University of Southern California Dissertations and Theses

Mechanisms of ovarian cancer predisposition in BRCA1 mutation carriers

(USC Thesis Other)

Mechanisms of ovarian cancer predisposition in BRCA1 mutation carriers

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
MECHANISMS OF OVARIAN CANCER PREDISPOSITION IN BRCA1 MUTATION
CARRIERS

by
Hao Hong

____________________________________________________________

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
(PATHOBIOLOGY)

August 2008

Copyright 2008 Hao Hong
ii

Table of Contents

List of Tables v
List of Figures vi
Abstract viii
PART I Background 1
Chapter 1 Introduction to Ovarian Cancer 1
1.1 Classifications 2
1.2 Etiology and risk factors 4
1.3 Current theories about the cell origin of epithelial 10
ovarian cancer
1.4 Models of ovarian cancer 12

Chapter 2 Breast Cancer Susceptibility Gene1 (BRCA1) 19
2.1 Biological functions of BRCA1 19
2.2 Possible mechanisms of tissue-specific tumor suppression 21
2.3 Unusual properties as a tumor suppressor gene 24

Chapter 3 Introduction to Menstrual cycle 25
3.1 Ovarian folliculogenesis, ovulation, and luteinization 25
in humans
3.2 Ovarian granulosa cells functions 27
3.3 Menstrual cycle 28
3.4 Cyclic steroid hormone changes during menstrual cycle 29
3.5 Regulation of menstrual cycle 30

Chapter 4 A Novel Mouse Model for Epithelial Ovarian Cancer 32
4.1 A Mouse model of epithelial ovarian cystadenoma 32
via Brca1 mutation
4.2 Rationale and working hypothesis 35

PART II Studies on the mechanisms of ovarian tumor 38
predisposition in conditional Brca1 knockout mice
iii

Chapter 5 Role of Estrous Cycle Abnormalities in Ovarian Tumor 38
Predisposition
Chapter 5 Abstract 38
Introduction 41
Materials and Methods 45
Results 48
Discussion 68

Chapter 6 The Effect of Brca1 Inactivation on Circulating Level 73
of Estradiol
Chapter 6 Abstract 73
Introduction 74
Materials and Methods 75
Results 76
Discussion 78

Chapter 7 Cre Expression in Pituitary Gland 80
Chapter 7 Abstract 80
Introduction 81
Materials and Methods 83
Results 84
Discussion 89

Chapter 8 Distinguish between Effects of Brca1 Inactivation in the 91
Pituitary versus Granulosa Cells via Ovarian Transplantation
Chapter 8 Abstract 91
Introduction 93
Materials and Methods 95
Results 97
Discussion 114

PART III Germline rearrangement of a conditional allele 115
via parental transmission of Cre recombinase

Chapter 9 Germline Recombination caused by Diffusion of Cre 115
Recombinase Protein from Gonadal Interstitial Cells
to Germ Cells
Chapter 9 Abstract 115
Introduction 117

iv

Materials and Methods 119
Results 120
Discussion 123

PART IV Epilogue 125
Summary 125
Future directions 127

BIBLIOGRAPHY 132
v

LIST OF TABLES

Table 1. Factors influencing ovarian cancer risk 9
Table 2. Classification of stages of estrus cycle by cellular morphology in 50
Vaginal smears
Table 3. Comparison the length of each phase of estrus cycle in mice receiving 108
ovarian transplants
vi

LIST OF FIGURES

Figure 1. Mutant mice develop ovarian cystadenomas 37
Figure 2. Classification of estrous cycle stages by cellular morphology in vaginal 49
Smears
Figure 3. Histological evaluation of ovaries collected from wild type mice at 52
various stages of natural estrous cycle determined by vaginal cytology
Figure 4. Serum levels of 17{ß}-estradiol (E2) as a function of the mice estrous 54
cycle
Figure 5. Effects of Brca1 inactivation on duration of Proestrus stage 56
Figure 6. Effects of Brca1 inactivation on duration of Metestrus stage and 59
Diestrus stage combined length
Figure 7. Effects of Brca1 inactivation on length ratios of 60
Proestrus/Mestrus+Diestrus
Figure 8. Effects of Brca1 inactivation on duration of Metestrus stage 61
Figure 9. Effects of Brca1 inactivation on Proestrus/Metestrus length ratios 62
Figure 10. Comparison of Proestrus stage length in mutants that developed 65
Tumors versus those that didn’t
Figure 11. Comparison of the length ratio of Proestrus/Metestrus+Diestrus in 66
mutant mice that developed tumors versus those that didn’t
Figure 12. Comparison of the length ratio of Proestrus/Metestrus in mutant mice 67
that developed tumors versus those that didn’t.
Figure 13. Significant enlarged dilated ducts in the mammary glands of 18 to 20 85
months old mutant virgin mice
Figure 14. Cre expression in pituitary gland 88
vii

Figure 15. Ovarian transplantation into renal capsule 100
Figure 16. Determination of estrus cyclicity in mice receiving ovarian transplants 101
from external vaginal examination
Figure 17. Validation of ovarian graft function by vaginal cytology 102
Figure 18. Follicular development in a transplanted ovarian graft 103
Figure 19. Comparison of the Proestrus stage, Proestrus and estrus combined 109
length in mice receiving ovarian transplants
Figure 20. Ovarian and uterine cystadenomas developed in mice receiving 113
Transplanted ovaries
Figure 21. Genotyping analyses of 20 Fshr-Cre; Brca1
flox/flox
mice 122
viii

ABSTRACT

The exact mechanism by which Breast Cancer Susceptibility Gene1 (BRCA1) mutations
predispose to breast and ovarian cancer remains poorly understood.

Our laboratory developed a mouse model for epithelial ovarian cystadenoma based on
inactivation of Brca1 specifically in ovarian granulosa cells which play a critical role in
regulating menstrual cycle and a subset of cells in pituitary via Cre-loxP technology.

My project was to study the possible mechanisms of ovarian tumors development in this
mouse model. I characterized dynamics of ovulatory cycles of these mice in a large scale
and over a long period time. Mutant mice showed statistically significant increase in the
length of proestrus phase, increases in the length ratio of proestrus/metestrus and diestrus
combined, and increases in the length ratio of proestrus/metestrus. Moreover, these estrus
cycle abnormalities were associated with tumor predisposition.

These results suggest mutant mice have a relative increase in estrogen growth stimulation
unopposed by progesterone. Given the well known influence of the menstrual cycle and
estrogen as important risk factors for ovarian cancer and progesterone as a protective
factor, my studies raise the possibility that predisposition to ovarian and breast cancers in
ix

individuals bearing germline BRCA1 mutations may be due, at least in part, to menstrual
cycle changes driven by these mutations.

My studies also strongly supported our hypothesis that BRCA1/Brca1 controls tumor
development in a cell non-autonomous manner as opposed to acting as a classical tumor
suppressor gene by indirectly modulating the estrus cycle. This novel role for Brca1
suggests a new possible mechanism for tumor tissue specificity in BRCA1 mutations
carriers.

In order to distinguish between any effects of Brca1 inactivation in the pituitary versus
granulosa cells, I performed ovarian transplantation. My studies showed the estrus cycle
changes in the mutant mice were driven by granulosa cells as opposed to pituitary cells.

Lastly I found that germline transmission of a mutant Brca1 allele had taken place in
some of the mutant mice despite the tissue specificity of the promoter we used to drive
Cre. Such germline rearrangement occurred through parental transmission of Cre
recombinase but the transmission frequency is lower via female transmission.
1

PART I

BACKGROUND

Chapter 1

Introduction to Ovarian Cancer

Ovarian cancer is a relatively rare disease and has less than a 1% life-time risk, yet is the
most lethal gynecologic malignancy and is the fourth most common cause of

death from
cancer in women [American Cancer Society, 2003]. The unusual high mortality rate
associated with this disease is largely due to the asymptomatic nature of early stages of
this disease. Ovarian cancers are highly insidious. In contrast to breast cancer, where
most cases are detected at an early stage, approximately 70% of women with ovarian
cancer are diagnosed at advanced stages when the cancers have spread beyond the ovaries.
Treatments with aggressive surgery followed by chemotherapy rarely result in cure, and
most patients eventually succumb to their disease.

2

Unfortunately, however, effective and sensitive screening tests for early stage detection
are still unavailable despite intensive clinical and experimental research. This progress
was hampered due to lack of understanding of the nature of the pre-invasive or precursor
lesions and genetic alterations in these lesions for ovarian carcinoma, and uncertainty
about the cell origin. Another significant obstacle in this field is the lack of a reliable
animal model that can precisely recapitulate the early events in ovarian tumorigenesis.
Therefore, epithelial ovarian cancer pathogenesis remains among the least understood of
all major cancers.

This chapter highlights some basic epidemiological aspects of ovarian cancer, current
controversy about cell origin of this disease with special consideration of the strengths
and weaknesses of some of the current models of ovarian cancer including genetically
engineered mice models.

1.1 Classification

Ovarian cancer is a broad term used for a wide range of neoplasms that originate in the
ovary. Ovarian neoplasms are classified into three broad groups according to the most
probable tissue of origin.
a) Germ Cell Tumors
3

Account for less than 5% of ovarian neoplasms. These tumors develop from an
oocyte. They typically occur in teenagers and young women. Ninety five percent
are benign mature cystic teratomas and often contain a variety of differentiated
tissues such as skin, teeth, and hair etc. The remainder has a higher incidence of
malignant behavior and poses problems in histologic diagnosis and in therapy.
b) Sex Cord–Stromal Cell Tumors
Account for less than 5% of ovarian neoplasms. These originate from granulosa
cells, theca cells and their luteinized derivatives, Sertoli cells, Leydig cells, and
fibroblasts of stromal origin. Most clinically malignant sex cord–stromal tumors
are of the granulosa cell type.
c) Epithelial ovarian tumors
The most prevalent type, account for more than 90% of all ovarian cancers.
Histologicaly, epithelial ovarian carcinomas are a very heterogeneous group of
tumor. There are five distinct histological subtypes and interestingly these
subtypes morphologically resemble those of the epithelia of the reproductive tract
that embryonically derived from the Mueller Ian ducts:
1. Serous tumor (fallopian tube-like): comprise approximately 80% of all
epithelial ovarian cancers. Resemble the glandular epithelium lining the
fallopian tube.
4

2. Endometrioid carcinomas: These are the second most common forms. At the
morphological level, these tumors resemble endometrium
3. Mucinous tumor (endocervical-like): These tumors resemble endocervical
glands
4. Clear cell carcinomas: These are high-grade aggressive tumors and resemble
mesonephros
5. Transitional cell/Brenner tumors

It is generally accepted that each subtype differs in etiology and genetic alterations
[Chatzistamou et al., 2002 and Tsukagoshi et al., 2003]

Based on malignant potential, ovarian epithelial cancer is subdivided into benign tumors
called cystadenoma, low malignant potential (also called tumors of borderline
malignancy), and high-grade malignant tumors (also called carcinomas). The extent of
applicability of the tumor progression model first advanced by Fould to ovarian
carcinoma development is still unclear. It is not known if the ovarian carcinomas arise de
novo or develop by multistep clonal expansion from preexisting precursor benign tumors
[Zheng et al.,1993].

1.2 Etiology and risk factors

5

The etiology of the epithelial ovarian carcinomas is poorly understood. The major risk
factors are summarized in Table 1.

1.2.1 Family history

At present, family history is the strongest risk factor for ovarian cancer. Compared with a
1 % lifetime risk for developing ovarian cancer in the general population, women with
one first-degree relative with ovarian cancer have a 5% risk, and women with two
first-degree relatives have a 7% risk.

1.2.2 Ovulation

Numerical epidemiological studies have consistently demonstrated that the accumulated
number of menstrual cycles correlates with the risk of ovarian cancer [Riman et al., 1998]
and consequently reproductive factors that decrease the number of lifetime ovulations
lower the risk of developing ovarian cancer such as full-term pregnancies, increasing
parity, and contraceptive pills use have been convincingly shown to have strong
protective effect on epithelial ovarian cancer risk. In contrast, it has long been known that
there is an increased risk of ovarian cancer in nulliparous women.

6

All these observations serve as the basis for the famous “incessant ovulation hypothesis”
of ovarian cancer etiology proposed by Fathalla in 1971. Briefly, he hypothesized that the
repeated rupture followed by rapid proliferation to repair the ovarian surface epithelium
at the site of ruptured follicle as occurs with repetitious ovulation during menstrual cycle
provide an opportunity for genetic aberrations.

Sine this hypothesis was proposed, it has been accepted almost as a dogma in etiology of
ovarian cancer. However, now this theory has been abandoned by epidemiologists
because many observations can not be accounted for by this theory such as:

Ethnic variations in ovarian cancer incidence do not correlate with differences in
number of ovulations
Pregnancy is more protective than one year of oral contraceptive use
Risk of ovarian cancer continues to rise after the menopause
Age of menarche does not influence ovarian cancer risk
Late pregnancy is more protective

Prophylactic removal of ovaries at high risk for developing ovarian cancer is an effective
prevention strategy and is especially applicable to women with strong family histories of
ovarian cancer.

7

1.2.3 Hormonal factors

Additional mechanisms must also be involved in the etiology of ovarian cancer. These
mechanisms are probably hormonal.
a. Gonadotrophin hypothesis
The “gonadotrophin hypothesis” proposes that high levels of pituitary gonadotrophins
increase cancer risk by stimulating the ovarian surface epithelium. Historically, the
gonadotrophin hypothesis arose from the observation that ovarian tumors occurred in the
rodents following bilateral oophorectomy and ovarian transplantation under the kidney
capsule [Biskind MS and Biskind GR 1944].

Serum FSH and LH reach maximal values in the postmenopausal years with the depletion
of oocytes and remain highly elevated thereafter. This suggests that a relationship may
exist between the rise in gonadotrophins and the later peak in ovarian cancer incidence.
Epidemiological evidence seems to suggest that the gonadotrophins may not themselves
be responsible for alterations in ovarian cancer risk but could reflect certain hormonal
circumstances that are related to risk.
b. Estrogen
Epidemiologic evidence bears on the role of estrogen in the pathophysiology of ovarian
cancer. A number of recent large studies have consistently demonstrated that risk of
ovarian cancer increases with duration of postmenopausal estrogen replacement therapy
[Riman et al., 1998；Rodriguez et al., 2001； Purdie et al., 1996; Risch et al., 1996].
8

Conversely, the well-known strong protective effect of oral contraceptive bill use against
ovarian cancer is due to not only suppression of ovulation but also may be due in part to
reduction in endogenous estradiol production [Killick et al., 1987].
c. Progesterone
In contrast, progesterone has been suggested to have a protective role in the etiology of
ovarian cancer based on epidermiological oberservations. Consistent with these
observatiions, many studies including work from our laboratory have demonstrated the
role of progesterone in inhibiting cell proliferations, decreasing invasiveness, suppressing
transformation and inducing apoptosis [Bu, Yin et al., 1997; Keith Bechtel and Bonavida
2001; McDonnel and Murdoch 2001; Yu, Lee et al., 2001; Zhou, Luo et al., 2002;
Blumenthal, Kardosh et al., 2003; Syed V, Ho SM.2003].

9

Table1.1 Factors Influencing Ovarian Cancer Risk

Increased risk____________________________________

Family history of ovarian cancer
Nulliparity
Infertility
Early menarche, late menopause
________________________________________________

Decreased risk____________________________________

Oral contraceptive pill
Multiparity
Breast feeding
Tubal ligation
Hysterectomy
Prophylactic oophorectomy
______________________________________________________

10

1.3 Current theories about the cell origin of epithelial ovarian cancer

Searching for the precursor lesion of ovarian epithelial tumors implies knowledge of their
actual site of origin. However, the exact nature of cell origin from which ovarian
epithelial tumors originate is controversial.

a. Ovarian surface epithelium origin

The widely held assumption is that ovarian carcinomas arise from the ovarian surface
epithelium (OSE), a single cell layer of simple epithelial cells that covers the entire
surface of the ovary. However, it is difficult to explain the diverse histological types of
ovarian carcinomas that arise from this relatively homogenous tissue.

Why epithelial ovarian cancers show müllerian-like features when OSE is not of
müllerian duct origin? The common assumption is that OSE first undergoes metaplasia to
become müllerian-like before it becomes malignant transformation. However, the finding
of müllerian–like epithelium on the ovarian surface is rare. Moreover, preneoplastic or
early neoplastic lesion was almost never detected in OSE.

11

Another intriguing feature is that tumors that are histologically and clinically identical to
ovarian carcinomas, referred to as primary peritoneal carcinomas, are sometimes found
outside of the ovaries. These tumors are confined to the women whose ovaries were
surgically removed several years earlier due to the reasons other than cancers.

b. Secondary müllerian duct origin

Given the resemblance of ovarian epithelial tumors to those arising from
müllerian-derived organs, an alternative hypothesis that these cancers arise from
secondary müllerian system was proposed by Dr. Dubeau [1999]. The term “secondary
müllerian system” is used to designate structures lined by Müllerian system epithelium
found outside the uterus, cervix and fallopian tubes [Lauchlan 1994]. These structures are
often found in paratubal, paraovarian areas, and rete ovarii which consist of numerous
müllerian rests lined by müllerian epithelium.

The above mentioned difficulties with OSE origin can be easily explained by secondary
müllerian duct origin hypothesis. Firstly, the müllerian-like histological features of EOC
can be readily explained without the need of the intermediate metaplasia step. Secondly,
the presence of secondary müllerian structures away from the ovary can account for the
presence of ovarian-like tumors arising outside of ovary.

12

Although this hypothesis has not been tested directly, the importance of müllerian tract in
the histogenesis of ovarian epithelial tumors is supported by a recent study demonstrating
that the HOX proteins expressed by ovarian tumors are similar to those expressed in
müllerian epithelia and different than those expressed in the ovarian surface [Wenjun et
al., 2005].

1.4 Models of ovarian cancer
a. Xenograft models
Human ovarian cancer cells xenografts into mouse are extensively used research models
for the study of ovarian caner as they are derived from naturally occurring clinical cases
such as surgically excised solid tumor specimens

or cells isolated from the peritoneal
ascite fluid, and therefore most representative of the diverse disease.

Implantation or injection of different human ovarian cancer cell lines subcutaneously or
intraperitoneally in the mice have formed ovarian tumors with different histologies. For
instance, OVCAR3 cell lines xenografts form poorly differentiated serous
adenocarcinomas [Molthoff et al., 1991]; HEY cell lines form undifferentiated
carcinomas [Shaw et al., 2004]; OV-MZ 1-6 form serous cystadenocarcinomas [Mobus et
al., 1992]; and SKOV3 tumors were clear cell carcinomas [Shaw et al., 2004].

13

Xenograft mouse models have been used extensively for initial evaluation therapeutic
[Elkas et al., 2002; von Gruenigen et al., 1999; Nilsson et al.,

2002] strategies against the
ovarian tumor cells and also allow a detailed dissection of the roles that different
signaling pathways play in the malignant phenotype of these cell lines.

The major drawback of xenograft models is that tumor development is only achieved in
immune-compromised mice therefore the tumor–host interactions and anti-tumor immune
response can not be stimulated. Studies have shown immune system is an important
mediator of cancer progression and metastasis.

Another disadvantage of xenograft models is the majority of ovarian cancer cell lines
have been established from tumors or ascites from patients with advanced ovarian
cancers that have already accumulated numerous genetic changes, thus they shed very
little light on the earlier steps of ovarian cancer and tumor progression.
b. In vitro transformed ovarian surface epithelium (OSE) models
The ability to isolate and culture pure populations of OSE cells from different species
including humans [Tsao et al., 1995 , Gregoire et al., 1998, Nitta et al., 2001] has allowed
the researchers to genetically inactivate and or activate genes in order to induce the
immortalization and transformation of these cells and followed by implantation the
transformed OSE cells back to hosts.

14

Orsulic and colleagues [Orsulic et al., 2002] used the RCAS retroviral vector to introduce
oncogenes into mouse OSE cells and they found that a p53-null background in
combination with at least two oncogenes from among C-MYC, K-RAS, or AKT were
required for efficient tumor development in immune-compromised mice and transformed
OSE cells by p53 deficiency in combination with all the above three oncogenes were able
to produce tumors in immune-competent syngeneic mice with longer latency.

In order to use human OSE for ovarian cancer models, Kusakari [Kusakari et al., 2003]
immortalized and transformed human OSE cells by transfection of the early region of
SV40 (LT) and the catalytic subunit of telomerase (hTERT) followed by subsequent
expression of activated Ha-ras or c-erbB2. Subcutaneous injection of nude mice with
OSE/LT/hTERT/Ha-ras cells or OSE/LT/hTERT/c-erbB2 cells resulted in the
development of carcinomas in 40 and 50% of mice, respectively.

Establishment of in vitro models of normal and transformed OSE cells has provided the
possibility to identify differential gene expression patterns that can distinguish normal
OSE and ovarian cancer cells. Mouse OSE cells models allow for studying the disease
progression in immune-competent animals and human OSE cell lines, especially ones
derived from the ovaries of women with family histories of breast/ovarian cancer serve as
a reliable, convenient, and clinically relevant model system for further delineation of
15

molecular pathways involved in the early stages of ovarian oncogenesis. Ultimately these
studies will be useful for the elucidation of molecular events associated with OSE cell
transformation and can help to generate transgenic models in which tumors initiate and
progress in situ.

However, how faithfully these in vitro cell cultures and transformation of human and
mouse OSE cells will represent the natural tumor development in vivo remains unknown.
Are cultures of OSE cells representative of the normal epithelium resident on the ovarian
surface? Does the malignant transformation of OSE in culture bear any relevance to the
malignant transformation that occurs in clinical disease?
c. Transgenic mouse models
Whereas epithelial cancers of many organ sites have been modeled successfully in
genetically engineered mice such as prostate, mammary gland and skin cancers, efforts to
model human ovarian cancer over the last decades have been quite disappointing. The
general lack of understanding of the biology and genetics underlying precursor lesions of
ovarian cancer is a stumbling block in creating mouse models for this disease. Therefore,
it is hard to determine which genetic pathways to target.

To date, most mouse models resulted in the rare sex cord–stromal tumors or granulosa
cell tumors but not epithelial tumors such as inhibin α

[Lau and Matzuk 1999], follitropin
16

receptor

[Danilovich et al., 2001], and Lats1 [St. John et al., 1999] knockout mice and
also luteinizing hormone (LH) [ Risma et al., 1995; Keri et al., 2000] , follicle
stimulating hormone (FSH)

[Rahman and Stratton 2001, Kumar et al., 1999]
progesterone

[Rahman and Huhtaniemi 2001], and AMH

[Dutertre et al., 2001] transgenic
mouse models.

Only during the past six years a few genetically engineered mouse models of epithelial
ovarian carcinoma were reported [Connolly et al., 2003; Flesken-Nikitin et al., 2003;
Orsulic et al., 2002]

The first successful genetically defined model of sporadic epithelial ovarian cancer
developing in immunocompetent mouse was reported by Connolly et al., [2003). They
generated transgenic mice expressing the early region of SV40 under the control of
themüllerian inhibiting substance receptor type II promoter. Approximately 50% of the
transgenic females developed bilateral ovarian tumors within 6–13 weeks of age.

Instead of introducing strong oncogene such as SV40 T large antigen, Flesken-Nikitin et
al., [2003] conditionally inactivated endogenous tumor suppressor genes pRb and p53
concurrently in the surface epithelium of the mouse ovary. Mice in which both the p53
and Rb genes had been inactivated developed epithelial ovarian tumors at 7 months old
17

and many of the mice developed ascites and metastatic disease in the liver and lungs.
Tumors were either well-differentiated serous neoplasms or poorly differentiated.

These models are useful to determine the functional contributions of individual pathways
to development of ovarian cancer and more importantly will be useful to test the efficacy
of interrupting specific pathway as a means to treat ovarian cancers so that development
pathway-targeted therapies.

However, most of the models successfully recapitulate the later stages of the disease and
the metastatic spread in the peritoneal cavity but they failed to accurately model the early
steps in transformation of the ovarian cells, and therefore do not permit the investigation
of EOC progression.
d. Cancer stem cells models
Recent evidence suggests that tumor growth is driven by a small subset of
cancer-initiating cells or cancer stem cells.

Researchers have succeeded in isolating subpopulations of tumor cells with stem cell-like
features and with tumorigenic capabilities from many solid tumor types including breast,
brain, prostate, colon, pancreatic, and hepatic carcinomas, melanoma, and a few other
tumor types [Al-Hajj et al 2003, Dalerba et al., 2007, Wicha et al., 2006].

18

The fact that the majority

ovarian cancer patients eventually recur with chemoresistance
and die of metastatic disease although they initially respond well to surgical resection
followed by platinumtaxane based chemotherapy and the multiple

histologic phenotypes
of ovarian cancer suggest that cancer stem cells may also be involved in ovarian tumors.
Multipotent cancer stem cells may explain the

histologic heterogeneity often found in
tumors.

Recent studies did suggest the presence of a side population of self-renewing cells with
differentiation potential and high tumorigenicity capacity in ovarian cancer [Bapat et al.,
2005; Szotek et al., 2006].

The identification of stem cells in ovary, and targeted genetic alterations in these cells,
should provide a new model to test the role of these cells in tumorigenesis of EOC and in
the development of chemoresistance.
19

Chapter 2

Breast Cancer Susceptibility Gene1 (BRCA1)

The vast majority of epithelial ovarian cancers are sporadic, while approximately 15% of
ovarian cancers are due to familial predisposition [Randall et al., 1998]. In 1994, a
specific gene on chromosome 17, named breast cancer susceptibility genes BRCA1, was
identified. It is the first major gene found to be associated with increased susceptibility to
development of familial breast and ovarian cancers. Approximately 90% of women with
familial breast and ovarian cancer carry germline mutations in the BRCA1 gene [Futreal
et al., 1994]. The estimated lifetime risk of ovarian cancer development by age 60 in a
BRCA1mutation carrier is 60% [Rahman and Stratton1998].

2.1 Biological Functions of BRCA1
Multitude of important biological functions have been identified for Brca1[Scully et al.,
2000; Weinert et al., 1998; Chu-Xia Deng et al., 2000; Hashizume et al., 2001; Ruffner et
al., 2001] such as:
DNA damage sensing and dsDNA damage repair
The involvement of BRCA1 in DNA-damage response pathways is well-documented.
BRCA1 interacts with various different proteins that are implicated in homologous
20

recombination and recognition of various different aberrations in DNA structure. For
instance: RAD51, a homolog of yeast RecA, which functions in homologous
recombination and DNA damage repair [Ogawa et al., 1993]; the mismatch-repair
enzymes MSH2 and MSH6, and the RAD50–MRE11–p95 complex, an essential
component of recombination-mediated repair of DNA double-stranded break [Zhong
1999]. Interactions with these proteins trigger downstream effector pathways leading to
either double stranded DNA break repair or cell-cycle checkpoint activation.
Checkpoint control of the cell cycle
BRCA1 works downstream of the checkpoint mechanisms that sense and signal either
DNA damage or problems with DNA replication during S phase. There is now good
evidence that BRCA1 is rapidly phosphorylated after DNA damage in dividing cells by
several protein kinases such as ATM , ATR or CHK2 [Shiloh, 2003]. One functional
consequence of this phosphorylation is the checkpoint control of cell cycle progression.
BRCA1 has been implicated in several different checkpoint events in the presence of
different kinds of DNA lesions.
Transcription regulation
While BRCA1 itself is not a sequence-specific DNA binding transcription factor, it can
interact with a variety of sequence-specific transcription factors, either stimulate or
inhibit their activity (e.g. c-myc, p53, STAT1 and others). BRCA1 can also interact with
components of the basal transcriptional machinery (e.g. RNA polymerase II),
21

transcriptional coregulators and chromatin-modifying proteins (e.g. p300/CBP), histone
deacetylases (HDACs) and other transcriptional regulatory proteins
Maintenance of chromosomal stability and genome integrity
Ample findings establish that BRCA genes are essential for preserving chromosome
structure, suggesting that they behave as caretakers, suppressing genome instability.

2.2 Possible mechanisms of tissue-specific tumor suppression

However, the above mentioned functions by themselves do not explain the strong
predisposition of mutations in this ubiquitously expressed gene to breast and ovarian
tissue-specific cancers, as opposed to a more generalized spectrum of cancers. Several
studies have proposed some possible mechanisms regarding the tissue-specific tumor
suppression:
a. Estrogen metabolites
Both breast and ovary are estrogen-responsive tissues. Some oestrogen metabolites can
adduct DNA, and so could act as tissue-specific carcinogens (so-called "remote
carcinogenesis")

[Fishman et al., 1995]. BRCA1 disruption may make breast or ovarian
cells more sensitive to estrogen metabolites.
b. Inhibitor of estrogen receptor (ER) signaling
Both breast and ovarian cancer are related to estrogen exposure. Consequently, many
22

reproductive, environmental, and lifestyle factors possibly affecting estrogen levels might
interfere with breast and ovarian cancer development. Estrogens stimulate BRCA1
expression [Spillman and Bowcock 1996; Marquis et al., 1995; Gudas et al., 1995], and
conversely accumulating evidence indicate that the BRCA1 protein inhibits Era-mediated
transcriptional pathways related to cell proliferation.mediated signaling [Xu et al., 2005;
Fan 2001]. During puberty and pregnancy, when estrogen levels are dramatically
increased, BRCA1 expression is also very high. This high cell proliferation may favor
accumulation of mutation in the absence of BRCA1. It was suggested that in addition to
repairing genetic damage and maintaining genomic stability during periods of rapid
cellular division and multiplication, BRCA1 may also protect the breast from
estrogen-induced genetic instability by suppressing ERα mediated pathways, inducing
differentiation. Therefore, it was suggested that estrogen signaling might be important in
BRC1A-associated tumorigenesis.BRCA deficiency may promote cell outgrowth or
survival, thus providing a possible explanation for the observed tissue specificity.
c. High proliferative activity in ovary and breast tissues.
Another plausible explanation involves the high proliferative activity in ovary and breast
tissues. In these two tissues, there are prolonged proliferative quiescence alternates with
periodic bursts of proliferation. The breast epithelium proliferates rapidly during puberty
and under the influence of oestrogenic hormones. Unlike many other rapidly proliferating
epithelial cells, such as those of the intestine or of the uterine endometrium, progeny of
23

this proliferative burst are retained within the breast epithelium. This is demonstrated by
the finding that breast lobules are clonal and an entire functioning mammary gland may

develop from a single cell [Kordon et al., 1998]. If, before lobular development, a lobular
precursor cell has acquired a cancer-predisposing mutation, the entire lobule would then
carry that mutation. This, in turn, could amplify the risk of acquiring a cancer-causing
mutation, and subsequent neoplastic progression.
d. Increased survival in breast and ovarian tissue
Elledge and Amon [2002] proposed that loss of BRCA1 leads to cell death by DNA
damage checkpoint or a severe decrease in proliferation in tissues other than the breast
and ovary, thus reducing the likelihood that additional mutations will occur that allow
tumor formation. Only breast and ovary tissues are able to survive for a prolonged period
of time in the absence of BRCA1. This hypothesized increased survival may be due to
either genetic factors or physiological environment that is unique to these particular
tissues. For example, the presence of survival factors in the form of hormones may have a
protective effect on individual cells. Other genetic alteration(s) occur that may overcome
cell-cycle arrest induced by BRCA1 deprivation.
e. Modulator of aromatase expression
Aromatase is the rate-limiting enzyme in estrogen biosynthesis. One recent study
revealed that BRCA1 inhibits estrogen biosynthesis through modulating aromatase
expression in ovarian granulosa cells and primary adipocytes (Hu et al., 2005).
24

f. Mammary stem cell regulator
The most recent study [Liu et al., 2008] suggested BRCA1 plays an important role in
regulating the differentiation of ER-negative stem/progenitor cells into ER-positive
epithelial cells. Therefore loss of BRCA1 function results in blocked epithelial
differentiation leading to the accumulation of genetically unstable undifferentiated breast
stem cells that served as prime targets for further carcinogenic events. Further studies are
needed to elucidate the merit of each explanation.

2.3 Unusual properties as a tumor suppressor gene
Despite strong evidence to support the view that BRCA1 is a classical tumor suppressor,
many observations seemed to be incompatible with this notion [Hakem et al., 1996;
Gowen et al., 1996]. Mice deprived of BRCA1 or BRCA2 succumb to early embryonic
lethality, and cells from these embryos exhibit a severe proliferative defect and induction
of p53-dependent cell cycle arrest [Ludwig et al., 1997; Sharan et al., 1997; Suzuki et al.,
1997]. Tumor cell lines lacking a functional BRCA1 gene have been extremely difficult to
establish. These cell lines surprisingly have very long doubling times. Brca1-deficient
mouse embryonic fibroblasts failed to grow, even when p53 gene was disrupted [Shen et
al., 1998].

25

Chapter 3

Introduction to Menstrual Cycle

3.1. Ovarian folliculogenesis, ovulation, and luteinization in humans

The Ovaries are the primary female reproductive organs. The fundamental reproductive
unit in the ovary is the ovarian follicle. Oocyte is enclosed within a follicle surrounded by
a cluster of endocrine cells. Follicles at various stages of development and regression are
present throughout the ovary during the reproductive years.

During early fetal development, primordial germ cells migrate into the ovarian cortex
where they divide rapidly and differentiate into oogonia. These oogonia become primary
oocytes by starting the first meiotic division, followed by an arrest in the prophase. When
a single layer of flattened pregranulosa cells forms around the naked primary oocyte, this
structure is called a primordial follicle. The primordial follicle pool forms the resting
stockpile of follicles in the ovary. Throughout reproductive life, dormant primordial
follicles are continuously recruited from the resting pool into the pool of growing follicles
to start growing.

26

Following recruitment of primordial follicles, the surrounding cells transform from
flattened into cuboidal granulosa cells leading to the formation of a primary follicle. As
the granulosa cells proliferate and finally form multiple cell layers surrounding the oocyte,
a secondary follicle is formed. The secondary follicle continues to grow and at this point
the oocyte has reached its maximal size. The first stage of follicular development is now
complete. This is the maximal degree of development found in the prepubertal ovary.

The second stage of follicular development takes place mainly after menarche. Past the
midpoint of each menstrual cycle, a small cohort of secondary follicles is recruited to
enter the next sequence. An antral follicle is formed when the follicular fluid secreted by
the granulosa cells form separate extracellular spaces and these separate spaces coalesce
into a single fluid-filled chamber, the antrum. The preovulatory follicle or antral follicle
represents the final stage of follicular development before ovulation. Such follicles have a
large antrum. The granulosa cells continue to proliferate and displace the oocyte into an
eccentric position forming a hillock. The oocyte now is surrounded by specialized
granulose cells, the cumulus oophorus which communicate with and nourish the oocyte
through gap junctions.

Now, the third and final stage of follicular development starts. Five to seven days after
the onset of menses, only one antral follicle is selected from its cohort of recruited
27

follicles and achieves dominance and continues growth while the remaining subordinate
follicles undergo atresia. LH surge initiates rupture of follicle and release of the oocyte
into the peritoneal cavity, termed as ovulation.

After ovulation, the wall of the ruptured follicle is converted into an endocrine structure
called corpus luteum which has a primary function of producing large amounts of
progesterone. If fertilization takes place, the corpus luteum persists and continues its
production of progesterone until the placenta is developed sufficiently to produce the
necessary hormones. Ultimately, the corpus luteum degenerates.

3.2. Ovarian granulosa cells functions

The major functions of granulosa cells include the production of steroids, as well as a
myriad of growth factors thought to interact with the oocyte during its development.

The steroid hormone estradiol is

produced primarily by ovarian granulosa cells. In
response

to the combined actions of the gonadotrophins FSH and LH released from the
pituitary gland, the ovarian granulosa cells undergo extensive proliferation and convert
androgens from the thecal cells to estradiol by aromatase to modulate the development
and release of mature oocytes.

28

In response to the surge of LH that triggers ovulation, granulosa

cells of preovulatory
follicles cease dividing and initiate

a program of terminal differentiation. These cells are
irreversibly

c converted into highly steroidogenic luteal cells that synthesize and secrete
high levels of progesterone.

3.3. Menstrual cycle

In both humans and other mammals, ovarian follicles development and ovulation is a
cyclic process. In primates this involves a menstrual cycle, while in non-primates an
estrous cycle is apparent. In humans, the menstrual cycle can be further divided into the
following three sequential phases:

Follicular phase, also known as proliferative phase, begins with the onset of menstrual
bleeding. This phase is characterized by the rapid growth and maturation of ovarian
follicles. As they mature, the follicles secrete increasing amounts of estradiol, which
stimulate the extremely rapid growth of endometrium in the uterus.

Ovulatory phase: one dominant follicle achieves the maximal degree of maturity and
ruptured and releases the egg.

29

During luteal phase, the oocyte travels down the fallopian tubes to the uterus and the
remains of the dominant follicle becomes a corpus luteum. Under the influence of
progesterone, the endometrium changes to prepare for potential implantation. The
secretory activity of endometrium gland is most prominent during this phase.

If implantation does not occur within approximately two weeks, the corpus luteum will
disintegrate causing sharp drops in levels of both progesterone and estrogen subsequently
causing the uterus to shed its lining in a process termed menstruation. Luteal phase ends
with the onset of menstrual bleeding.

3.4. Cyclic steroid hormone changes during menstrual cycle

A series of cycling changes in gonadal steroid and protein hormones production
characterize menstrual cycle. During the early follicular phase, the ovarian granulosa
cells gradually secrete increasing amount of estrogens and during the late follicular phase,
the estradiol and estrone production and plasma levels rise sharply, and just before the
ovulatory phase, they reach peaks. Progesterone remains at relatively low, constant levels.
During the succeeding ovulatory phase, plasma estradiol levels decrease from their peak.
During the luteal phase, levels of estradiol increases again but the most distinctive and
important feature of the luteal phase is a tenfold increase in the progesterone level.

30

3.5. Regulation of menstrual cycle

The endocrine control of the estrous cycle represents a complex interplay of positive and
negative feedback mechanisms by ovarian steroids acting at both the hypothalamic and
the anterior pituitary level. The key neuropeptide controlling reproductive function in all
vertebrate species is the pulsatility of gonadotropin-releasing hormone (GnRH) that is
synthesized by the hypothalamus and released into the hypophyseal-portal circulation.
GnRH drives the cyclic synthesis and secretion of LH and FSH from pituitary gland into
the systemic circulation.

One or two days before the onset of menses, FSH levels begin to rise. This period of FSH
drive is critical for stimulating the growth of recruited ovarian follicles. FSH enhances
the synthesis of estradiol. The increase in estradiol and most importantly inhibin
production by dominant follicles exerts negative feedback on FSH release, thereby
possibly limiting the growth of its sister follicles. The sharply increasing estradiol release
is the primary trigger for the gonadotropin LH surge, leading to ovulation.
A critical high and sustained level of plasma estradiol is required to elicit this positive
feedback on LH surge. The regulation of this positive feedback occurs at both the
pituitary and the hypothalamus.

31

After the LH surge, estradiol levels decline rapidly. LH stimulates luteinization of the
granulosa cells and formation of corpus luteum followed by steadily increasing
production of progesterone which together with inhibin exerts a negative feedback on the
pituitary gland. As a result, the levels of LH and FSH gradually go down. If pregnancy
does not occur, the corpus luteum disintegrates, and its secretion of progesterone and
estradiol stops completely in 14 days. By the twelfth postovulatory day, progesterone,
estradiol, and inhibin levels have fallen to levels low enough to release the negative
feedback form the pituitary gland so that FSH then begins to rise again and the next
whole cycle is repeated.

32

Chapter 4

A Novel Mouse Model for Epithelial Ovarian Cancer

4.1 A Mouse Model of Epithelial Ovarian Cystadenoma via Brca1 Mutation

A mouse model of epithelial ovarian cancer that can allow us to study the early stage of
this disease is highly needed and also the exact role of Brca1 plays in ovarian cancer
remains poorly understood. Our initial attempt was to generate an inherited mouse model
for epithelial ovarian cancer based on inactivation of Brca1 [Chodankar et al., 2005]. This
project was initiated by previous two graduate students Dr. Kwang and Dr. Chodankar
and I took over this project after they graduated.

Due to the association of Brca1 mutation and high risk of familial breast and ovarian
cancer, many researchers have disrupted Brca1 in the mouse using gene targeting [Gowen
et al., 1996; Hakem et al., 1996; Liu et al., 1996; Ludwig et al., 1997] in order to model
human diseases. Unfortunately, however, embryos homozygous for these mutations died
between embryonic days (E)5.5 -13.5.

33

In an attempt to circumvent embryonic lethality associated with germline mutation of
Brca1, we decided to utilize Cre-LoxP technology to conditionally mutate Brca1 in a
tissue-specific manner. Cre-loxP based endogenous gene inactivation is a powerful
approach to examine the function of a specific gene in a temporally and spatially
restricted manner. This approach is especially useful to overcome the embryonic lethality
that may result from gene knockouts targeting the germline [Nagy A. 2000].

4.1.1 Rationale of choosing target cells for Brca1 mutation

The first important issue needs to be considered before generating this mouse model was
we needed to decide which cells are the target cells for Brca1 mutation.

One possible mechanism of ovarian cancer predisposition in BRCA1 mutation carrier
proposes that inactivation of second allele in individuals with BRCA1 mutation leads to
tumor. This scenario is based on the assumption that BRCA1 is a classical tumor
suppressor gene which will need two hits in Knudsen’s hypothesis to get tumor
development. However, Brca1 is a very unusual tumor suppressor gene (TSG), unlike all
the other classical TSG, Brca1 is essential for cell survival. Many observations are at odd
with the notion that this gene is a TSG and raise questions about the exact mechanism by
which BRCA1 mutations cause tumor formation.

34

Therefore, we developed this mouse model based on the idea that BRCA1 acts in a cell
non-autonomous manner as opposed to acting as a classical tumor suppressor. We
proposed that Brca1 mutation in a different cell type from the cell of origin of epithelial
ovarian cancers (EOC) is responsible for influencing the growth and neoplastic
transformation of the cell of origin.

Based on two clues, one that epidemiological observations showing the strong association
of menstrual cycle activity with increased ovarian cancer risk, another that the
implication of Brca1 mutation in familiar ovarian cancer, we hypothesized that genetic
predisposition (Brca1) to epithelial ovarian cancer targets the same pathways involved in
the association between menstrual cycle activity and ovarian cancer risk.

Therefore, we chose ovarian granulosa cells as our target cells for Brca1 mutation given
the fact that these cells are essential for regulating menstrual cycle progression and the
role of this cycle in predisposition to ovarian cancer. We proposed granulosa cells
interact with the cell of origin of EOC and such interactions are regulated by Brca1. Thus
one of the consequences of Brca1 mutation would be disruption of these normal
intercellular interactions resulting in ovarian tumor predisposition.

4.1.2 Mouse models of ovarian granulosa cells specific Brca1 mutation

35

The mouse model carrying ovarian granulosa cells specific BRCA1 mutation by driving
Cre recombinase expression under the control of follicle-stimulating hormone receptor
promoter (Fshr) was established by previous graduate students. This trangenic line was
crossed with the mice carrying Brca1 alleles flanked by LoxP sites to achieve Brca1
knockout.

About 60% of mutant mice developed ovarian and/or uterine cystadenoma (Fig.1). These
tumors histologically resemble human serous cystadenomas, which are benign tumors
made up of the same cell type as ovarian serous carcinomas, the most common
histological subtype of epithelial ovarian cancer. These tumors did not originate from
granulosa cells but instead they were epithelial by nature. Moreover, these tumor cells
contain wild-type non-recombined Brca1 allele indicating the tumors were the indirect
results of Brca1 mutation [Chodankar et al., 2005].

4.2 Working hypothesis

We hypothesized that inactivation Brca1 gene in granulosa cells act at a distance to
influence tumorigenesis in cells from which ovarian epithelial tumors originate via a
mechanism(s) regulated by Brca1. One of the possible explanations for ovarian tumor
predisposition in this mouse model based on cell non-autonomous scenario is Brca1
36

inactivation in granulosa cells results in alterations in the dynamics of ovulatory cycle
which in turn influences ovarian epithelial tumor development. I verified this possibility
through the two following specific aims:

Aim1: to test the effect of Brca1 mutation on ovulatory cycle activities and if any of these
alterations predispose to tumor development

Aim2: Given that Brca1 mutation was targeted to hormone producing ovarian granulosa
cells, I investigated if there could be also changes in the hormone profile

37

Fig.1. Mutant mice develop ovarian cystadenomas. (A) normal ovaries and uterus
from a wild type mouse (B) mutant mice developed ovarian cystadnoma which is grossly
seen as a large translucent cyst filled with clear serous (C) Occasionally the mice
developed bilateral ovarian cysts (D) a hemorrhagic ovarian cyst. The contralateral ovary
is normal.
A B
C D
ovary
uterus
38

PART II

STUDIES ON THE MECHANISMS OF OV ARIAN
TUMOR PREDISPOSITION IN CONDITIONAL
BRCA1 KNOCKOUT MICE

Chapter 5

Role of Estrus Cycle Abnormalities in Ovarian Tumor
Predisposition

Abstract

One possible explanation could account for tumor predisposition in our mouse model
based on cell non-autonomous scenario is that Brca1 inactivation in granulosa cells
results in alterations in the dynamics of ovulatory cycle which in turn influences ovarian
epithelial tumor development.

39

The purpose of this chapter is to investigate whether BRCA1 disruption in ovarian
granulosa cells would affect reproductive cyclicity and if any alterations in this cycle play
a role in the predisposition to epithelial ovarian cancer.

The average length of each phase of estrus cycle was measured and compared in a group
of age-matched mutant versus normal mice when the mice were 3 months and 8 moths
old.

Brca1 mutant mice showed statistically significant increase in the length of proestrus
phase at 3 months old (1.64±0.11 (N=20) versus 1.33±0.09 (N=19), P=0.05) and more
pronounced at 8 months old (1.41±0.10 (N=16) versus 0.93±0.05 (N=14), P=0.001).

There was also an increase in the length ratio of proestrus/metestrus and diestrus
combined. This increase in the ratio is of borderline statistical significance at 3 months
old (Mutants 0.29 ± 0.029 (N=20) versus Wild type 0.22 ± 0.02 (N=19), P=0.068) and
statistically significant at 8 months old age group (0.34 ± 0.06 (N=16) versus 0.17 ± 0.02
(N=14), P=0.012)

Further analysis of metestrus and diestrus revealed that the shorter combined length was
mainly due to the decreases in the length of metestrus. As expected, the length ratio of
proestrus/metestrus was also bigger in mutant mice and statistically significant at 8
40

months old (1.12 ± 0.11 (N=16) versus 0.61 ± 0.09 (N=14), P=0.01). Moreover, these
estrus cycle abnormalities were associated with tumor predisposition. The proestrus/
metestrus and diestrus combined length ratio was increased in the mutant mice that
developed tumors and was of borderline statistical significance at both 3 months old (0.30
± 0.05 (n=4) versus 0.45 ± 0.03 (n=5), P=0.059) and 8 months old (0.29 ± 0.04 (n=5)
versus 0.59 ± 0.11 (n=5), P=0.056). This ratio was statistically significant higher when
the mice were 14 months old (0.29 ± 0.04 (n=5) versus 0.59 ± 0.11 (n=5), P=0.056)

Proestrus phase conresponds to estrogen-dominated follicular phase and metestrus and
diestrus pair is together considered as progesterone-dominated luteal phase in humans.

Given the well known influence of the menstrual cycle on risk of ovarian cancer and the
role of estrogen as an important risk factor and progesterone is a protective factor against
ovarian cancer, my studies suggested that mutant mice have a relative increase in
estrogen stimulation unopposed by progesterone. This raises the possibility that
predisposition to tumors of the reproductive tract and breast in individuals bearing
germline BRCA1 mutations may be due, at least in part, to menstrual cycle changes
driven by these mutations.

41

Introduction

5.1. Rodent estrous cycle

Rodents have similar follicular development as humans. Due to the ease of manipulation,
genetic tractability, mouse estrous cycle has been studied extensively as a model to
understand human reproductive cycle regulation.

By 6 to 8 weeks of age, young female mice are reproductively mature and exhibit estrous
cycles. Female mice are polyestrous. Duration of mice estrous cycle is much shorter
compared to humans. The mice estrous cycle occurs at an interval of 4 to 6 days
throughout the year [Nelson et al., 1982]. This cycle is divided into proestrus, estrus,
metestrus and diestrus. Proestrus is the period during which ovarian follicular
development occurs. As the mice enter estrus, the ovulation occurs. During metestrus, the
released eggs move through the oviduct into the uterus, and the corpora luteum form. If
pregnancy does not occur, the metestrus phase is ultimately followed by the last phase of
the estrus cycle, diestrus. Unfertilized eggs are eliminated, and new follicles begin to
undergo a rapid growth for the next ovulation. The proestrus and estrus phases together
constitute the follicular phase in humans; the metestrus and diestrus phases together
correspond to the post-ovulatory or luteal phase.
42

5.2 Cyclic steroid hormone changes during estrous cycle

Similar to human beings, circulating levels of 17b-estradiol (E2) and progesterone (P4)
produced by the ovaries fluctuate as a result of the estrous cycle. The increasing amounts
of E2 reach a peak near the end of proestrus, whereas during metestrus and diestrus, E2
levels subside and P4 levels rise up and then decline from proestrus to estrus [Walmer et
al.,. 1992; Fata et al.,. 2001; David et al., 1992] Another recent study showed E2 levels
are significantly higher during estrus than all other stages [Geoffrey et al., 2007]

5.3 Determination of estrous cycle stage by vaginal cytology

The vagina canal is lined with a stratified squamous epithelium that is highly sensitive to
estradiol thus undergoes typical cyclic histological changes i.e. proliferation, superficial
mucification, keratinization, and desquamation under influence of the cyclic fluctuation
of estrogen during estrous cycle [Putti and Varano 1979].

In proestrus phase, the superficial layers of stratified squamous epithelium of the vagina
undergo extensive mucification and consist of cuboidal mucinous cells. The squamous
epithelial cells underlying the mucinous cells show some cornification in late proestrus.
43

Estrus phase is marked by superficial cornification of the vagina squamous epithelium.
Metestrus phase is conventionally divided into two stages: metestrus I and metestrus II.
In metestrus I, the predominant feature is exfoliation of the cornified cells and the
appearance of leukocytes in the stroma. In metestrus II many more leukocytes appear.
During diestrus, the vaginal lumen is narrow and contains mainly leukocytes.

These standard changes in the cellular types are observable in the vaginal smears. Vaginal
smear cytology is now widely used to stage female animals and estimate the duration of
the estrous cycle in various mammalian species.

5.4 Papanicolaou stain (also PAP stain)

Pap stain is a polychrome staining histological technique developed by George
Papanikolaou, the father of cytopathology in 1930s. Originally it was used to detect the
presence of cancer cells from uterine cervical mucosa smear (PAP smear) slides.

Now PAP staining is widely used in cytology to stain smears prepared from various
bodily secretions; such as gynecological smears (Pap smears), washings, cerebrospinal
fluid etc to screen cancer cells. PAP smear is now routinely used in clinic to screen
cervical cancer. The classic form of PAP stain involves the following dyes:
44

• Haematoxylin: stains nuclear blue.
• First OG-6: counterstain
• Orange G: stains keratin.
• Eosin Azure 50 (EA50) : counterstain, comprising of three dyes:
o Eosin Y: stains the superficial epithelial squamous cells, nucleoli, cilia and
red blood cells
o Light Green SF yellowish: stains the cytoplasm of all other cells.
o Bismarck brown Y: stains nothing and in contemporary formulations it is
often omitted.
The combination of Orange G and EA50 give the subtle range of green, blue and pink
hues to the cell cytoplasm accroding to the degree of keratinization situation of the cell.
On a well prepared specimen, the cell nuclei are blue to black and the cytoplasm is
stained a subtle range of green, orange and pink depending on the cell maturity.
45

Materials and Methods
Animals
The mice were housed under standard 12 h light: 12h dark lighting conditions with
lighting automatically changes occurring at 0600 and 1800 h. They were fed mouse chow
and tap water ad libitum. Mice were cared for in accordance with institutional guidelines
under the protocols approved by the University of Southern California Institutional
Animal Care and Use Committee.

Genotyping
Total genomic tail DNA was extracted using standard Proteinase K protocol and
genotyped by PCR as previously described [Chodankar et al., 2005].

Collection of vaginal samples
Vaginal smears were taken daily at the same hour between 1000 and 1200h. Disposable
Pasteur pipets loaded with small amount of (about 100ul) sterile PBS were used to flush
the mouse vagina to get lavages. PBS was gently expelled into the vagina and aspirated
back into the tip twice, and then the cells were smeared on glass slides and allowed to air
dry followed by Papanicolou (PAP) staining. Daily vaginal lavages were obtained over a
continued 30-days period spanning several consecutive cycles

46

PAP Staining
The air dried slides are fixed in 95% alcohol for 1 minute then rehydrated through 80%
alcohol for 30 second, then water rinsed, Harris hematoxylin stain for 1 minute, , Bluing
agent for 1 minute, water rinsed. The slides are then dehydrated again through 80% and
95% alcohol solution, then counterstained in OG-6 for 1 minute, then 95% alcohol again
twice, then counterstained again in EA-50 solution. The slides are rinsed through 95%
and 100% alcohol, finally cleared with clearite 3 solution. The slides are mounted with
secure mount mounting medium and allowed to air dry in the hood.

Estrous cycle stage classification
Stages of estrous were determined by cytological evaluation

of vaginal smears slides
under a microscope. The following criteria were used for identification of cycle stages:
Proestrus, mixed populations of nucleated and cornified cells; Estrus, exclusively
cornified cells; Metestrus, leukocytes and cornified cells; Diestrus, predominantly
leukocytes.

Estrous cycle stage duration estimation
For comparing the length of each phase in a group of age-matched mutant versus normal
mice, the average length of each phase was calculated for each individual mouse and used
to calculate the average length ± SEM for the entire group. Only those females exhibiting
47

at least two cycles were used in the studies. Those mice which did not show any cyclic
changes in vaginal smears were excluded.

Preparation and histological evaluation of tissues
Tissues were collected at necropsy and were fixed in 10% (v/v) Neutral Buffered
Formalin overnight, transferred to 70% ethanol, and paraffin-embedded. 5μm sections
were cut and routinely stained with H&E.

Statistical analysis
Comparisons between averages were made using an unpaired, two-tailed Student's t test
assuming unequal variance. P≤ 0.05 was considered statistically significant.

48

Results

Determination of estrous cycle stages by cellular morphology in vaginal smears
stained by Papanicolaou (PAP) stain procedure
In order to study the effect of Brca1 inactivation in ovarian granulosa cells on estrous
cyclicity, I firstly needed a protocol to identify and differentiate the various stages of this
cycle. Cytological evaluation of vaginal smears is widely used to stage the estrous cycle
in various species. This technique involves an identification of three different cell types
i.e. leucocytes, intermediate immature epithelial cells, and cornified epithelial cells and
their relative quantities present at different phases of estrous cycle. Instead of using the
traditional staining technique, I used Papanicolaou (PAP) stain procedure. This method
has the advantage to allow the clear visualization of the degree of cellular differentiation
and maturity according to the degree of keratinization situation of the cell.

My first goal

was to establish a normal pattern of vaginal cytology for each

stage of the
estrous cycle.

To accomplish this goal, 3 moths old female wild-type mice

(n = 5) were
monitored daily by vaginal lavages for 2 weeks, and the slides were examined under
microscope. As shown in Fig. 2, different types of cells appear at each phase of estrus
cycle. These distinctive morphological features easily allow me to distinguish the various
stages. The relative ratios of these cells types are summarized in Table 2.
49

Fig. 2. Classification of estrous cycle stages by cellular morphology in vaginal
smears. A normal pattern of vaginal cytology for each stage of the estrus cycle was
established by obtaining daily vaginal smears from 3 month old female wild-type mice.
The vaginal smear slides were subsequently stained using the PAP stain procedure.
Diestrus smear contains predominantly leucocytes stained dark blue. Proestrus is marked
by the appearance of immature intermediate epithelial cells stained turquoise with dark
stained nuclei. The next proestrus/estrus transition stage contains an equal proportion of
intermediate immature and mature cornified epithelial cells stained pink to orange. Estrus
stage is characterized exclusively of mature cornified epithelial cells which stain orange.
Metestrus stage shows mature epithelial cells admixed with an increasing number of a
fresh batch of leucocytes. Metestrus is followed by diestrus and the cycle is started over
again.
Metestrus Estrus
Proestrus/Estrus
Diestrus
50

Table 2. Classification of stages of estrus cycle by cellular morphology in vaginal
smears
Stage of cycle Cell type
____________________________________________________

Leukocytes Nucleated epithelia Cornified epithelia
________________________________________________________________________

Proestrus 0 to + + to +++ (predominant) 0 to +
Proestrus/Estrus 0 + to ++ ++ to +++
Estrus 0 0 ++ to +++
Metestrus I + to ++ 0 ++ to +++
Metestrus II ++ to +++ + to ++ + to ++
Diestrus +++ 0 to + 0

51

Histological evaluation of ovaries at various stages of estrous cycle
To complement the PAP smear interpretation of staging estrous cycle, I sought to
correlate vaginal cytology in cycling female mice with ovarian histology and natural
circulating level of estradiol. During each stage of the cycle in a group of 6 to 7 month
old wild type mice, blood were collected and ovaries were dissected out followed by
H&E staining. This allowed vaginal cytology determination concurrently with serum
hormone measurement and ovary histology evaluation.
The

ovaries obtained from sexually mature cycling mice at any stages of cycle contain a

heterologous population of follicles at different stages of

development, among which
primary and secondary follicles, various

sizes of antral follicles, and corpus luteum could
all be observed [Pedersen and Peters, 1968; Butcher et al., 1984; Albina et al.,1999].
Hence there are no obvious estrous cycle stage-specific histological features of mouse
ovary. However, more large follicles are recruited between proestrus and estrus for the
next ovulation [Welschen and Dullaart, 1976; Hoak and Schwartz, 1980; Hirshfield and
DePaola, 1981] whereas diestrus

is marked by increased numbers of small primary and
secondary

follicles in both rats and mice [Pedersen and Peters, 1968; Albina et al.,1999].
Histological sections of ovary collected at preestrus and estrous phase (Fig. 3A and 3B,
respectively) revealed the presence of the characteristic large antral

preovulatory follicles
with its oocyte surrounded by a well-formed

cumulus.

52

Fig.3. Histological evaluation of ovaries collected from wild type mice at various
stages of natural estrous cycle determined by vaginal cytology. (A) Proestrus (B)
Estrous (C) Metestrus (D) diestrus (A, B) arrow head indicate large antral preovulatory
follicles. Such follicles are large and distended with considerable amounts of follicular
liquid forming a large antrum. The oocyte is displaced into an eccentric position and
surrounded by the cumulus oophoru (C, D) arrowhead indicate the abundance of small
follicles. In all figures, mixed populations of corpus lutea at different ages are present.
A B
C D
CL
CL
CL
CL
CL
CL
CL
53

Serum 17{ß}-estradiol (E2) levels at each stage of the estrous cycle
Blood were collected from 6 to 7 month old wild type mice at 4 stage of the cycle
determined by vaginal cytology and natural circulating serum concentration of estradiol
was measured by quantitative radioimmunoassay technique. As shown in Fig. 4, serum
estradiol level varies throughout the estrous cycle. The highest E2 concentration occurred
in proestrus and gradually declined in estrus and remained low at metestrus and diestrus.
This result is consistent with previous studies [Walmer et al., 1992; Fata et al., 2001;
David et al., 1992].
Taken together, both histological evaluation of ovaries and serum estradiol measurement
confirmed the reliability of my protocol of evaluation vaginal cytology via PAP stain in
staging mice estrous cycle. Thereafter I relied on PAP stain in my following further
studies.

54

0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Diestrus Proestrus Estrus Metestrus
Serum levle of E2 (pg/ml)

Fig.4. Serum levels of 17{ß}-estradiol (E2) as a function of the mice estrous cycle.
Blood was collected from wild type mice at 4 stages of natural estrous cycle determined
by vaginal cytology stained by PAP procedure. Serum concentration of E2 was measured
Data are presented as mean + SEM with 3 animals per stage.
55

Brca1 mutation in ovarian granulosa cells causes prolonged proestrus phase

I sought to measure and compare the length of each stage of the estrous cycle in a group
of age-matched mutant versus normal mice when they were 3 months. Their estrous
cycles were examined again when they grew up to 8 months in order to determine
whether the consequences of Brca1 mutations increase in severity with increasing age.
Daily vaginal smears were collected for about 30 consecutive days for all 3 age groups in
order to allow the mice to progress through at least three complete consecutive estrous
cycles.

I observed Brca1 mutant mice showed statistically significant increase in the length of
proestrus phase at both 3 months (1.64±0.11 (N=20) versus 1.33±0.09 (N=19), P=0.05)
and this increases became more pronounced at 8 months old (1.41±0.10 (N=16) versus
0.93±0.05 (N=14), P=0.001) (Fig. 5). No significant differences in the duration of other
phases or ovulatory frequency were observed at both age groups.

56

0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
3 months 8 months
Proestrus Stage Length (Days)
Wild type
Mutants

Proestrus Length (days) 3 months 8 months
Wild type 1.33 ± 0.09 (n=19) 0.93 ± 0.05 (n=14)
Mutants 1.67 ± 0.13 (n=20) 1.41 ± 0.09 (n=16)
P value 0.05 0.001

Fig. 5. Effects of Brca1 inactivation on duration of Proestrus stage. Data are
displayed as Mean ± SEM. Asterisks indicate significant differences (P <0.05) between
genotypes.
57

Brca1 mutation in ovarian granulosa cells causes increased Proestrus/Metesrus +
Diestrus and Proestrus/Metestrus length ratios

When comparing the length of metestrus or diestrus, initially no attempt was made to
distinguish metestrus from diestrus because the pair is together considered as
post-ovulatory phase, so length of metestrus and diestrus combined was compared. A
trend was found that the mutant mice showed shorter combined length. Although this
difference was not statistically significant, this trend was consistently found at both 3
moths and 8 months old age groups (Fig. 6).

So this led me to look at the ration of the average length of proestrus over metestrus and
diestrus combined length. As expected, this ratio was bigger in mutant mice and was of
borderline statistical significance at 3 months old (Mutants 0.29 ± 0.029 (N=20) versus
Wild type 0.22 ± 0.02 (N=19), P=0.068) and statistically significant at 8 months old age
group (0.34 ± 0.06 (N=16) versus 0.17 ± 0.02 (N=14), P=0.012) (Fig. 7)

Further analysis of metestrus and diestrus revealed that the shorter combined length was
mainly due to the decreases in the length of metestrus (Fig. 8). So I calculated the ration
of the average length of proestrus over metestrus phase alone. As expected, this ratio was
also bigger in mutant mice and statistically significant at 8 months old age group (Fig. 9.
58

3 months old: 1.13 ± 0.11 (N=20) versus 0.86 ± 0.1 (N=19), P=0.08; 8 months old: 1.12 ±
0.11 (N=16) versus 0.61 ± 0.09 (N=14), P=0.01).

59

0
1
2
3
4
5
6
7
8
3 months 8 months
Combined Metestrus and Diestrus Length
Wild type
Mutants

Metestrus + Diestrus
(Days)
3 months 8 months
Mutants 6.14 ± 0.34 (n=19) 5.31 ± 0.65 (n=16)
Wild type 6.39 ± 0.35 (n=20) 5.97 ± 0.55 (n=15)
P value 0.615 0.54
Fig. 6. Effects of Brca1 inactivation on duration of Metestrus stage and Diestrus stage
combined length. Data are displayed as Mean ± SEM.

60

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
3 months 8 months
Length ration of
Proestrus/Metesturs+Diestrus
Wild type
Mutants

Proestrus/Metestrus+Diestrus 3 months 8 months
Mutants 0.29 ± 0.029 (n=20) 0.34 ± 0.06 (n=15)
Wild type 0.22 ± 0.02 (n=20) 0.17 ± 0.02 (n=16)
P value 0.068 0.012

Fig. 7. Effects of Brca1 inactivation on length ratios of Proestrus/Mestrus+Diestrus.
Data are displayed as Mean ± SEM. P <0.05 indicates significant differences between
genotypes.
61

0
0.5
1
1.5
2
2.5
3 month 8 month
Metestrus Stage Length
Wild type
Mutants

Metestrus Length (Days) 3 month 8 month
Wild type 1.9 ±0.25 (N=19) 1.9 ± 0.25 (N=14)
Mutants 1.7 ±0.20 (N=20) 1.4 ± 0.15 (N=16)
P value 0.61 0.11

Fig. 8. Effects Brca1 inactivation on duration of Metestrus stage. Data are displayed
as Mean ± SEM.

62

0
0.2
0.4
0.6
0.8
1
1.2
1.4
3 months 8 months
Proestrus/Metestrus Length Ratios
Wild type
Mutant

Proestrus/Meterus 3 months 8 months
Wild type 0.86 ± 0.1 0.61 ± 0.09
Mutant 1.13 ± 0.11 1.19 ± 0.12
P value 0.07 0.01

Fig. 9. Effects of Brca1 inactivation on Proestrus/Metestrus length ratios. Data are
displayed as Mean ± SEM. P < 0.05 indicates significant differences between genotypes.

63

Altered estrous cycle is correlated with ovarian tumor predisposition in Brca1
mutant mice

Apparently, Brca1 mutation deregulated estrous cycle in our mouse model. The next
question I sought to address was whether these estrous cycle abnormalities were
associated with ovarian tumor predisposition seen in mutant mice.

After examination of the estrous cycle status in the group of 14 months old mice, they
were euthanatized. Ovaries & uterus were dissected out at necropsy for both gross
observation and histological evaluation. I separated the mutant mice into two groups
based on their phenotype: the group that developed ovarian and/or uterine tumors and the
group that didn’t, and I compared these two groups estrous cycle.

Similar to the comparisons made between mutant and wild type mice mentioned above,
the mutant mice that later developed tumors showed an increase in their proestrus stage
length, shorter metestrus phase, and shorter combined length of metestrus and diestrus
compared with those that didn’t develop tumors (Fig.10.) Although these differences
were not statistically significant, this general trends have been so consistently observed at
all three age groups and therefore they still merit being mentioned.

64

The proestrus/ metestrus and diestrus combined length ratio was increased in the mutant
mice that later developed tumors and was of borderline statistical significance at both 3
months old (mutants without cysts: 0.30 ± 0.05 (n=4) versus mutants with cysts: 0.45 ±
0.03 (n=5), P=0.059) and 8 months old (mutants without cysts: 0.29 ± 0.04 (n=5) versus
mutants with cysts: 0.59 ± 0.11 (n=5), P=0.056). This ratio was statistically significant
higher when the mice were 14 months old (mutants without cysts: 0.29 ± 0.04 (n=5)
versus mutants with cysts: 0.59 ± 0.11 (n=5), P=0.056) (Fig. 11)

The proestrus/ metestrus length ratio was also consistently higher in the mutant mice that
later developed tumors and was statistically significant at 3-6 months (1.17 ±0.16 (N=4)
versus 0.66 ±0.13 (N=5), P=0.03) (Fig. 12).

65

0
0.5
1
1.5
2
2.5
3
3 months 8 months 14 months
Proestrus Stage Length (Days)
MT
with
cysts
MT
withou
t cyst

Proestrus Length (Days) 3 months 8 months 14 months
Mutants with cysts
2.1 ± 0.34
(N=4)
1.6 ± 0.19
(N=5)
1.8 ± 0.16
(N=7)
Mutants without cyst
1.9 ± 0.28
(N=5)
1.5 ± 0.35
(N=5)
1.3 ± 0.08
(N=8)
P value 0.90 0.88 0.03

Fig. 10. Comparison of Proestrus stage length in mutants that developed tumors
versus those that didn’t. Data are displayed as Mean ± SEM.

66

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
3 months 8 months 14 months
Length Ratios of
Proestrus/Metestrus+Diestrus
Mutants
without
cysts
Mutants
with cysts

Proestrus/Metestrus+Diestrus 3 months 8 months 14 months
Mutants with cysts
0.45 ± 0.03
(n=5)
0.59 ± 0.11
(n=5)
0.27 ± 0.03
(n=7)
Mutants without cysts
0.30 ± 0.05
(n=4)
0.29 ± 0.04
(n=5)
0.19 ± 0.01
(n=8)
P value 0.059 0.056 0.04

Fig. 11. Comparison of the length ratio of Proestrus/Metestrus+Diestrus in mutant
mice that developed tumors versus those that didn’t. Data are displayed as Mean ±
SEM. P < 0.05 indicates significant differences between genotypes
67

0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
3 months 8 months 14 months
Proestrus/Metestrus length ratios
Mutants
with
cysts
Mutants
without
cysts

Proestrus/Meterus 3 months 8 months 14 months
Mutants with cysts
1.13 ± 0.13
(N=4)
1.24 ± 0.2
(N=5)
0.7 ± 0.08
(N=7)
Mutants without cysts
0.68 ± 0.09
(N=5)
0.83 ± 0.2
(N=5)
0.53 ± 0.05
(N=8)
P value 0.033 0.19 0.125

Fig. 12. Comparison of the length ratio of Proestrus/Metestrus in mutant mice
that developed tumors versus those that didn’t. Data are displayed as Mean ±
SEM.

68

Discussion

Despite extensive epidemiological evidence to support that menstrual cycle activity is a
strong risk factor for ovarian cancer, aside from ages at menarche and menopause, few
studies have examined the role and impact of other factors related to menstrual cycle in
relation upon ovarian cancer risk because many other factors besides physical factors can
also disturb the regulation of this cycle such as emotional or physical stress [Barsom et
al., 2004; Fenster et al., 1999 ], calorica deprivation，climate change, some chronic
inflammatory diseases and even lifestyle factors such as excess exercise, obesity,
smoking, and diet [Rowland et al., 2002; Solomon et al., 2001] and physical
translocation.

Moreover, there is great natural variability in the length of menstrual cycle between and
even within regularly cycling women in good health [Fehring et al., 2006]. These
variability are largely determined by variation in follicular phase whereas the length of
luteal phase is relatively more constant. However, the causes of this variability are
incompletely understood. Information on this issue is very scanty. Another factor that
hampers the investigation of the contribution of menstrual cycling to ovarian
carcinogenesis is the lack of appropriate experimental model systems.

69

Our mouse model is a very good model to study the effect of menstrual cycle on ovarian
cancer experimentally. Although the timing of estrus cycle in mice is substantially
compressed compared to human’s menstrual cycle, important landmarks of ovarian
development in mice are very similar

to those in the human.

Proestrus in mice is equivalent to the estrogen-dominated follicular stage in humans and
the pair of metestrus and diestrus is together considered as progesterone-dominated luteal
stage in humans [Smith et al., 1975; Schedin et al., 2000]. At late follicular phase right
before ovulation, follicular fluid that may contain estradiol in concentrations some 10,
000 times higher than circulating levels [Clement et al., 1987]. In mice, estrogen
gradually increases and reaches peak close to estrus [Geoffrey et al., 2007; Walmer et al.,
1992], whereas during metestrus and diestrus, estradiol levels subside and progesterone
levels significantly rises up due to the formation of corpus luteum and then decline from
proestrus to estrus.

In our mouse model, Brca1 conditional mutant mice displayed longer proestrus phase and
they also spend relatively smaller proportions of their reproductive years in either
metestrus phase alone or metestrus and diestrus combined, therefore increased
proestrus/metestrus and diestrus combined length ratio and increased proestrus/metestrus
length ratio.
70

Moreover, these estrous cycle abnormalities were associated with tumor predisposition.
These results implied that mutant mice had somewhat longer periods of exposure to
extremely high level of estradiol production unopposed by progesterone production.

The comparatively great variability of the estrous cycle emphasizes the obvious fact that
the reliability of any mean value depends on the amount and constancy of the data from
which it is derived. For this reason, I performed these studies in a large scale and over a
long period of time spanning 3 different age groups. Approximately 2670 slides in total
were made. The overall trend has been consistently observed throughout the whole
studies lending strong support to the observations made.

Epidemiologic evidence bears on the role of estrogen in the pathophysiology of ovarian
cancer. A number of recent large studies have consistently demonstrated that risk of
ovarian cancer increases with duration of postmenopausal estrogen replacement therapy
[Riman et al., 1998；Rodriguez et al., 2001； Purdie et al., 1996; Risch et al., 1996].
Conversely, the well-known strong protective effect of oral contraceptive bill use against
ovarian cancer is due to not only suppression of ovulation but also may be due in part to
reduction in endogenous estradiol production [Killick et al., 1987].

In contrast, progesterone has been suggested to have a protective role in the etiology of
ovarian cancer based on epidermiological oberservations. Consistent with these
71

observatiions, many studies including work from our laboratory have demonstrated the
role of progesterone in inhibiting cell proliferations, decreasing invasiveness, suppressing
transformation and inducing apoptosis [Bu et al., 1997; Keith Bechtel and Bonavida 2001;
McDonnel and Murdoch 2001; Yu et al., 2001; Zhou et al., 2002; Blumenthal, Kardosh et
al., 2003; Syed V, Ho et al., 2003].

Estrus cycle in laboratory mice occurs throughout the reproductive years and we have
observed some of our mice still underwent regular estrus cycle changes up to 22 months
old. My studies suggest that BRCA1 mutation causes lifelong disrupted estrous cycle
pattern throughout the whole reproductive life span. This longterm cumulative exposure
to long duration of high level of estrogen and relatively shorter duration of progesterone
along with other potential mechanisms culminates in ovarian cancer.

All taken together, given the well known influence of the menstrual cycle on risk of
ovarian cancer and the role of estrogen as an important risk factor in breast and ovarian
tumorigenesis and progesterone is a protective factor against ovarian cancer, this mouse
model suggested predisposition to tumors of the reproductive tract and breast in
individuals bearing germline BRCA1 mutations may be due, at least in part, to menstrual
cycle changes driven by these mutations.

72

These studies also strongly supported our hypothesis that BRCA1/Brca1 controls the
development of tumors of the reproductive tract in a cell non-autonomous manner as
opposed to acting as a classical tumor suppressor by indirectly altering the estrus cycle.

The underlying biological mechanisms which result in estrus cycles disruption by Brca1
mutation in ovarian granulosa cells are currently unclear. There are two theories
regarding the variables which control the duration of the follicular phase. One theory
states that the number of non-dominant small follicles which produce gonadotrophin
surge inhibiting or attenuating factor (GnSIF) [Fowler et al., 1996] may shorten or
lengthen the follicular phase by advancing or retarding the LH surge. Another theory
states that GnSIF doesn’t seem to play a role in the inhibition of the LH surge. [Levran et
al., 1995]. It is the decline or leveling off of estradiol production induces the LH surge.
The second theory more fits our mouse model. The Brca1 mutation causes sustained high
level of estradiol and therefore will take longer to decline which will delay the LH surge.
The exact mechanism is yet to be elucidated.

