas from n on-
malignant lesions. Nat Med 1: 1257-1260.
183. Whittemore, A. S., Gong, G., and Itnyre, J. (1997). Prevalence and
contribution of BRCA1 mutations in breast cancer and ovarian cancer:
results from three U.S. population-based case-control studies of
ovarian cancer. Am J Hum Genet 60: 496-504.
184. Wiesen, J. F., Young, P., Werb, Z., and Cunha, G. R. (1999).
Signaling through the stromal epidermal growth factor receptor is
necessary for mammary ductal development. Development 126: 335-
344.
185. Yan, G., Fukabori, Y., McBride, G., Nikolaropolous, S., and
McKeehan, W. L. (1993). Exon switching and activation of stromal and
embryonic fibroblast growth factor (FGF)-FGF receptor genes in
prostate epithelial cells accompany stromal independence and
malignancy. Mol Cell Biol 13: 4513-4522.
186. Zammit, C., et al. (2001). Altered intracellular localization of fibroblast
growth factor receptor 3 in human breast cancer. J Pathol 194: 27-34.
187. Zhan, X., Culpepper, A., Reddy, M., Loveless, J., and Goldfarb, M.
(1987). Human oncogenes detected by a defined medium culture
assay. Oncogene 1: 369-376.
188. Zhang, J. D., Cousens, L. S., Barr, P. J., and Sprang, S. R. (1991).
Three-dimensional structure of human basic fibroblast growth factor, a
structural homolog of interleukin 1 beta. Proc Natl Acad Sci U S A 88:
3446-3450.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Cellular factors involved in mouse hepatitis virus replication

PDF

Functional analysis of MSX2 and its role in skull patterning

PDF

Immunogenicity of acute lymphoblastic leukemia: In vivo and in vitro studies of Bcr -Abl specific immune responses

PDF

A new model for hepatitis delta virus transcription and replication

PDF

Analysis of multistep mammary tumorigenesis in Wnt1 transgenic mice: The role of fibroblast growth factor-8 in oncogenesis and apoptosis

PDF

A Study Of The Mechanism Of Lead Chromate Induced Neoplastic Transformation Of C3H/10T1/2 Mouse Embryo Cells

PDF

Biochemical characterization of hydrogen,potassium-ATPase-rich membranes from the gastric parietal cell

PDF

Cloning, characterization, anti -apoptotic molecular mechanism, transgenic and protein binding studies of mouse TOSO gene

PDF

Cloning and characterization of Grb4: A novel adapter protein interacting with BCR -Abl

PDF

A transgenic mouse model for small cell lung cancer

PDF

Expression of the RGR opsin and its function in the photic visual cycle

PDF

The androgen receptor: Its role in the development and progression of cancers of the prostate and breast

PDF

Wnt factors: Genes, functions and cooperative signaling in mammary cell transformation

PDF

Characterization of new factors involved in feline leukemia virus (FELV)-mediated leukemogenesis

PDF

Analysis of the ALOX5 gene in atherosclerosis

PDF

Identification and cloning of developmentally regulated genetic loci in transgenic mice by screening for novel expression pattern of the lacZ reporter gene

PDF

Androgens and breast cancer

PDF

Characterization of target cell entry by murine leukemia viruses

PDF

Studies Of Topoisomerase Ii As The Target Of Anti-Cancer Agents

PDF

Breast cancer susceptibility gene 1: A role in transcriptional regulation

Asset Metadata

Creator Lopez-Diego, Rocio Sagrario (author)

Core Title Identification of oncogenes cooperating in murine mammary tumorigenesis and transgenic mouse models of breast cancer

School Graduate School

Degree Doctor of Philosophy

Degree Program Molecular Microbiology and Immunology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, microbiology,biology, molecular,health sciences, oncology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Shackleford, Gregory M. (committee chair), [illegible] (committee member), Ou, Jing-Hsiung James (committee member), Tahara, Stanley (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-253330

Unique identifier UC11339291

Identifier 3093785.pdf (filename),usctheses-c16-253330 (legacy record id)

Legacy Identifier 3093785.pdf

Dmrecord 253330

Document Type Dissertation

Rights Lopez-Diego, Rocio Sagrario

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA