University of Southern California Dissertations and Theses

Corisol's role in breast-to-brain metastasis

(USC Thesis Other)

Corisol's role in breast-to-brain metastasis

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

CORTISOL’S ROLE IN BREAST-TO-BRAIN METASTASIS
by
Robert A. Herrera

A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDINCE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(MOLECULAR MICROBIOLOGY AND IMMUNOLOGY)

August 2020

Copyright 2020 Robert A. Herrera
ii

Acknowledgments
Foremost, I would like to thank Dr. Josh Neman for giving me the opportunity to work in
his lab and allowing me to take on an exciting new project. His mentorship has helped me grow
during my time here at USC. From the first day I joined his lab, he has guided me in the right
direction while encouraging me to be an independent thinker. I can’t imagine having a better
advisor and mentor for my master’s study. I will continue to practice all the lessons he has taught
me.
I would also like to thank Dr. Axel Schonthal, for introducing me to the Neman lab and
for all of his constructive comments, suggestions, and thought provoking questions. I’d also like
to thank the rest of my thesis committee, Dr. Stan Tahara and Dr. Florence Hoffman, for their
advice and constructive criticism.
Additionally, I would like to thank my lab mates, Krutika Deshpande and Vahan
Martirosian, for all their help, encouragement, and, most importantly, their patience. Although
we were lab mates, I looked up to them. The productive work environment they created in the
lab is what truly drove me to emulate their work ethic and also strive to contribute my part to the
never ending important research of our team.
Finally, and most importantly, I would like to thank my family for being my biggest
source of support. My brother Andrew has always believed in me and has given me the
confidence needed to realize my potential. Without the support and sacrifices my parents Dino
and Carol have made, I wouldn’t have had the opportunity to continue my education and achieve
my life goals. I am extremely grateful to have them as my family, and I know they will continue
to support me into my next life’s chapter.
iii

Contents
Acknowledgements…………………..…………………………………………..………………ii
List of Figures…………………………………………………………………...………………..v
List of Abbreviations…………………………………………………………...………………..vi
Abstract…………………………………………………………………………………….…...vii
Chapter 1 – Introduction…………………………………………………………………….…..1
1.1 Brain Metastasis……………………………………………………………………….1
1.2 Blood CNS Barriers……………………………………………………………......…..2
1.3 Tight Junctions………………………………………………………………...…..….3
1.4 Stress Hormone Cortisol………………………………………………………………6
1.4.1 Introduction…………………………………………………………….….6
1.4.2 Cortisol’s role in metastasis………………………………………….…....7
1.4.3 Cortisol’s protective function…………………………………….………..8
1.5 Hypothesis……………………………………………………………………….…….9

Chapter 2 – Materials and Methods……………………………………………………….…...10
2.1 Cell lines and maintenance……………...………… …………………………….…..10
2.2 BBB and BCSFB in vitro Model……………...………………….………….....……11
2.3 Treatments …………………………….………………………………………....…..12
2.4 Concentrating Conditioned Medium………………………….………….….…....….12
2.5 Permeability Assay………………………………………….………………...…..….13
2.5.1 Trans Epithelial Electrical Resistance Assay…………………..…….…..13
2.5.2 Permeability Assay………………….……………………...…..………...13
2.6 Tumor Transmigration……………………………………………………….…...…14
2.7 Immunocytochemistry ……………………………………………….. …….……….15
2.8 qPCR…………………………………………………………………………………15

Chapter 3 – Results …….………………………………………………………………………18
3.1 Hydrocortisone and tumor conditioned medium influence barrier permeability of the
BBB and BCSFB………..………………….…………………………………...…....18
3.1.1 Purpose of the Study …………………………………….……..…….18
3.1.2 Permeability Assay Results…………………………………….…….18
3.2 Breast tumors migrate more effectively through the BCSFB than the BBB................21
3.2.1 Purpose of the study……………………………………..….………..21
iv

3.2.2 BBM 3.1 Tumor Migration Progression ……………………….…….21
3.2.3 Transmigration Results …………………………………...…..……...23
3.3 Breast tumor environments degrade BCSFB tight junctions …………………..…….26
3.3.1 Purpose of the study……………………………………….........……26
3.3.2 Tight Junction Results ……………………………………….………27
3.4 Effects of hydrocortisone on tumor cell migration…………………………………...30
3.4.1 Purpose of the study…………………………………….…...……….30
3.4.2 Tight Junction Results ………………………………………....…….30

Chapter 4 – Discussion.…………………………………………………………………….…..32
References …………………………………………………………………………………..…..37

v

List of Figures
Table1. Main tight junction proteins found in BBB and BCSFB…………………………………4
Figure 1. MDA-MB-231 and BBM 3.1 induce leakiness of the BCSFB even in the presence
of HC……………………………………………………………………………………………..20
Figure 2. 3D images of increasing migration progression……………………………………….22
Figure 3. MDA-MB-231 and BBM 3.1 migrate more efficiently through the BCSFB than the
BBB even in the presence HC……………………………………………………………..….....25
Figure 4. HC downregulates main barrier function TJs in CP cells while upregulating
CLDN-5 and ZO-1 on the mRNA and protein level…………………………..……………..…..27
Figure 5. Breast tumor co-cultures do not change tight junction expression……………….……28
Figure 6. Beast tumor co-cultures decrease CP claudin-5 protein in the presence of
hydrocortisone.…………………………………………………………………….……………..29
Figure 7. HC increases migration independent of barrier resistance in LuBM5 and BBM3.1…..31

vi

List of Abbreviations
BBB – blood-brain-barrier
BBB – blood-brain-barrier
BEC – brain endothelial cells
CLDN – claudin
CNS – central nervous system
CPE – choroid plexus epithelial
CSF – cerebrospinal fluid
FA – Fluorescein Absorbance
HC – hydrocortisone
HCMEC – human cerebral microvessel endothelial cells
HCVP – human cerebral vascular pericytes
ICC – immunocytochemistry
JAM – junction adhesion molecules
TEER – transepithelial electrical resistance
ZO – zonal occludens

vii

Abstract
More than 90% of brain tumors are the result of metastasis. Lung cancer is the most
common source of brain metastasis for men and for women it’s breast cancer. For brain
metastasis to occur a tumor needs to successfully breach the central nervous system (CNS)
barriers, which are made of the blood-brain-barrier (BBB) and blood-cerebrospinal fluid barrier
(BCSFB).
Studies have shown that women with breast cancer have elevated basal cortisol levels
when compared to healthy women. Cortisol has two counterintuitive properties with respect to its
potential effects of cortisol on the evolution of breast-to-brain metastases. First, cortisol has pro-
metastatic properties in breast cancer. Second, cortisol strengthens the BBB thus protecting the
brain from microbes and peripheral immune cells. In this study, we determine whether cortisol’s
role in tumor invasiveness outweighs cortisol’s supporting role in strengthening the CNS
barriers. We expanded our study to include the BCSFB, which has been an underestimated site
of tumor entry. We provide evidence that hydrocortisone (HC), a cortisol analog, increased the
tightness of the BCSFB similar to BBB by upregulating a tight junction protein reported being
exclusive to the BBB.
To determine whether extracellular components released from lung or breast tumor cells
influence barrier properties, we treated our in vitro barriers with tumor conditioned medium
(CM). Our results showed that under HC treatments, lung tumor CM had no effect on either
barrier, but breast tumor CM was able to break down the BCSFB. We hypothesize extracellular
components released from breast tumors influence breast-to-brain metastasis through the
BCSFB. In vitro transmigration results confirmed this hypothesis by showing breast tumor
migration occurred significantly more through the BCSFB than the BBB, and lung tumors
viii

showed equal migration across both barriers. This preference became more apparent under HC
treatments and showed no breast tumor migration through the BBB but still able to cross the HC-
treated BCSFB. We also showed cortisol actually increased breast tumor cell migration through
the BCSFB. We conclude increased cortisol levels facilitate breast-to-brain metastasis through
the BCSFB. This suggest that cortisol plays a pro-metastatic role in breast-to-brain metastasis
and caution is needed when using glucocorticoids to treat breast cancer patients.
1

Chapter 1 – Introduction

1.1 Brain Metastasis
Metastasis is when cancer spreads from its original tissue location to a distant secondary
tissue location. Metastasis initiates when primary tumor cells migrate and enter the circulatory
system. Once in the circulatory system, tumor cells will continue to travel throughout the body
until they come to rest at another site, where they will extravasate out of the vasculature and
implant themselves at the new site (Achrol et al., 2019). Depending on where metastasis occur
will determine the outlook of the patient’s survival. One of the most hazardous and life-
threatening locations for metastases to develop is the brain (Wang et al. 2017).
Although primary brain tumors do occur, more than 90% of brain tumors are the result of
metastases. The most common cancers involved in brain metastases are lung (50%) breast
(20%), melanoma (15%) and gastrointestinal cancer (< 5%). Lung cancer is the most frequent to
develop brain metastases in men and breast cancer is the most frequent to develop brain
metastases in women (Achrol et al., 2019).
Patients that develop lung to brain metastases usually are diagnosed at the same time as
the initial lung cancer diagnosis. However, breast to brain metastases arise approximately 3 years
after initial breast cancer diagnosis (Achrol et al., 2019). The lag time for breast to brain
metastases provides a possible window for preventative treatments to inhibit the progression of
brain metastases. Studying the mechanisms involved in brain metastasis may offer insight to
prevent its development.

2

1.2 Blood CNS Barriers
A tumor cell has no direct point of entry to invade the central nervous system (CNS). The
brain is tightly protected and sealed off by its CNS barriers where their main functions are to
sequester the brain from circulating pathogens. The two main barriers of the CNS are the blood
brain barrier (BBB) and the blood cerebrospinal fluid barrier (BCSFB). The BBB is a stronger,
larger, and less permeable barrier than the BCSFB. It has been shown in vitro TEER
(transendothelial / epithelial electrical resistance) across the BBB (80-100 ohms x cm
2
) is
relatively higher than the BCSFB (30-40 ohm x cm
2
) (Klas et al., 2010, Förster et al., 2008 ).
The BBB is found throughout the brain parenchyma and makes up 99% of all brain
capillaries (Metzger et al., 2017). The three cell types that make up the BBB are brain
endothelial cells (BECs), astrocytes and pericytes. BECs are in the luminal surface of the
capillary wall. Unlike peripheral endothelial cells, BECs are non-fenestrated and have virtually
no paracellular diffusion of solutes.(Engelhardt et al., 2009). The movement of particles from the
blood to the brain is tightly regulated by active transcellular transport properties. (Daneman and
Prat, 2015).
Astrocytes and pericytes surround the abluminal endothelial tube. Astrocytes are glial
cells which utilize their end feet to regulate the BBB’s vascular functions such as contraction,
dilation, water uptake, and permeability (Engelhardt and Sorokin., 2009). Astrocytes provide
direct link to neuronal signals and the brain’s vasculature (Daneman and Prat, 2015). Pericytes
are mural cells and are associated with endothelium throughout the body. However, the ratio of
pericytes to BECs (1:1) is higher than the ratio of pericytes to endothelial cells throughout the
body (1:100) (Daneman and Prat, 2015). Along with providing support for BBB properties,
3

pericytes also play a role in angiogenesis, wound healing, and modulation of the extracellular
matrix (Abbot et al., 2006).
The BBB maintains a strong “physical barrier” throughout the brain, making it difficult
for circulating tumor cells to breach. However, blood vessels located in the ventricles of the
brain do not have the same characteristics of the BBB. In fact, these blood vessels are absent of
astrocytes and contain fenestrated endothelial cells which allow the free flow of ions,
macromolecules, and peripheral immune cells. However, blood circulation in these fenestrated
endothelial cells is separated by the BCSFB (Engelhardt et al., 2009).
The BCSFB is formed by choroid plexus epithelial cells (CPEs) and is located in the
ventricles of the brain. It is responsible for producing cerebrospinal fluid (CSF), which is
secreted into the ventricles and dispersed throughout the CNS. The CSF provides protective
nutrients, a protective shock cushion, and also facilitates the removal of metabolic waste (Redzic,
2011). Along with producing CSF, the BCSFB’s other important role is its barrier function.
Unlike the endothelium of the BBB, the BCSFB’s protective properties lie within a much leakier
epithelium (Redzic, 2011). The tight junctions involved in the BBB and BCSFB determine their
unique properties.

1.3 Tight Junctions
Tight junctions (TJ) serve as functional barriers throughout the body such as the
gastrointestinal tract, kidneys, liver, lungs, eyes and brain (Förster et al., 2008 ). TJs are formed
between cells and are maintained by specific proteins. TJ proteins interconnect cells and create a
tight seal that limits paracellular diffusion. The amount of paracellular diffusion across tight
4

junctions will depend on the organ. For instance, TJs in the intestinal epithelium allow for
paracellular transport of water and ions while also limiting diffusion of antigens and microbes
(Lee et al., 2018). However, under normal conditions in the brain, there is no paracellular
diffusion through the BBB, there is only transcellular transport (Redzic et al., 2011).
The protective barrier properties of BECs and CPEs rely solely on the production and
formation of TJs. TJ proteins have an extracellular binding domain that binds to adjacent TJ
proteins of neighboring cells, a transmembrane domain, and a cytoplasmic domain (Lee et al.,
2018). The cytoplasmic domain interacts with tight junction-associated scaffold proteins which
anchor to the actin cytoskeleton components and allow the cell to alter its morphology or induce
cell signaling (Weber 2012).
Transmembrane TJ proteins include occludin, claudins and junction adhesion molecules
(JAMs). On the cytoplasmic side, zona occludens make up the associated TJ scaffold proteins.
These proteins are found throughout the CNS barriers and differ between expression levels and
types among the BBB and BCSFB cells (Bauer et al., 2014) The most noted tight junction
proteins of each barrier are found in Table 1.
BBB BCSFB
Claudins Claudin-3, -5, and -12 Claudin-1, -2, -3, and -11
Occludin Occludin Occludin
JAM JAM-A, -B and -C JAM-C
Zona occludens (ZO) ZO-1, -2, and -3 ZO-1, -2, and -3
Table 1. Main tight junctions found in BBB and BCSFB
5

In the BBB claudin-5 is the most abundant tight junction protein and has been shown to
be the BBB’s primary TJ for barrier formation (Greene et al., 2019). Expression levels of
claudin-5 have been shown to be 500 times that relative to other claudins in BBB.
Downregulation of claudin-5 in mice has shown increased leakiness in the brain, peripheral
lymphocyte invasion, and depression like behaviors (Menard et al., 2017).
Claudin-1,2,3 and 11 are shown to be expressed at the mRNA and protein level in the
BCSFB (Goncalves et al. 2012). Knockout mice models have shown that claudin-1 and claudin-3
are primarily responsible for BCSFB’s barrier formation and strength. Claudin-2 and claudin-11
have been shown to function more as paracellular water and cation channels. Claudin -2 and -11
are considered as pore-forming tight junction proteins. Claudin-5 has been reported to only be
specific to the BBB and not present in the BCSFB (Goncalves et al. 2013, Steinneman et al.,
2016, Kratzer et al. 2012)
A different ratio of tight junction species being expressed will determine the barrier
strength as well as function. The BBB’s primary function is to inhibit all paracellular diffusion
and protect the brain from outside pathogens. The BCSFB needs to be more permeable, so that
water uptake can occur more readily into the CSF, but it still needs to protect the brain from
outside pathogens so tight junction characteristics seem to be more complex (Kratzer et al.,
2012). In general, TJs in BBB and BCSFB are similar, but they express different claudins. This
is most likely an important structural difference leading to lower TEER in the BCSFB relative to
the BBB as mentioned above.

6

1.4 Stress Hormone Cortisol
1.4.1 Introduction
In times of stress, the body will respond by changes in hormone production. These
hormones include glucocorticoids, catecholamines, growth hormone and prolactin. These
changes increase cardiac output, blood pressure, glucose uptake, metabolic responses, and other
survival mechanisms (Ranabir and Reetu, 2011). Growth hormone secretion can only respond to
physical stress such as physical damage or hypoglycemia (Delitala et al., 1987, Skuse et al.,
1996). Catecholamin and glucocorticoid production respond to physical and emotional stress.
Catecholamines are released upon immediate stress and cause the body to increasing heart rate,
respiration rate and rate of reabsorption of water which facilitates the body’s flight-or-flight
response (Ranabir and Reetu, 2011). Glucocorticoids, such as cortisol, do not make immediate
changes to the body physiology. Cortisol prepares the body for long-term stress. To focus the
body’s energy on self-survival, cortisol inhibits reproductive functions and suppresses the
immune system (Whirledge et al., 2010). It stimulates gluconeogenesis, increases blood sugar,
and increases glucose uptake in the brain. Because of cortisol’s vast changes it makes on the
body in times of stress, it is considered to be the primary stress hormone (Perogamvros et al.,
2012).
Cortisol is a glucocorticoid steroid hormone produced by the adrenal cortex and released
into the blood stream. Under normal conditions, cortisol levels follow a circadian rhythm where
its concentration in blood plasma peaks (10-20 g/dL) upon waking up in the morning, declines
throughout the day, and is lowest (1-5 g/dL) at night just before sleeping (Nicoles-Robin et al.,
7

2011). It has many different physiological functions which are immunosuppressive, anti-
inflammatory, metabolic and homeostatic (Perogamvros et al,. 2012)
Because of cortisol’s vast ability to produce pleiotropic physiological responses, it has
been widely used in the clinic, and is one of the more prescribed drugs (Rhen and Cidlowski.,
2005). Dexamethasone or hydrocortisone, cortisol’s synthetic analog, is used to treat many
autoimmune diseases such as multiple sclerosis, asthma, eczema or other allergic reactions (Shih
and Jackson, 2007). It is also used to treat cancers involved in the immune system such as
leukemia, lymphoma and multiple myeloma (Pufall, 2015). Frequently, it is used to combat
against the side-effects of chemotherapy (Shih and Jackson, 2007).
1.4.2 Cortisol’s role in metastasis
Cortisol is a lipid soluble hormone which diffuses across the membrane of a target cell
and binds to a glucocorticoid receptor in the cytoplasm (Perogamvros et al., 2012). Once bound,
the complex will translocate into the nucleus and induce or repress the transcription of target
genes which comprise up to 10-20% of the human genome (Oakley and Cidlowski, 2013).
Aside from stimulating gluconeogenesis and its ant-inflammatory responses, cortisol’s
role in tumor metastases has not been thoroughly investigated. It has been shown that women
with early stage breast cancer have elevated basal cortisol levels (0.49 mol/L) when compared
to healthy women (0.29 mol/L), and levels are even higher with metastatic breast cancers (0.70
mol/L). Breast cancer patients also have arrhythmic cortisol levels which maintain a high level
of cortisol regardless of the circadian rhythm (G. van der Pompe et al., 1996). This study implies
increased cortisol levels in these patients are mainly associated with the pathology of cancer and
not the psychological stresses involved in the illness.
8

Recently, a study has expressed caution when treating breast cancer patients with
glucocorticoids (GCs). Glucocorticoids such as dexamethasone, actually induce breast cancer
metastasis in vivo (Obradovic et al., 2019). Without GC treatments in mice the basal cortisol
levels increased upon tumor formation and even more so among tumor metastases. These
findings display a positive feedback loop whereby tumor presence increases cortisol levels, and
as a consequence, high cortisol levels trigger metastases which then leads to even higher cortisol
production. The mice used in this study developed metastases in multiple locations including the
lungs, liver, spleen, and ovaries (Obradovic et al., 2019). However, metastases to the brain were
not mentioned in this study. This could be due to the CNS barrier’s inherent protective
properties, but also may be due to the fact that cortisol increases the BBB’s barrier strength
(Förster et al., 2008 ).
1.4.3 Cortisol’s role in strengthening the BBB
Another physiological property cortisol contributes to is creation of a tighter blood brain
barrier. Cortisol targets endothelial cells of the BBB and increases expression of claudin-5
ultimately leading to a tighter barrier (Förster et al., 2008 ). Because of this unique property,
glucocorticoids have been used in the clinic for treatment of multiple sclerosis by way of
preventing peripheral lymphocyte infiltration to the brain (Greene and Campell, 2016). The
relationship between cortisol and brain metastasis has yet to be studied along with effects of
cortisol on the BCSFB.

9

1.5 Hypothesis
Because cortisol reinforces the blood brain barrier as well as induces breast cancer
metastasis we decided to study cortisol’s role in tumor migration through the CNS barrier. Our
study also expands our understanding of the lesser known CNS barrier the BCSFB and its
relationship to cortisol and tumor migration. Therefore, we hypothesize elevated levels of
cortisol, induced by tumor formation, decrease breast-to-brain metastasis by reinforcing the CNS
barrier properties.

10

Chapter 2 – Methods

2.1 Cell maintenance and growth

Cells used for the blood brain barrier (BBB) were human cerebral microvessel
endothelial cells (HCMEC), human cerebral vascular pericytes (HCVP), and human astrocytes
(SciencCell Research Laboratories, Carlsbad, CA). BBB cells were cultured separately in
Artificial BBB medium containing 50% Advance DMEM/F12 (Gibco Life Technologies –
Waltham, Massachusetts, USA), 50% Neurobasal -A Medium (Gibco Life Technologies –
Waltham, Massachusetts, USA), 1% Anti-Anti (Gibco Life Technologies – Waltham,
Massachusetts, USA), 1% GlutaMAX (100X) (Gibco Life Technologies – Waltham,
Massachusetts, USA), 5% Fetal Bovine Serum (Omega Scientific – Tarzana, California, USA),
and 1% B-27 Supplement (50X) (Thermo Fisher Scientific – Waltham, Massachusetts, USC).
The cell line used for the blood cerebral spinal fluid barrier (BCSFB) was human choroid plexus
epithelial (CPE) cells (SciencCell Research Laboratories, Carlsbad, CA). The tumor cell lines
used were lung adenocarcinoma (A549), breast cancer (MDA-MB231), patient-derived lung to
brain metastasis (LuBM5), and patient derived breast to brain metastasis (BBM3.1). Tumor cells
and CPE cells were cultured in Advanced DMEM/F12, supplemented with 1% Anti-Anti, 1%
GlutaMAX, and 10% FBS. Tumor cells and HCMECs were grown on plastic while CPEs,
astrocytes, and pericytes were grown on collagen I, rat tail (Thermo Fisher Scientific – Waltham,
Massachusetts, USA). All cell lines were grown and maintained in a humidified incubator at
37
o
C and 5% CO2.

11

2.2 BBB and BCSFB in vitro model
Twelve-well plate transwell inserts (665635 - Greiner Bio-One – Monroe North Carolina,
USA) with 8.0 m
2
pores were used to measure barrier integrity of the BBB and BCSFB in vitro.
The transwell inserts were inverted and placed into a 6-well plate where the exterior side of the
transwell was coated with rat tail collagen I.
For BBB assays, the goal was to culture HCMECs on top of the transwell with astrocyte
pericyte mixture on the bottom. In order to achieve this, a 1:1 ratio of 100,000 astrocytes and
pericytes were seeded onto the exterior side of the inverted transwells. After 24 hours,
transwells were re-inverted and 150,000 HCMECs in 1 mL of medium were seeded into the top
chamber. BBB models were used five days after this setup.
For BCSFB assays, 80,000 CPE cells were seeded similarly onto inverted transwells.
Cells were seeded with 300 L of their appropriate culture medium. Lids were placed onto the 6-
well plates making contact with the 300 L of medium on the inverted transwells which prevents
evaporation. The cells were allowed to settle and adhere to the transwells overnight. The
following day the transwells were re-inverted to their normal positioning and placed into a 12-
well plate containing 1 mL of their appropriate medium. 1 mL of medium without cells was
placed into the top chamber of the BCSFB transwells. BCSFB models were used 5 days after
this setup.

12

2.3 Treatments
After transwell cultures reached confluency, the medium was changed to a serum-
reduced barrier medium consisting of 0.25% FBS, DMEM, 1% Anti-Anti, and 1% GlutMAX
with or without 550 nM hydrocortisone (HC) (cat# H0888 - MilliporeSigma – Burlington
Massachusetts, USA). After 48 hours of HC pretreatments, 15 L of concentrated tumor
conditioned medium was added to the top chamber of the transwell for a 24 hour treatment with
or without 550 nM HC if samples were initially pretreated with HC.

2.4 Concentrating conditioned medium
Tumor cells were seeded onto 3 wells of a 6-well plate at 150,000 cells/well and 3
mL/well of their appropriate culture medium. The conditioned medium was collected after 48
hours. The collected conditioned medium was transferred to a 15 mL conical tube and
centrifuged at 2000 xg for 2 minutes. The supernatant was transferred to an Amicon Ultra-15
Centrifugal Filter Unit (MilliporeSigma – Burlington Massachusetts, USA) where it was
centrifuged at 5000 xg for 20 minutes. The filtrate was discarded and the retentate (~400 L)
was used for treatments or stored at -80
o
C for later use.

13

2.5 Permeability Assay
2.5.1. Trans Epithelial Electrical Resistance (TEER) assay
Transwell cultures were medium changed after two days of initial seeding. According to
the MilliCell ® ERS-2 User Guide, the TEER was measured in Ω x cm
2
daily to determine
when maximum resistance was observed – operational criterion for confluency. (MilliporeSigma
– Burlington Massachusetts, USA). Confluency was determined when resistance peaked and
leveled off ( > 40 Ω x cm
2
for BCSFB, and >70 Ω x cm
2
for BBB) which occurred after 4 days of
initial seeding.
2.5.2 Permeability Assay
After BBB or BCSFB models were setup and treated, transwells were washed twice with
diffusion buffer which consists of 10 mM HEPES, 4.5% glucose, and 0.1% bovine serum
albumin (MilliporeSigma – Burlington Massachusetts, USA). 900 L diffusion buffer was
placed in the top chamber (transwell insert) and 1 mL diffusion buffer placed into the bottom
chamber. 100 L of 1 mg/mL sodium-fluorescein (MilliporeSigma – Burlington Massachusetts,
USA) was added to the top chamber and was allowed to diffuse to the bottom chamber (12-well
plate). After 30 minutes, the bottom chamber was gently mixed using a pipette, and 100 L was
removed and placed into a clear bottom 96 well plate. 100 L was removed again from the
bottom chamber after an additional 30 minutes. The fluorescence (FL) of the 96-well plate
samples were measured using a FLUOstar Omega filer-based multi-mode microplate reader
(BMG Labtech – Cary, North Carolina, USA) with excitation and emission at 460 nm and 515
nm, respectively. FL is directly proportional to permeability.
14

2.6 Tumor Transmigration
BBB and BCSFB transwell cultures were set up as mentioned above on a 12-well culture
dish. Once the barriers have reached confluency and pretreated with or without 550nM HC, a
tumor migration gradient was formed by placing 1.5 mL of serum free media on the top chamber
and 2.5 mL of 10% FBS media in the bottom chamber. (Tumor migration gradient will have
550nM HC present, if sample was previously pretreated with HC). 150,000 GFP positive tumor
cells, previously cultured for 24 hours in serum free medium, were seeded into the top chamber
of the transwell containing the tumor cell migration gradient. Tumor cells were allowed to
migrate for 24 or 48 hours through the transwell mesh. 500 L of medium was replaced after 24 h
then incubation continued for another 24 h. After the designated periods of incubation transwell
cultures were removed and tumor cells which did not cross the transwell barrier and adhered to
the inner side wall were removed using a wet cotton swab. The transwells were then washed
twice gently with 1x PBS and then placed into a new 12-well plate containing 1 mL of 1%
paraformaldehyde and allowed to fix for 10 minutes. After fixation, the transwells were washed
twice in 1x PBS and allowed to air-dry for 3 minutes. The transwell mesh was cut out of the
insert using a razor and placed onto a microscope slide with the cells facing upward. 10 L of
ProLong Gold Antifade Mountant with DAPI (Cat no. 369 Thermo Fisher Scientific – Waltham,
Massachusetts, USA) was added to the fixed sample, and a 22 mm square cover glass was
pressed onto the sample. Clear nail polish was used to seal the cover glass.
Seven fields from each mesh filter were imaged using a confocal microscope at 20x
magnification. Images were acquired using only the GFP channel. GFP positive cells were
counted and averaged among the 7 fields. All fluorescent imaging was done in the Cell and
Tissue Imaging Core of the USC Research Center for Liver Diseases.
15

2.7 Immunocytochemisty (ICC)
To verify tight junction expression, CP cells or HCEMCs were stained using the standard
ICC immunofluorescence protocol. Cells were grown on 15mm circular coverslips in a 24 well
plate and fixed in 4% PFA for 10 minutes. After two washes with PBS the cells were blocked
with blocking buffer (50% seablock and 50% 1X – PBS) for 1 hour. The cells were then
incubated overnight at 4
o
C with primary antibodies ZO-1 (cat# 33-9100 Thermo Fisher Scientific
– Waltham, Massachusetts, USA) and Claudin-5 (cat# 34-1600 Thermo Fisher Scientific –
Waltham, Massachusetts, USA) at dilutions of 1:200 and 1:100 respectively. The following day,
the cells were washed 3 times in 1X-PBS for 5 minute per wash. Samples were then incubated
with secondary antibodies cy3 (1:300, Cat# 111-165-144, Jackson ImmunoResearch
Laboratories – West Grove, Pennsylvania, USA) and AlexaFluor 488(1:300, Cat# 115-545-146,
Jackson ImmunoResearch Laboratories – West Grove, Pennsylvania, USA) for one hour
combined with AlexaFluor 647 conjugated Phalloidin (1:200, Cat# A12379 Thermo Fisher
Scientific – Waltham, Massachusetts, USA). Afterwards samples were washed again 3 times for
5 minutes and then were mounted onto microscope slides containing ProLong Gold Antifade
Mountant with DAPI. The coverslips were then sealed with nail polish and allowed to dry.

2.8 qPCR
Quantitative PCR was used to determine relative mRNA levels of target genes. The
procedure from the RNeasy Plus Mini Handbook was followed (Bioline – Taunton,
Massachusetts, USA). 350 L Buffer RLT Plus combined with 3.5 L -mercaptoethanol were
added to cells and vortexed until completely homogenized. Lysate was then transferred to
16

QIAshredder and centrifuged for 2 minutes at maximum speed. The flow-through was then
transferred to the gDNA eliminator spin column and centrifuged for 1 minute at 9000 x g. 350 L
of 70% ethanol was added to the flow-through and the 700 L solution was mixed and transferred
to an RNeasy spin column and centrifuged for 15 seconds at 9000 x g. The flow-through was
discarded and 700 L of Buffer RW1 was added to the same RNeasy spin column and
centrifuged for 15 seconds at 9000 x g. The flow- through was discarded, the column was
washed with Buffer RPE, and centrifuged for 1 minute at 9000 x g twice more. The RNeasy spin
column with sample was placed into a new collection tube and spun for 1 minute at maximum
speed to dry the membrane. 35 L of RNase-free water was added directly to the spin column and
allowed to sit for 5 minutes before centrifuging for 1 minute at 9000 x g, and the final flow-
through contained the eluted RNA.
The Nanodrop 8000 (Thermo Scientific – St. Louis, Missouri, USA) was used to
determine the RNA concentration and purity. The RNA was then used to make cDNA using
Maxima First Strand cDNA Synthesis Kit (Cat # K1672, Thermo Scientific – St. Louis, Missouri
USA). The reaction contained 4L of 5x Reaction Mix, 2 L of Maxima Enzyme Mix, volume
equal to 1 ug of RNA, and 14 L nuclease free water were combined into a PCR tube. A Vertiti
96 well Fast Thermal Cycler (Applied Biosystem – Waltham, Massachusetts, USC) was used to
synthesize cDNA.
A 2L aliquot of cDNA was placed into a 96-well PCR plate containing a mixture of 1L
qPCR primers (Integrated DNA Technologies – Coralville, Iowa, USA), 7 L of nuclease free
water, and 10 L of SYBR Green Supermix ( Bio-Rad – Hercules, California, USA). Samples
were done in triplicates. The plates were run using a StepOnePlus Real-Time PCR System
17

(Applied Bioststems – Waltham, Massachusetts, USA. The results were analyzed using the CT
values and GraphPad Prism (GraphPad Software – San Diego, California, USA).

18

Chapter 3 – Results

3.1 Hydrocortisone and tumor conditioned medium influence