73

Chapter 6

The Effect of Brca1 Inactivation on Circulating
Level of Estradiol

Abstract

In the previous chapter, my studies show that the estrus cycles in the mutant mice were
deregulated. Given that BRCA1 mutation was targeted to estrogen hormone producing
tissue-ovarian granulosa cells, the purpose of this chapter is to determine if there could be
also alterations in the natural circulating levels of estradiol.

Proestrus phase is the one that displayed significantly changes in mutant mice and also
the phase when estrogen level increases dramatically and reaches its peak. Therefore,
blood was specifically collected at proestrus phase of estrus cycle as determined by
vaginal cytology. Using a quantitative radioimmunoassay technique, serum concentration
of estrogen was measured. No significant changes were observed between mutant and
wild type mice (wild type: 10.4 ± 2.1 pg/ml, N=12 versus mutants: 9.0 ± 1.1 pg/ml,
N=12). In conclusion, the natural circulating level of estradiol was not affected by Brca1
mutation in our mouse model.

74

Introduction

Ovarian granulosa cells are the principle source of the female sex hormones production
including estradiol and progesterone which control the estrus cycle and reproduction. The
circulating levels of estradiol and progesterone fluctuate in

dramatic patterns during
folliculogenesis and

ovulation.

Estradiol has been shown to stimulate proliferation of ovarian and breast carcinoma cells
and concurrently up-regulate BRCA1 mRNA and protein [Romagnolo et al, 1998].

75

Materials and Methods

Blood collection
Vaginal lavages were obtained in the morning at 10:00am and the slides were stained the
same day when the samples were collected and right after the slides were dried. Once a
mouse was identified at proestrus phase of the estrus cycle as indicated by vaginal
epithelial cytology, it was anesthetized immediately with ketamine and xylazine and
terminally bled by cardiac puncture under anesthesia. About 700μl to 1000μl blood were
collected and serum were separated and stored at -80˚C until assay.

Serum hormone measurement
Serum concentration of estradiol was measured at Clinical Endocrinology Laboratory at
USC using a quantitative radioimmunoassay technique. All final assays were

quantified
using a -counter. Average coefficient of variation is less than 15%.
76

Result

In order to determine whether the estrous cycle abnormalities also correlate with
alterations in the circulating level of reproductive hormone, I measured the natural
endogenous circulating level of hormone at 8 months old mice. Considering the great
physiological fluctuation of these hormones during estrus cycle, the blood needed to be
collected at the same estrus cycle stage in order to normalize the variations due to the
estrus cycle.

I was particularly focused on the proestrus stage. The first apparent reason was that this
was the stage that showed significantly changes in the mutant mice. The second reason is
that this is he stage that estradiol level is high and reaches its peak and therefore would be
easier to be measured. So proestrus is the best stage for estradiol measurement.
Progesterone is another hormone that is also relevant in this study, however, during
proestrus, progesterone drops precipitously to low basal level and the assay is not
sensitive enough to be able to detect such low level. In comparison, metestrus phase
would be a better phase because progesterone reaches its maximal level during this phase.

77

Collecting blood from both proestrus and metestrus phase for estradiol and progesterone
measurement respectively would be ideal. However, due to the large amount of serum
required for each individual hormone assay, I had to collect blood by cardiac puncture
which is a terminal procedure. The mice were not able to survive to allow blood
collection again. So I was only able to collect blood from one phase.

I decided to draw blood specifically at proestrus phase because many studies indicated
estrogen is implicated in the etiology of ovarian cancer. No significant changes in serum
concentration were observed between mutant and wild type mice (wild type: 10.4 ± 2.1
pg/ml, N=12 versus mutants: 9.0 ± 1.1 pg/ml, N=12).

78

Discussion

My results indicated the natural circulating level of estradiol was not affected by Brca1
mutation in our mouse model.

There are two possibilities. The first possibility is the changes in the level of circulating
reproductive hormone during one single estrous cycle might be too minimal to be
detected by our assay considering the relatively long tumor latency period of this mouse
model,. However, the cumulative gradual build-up throughout the whole reproductive life
of the mutant mice might play a role in tumor predisposition.

The second possibility is that the staging of estrus cycle was based on vaginal cytology
evaluation which although can tell the proestrus stage apart from the other phases, it can
not accurately differentiate the initial beginning proestrus, or middle proestrus or late
proestrus. The estradiol level can vary greatly during these minute intra-phase transitions.
The more accurate staging would require more frequent vaginal lavages sampling which
is not very practical considering the time involved in getting the smears, staining the
slides and reading the slides.

79

In order to rule out the variations due to estrus cycle stages, an alternative way is to
measure the circulating level of estradiol after exogenously administering pregnant mare
serum gonadotrophin (PMSG), a follicle stimulating hormone (FSH) analog which is
commonly used to induce superovulation. The super-physiological dose of PMSG can
override the recipient’s endogenous FSH to rapidly stimulate follicles growth, stimulate
estrogen production and induce superovulation. Although administration of PMSG is an
artificial manipulation and the hormone level measured doesn’t represent physiological
level, it has the advantages of more strictly synchronizing the estrus cycle and strongly
stimulate the burst of estrogen production therefore might be able to amplify the small
differences which otherwise is unable to be detected. It can also reflect the ovarian
response to gonadotrophin stimulation. Our preliminary data did show mutant mice
showed elevated level of estrogen after PMSG injection in comparison to wild type mice.
These studies will be performed in larger amount of mice before a solid conclusion can
be drawn.

80

Chapter 7

Cre expression in Pituitary Gland

Abstract

It is well known that BRCA1 mutations predominately predispose to breast cancer. In this
chapter, I investigated whether our mutant mice were predisposed to abnormalities in the
mammary glands in addition to the abnormalities in the reproductive organs mentioned
above. Mutant mice did display abnormal morphological features reminiscent of the
dilated ducts seen in the human fibrocystic disease. This observation led us to find out the
285bp regulatory region from FSHR promoter originally used to drive Cre in our Brca1
knockout mice is transcriptionally active in the mouse pituitary gland and causes Brca1
mutation in pituitary gland.

81

Introduction

Pituitary gland secrets a variety of hormones such as growth hormone (GH),
thyroid-stimulating hormone (TSH), adrenocorticotropic hormone (ACTH), follicle
stimulating hormone (FSH), luteinizing hormone (LH) to regulate many endocrine organs
throughout the body. The three hormones that are relavent in this study are:

1. FSH: essential for follicular growth until the antrum develops. FSH induces estrogen
and progesterone secretion at the level of the ovary by activating aromatase and p450
enzymes. FSH further induces the proliferation of granulosa cells and expression of
LH receptors on granulosa cells

2. LH: required for both growth of preovulatory follicles and luteinization and
ovulation of the dominant follicle. During the follicular phase of the menstrual cycle,
LH stimulates proliferation, differentiation, and secretion of follicular thecal cells;
and increases LH receptors on granulosa cells. The preovulatory LH surge drives the
oocyte into the first meiotic division and initiates luteinization of thecal and
granulosa cells.

3. Prolactin: initiates milk synthesis in the mammary glands.

82

The cyclic secretion of ovarian steroids is determined by cyclic secretion of pituitary
gonadotrophins FSH and LH. The same ovarian steroids also control the secretions of
gonadotrophins from the pituitary via positive and negative feedback mechanisms.

83

Materials and Methods

Histology evaluation of mammary glands
The fourth inguinal pair of mammary glands was removed, fixed in formalin, embedded
in paraffin. Sagittal sections were cut and stained by standard protocol of H&E

X-gal staining
Tissue was fixed in 4% paraformaldehyde in phosphate buffered solution (PBS, pH 7.2)
for 1 h or 2–3 h at 4˚C, washed in wash buffer (0.1 M NaH2PO4 (pH 7.3),2 mM MgCl2,
0.02%NP-40), incubated at 37˚C overnight in an X-gal stain solution (1 mg/ml X-gal , 5
mM K
3
Fe(CN)
6
,and 5 mM K
4
Fe(CN)
6
3H2O in wash buffer). Specimens were dehydrated
and embedded in paraffin sections (5 uM) were prepared and counterstained with Nuclear
Fast Red (Vector, Burlingame, CA) for 4 min.

84

Results

Mammary gland in virgin Brca1 conditional mutant mice displayed significant
enlarged dilated ducts

It is well known that BRCA1 mutations predominately predispose to breast cancer. Next I
investigated whether our mutant mice with the Brca1 conditional knockout were
predisposed to abnormalities in the mammary glands in addition to the abnormalities in
the reproductive organs mentioned above.

I examined the fourth mammary gland pair in mice with ages ranging between 18 and 20
months. All mice used in this study were virgins. Mammary glands from wild type mice
of this age consisted predominantly of the adipose tissue which replaces the interlobular
stroma seen in younger mouse and small inactive ducts plus few atrophic lobules. In
comparison, 3 out of 10 Brca1 mutant mice showed significant large dilated ducts
alternating with areas showing small, inactive ducts. These dilated ducts were filled with
lots of proteinaceous fluid. Representative sections are shown in Fig. 13. These
morphological features are reminiscent of the dilated ducts seen in the human fibrocystic
disease. These results are still preliminary because of the small number of mammary
glands so far examined.
85

Fig. 13. Significant enlarged dilated ducts in the mammary glands of 18 to 20
months old mutant virgin mice. (A) A wild type mouse mammary gland: Consist of
small ducts (arrow) plus few atrophic lobules (arrow) and the predominant adipose tissue
which replace the interlobular stroma seen in younger mouse. (B) In comparison, mutant
mice showed significant large dilated ducts.
WT
A
Normal ducts
Dilated ducts
MT
B
86

Cre recombinase is ectopically expressed in pituitary gland

The abnormalities in mammary gland can not simply be explained by Brca1 mutation in
ovarian granulosa cells only. This raised the possibility that pituitary gland might also be
implicated in addition to ovarian granulosa cells. This led me to speculate that the FSHR
promoter we chose to drive the Cre recombinase expression might also be expressed in
pituitary gland by a feedback mechanism.

To investigate the promoter activity in pituitary gland, we utilized Rosa26 (lacZ) reporter
strain carrying LacZ transgene [Soriano et al., 1999]. LacZ transgene is under the
control of ROSA26 promoter which is expressed ubiquitously. However, LacZ is
separated from this promoter by transcriptional stop sequences flanked by two lox sites,
therefore, lacZ remains silent. The lacZ can only be activated upon deletion of the floxed
stop sequences by Cre recombinase-mediated recombination. Therefore, the activation of
lacZ gene will mark Cre expressing cells. Once lacZ is activated, cells will express
β-galactosidase and will stain blue with X-gal. Male mice carrying the Fshr-Cre transgene
were crossed with female ROSA26(LacZ) reporter mice. Offspring of these crosses,
which carried both the Cre transgene and the ROSA26 LacZ transgene were analyzed by
X-gal staining for β-galactosidase activity.
87

The blue staining in both whole mount and tissue section of pituitary gland clearly
demonstrated the expression of functionally active FSHR-Cre (Fig. 14A and 14B). This
experiment was performed by a graduate student Hai-Yun Yen at Dr. Robert Maxson
laboratory at the University of Southern California.

To further determine if Brca1 is recombined by Cre recombinase, I extracted DNA from
pituitary gland and performed a recombination analysis by PCR using primers specific
for recombined Brca1. A recombined Brca1 PCR product of expected size was detected
(Fig. 14C – 14F).

88

684bp
530bp
470bp
Ovary
Pituitary
Gland
Spinal
cord
Ovary H
2
O
Pituitary
Gland
A B
C
E
D
F
1 2 3
1 2 3
1 2 3
1 2 3
Fcre,
Brca1-/-
Fcre,
Brca1-/-
Fcre,
Brca1-/-
Fcre,
Brca1+/-
Fcre,
Brca1+/-
Fig.14. Cre expression in pituitary gland. X-gal staining was used to detect Cre
expression in whole mount (A) and tissue section (B) of the pituitary gland of
FSHRCre/Rosa26 double transgenics. Cre-mediated Brca1 recombination was analyzed
by PCR (C-F). The 684 bp fragment represents the rearranged allele while the 470 bp
fragment represent the wild type allele and the 530 bp fragment represents the floxed, but
unrearranged allele. (C; D) Representative mice analyzed with primers for the rearranged
allele. (E; F) Representative mice analyzed with primers for the non-rearranged allele.

+/+: mice carrying 2 wild type alleles
-/- : mice carrying 2 floxed alleles
+/- : mice heterozygous for the floxed allele
89

Discussion

My preliminary studies demonstrated that the mammary glands in Brca1 conditional
knockout mice displayed abnormal morphological features reminiscent of the dilated
ducts seen in the human fibrocystic disease.

We can make crosses between Fshr-cre mice and R26R mice in order to determine
whether the truncated Fshr receptor used to drive Cre in our Brca1 knock out mice is
active in the breast. I also collected more mammary glands from additional mutant mice.
Laser microdissection of dilated ductal structures in these mammary glands followed by
PCR recombination assay will allow us to investigate whether they harbor a mutant Brca1
allele. If, as we anticipate, the results show that those structures only contain the wild
type allele, this will provide further support for our conclusion that loss of Brca1
expression in granulose cells can influence specific epithelial tissues cell
non-autonomously in addition to providing further support for the relevance of our model
to humans. This would be consistent with a recent report demonstrating that prophylactic
oophorectomy can protect women with germline BRCA1 mutations against the
development of breast cancer [Narod et al., 1998].

Our data are the first in vivo experiment to demonstrate that the 285bp regulatory region
from FSHR is transcriptionally active in the mouse pituitary gland. Cre functionally
90

mediates Brca1 recombination in pituitary gland. We currently don’t know which cells
within the pituitary gland expressed Cre because the distinction between 5 different cell
types is very weak by traditional histological methods. Specific immunohistochemical
analysis are needed according to the specific anterior pituitary secreted hormones.

91

Chapter 8

Distinguish between Effects of Brca1 inactivation in the
Pituitary versus Granulosa Cells via Ovarian Transplantation

Abstract

Our original intension was to knock out Brca1 exclusively in ovarian granulosa cells by
driving Cre recombinase expression under the control of ovarian granulosa cells specific
FSHR promoter. But recently, we realized there is a limited leakiness for this promoter
because we found ectopic expression of Cre in a discrete subset of pituitary cells.

The purpose of the present chapter is to distinguish any effect of Brca1 inactivation in the
pituitary from effects of inactivation in granulosa cells. We performed the ovary
transplantation procedure underneath the kidney capsule.

We transplanted ovaries from mutant mice into the renal capsule of wild type mice and
vice versa. Mutant ovaries transplanted into mutant mice and wild type ovaries
transplanted into wild type mice were also performed as methodological control.

92

All recipient mice were observed for cyclicity when they reached 3 to 6 months old. The
length of each phase was measured and compared between all different groups. Mutant
mice harboring grafted mutant ovaries showed increases in the length of proestrus and the
combined length of proestrus and estrus compared to wild type mice harboring grafted
wild type ovaries. These results were similar to those obtained in the mice that had not
undergone any transplantation procedure described as above. Transplantation of wild-type
ovaries into mutant mice result in rescues the abnormalities of estrus cycle. When wild
type mice were grafted mutant ovaries, their proestrus phase and the length of proestrus
and estrus combined became longer.

These results clearly demonstrated that a mutant grafted ovary alone could cause the
abnormalities of estrus cycle whereas mutant pituitary alone did not. These results not
only in an alternative way confirmed the estrus cycle changes in the mutant mice I
observed previously were true but also demonstrated that these changes were driven by
Brca1 mutation in granulosa cells as opposed to anterior pituitary cells.

These mice were sacrificed at the age of 14 to 17 moths old to allow phenotype
characterization. Ovarian grafts developed cystic tumors in 2/10 wild type mice receiving
mutant ovarian grafts and 2/11 mutant mice receiving mutant ovarian grafts but not in
wild type ovarian grafts.
93

Introduction

Female cancer patients are at risk of suffering infertility or endocrine disruption after
chemotherapy and/ or radiation therapy. Cryopreservation of ovarian tissue prior to
initiation of cancer therapies, followed by autologous transplantation is a promising
method for restoring fertility [Radford et al., 2001; Oktay et al., 2003]. This method has
important implications especially for pediatric cancer patients, who at present have no
other options available to them to preserve their fertility. These patients could have their
ovarian tissue cryopreserved and autotransplanted back years later at child-bearing age.

Ovarian transplantation is also currently used to prevent the extinction of endangered
species such as snow leopard et al. In the animal laboratory, ovarian transplantation is a
powerful tool to save valuable mutants and transgenic animals that have infertility not
caused by germ cells but may be due to inability to mate or incapacity to complete a
successful fertilization, gestation or parturition. In these cases, this technique has been
successfully used to transplant the ovaries from the affected female into healthy recipient
females to maintain mouse strains. The ovaries are commonly transplanted into the
following three sites:

1. Normal anatomical site. The advantage of this orthotopic transplantation is the
preserve of fertility following transplantation. However, the disadvantage is that it is
94

a less densely vascularised site. Thus revascularization of ovary graft is lower and
fewer follicles can survive compared to other site.

2. Heterotopical transplantation i.e. placing ovary in a different site in the body of the
recipient than its normal anatomical position.
a) Subcutaneous transplantation. Apparently this site has the most convenient
access to perform surgical procedure compared to other sites. But its
disadvantage is also very obvious. Skin has the lowest revasculariztion rate
therefore very few follicles can survive.
b) Kidney capsule. This site is most commonly used for research purpose because
kidney is rich of extensive capillary bed thus favors rapid graft
revascularization. Therefore ovarian follicles survival rates are greatly
increased.
95

Materials and Methods

Animals
All procedures were performed in accordance with institutional guidelines under the
protocols approved by the University of Southern California Institutional Animal Care
and Use Committee.

Ovarian transplantation under kidney capsule
At an average of 35 days, the recipient mice were anaesthetized by I.P. injection of
ketamine and xylazine. One dorsal incision was made into the skin and two dorsalateral
incisions were made into peritoneum to expose the ovaries and kidney. The recipients
were ovariectomized bilaterally and these ovaries were used as

donor tissue for grafting
to another recipient. Then a small incision was made into the kidney capsule and then
one whole intact fresh donor ovary was inserted beneath the capsule through the small
incision. The kidney was returned back to abdominal cavity. Finally the peritoneum and
skin incision was closed with suture and the animals were transferred back to the cages
for recovery. We transplanted mutant ovaries into wild type mice and vice versa. Mutant
ovaries transplanted into mutant mice and wild type ovaries transplanted into wild type
96

mice were also performed as methodological controls. For practical reasons, we only
transplanted one ovary.

Histological examination of ovarian graft and uterine tissues
Ovarian grafts and uterine horns were removed from recipient mice and were fixed in
10% neutral buffered formalin for 24 hours, embedded in paraffin wax, serial sectioned at
5μm thickness, and stained by standard protocols with hematoxylin and eosin.

97

Results

Validation of transplantation procedure
For my studies, I choose to transplant the entire intake ovarian tissue under the kidney
capsule (Fig. 15) because we studied the ovarian tumor formation in virgin but not
pregnant mice and achievement of the maximal ovarian graft survival is most critical for
our purpose.

I not only needed to ensure the mice could survive the surgical stresses but also more
importantly I needed to verify the ovarian graft was able to recover its function fully after
surgical procedures because ischaemia can still be a serious problem despite the natural
advantage of highly vascularization in kidney. Tissue ischameia and other factors such as
inflammation, tissue adhesion etc can potentially result in transplantation failure.
Therefore, I firstly needed to establish an optimal transplantation protocol trying to avoid
post-grafting ischemia which is the key factor influencing graft survival. To evaluate my
transplantation method, I used the three following indicators of functional recovery of
ovary graft:

1. The occurrence of vaginal opening.
It is an external indicator of estradiol response originating from a functional ovary.
98

As shown in Fig. 16A, vaginal opening occurred in this mouse after ovarian
transplantation indicating regained endocrine function of the ovarian transplants. In
contrast, the vagina of the mouse in Fig. 16B remained closed suggesting an
unsuccessful surgery.
2. Re-established estrus cycle.
If a mouse was able to reinitiate estrous cycle assessed by vaginal cytology after
operation (Fig. 17), I would consider it resumed normal uterine response to estrogen
fluctuation during ovarian cycle.
3. Presence of ovarian follicles in ovarian graft.
To further evaluate the developmental potential of ovarian graft, a recipient mouse
that displayed estrous cyclicity after transplantation was sacrificed and recovered
ovarian graft was processed for histological examination of the presence of follicles
and their developmental stages. As shown in Fig. 18. growing ovarian follicles at
different developmental stages including multiple corpora lutea were present. One
follicle had proceeded to apparently morphologically normal antral stage which
implies ovulation would occur shortly. The presence of corpus luteum indicates the
occurrence of ovulation. The full follicular development and corpora lutea formation
displayed by ovarian graft was comparable to those in situ mouse ovaries.

99

Taken together, fresh ovarian tissue fully functionally recovered its normal function after
transplantation into ovariectomized recipient mouse. Estrous cyclicity, folliculogenesis,
and ovulation were restored. Primordial follicles in ovarian graft can be recruited and
were able to develop into primary, antral stages, ovulatory stages and finally formed
corpora lutea under the capsule of kidney. These clearly demonstrated the feasibility and
success of my transplantation procedure.

100

Fig 15. Ovarian transplantation into renal capsule. Entire intact ovary from donor in
(A) was transplanted underneath the recipient’s renal capsule in (B)
A
B
101

Fig 16. Determination of estrus cyclicity in mice receiving ovarian transplants from
external vaginal examination. (A) The vagina is gaping indicating cyclicity and
subsequent vaginal cytology confirmed this recipient mouse was cycling. (B) The orifice
of the vagina remained tightly closed during one-month period of examination in this
mouse after transplantation. Subsequent vaginal cytology demonstrated this mouse lacked
cyclicity.
A cycling mouse A non-cycling mouse
A B
102

Fig. 17. Validation of an ovarian graft function by vaginal cytology. After 20 days of
postsurgical recovery, the daily vaginal smears were performed for 30 consecutive days
and then stained by PAP staining. The mouse went through 4 complete estrous cycles
which are quite comparable to the similar age of normal mice which have not undergone
surgeries. This demonstrated this mouse has successfully restored estrogenic activity.
These pictures were taken from one of these fours cycles: it had all the 4 stages of a
normal estrus cycle
Estrus Metestrus
Diestrus Proestrus
103

Fig. 18. Follicular development in a transplanted ovarian graft. One ovary graft
was recovered one month after transplantation and processed for H&E staining. (A)
Gross observation illustrating the appearance of a grafted ovary underneath kidney
capsule (B) Low magnification of section (C-E) 10 times magnification (F-H) 40
times magnification corresponding to C, D and E (C) A primary follicle (D) one large
antral follicle and enclosed oocyte are clearly visible and a corpus luteum (E) One
early antral follicle and two corpus luteum. P, primordial follicle; Pr, primary follicle;
A, antral; CL, corpus luteum.
A
B
H&E
C
D
E
H&E
H&E
H&E
F
G
H
CL
CL
kidney
Ovary graft
X 10 magnification X 40 magnification
ovary graft
kidney
uterus
oocyte
A
CL
Pr
A
A
104

Experimental design

Once I was confident with my protocol and surgical skills, I started performing ovarian
transplantation in a large amount of mice.

I chose to perform the surgeries when both the donor and recipient mice were only
around 30 to 35 days old. Even though performing transplantation on such young mice
caused greater technical difficulties compared to older mice, it is imperative to do so
because at this age the female mice are not sexually mature so they do not have ovarian
cycles yet. This means there should be no recipient’s own ovaries estrus cycle influence
prior to receiving ovary graft. Thus, any changes in the phenotype after transplantation
can be only attributed to the presence of the ovarian graft from the donor but not from
receipt’s own ovaries carried through before surgery.

We transplanted mutant ovary into wild type mice and vice versa. Mutant ovaries
transplanted into mutant mice and wild type ovaries transplanted into wild type mice
were also performed as methodological control. This allowed us to have 4 different
combinations regarding Brca1 mutation status in ovarian granulosa cells and pituitary
gland, i.e. Brca1 was mutated in both, in neither and in either tissue. Mutant mice
harboring grafted mutant ovaries have Brca1 mutation in both tissues, wild mice
105

harboring grafted wild type ovaries have normal Brca1 in both tissues whereas mutant
mice harboring grafted wild type ovaries have mutant pituitary glands but normal ovaries
and finally wild mice harboring grafted mutant ovaries have Brca1 mutation only in
ovarian granulosa cells. We allowed recipient mice adequate time to recover from surgery
and to restore functions of grafted ovary. The overall successful recovery rate of
transplanted ovaries varied from 48% to 64.3%.
106

Ovarian Transplant Experiment Design Outline

Recipient mice were bilaterally ovariectomized
(30 to 35 days old)

Donor ovary was transplanted under the kidney capsule

Mice were allowed to recover from surgery procedures

The length of estrous cycle stages was measured and compared
The mice lacking of cyclicity post surgery were eliminated from further studies
(3 to 6 months old)

Ovaries and uterus were collected and processed
for histology examination of tumors formation
(14 to 17 months old)

Brca1 mutation
Ovary Pituitary
WT with WT ovary graft - -
MT with MT ovary graft + +
WT with MT ovary graft + -
MT with WT ovary graft - +
107

Estrous cycle Characterization in transplanted ovaries

All recipient mice were observed for cyclicity by vaginal cytology as described in above
chapter when they reached 3 to 6 months old. The vaginal saline lavage was taken daily
over a month period. The mice that lacked evidence of cyclicity i.e. absence of any
periodic cornification of vagina indicating an unsuccessful surgery were eliminated from
further studies. The length of each phase was measured and compared between each
different groups and are presented in Table 3 and Fig. 19.

108

Days Proestrus Estrus PRO+E Metestrus Diestrus
WT with WT ovary graft
(N=15)
1.04 ±
0.08
0.98 ±
0.09
2.02 ±
0.11
1.53 ±
0.19
4.70 ±
0.67
MT with MT ovary graft
(N=14)
1.37 ±
0.08
1.24 ±
0.18
2.61 ±
0.17
1.57 ±
0.23
4.74 ±
0.6
MT with WT ovary graft
(N=10)
1.12 ±
0.06
0.90 ±
0.08
2.02 ±
0.07
1.63 ±
0.17
4.63 ±
0.75
Mean ±
SEM
WT with MT ovary graft
(N=9)
1.27 ±
0.07
1.23 ±
0.15
2.49 ±
0.17
1.57±
0.14
3.69 ±
0.52
WT with WT ovary graft vs
MT with MT ovary graft
0.01 0.21 0.01 0.90 0.78
MT with WT ovary graft vs
WT with WT ovary graft
0.47 0.50 0.96 0.69 0.95
MT with WT ovary graft vs
MT with MT ovary graft
0.02 0.10 0.01 0.83 0.91
WT with MT ovary graft vs
WT with WT ovary graft
0.35 0.96 0.63 0.98 0.22
WT with MT ovary graft vs
MT with MT ovary graft
0.05 0.18 0.03 0.85 0.22
P value
WT with MT ovary graft vs
MT with MT ovary graft
0.15 0.076 0.025 0.803 0.327
Table 3. Comparison the length of each phase of estrus cycle in mice receiving ovarian transplants

109

Fig. 19. Comparison of the Proestrus stage, Proestrus and Estrus combined length
in mice receiving ovarian transplants. * Indicates statistically significant difference.
0
0.5
1
1.5
2
2.5
3
3.5
Proestrus Proestrus+Estrus
WT with WT
ovary graft
MT with MT
ovary graft
MT with WT
ovary graft
WT with MT
ovary graft
*
*
*
*
*
*
*
110

I first compared the two control groups: mutant mice harboring grafted mutant ovaries
(MT↔MT) in which Brca1 is inactivated in both ovarian granulosa cells and pituitary
gland and wild type mice harboring grafted wild type ovaries (WT↔WT) in which Brca1
is intact in both tissues. MT↔MT mice showed increases in the length of proestrus and
the combined length of proestrus and estrus compared to WT↔WT mice. These results
were consistent to those obtained in the mice that had not undergone any transplantation
procedure described in previous chapter.

Next, I looked at the group of mutant mice harboring grafted wild type ovaries. So in this
group of mice, Brca1 is normal in their ovarian granulosa cells but inactivated in pituitary
gland. Transplantation of wild-type ovaries into mutant mice resulted in rescues the
abnormalities of estrus cycle because they showed no differences in the average length of
any phase of the cycle compared to the control group of wild type mice harboring
transplanted wild type ovaries. However, they did show a shorter proestrus phase and the
combination of proestrus and estrus compared to mutant mice harboring grafted mutant
ovaries. This result suggested that Brca1 inactivation in pituitary gland alone is not
sufficient to cause the abnormality of estrous cycle.

Next, I examined the group of wild type mice harboring transplanted mutant ovaries.
These mice have Brca1 mutation in the ovary but normal Brca1 in pituitary. This group
111

did not display differences in the average length of any phase of the cycle compared to
the control group of mutant mice harboring transplanted mutant ovaries but their
proestrus phase and the length of proestrus and estrus combined became longer than wild
type mice with grafted wild type ovaries. This implied that mutant ovary alone can
sufficiently cause the alteration in the estrus cycle

Finally, the transplantation of mutant ovaries into the wild type mice resulted in increases
in the length of both proestrus and estrus phase in comparison to mutant mice harboring
wild type ovary grafts. Although the differences for these two individual phases are not
statistically significant, the difference for the combined length of these two phases is.

Tumor development in transplanted ovaries

These mice were sacrificed when they reached 14 to 17 months old. Ovaries and uterus
were dissected out for phenotype characterization. Ovarian grafts developed cystic
tumors in 2/10 wild type mice receiving mutant ovarian grafts and 2/11 mutant mice
receiving mutant ovarian grafts (control group) but in all the other groups. As shown in
Fig. 20. These tumors are morphologically identical to the tumors that developed in the
mice that had not undergone any surgeries.

112

Another more interesting part that I wanted to look at is to see if uterine tumors
developed in these mice receiving grafted ovaries. If yes, as we anticipate, this will
provide strong support to our cell non-autonomous role of Brca1 mutation in ovarian
tumor predisposition in our mouse model. Although the uterus of the majority mice was
atrophied, a few mice did develop uterine cystadenoma. This atrophy of uterine was most
likely caused by less cellular proliferation stimulation they received from ovarian graft in
kidney compared to the ovary in normal anatomy site because the proliferation changes
from paracrine to endocrine. These results strongly support our cell non-autonomous role
of Brca1 because the uterus doesn’t contain the Brca1 mutant cells.
113

A B C
D E F
Fig. 20. Two representative wild type mice that received mutant ovary grafts
developed ovarian cystadenoma. (A) Gross observation of a huge ovarian cyst in one
mouse. (B) low magnification of H&E staining of cross section through the large
ovarian cyst underneath the kidney capsule (C) higher magnification of figure B.
(D-F) gross observation and histology examination of another representative mouse.
114

Discussion

A mutant grafted ovary alone could cause increases in the length of proestrus plus estrus
phase and/or proestrus stage alone compared to a wild type grafted ovary and the mutant
pituitary alone was not able to cause this alteration. These results not only confirmed the
estrus cycle changes in the mutant mice we observed previously were true in an
alternative way but also demonstrated that these changes were driven by Brca1 mutation
in granulosa cells as opposed to anterior pituitary cells.

115

PART III

GERMLINE REARRANGEMENT OF A
CONDITIONAL ALLELE VIA
PARENTAL TRANSMISSION
OF CRE RECOMBINASE

Chapter 9

Germline Recombination Caused by Diffusion of Cre
Recombinase Protein from Gonadal Interstitial Cells to Germ
Cells

Abstract

We previously crossed a transgenic line expressing Cre recombinase driven by a
truncated Fshr receptor promoter specific for ovarian granulosa cells in females with a
line carrying loxP sites flanking exon 11 of the Brca1 gene in order to knockdown this
gene in a granulosa cell specific manner.

116

In the present study, I show that maternal transmission of the Cre recombinase transgene
resulted in germline recombination at the loxP sites in 4 of 56 (7%) of the progeny. The
most likely explanation is that the Cre recombinase protein diffused from the maternal
granulosa cells to the adjacent oocytes. Similarly, diffusion of Cre recombinase from
testicular Sertoli cells to spermatocytes could result in paternal transmission of germline
Brca1 rearrangement, albeit at lower frequencies than maternal transmission.

117

Introduction

Cre-loxP based endogenous gene inactivation has become a powerful approach to
examine the function of a specific gene in a temporally and spatially restricted manner.

This is usually achieved by generating two genetically altered mouse lines, one that
harbors the Cre recombinase transgene under the control of a tissue-specific or
developmental stage-specific promoter and another that carries a gene sequence of
interest in which a critical functional region is flanked by two Cre-recognition-sequences
termed LoxP. Crosses between these two lines lead to recombination at the 2 LoxP sites
and deletion of the intervening sequence subsequently inactivation of this gene only in
cells expressing the tissue specific promoter.

Cre-mediated genetic recombination not following the expected pattern of inheritance or
tissue distribution has occasionally been observed. In these cases, rearrangement of a
conditional allele was observed in the absence of inheritance of a Cre recombinase
transgene [Cochrane et al., 2007; Hayashi et al., 2003; Ramirez et al., 2004; Sakai et al.,
1997; Vincent et al., 2003] Although the exact underlying mechanisms have not been
established, expression of Cre in either oocytes or gamete precursors has been suggested
as the most likely explanations.

118

Here we report on constitutional recombination of Brca1 allele at loxP sites driven by a
promoter expressed specifically in gonadal stromal cells, either ovarian granulosa cells in
females or testicular Sertoli cells in males.

119

Materials and Methods

Mouse Lines and Mating Scheme
The generation of the Fshr-Cre transgenic line was described previously [Chodankar et
al., 2005]. Mice carrying LoxP sequences flanking exon 11 of the Brca1 gene were
obtained from Dr. Chuxia Deng, National Institutes of Health [Xu et al., 1999]. Mating
units were arranged such that either only the male or only the female breeders carried the
Cre transgene.

Genotyping Analyses
Total genomic tail DNA was extracted using standard Proteinase K protocols. Offspring
of crosses between Fshr-Cre and Brca1
flox/flox
mice were genotyped for Cre-mediated
rearrangement as described [Chodankar et al., 2005].
120

Results

We used a truncated form of the follicle stimulating hormone receptor (Fshr) promoter
previously shown to be exclusively expressed in ovarian granulosa cells in
females .[Heckert LL, Griswold MD.1991.] in order to inactivate the Brca1 gene in those
cells [Chodankar et al.， 2005]. The specificity of FSHR-Cre expression was confirmed by
using Rosa26 Cre reporter mice [Soriano P. 1999] as well as by RT-PCR.

I came to suspect that germline transmission of a mutant Brca1 allele had taken place in
some of the mutant mice when I realized that the mutant allele was occasionally present
in tissues in which the Fshr promoter was previously shown not to be active, such as the
tails (Fig. 21). All viable pups in which such germline rearrangement had taken place
were Brca1
+/-
because of the embryonic lethality associated with total loss of Brca1. A
total of 4 mice out of 56 born from mothers carrying the Cre transgene showed germline
inactivation of one Brca1 allele.

This result led me to also test the possibility that recombination could be transmitted
through the male germline in this mouse model given that the Fshr promoter is also active
in testicular Sertoli cells. Such cells are closely associated with germ cells in testicular
seminiferous tubules.
121

Male transmission was indeed observed in a single case out of a total of 100 examined.
The difference in transmission frequency through female versus male germlines (4/56
versus 1/100) was of borderline statistical significance (2-sided P = 0.06, Fisher’s exact
test).

122

684bp
530bp
1 2 3 4 5 6 7 8 9 1011 12 13 14 15 16 17 18 19 20 21
A
B

Fig. 21. Genotyping analyses of 20 Fshr-Cre; Brca1 flox/flox mice. DNA was
extracted from the tail of 20 selected Fshr-Cre; Brca1 flox/flox mice. In each case, the
Fshr-Cre transgene was transmitted through the female germline. Approximately
100ng of each DNA sample was amplified enzymatically using primers specific for the
Brca1flox/flox allele in which recombination at the two LoxP sites either occurred (A)
or did not occur (B). DNA extracted from the ovary of a mutant mouse was used as
positive control in lane 1 and 21. The presence of 684 bp products in cases number 9,
13, 16, and 18 attests to the fact that germline recombination had taken place in these 5
mice. The presence of a 530 bp fragment representing the unrearranged allele attests to
the presence of amplifiable DNA in all cases.
123

Discussion

Although, the exact mechanism of germline deletion of Brca1 is still not clear, we have
excluded the possibility that the FSHR-Cre transgene integrates into an as yet
unidentified gene site active in oocyte resulting in Cre expression in female germline

The Fshr promoter used to drive Cre recombinase expression in this mouse model is
strongly expressed in ovarian granulosa cells, which literally surround oocytes within
ovarian follicles. Interactions between oocytes and granulosa cells through cytoplasmic
channels such as gap junctions have been described [Albertini et al., 1974; Amsterdam et
al., 1976]. Given that such gap junctions only allow passage of small molecules, it is
unlikely that the Cre protein, with a molecular weight of 38 kDa, entered the oocyte via
such junctions. However, this protein can readily diffuse through the cytoplasmic
membrane [Will et al., 2002] and could therefore have reached the intracellular
compartment of the oocyte by this mechanism. Thus, the most straightforward
explanation for our observation of female germline transmission of mutant Brca1 allele is
that molecules of Cre recombinase synthesized in granulosa cells diffused to the oocyte,
which is literally enclosed within a mass of granulosa cells with which it closely interacts
through intercellular gap junctions.

124

Potential reasons for apparent lower rate of diffusion from Sertoli cells compared to
granulosa cells include the fact that while a single oocyte is surrounded by multiple
granulosa cells in the ovary, the ratio of germ cells to stromal cells involved in
spermatocyte : Sertoli cell interactions in the testis is much smaller. Another possibility is
that germ cells expressing Cre recombinase may produce spermatozoa that are at a
disadvantage when competing for an egg in light of the fact that expression of Cre
recombinase in spermatids has led to chromosomal rearrangements and sterility [Schmidt
et al., 2000].

My results suggest that any mouse model based on expression of Cre recombinase using
a promoter active in gonadal stromal cells, especially ovarian granulosa cells but also, to
a lesser extent, testicular Sertoli cells, can potentially lead to germline recombination
resulting from diffusion of Cre to adjacent germ cells. This potential problem is likely to
be missed unless specifically tested for by performing genotyping analyses of pups aimed
at the detection of recombined (mutant) alleles. The results also underscore the possibility
that a similar phenomenon could also take place between adjacent somatic cell types.
Although this would not result in germline rearrangements, this could lead to
Cre-mediated rearrangements in cells other than target cells regardless of promoter
specificity.
125

PART IV

EPILOGUE

Summary

This study took advantage of a mouse model for epithelial ovarian tumor developed in
our laboratory based on conditional Brca1 knockout to study the role of Brca1 mutation
in ovarian tumor development in an in vivo natural immuno-competent animal system.

These results strongly support the epidemiological observation that menstrual cycle
activity is a risk factor for ovarian cancer and also reveal the importance of the stromal
cells-ovarian granulosa cells in influencing tumor formation by closely interact with the
cells origin of ovarian cancer at a distance via regulating menstrual cycle. I identified
Brca1 plays a critical role in modulating these cycles. Thus one of the consequences of
Brca1 mutation was the abnormalities of these cycle dynamics resulting mutant mice had
a relative increase in the estrogen exposure unopposed by progesterone which might be
responsible for influencing the growth and neoplastic transformation of the cell of origin.

126

This novel finding could provide a partial explanation for the largely unknown area about
this ubiquitously expressed gene i.e. issue-specific tumor development in its mutation
carriers.

Lastly, I also found germline mutation of conditional allele of Brca1 occasionally
occurred despite the tissue specificity of the promoter used to drive Cre recombinase. My
results suggest that any mouse model based on expression of Cre recombinase using a
promoter active in gonadal stromal cells can potentially have this problem.
127

Future Directions

1. To investigate if estrogen accelerates ovarian tumor formation in the
BRCA1-mutant mice

My above studies suggest mutant mice had a relative increase in estrogen stimulation
unopposed by progesterone. It would be interesting to directly test if estrogen treatment
can promote tumor formation. We can ectopically supply estrogen by implanting estrogen
soaked beads underneath the skin of mice for long term tumorigenesis study and compare
the tumor frequency in estrogen- or placebo treated mice after treatment.

2. To determine the potential disrupted downstream pathways caused by Brca1
inactivation in ovarian granulose cells via cDNA microarray and miRNA
microarray profiling.

2.1 cDNA microarray

I have already super-ovulated a group of 12-15months old normal and mutant mice by
injecting with exogenous hormone PMSG (pregnant mare serum gonadotrophin)
specifically at diesturs stage of estrus cycle in order to synchronize the follicles
128

development. Ovaries were surgically removed and quickly frozen on dry ice. The pure
granulosa cells will be harvested by laser capture microdissection followed by RNA
extraction.

We will use Affymetrix GeneChips Mouse Exon 1.0 ST Array (HuExon array) which will
enable two complementary levels of analysis—gene expression and alternative splicing.
The results will be analyzed using the Genetrix software program which is designed to
detect differences in individual signaling pathways between cells, and to allow mapping
of these genes to both established and user-defined signaling pathways. Differences in
global gene expression between normal versus mutant granulosa cells will be analyzed
will be further confirmed by other approaches such as real-time PCR and western etc and
further investigated in functional studies for their potential roles in ovarian tumorigenesis.

This global assessment of molecular changes in this model will lead to identification of
key gene functions implicated in human ovarian cancer. We are particularly interested in
genes involved in reproductive hormone signaling pathways and biosynthesis because the
potential role of these hormones in controlling ovarian tumorigenesis.

2.2 miRNA microarray profiling

129

MicroRNAs (miRNA) are a recently discovered class of noncoding

RNAs, which
negatively regulate gene expression either by degradation of

target mRNAs or by
posttranscriptional

repression. Numerous evidences point to a role for miRNAs

in the
etiology and pathogenesis of cancer by targeting oncogenes

or tumor suppressors.

A recent study showed that miR-34b

and miR-34c are dramatically down-regulated in
P53-null epithelial ovarian surface epithelium. Importantly, miR-34b and miR-34c

cooperate in suppressing proliferation and soft-agar colony

formation of neoplastic
epithelial ovarian cells [Corney et al., 2007].

It would be also interesting to examine the miRNAome alteration in our ovarian
granulose cells carrying Brca1 mutation. Total RNA can be isolated using a mirVana
miRNA

isolation kit (Ambion) that is highly enriched for mature miRNA

species and
hybridized to Microarrays containing probes against mouse

miRNAs to perform miRNA
microarray profiling

3. To develop mouse model of serous ovarian carcinomas by incorporating P53
mutation in mice carrying Brca1 mutation targeted to ovarian granulosa cells.

130

The long tumor latency period and benign nature of the epithelial tumors developed in
our mice carrying Brca1 mutation in ovarian granulosa cells suggest additional genetic
changes will be required in order to have malignant transformation.

Several studies suggest p53 are associated with progression of benign ovarian epithelial
cysts to malignant tumors in the general human population. In order to test for a
cooperative interaction between p53 and Brca1 in ovarian tumorigenesis, we propose to
generate Brca1 and p53 double knockout mice. Our hypothesis is that the introduction of
p53 mutation will promote the cystic lesions in our Brca1 knockout mouse models to
malignant transformation.

We will use the flipase/frt system (which works exactly like cre/lox except that flipase is
used instead of Cre recombinase and frt sequences are used instead of loxP) to target a
p53 mutation to müllerian tissues specifically which we believe to be the cell of origin of
ovarian tumors. This can be achieved by using the MisIIr promoter to drive flipase. The
proposed p53 frt/frt; MisIIr-flipase mice will be crossed with my current mice with Brca1
mutation in ovarian granulosa cells resulting in animals that carry a Brca1 mutation in
their granulosa cells and a p53 mutation in müllerian tissues.

131

A malignant and metastatic model will be more representative of natural human disease
and will be particularly attractive for developing and testing therapeutic approaches
aimed at the advanced stages of ovarian epithelial cancer in humans. Moreover, this
model can be used in investigating one of the key unanswered questions in this field, if
the ovarian carcinomas arise de novo or develop from preexisting precursor benign
tumors such as cystadenomas.

132

BIBLIOGRAPHY

Albertini DF, Anderson E. 1974. The appearance and structure of intercellular
connections during the ontogeny of the rabbit ovarian follicle with particular reference to
gap junctions. J Cell Biol 63:234-250.
Albina Jablonka-Shariff, Sapna Ravi, Angeline N. Beltsos, Laura L. Murphy, and Lisa M.
1999. Olson Abnormal Estrous Cyclicity after Disruption of Endothelial and Inducible
Nitric Oxide Synthase in Mice Biology of Reproduction 61: 171-177
Al-Hajj M, Wicha MS, Benito-Hernandez A. 2003. Prospective identification of
tumorigenic breast cancer cells. Proc Natl Acad Sci USA 100: 3983-3988

ALLEN 1922. The oestrous cycle in the mouse. Amer. Journ. Anat. 30: 297-346

American Cancer Society, 2003. Cancer facts and figures 2003. 5008.03. Atlanta, GA

Amsterdam.A. Josephs R, Lieberman ME, Lindner HR. 1976. Organization of
intramembrane particles in freezecleaved gap junctions of rat graafian follicles:
optical-diffraction analysis. J Cell Sci 21:93-105.

Bapat SA, Mali AM, Koppikar CB, Kurrey NK. 2005. Stem and progenitor-like cells
contribute to the aggressive behavior of human epithelial ovarian cancer. Cancer Res
65: 3025-3029
Bayani J, BrentonBarsom , S. H. , Mansfield , P. K. , Koch , P. B. , Gierach , G. , & West ,
S. G . 2004. Association between psychological stress and menstrual cycle characteristics
in perimenopausal women . Women’s Health Issues 14: 235-241

Bell DA. 2005. Origins and molecular pathology of ovarian cancer. Mod Pathol 18:
S19-S32

Biskind MS, Biskind GR. 1944. Development of tumors in the rat ovary after
transplantation into the spleen. Proc Soc Exp Biol Med 55:1761-79
Blumenthal M, Kardosh A, Dubeau L, Borok Z, Schönthal AH. 2003. Suppression of the
transformed phenotype and induction of differentiation-like characteristics in cultured
ovarian tumor cells by chronic treatment with progesterone. Mol Carcinog. 38: 160-169

133

Bu SZ, Yin DL, Ren XH, Jiang LZ, Wu ZJ, Gao QR, Pei G. 1997. Progesterone induces
apoptosis and up-regulation of p53 expression in human ovarian carcinoma cell lines.
Cancer. 79:1944-50
Butcher RL, Collins WE, Fugo NW. 1974. Plasma concentration of LH, FSH, prolactin,
progesterone and estradiol-17beta throughout the 4-day estrous cycle of the rat.
Endocrinology. 94:1704-1708
Butcher RL, Kirkpatric-Keller D. 1984. Pattern of follicular growth during the four-day
oestrus cycle in the rat. Biol Reprod 3: 280-286
Chatzistamou I, Schally A V , Pafiti A, Kiaris H, Koutselini H. 2002． Expression of
growth hormone-releasing hormone in human primary endometrial carcinomas, Eur. J.
Endocrinol. 147: 381-386.
Chodankar R, Kwang S, Sangiorgi F, Hong H, Yen HY, Deng C, Pike MC, Shuler CF,
Maxson R, Dubeau L. 2005. Cell-nonautonomous induction of ovarian and uterine serous
cystadenomas in mice lacking a functional Brca1 in ovarian granulosa cells. Curr Biol.
15:561-565

Cochrane RL, Clark SH, Harris A, Kream BE. 2007. Rearrangement of a conditional
allele regardless of inheritance of a Cre recombinase transgene. Genesis 45:17-20.

Connolly et al., 2003 D.C. Connolly, R. Bao, A.Y . Nikitin, K.C. Stephens, T.W. Poole, X.
Hua, S.S. Harris, B.C. Vanderhyden and T.C. Hamilton, Female mice chimeric for
expression of the simian virus 40 TAg under control of the MISIIR promoter develop
epithelial ovarian cancer, Cancer Res. 63: 1389-1397.

Corney DC, Flesken-Nikitin A, Godwin AK, Wang W, Nikitin AY. 2007. MicroRNA-34b
and MicroRNA-34c are targets of p53 and cooperate in control of cell proliferation and
adhesion-independent growth. Cancer Res. 67: 8433-8438

Danilovich N, Roy I, Sairam MR. 2001. Ovarian pathology and high incidence of sex
cord tumors in follitropin receptor knockout (FORKO) mice. Endocrinol 142: 3673-36

Dalerba P, Cho RW, Clarke MF. 2007. Cancer stem cells: models and concepts. Annu Rev
Med 58: 267–284

134

David K. Walmer, Mar A. Wrona, Claude L. Hughes, and Karren G. Nelson. 1992.
Lactoferrin Expression in the Mouse Reproductive Tract during the Natural Estrous
Cycle: Correlation with Circulating Estradiol and Progesterone Endocrinology. 131:
1458-1466

Deng CX, Brodie SG. 2000. Roles of BRCA1 and its interacting proteins. Bioessays.
2000 22:728-737.

Dubeau Louis. 1999. The cell of Origin of Ovarian Epithelial Tumors and the Ovarian
Surface Epithelium Dogma: Does the emperor Have No clothes? Gynecol. Oncol. 72:
437-442
Dutertre M, Gouedard L, Xavier F et al. 2001. Ovarian granulosa tumors express a
functional membrane receptor for anti-Mullerian hormone in transgenic mice. Endocrinol
142: 4040-4046
Elkas JC, Baldwin RL, Pegram M, Tseng Y , Slamon D, Karlan BY . 2002. A human
ovarian carcinoma murine xenograft model useful for preclinical trials. Gynecol Oncol 87:
200–206

Elledge, S. J. & Amon, A. 2002. The BRCA1 suppressor hypothesis: an explanation for
the tissue-specific tumor development in BRCA1 patients. Cancer Cell 1: 129-132

Fata JE, Chaudhary V & Khokha R. 2001. Cellular turnover in the mammary gland is
correlated with systemic levels of progesterone and not 17b-estradiol during the estrous
cycle. Biology of Reproduction 65: 680 -688
Fan, S., Ma, Y .X., Wang, C., Yuan, R.Q., Meng, Q., Wang, J.A., Erdos, M.,Goldberg, I.D.,
Webb, P., Kushner, P.J., et al., 2001. Role of direct interaction in BRCA1 inhibition of
estrogen receptor activity. Oncogene 20: 77-87
Fehring RJ, Schneider M, Raviele K. 2006. Variability in the phases of the menstrual
cycle. J Obstet Gynecol Neonatal Nurs. 35: 376-384
Fenster , L. , Waller , K. , Chen , J. , Hubbard , A. E. , Windham , G. C. , Elkin , E ., et al .
1999. Psychological stress in the workplace and menstrual function . American Journal of
Epidemiology 149: 127 -134
Fishman, J. , Osborne, M. P. & Telang, N. T. 1995. The role of estrogen in mammary
carcinogenesis. Ann. NY Acad. Sci. 768: 91-100
135

Flesken-Nikitin et al., 2003 A. Flesken-Nikitin, K.C. Choi, J.P. Eng, E.N. Shmidt and A.Y .
2003. Nikitin, Induction of carcinogenesis by concurrent inactivation of p53 and Rb1 in
the mouse ovarian surface epithelium, Cancer Res. 63: 3459-3463

Futreal PA, Liu Q, Shattuck-Eidens D, Cochran C, Harshman K, Tavtigian S, Bennett LM,
Haugen-Strano A, Swensen J, Miki Y , et al. 1994. BRCA1 mutations in primary breast
and ovarian carcinomas. Science. 266: 120-122.
Geoffrey A Wood, Jimmie E Fata, Katrina L M Watson and Rama Khokha 2007.
Circulating hormones and estrous stage predict cellular and stromal remodeling in murine
uterus. Reproduction. 133: 1035-1044

Gregoire et al., 1998 L. Gregoire, A. Munkarah, R. Rabah, R.T. Morris and W.D. 1998.
Lancaster, Organotypic culture of human ovarian surface epithelial cells: a potential
model for ovarian carcinogenesis, In Vitro Cell Dev. Biol. Anim. 34: 636-639.

Hashizume, R., Fukuda, M., Maeda, I., Nishikawa, H., Oyake, D., Yabuki, Y., Ogata, H.,
and Ohta, T. 2001. The ring heterodimer brca1-bard1 is a ubiquitin ligase inactivated by a
breast cancer derived mutation. J. Biol. Chem. 276: 14537-14540

Heckert LL, Griswold MD. 1991. Expression of follicle stimulating hormone receptor
mRNA in rat testes and Sertoli cells. Molec Endocrinol 5:670-677.

Godwin et al., 1992 A.K. Godwin, J.R. Testa, L.M. Handel, Z. Liu, L.A. Vanderveer, P.A.
1992. Tracey and T.C. Hamilton, Spontaneous transformation of rat ovarian surface
epithelial cells: association with cytogenetic changes and implications of repeated
ovulation in the etiology of ovarian cancer, J. Natl. Cancer Inst. 84: 592-601

Gowen LC, Johnson BL, Latour AM, Sulik KK and Koller BH. 1996. Brca1 deficiency
results in early embryonic lethality characterized by neuroepithelial abnormalities. Nat.
Genet. 12: 191-194

Gudas, J. M., Nguyen, H., Li, T., and Cowan, K. H. 1995. Hormone-dependent regulation
of BRCA1 in human breast cancer cells. Cancer Res., 55: 4561-4565

HakemR, de la Pompa JL, Sirard C, Mo R, WooM,Hakem A, Wakeham A, Potter J,
Hayashi S, Tenzen T, McMahon AP. 2003. Maternal inheritance of Cre activity in a
Sox2Cre deleter strain. Genesis 37:51-53.

Hirshfield AN, De Paolo LV . 1981. Effect of suppression of the surge of
follicle-stimulating hormone with porcine follicular fluid on follicular development in the
rat. J Endocrinol. 88: 67-71.
136

Hoak DC, Schwartz NB. 1980. Blockade of recruitment of ovarian follicles by
suppression of the secondary surge of follicle-stimulating hormone with porcine follicular
field. Proc Natl Acad Sci U S A. 77: 4953-4956.

Hu Y, Ghosh S, Amleh A, Yue W, Lu Y, Katz A et al. 2005. Modulation of aromatase
expression by BRCA1: a possible link to tissue-specific tumor suppression. Oncogene 24:
8343-8348

Keith Bechtel M, Bonavida B. 2001. Inhibitory effects of 17beta-estradiol and
progesterone on ovarian carcinoma cell proliferation: a potential role for inducible nitric
oxide synthase. Gynecol Oncol. 82:127-38

Keri RA, Lozada KL, Abdul-Karim FW, Nadeau JH, Nilson JH. 2000. Leutinizing
hormone induction of ovarian tumors: oligogenic differences between mouse strains
dictates tumor disposition. Proc Natl Acad Sci USA 97: 383-387

Kordon, E. C. & Smith, G. H. 1998.An entire functional mammary gland may comprise
the progeny from a single cell. Development 125: 1921-1930

Kumar TR, Palapattu G, Wang P et al. 1999. Transgenic models to study gonadotropin
function: the role of follicle-stimulating hormone in gonadal growth and tumorigenesis.
Mol Endocrinol 13:851-65

Kusakari T, Kariya M, Mandai M, Tsuruta Y , Hamid AA, Fukuhara K, Nanbu K,
Takakura K, Fujii S. 2003. C-erbB-2 or mutant Ha-ras induced malignant transformation
of immortalized human ovarian surface epithelial cells in vitro, Br. J. Cancer 89:
2293-2298

Oktay K, Buyuk E, Rosenwaks Z & Rucinski J .2003. A technique for transplantation of
ovarian cortical strips to the forearm. Fertility and Sterility. 80: 193-198

Orsulic S, Li Y , Soslow RA, Vitale-Cross LA, Gutkind JS, Varmus HE. 2002. Induction
of ovarian cancer by defined multiple genetic changes in a mouse model system. Cancer
Cell. 1:53-62

Rahman NA, Huhtaniemi IT. 2001. Ovarian tumorigenesis in mice transgenic for murine
inhibin alpha subunit and promoter-driven Simian Virus 40 T-antigen: ontogeny,
functional characteristics and endocrine effects. Biol Reprod. 64: 1122-1130

137

Reitmair A, Billia F, Firpo E, Hui CC, Roberts J, Rossant J and Mak TW. 1996. Cell.
85: 1009-1023

Ruffner, H., Joazeiro, C.A., Hemmati, D., Hunter, T., and Verma, I.M. 2001.
Cancer-predisposing mutations within the RING domain of BRCA1: loss of ubiquitin
protein ligase activity and protection from radiation hypersensitivity. Proc. Natl. Acad.
Sci. USA 98: 5134-5139

Lauchlan SC. 1994. The secondary müellerian system revisited. Int. Gynecol Pathol. 13:
73-79

Liu, G. Yang, J.A. Thompson-Lanza, A. Glassman, K. Hayes, A. Patterson, R.T. Marquez,
N. Auersperg, Y . Yu, W.C. Hahn, G.B. 2004. Mills and R.C. Bast Jr., A genetically defined
model for human ovarian cancer, Cancer Res. 64: 1655-1663

Liu CY, Flesken-Nikitin A, Li S, Zeng Y and Lee WH. 1996. Inactivation of the mouse
Brca1 gene leads to failure in the morphogenesis of the egg cylinder in early
postimplantation development. Genes Dev. 10: 1835- 1843.

Ludwig, T., Chapman, D.L., Papaioannou, V.E., and Efstratiadis, A. 1997. Targeted
mutations of breast cancer susceptibility gene ho mologs in mice: lethal phenotypes of
Brca1, Brca2, Brca1/Brca2, Brca2/p53 nullizygous embryos. Genes Dev. 11: 1226-1241

Tirkkonen M, Johannsson O, Agnarsson BA, Olsson H, Ingvarsson S, Karhu R, Tanner M,
Isola J, Barkardottir RB, Borg A, Kallioniemi OP. 1997. Distinct somatic genetic changes
associated with tumor progression in carriers of BRCA1 and BRCA2 germ-line mutations.
Cancer Res. 57:1222-1227.

Lau A, Matzuk MM. 1999. Transgenic models in the study of reproduction. Gravitational
Space Biol Bull 12:75-82

Liu S, Ginestier C, Charafe-Jauffret E, Foco H, Kleer CG, Merajver SD, Dontu G, Wicha
MS. 2008. BRCA1 regulates human mammary stem/progenitor cell fate. Proc Natl Acad
Sci U S A. 105: 1680-1685

Maeda, KI.; Ohkura, S.; Tsukamura, H. 2000. Physiology of reproduction. In: Krinke GJ. ,
editor. The laboratory rat. Academic Press; San Diego, Calif 145-176

138

Marquis, S. T., Rajan, J. V., Wynshaw-Boris, A., Xu, J., Yin, G. Y., Abel, K. J., Weber,
B. L., and Chodosh, L. A. 1995. The developmental pattern of BRCA1 expression
implies a role in differentiation of the breast and other tissues. Nat. Genet 11: 17-26

McDonnel AC, Murdoch WJ. 2001. High-dose progesterone inhibition of urokinase
secretion and invasive activity by SKOV-3 ovarian carcinoma cells: evidence for a
receptor-independent nongenomic effect on the plasma membrane. J Steroid Biochem
Mol Biol. 78: 185-91

V . Mobus, C.D. Gerharz, U. Press, R. Moll, T. Beck, W. Mellin, K. Pollow, P.G.
Knapstein and R.1992. Kreienberg, Morphological, immunohistochemical and
biochemical characterization of 6 newly established human ovarian carcinoma cell lines,
Int. J. Cancer 52: 76-84

Molthoff, J.J. Calame, H.M. Pinedo and E. 1991. Boven, Human ovarian cancer
xenografts in nude mice: characterization and analysis of antigen expression, Int. J.
Cancer 47: 72-79

Montes GS, Luque EH. 1988. Effects of ovarian steroids on vaginal smears in the rat.
Acta Anat (Basel).133: 192-199

Nagy A. 2000. Cre recombinase: the universal reagent for genome tailoring. Genesis 26:
99-109

Narod SA, Risch H, Moslehi R, Dørum A, Neuhausen S, Olsson H, Provencher D, Radice
P, Evans G, Bishop S, Brunet JS, Ponder BA. 1998. Oral contraceptives and the risk of
hereditary ovarian cancer. Hereditary Ovarian Cancer Clinical Study Group. N Engl J
Med. 339:424-428.

NelsonJF, Felicio LS, Randall PK, Sims C, Finch CE 1982. A longitudinal study
ofestrous cyclicity in aging C57BL/6J mice. I. Cycle frequency, length and vaginal
cytology. Bioi Reprod 27: 327-339

Nilsson EE, Westfall SD, Mcdonald C, Lison T, Sadler-Riggleman I, Skinner MK. 2002.
An in vivo mouse reporter gene (human secreted alkaline phosphatase) model to monitor
ovarian tumor growth and response to therapeutics. Cancer Chemother Pharmacol. 49:
93-100

Nitta, H. Katabuchi, H. Ohtake, H. Tashiro, M. Yamaizumi and H. 2001. Okamura,
Characterization and tumorigenicity of human ovarian surface epithelial cells
immortalized by SV40 large T antigen. Gynecol. Oncol. 81: 10-17
139

Ogawa T, Shinohara A, Nabetani A, Ikeya T, Yu X, Egelman EH and Ogawa H. 1993.
Cold Spring Harb Symp Quant Biol. 58: 567 ± 576

Orsulic S, Li Y, Soslow RA, Vitale-Cross LA, Gutkind JS, Varmus HE. 2002. Induction
of ovarian cancer by defined multiple genetic changes in a mouse model system. Cancer
Cell. 1: 53-62
Pedersen RA, Peters H. 1968. Proposal for a classification of oocytes and follicles in the
mouse ovary. J Reprod Fertil; 83: 555-557

Radford JA, Lieberman BA, Brison DR, Smith AR, Critchlow JD, Russell SA, Watson
AJ, Clayton JA, Harris M, Gosden RG & Shalet SM 2001. Orthotopic reimplantation of
cryopreserved ovarian cortical strips after high-dose chemotherapy for Hodgkin’s
lymphoma. Lancet 357: 1172-1175e

Rahman N., Stratton M. 1998. The genetics of breast cancer susceptibility. Annu. Rev.
Genet. 32: 95-120

Ramirez A, Page A, Gandarillas A, Zanet J, Pibre S, Vidal M, Tusell L. 2004. A keratin
K5Cre transgenic line appropriate for tissue-specific or generalized Cre-mediated
recombination. Genesis 39:52-57.

Randall TC, Bell KA, Rebane BA, Rubin SC, Boyd J. 1998. Germline mutations of the
BRCA1 and BRCA2 genes in a breast and ovarian cancer patient.Gynecol Oncol. 70:
432-434.

Riman T, Persson I, Nilsson S 1998: Hormonal aspects of epithelial ovarian cancer:
review of epidemiological evidenc. Clin Endocrinol (oxf) 49: 695-707
Riman T, Persson I, Nilsson S 1995. Hormonal aspects of epithelial ovarian cancer:
review of Risma KA, Clay CM, Nett TM, Wagner T, Yun J, Nilson JH. Targeted
overexpression of leutinizing hormone in transgenic mice leads to infertility, polycystic
ovaries and ovarian tumors. Proc Natl Acad Sci USA. 92: 1322-1326
K.F. Roby, C.C. Taylor, J.P. Sweetwood, Y . Cheng, J.L. Pace, O. Tawfik, D.L. Persons,
P.G. Smith and P.F. Terranova. 2000 Development of a syngeneic mouse model for events
related to ovarian cancer. Carcinogenesis. 21: 585-591.
140

Rowland , A. S. , Baird , D. D. , Long , S. , Wegeinka , G. , Harlow , S. D. , Alavanja , M .,
et al . 2002 . Influence of medical of medical conditions and lifestyle factors on the
menstrual cycle . Epidemiology 13: 668 - 674

Sharan, S.K., Morimatsu, M., Albrecht, U., Lim, D., Regel, E., Dinh, C., Sands, A.,
Eichele, G., Hasty, P., and Bradley, A. 1997. Embryonic lethality and radiation
hypersensitivity mediated by Rad51 in mice lacking Brca2. Nature 386: 804-810

T.J. Shaw, M.K. Senterman, K. Dawson, C.A. Crane and B.C. 2004. Vanderhyden,
Characterization of intraperitoneal, orthotopic, and metastatic xenograft models of human
ovarian cancer, Mol. Ther. 10: 1032-1042

Sakai K, Miyazaki J-I. 1997. A transgenic mouse line that retains Cre recombinase
activity in mature oocytes irrespective of the cre transgene transmission. Biochem
Biophys Res Commun 237:318-324.

Schmidt EE, Taylor DS, Prigge JR, Barnett S, Capecchi MR. 2000. Illegitimate
Cre-dependent chromosome rearrangements in transgenic mouse spermatids. Proc Natl
Acad Sci USA 25:13702-13707.

Shen SX, Weaver Z, Xu X, Li C, Weinstein W, Guan XY, Ried T and Deng CX. 1998. A
targeted disruption of the murine Brca1 gene causes gamma-irradiation hypersensitivity
and genetic instability. Oncogene. 17: 3115-3124.

Shih Ie M, Kurman RJ. 2004. Ovarian tumorigenesis: A proposed model based on
morphological and molecular genetic analysis. Am J Pathol 164 1511-1518

Shiloh, Y . 2003. ATM and related protein kinases: safeguarding genome integrity. Nature
Rev. Cancer 3: 155-168
Solomon , C. G. , Hu , F. B. , Dunaif , A. , Rich-Edwards , J. , Willet , W. C. , Hunter , D.
J ., et al . 2001. Long or highly irregular menstrual cycles as a marker for risk of type 2
diabetes mellitus . Journal of the American Medical Association 286: 2421-2426

Soriano P. 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat
Genet 21:70-71
Spillman MA, Bowcock AM. 1996. BRCA1 and BRCA2 mRNA levels are coordinately
elevated in human breast cancer cells in response to estrogen. Oncogene 113:1639-1645
St. John MAR, Tao W, Fei X et al. 1999. Mice deficient of Lats1 develop soft-tissue
sarcomas, ovarian tumours and pituitary dysfunction. Nature Genet 21: 182-186
141

Scully R. 2000. Role of BRCA gene dysfunction in breast and ovarian cancer
predisposition. Breast Cancer Res. 2: 324-330.

Scully R, Puget N, Vlasakova K. 2000. DNA polymerase stalling, sister chromatid
recombination and the BRCA genes. Oncogene. 19: 6176-6183.

Suzuki, A., de la Pompa, J.L., Hakem, R., Elia, A., Yoshida, R., Mo, R., Nishina, H.,
Chuang, T., Wakeham, A., Itie, A., et al. 1997. Brca2 is required for embryonic cellular
proliferation in the mouse. Genes Dev. 11: 1242-1252

Stockard CR and Papanicolaou GN. 1917 A Rhythmical “Heat Period” in the guinea-pig.
Science. 46: 42-44

Syed V, Ho SM. 2003. Progesterone-induced apoptosis in immortalized normal and
malignant human ovarian surface epithelial cells involves enhanced expression of FasL.
Oncogene. 22:6883-6890

Szotek PP, Pieretti-Vanmarcke R, Masiakos PT, Dinulescu DM, Connolly D, Foster R,
Dombkowski D, Preffer F, Maclaughlin DT, Donahoe PK. 2006. Ovarian cancer side
population defines cells with stem cell-like characteristics and Mullerian Inhibiting
Substance responsiveness. Proc Natl Acad Sci USA 103:11154-11159

Testa, L.A. Getts, H. Salazar, Z. Liu, L.M. Handel, A.K. Godwin and T.C. 1994.
Hamilton, Spontaneous transformation of rat ovarian surface epithelial cells results in
well to poorly differentiated tumors with a parallel range of cytogenetic complexity,
Cancer Res. 54: 2778-2784

Tsao, S.C. Mok, E.G. Fey, J.A. Fletcher, T.S. Wan, E.C. Chew, M.G. Muto, R.C. Knapp
and R.S. 1995.Berkowitz, Characterization of human ovarian surface epithelial cells
immortalized by human papilloma viral oncogenes (HPV-E6E7 ORFs), Exp. Cell Res.
218: 499-507

Tsukagoshi, Y . Saga, N. Suzuki, A. Fujioka, F. Nakagawa, M. Fukushima and M. 2003.
Suzuki, Thymidine phosphorylase-mediated angiogenesis regulated by thymidine
phosphorylase inhibitor in human ovarian cancer cells in vivo, Int. J. Oncol. 22:961-967

Vincent SD, Robertson EJ. 2003. Highly efficient transgene-independent recombination
directed by a maternally derived SOX2CRE transgene. Genesis 37:54-56.

142

V on Gruenigen VE, O'Boyle JD, Coleman RL, Wilson D, Miller DS, Mathis JM. 1999.
Efficacy of intraperitoneal adenovirus-mediated p53 gene therapy in ovarian cancer. Int J
Gynecol Cancer 9: 365372

Walmer DK, Wrona MA, Hughes CL & Nelson KG 1992 Lactoferrin expression in the
mouse reproductive tract during the natural estrous cycle: correlation with circulating
estradiol and progesterone. Endocrinology 131 :1458-1466

Weinert T. 1998. DNA damage and checkpoint pathways: molecular anatomy and
interactions with repair. Cell. 94: 555-558.

Welschen R, Dullaart J. 1976. Administration of antiserum against ovine
follicle-stimulating hormone or ovine luteinizing hormone at pro-poestrus in the rat:
effects on follicular development during the oncoming cycle. J Endocrinol. 70: 301-306

Wenjun Cheng, Jinson Liu, Hiroyuki Yoshida, Daniel Rosen Honami Naora.2005.
Lineage infidelity of epithelial ovarian cancers is controlled by HOX genes that specify
regional identity in the reproductive tract. Nature medicine. 11: 531-537

Wicha MS, Liu S, Dontu G. 2006. Cancer stem cells: an old idea—a paradigm shift.
Cancer Res 66:1883-1890, discussion 1895-1886

Will E, Klump H, Heffner N, Schwieger M, Schiedlmeier B, Ostertag W, Baum C,
Stocking C. 2002. Unmodified Cre recombinase crosses the membrane. Nucleic Acids
Res 30:e59.

Xu J, Fan S, Rosen EM. 2005. Regulation of the Estrogen-Inducible Gene Expression
Profile by the Breast Cancer Susceptibility Gene BRCA1. Endocrinology 146: 2031-2047

Xu X, Wagner K-U, Larson D, Weaver Z, Li C, Ried T, Hennighausen L,
Wynshaw-Boris A, Deng C-X. 1999. Conditional mutation of BRCA1 in mammary
epithelial cells results in blunted ductal morphogenesis and tumor formation. Nat Genet
22: 37-43.

Yu S, Lee M, Shin S, Park J. 2001. Apoptosis induced by progesterone in human ovarian
cancer cell line SNU-840. J Cell Biochem.82: 445-451

Zheng J, Wan M, Zweizig S, Velicescu M, Yu MC, Dubeau L. 1993. Histologically
benign or low-grade malignant tumors adjacent to high-grade ovarian carcinomas contain
molecular characteristics of high-grade carcinomas.Cancer Res. 18: 4138-4142

143

Zhong, Q. et al. 1999.Association of BRCA1 with the hRad50-hMre11-p95 complex and
the DNA damage response. Science 285: 747-750

Zhou H, Luo MP, Schönthal AH, Pike MC, Stallcup MR, Blumenthal M, Zheng W,
Dubeau L. 2002. Effect of reproductive hormones on ovarian epithelial tumors: I. Effect
on cell cycle activity. Cancer Biol Ther. 1: 300-306

Abstract (if available)

Abstract The exact mechanism by which Breast Cancer Susceptibility Gene1 (BRCA1) mutations predispose to breast and ovarian cancer remains poorly understood.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Cell non-autonomous control of proliferation in tissues of elevated cancer risk in BRCA1 mutation carriers

PDF

Insights from mouse models on the origin of ovarian cancer and its predisposing mechanisms

PDF

Mitotic regulation in ovarian epithelial tumors approaching in vitro crisis

PDF

Understanding and controlling mitotic errors leading to aneuploidy in early ovarian cancer development

PDF

Genes and hormonal factors involved in the development or recurrence of breast cancer

PDF

Telomerase expression as a marker for minimal residual disease detection in patients undergoing second-look laparotomy for ovarian cancer: a follow-up study

PDF

Understanding the role of BRCA1 and Aurora A kinase in polyploidy development in ovarian carcinoma precursor cells

PDF

Analysis of Sonic hedgehog/Patched-1 downstream genes in embryonic feather morphogenesis and the development of novel biotechnologies thereof

PDF

Investigating the role of SASH1 gene located on chromosome 6 in ovarian cancer

PDF

Homologous cell systems for the study of progression of androgen-dependent prostate cancer to castration-resistant prostate cancer

PDF

Studies of murine prostate cancer stem / progenitor cells

PDF

Study of the role of bone morphogenetic proteins in prostate cancer progression

PDF

Bimodal effects of bone morphogenetic proteins in prostate cancer

PDF

Study of bone morphogenetic protein-2 and stromal cell derived factor-1 in prostate cancer

PDF

The discovery of a novel 105kDa complex comprised of BUB3, C-JUN and SUMO-1 and its relationship to ovarian cancer development

PDF

Cancer of the psyche: antenatal maternal stress accentuates functional brain responses and neuronal endangerment during conditioned fear in adult offspring

PDF

Jagged1 functions downstream of Twist1 in the specification of the coronal suture and the formation of a boundary between osteogenic and non-osteogenic cells

PDF

The role of endoplasmic reticulum chaperone protein GRP78 in breast cancer

PDF

Role of cancer-associated fibroblast secreted annexin A1 in generation and maintenance of prostate cancer stem cells

PDF

The mechanism by which extracellular Hsp90α promotes cell migration: implications in wound healing and cancer

Asset Metadata

Creator Hong, Hao (author)

Core Title Mechanisms of ovarian cancer predisposition in BRCA1 mutation carriers

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Pathobiology

Publication Date 07/01/2010

Defense Date 05/30/2008

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

Tag BRCA1,OAI-PMH Harvest,ovarian cancer

Language English

Advisor Dubeau, Louis (committee chair), Chuong, Cheng-Ming (committee member), Maxson, Robert E. (committee member), Roy-Burman, Pradip (committee member)

Creator Email haohong@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1309

Unique identifier UC1132429

Identifier etd-Hong-20080701 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-81768 (legacy record id),usctheses-m1309 (legacy record id)

Legacy Identifier etd-Hong-20080701.pdf

Dmrecord 81768

Document Type Dissertation

Rights Hong, Hao

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu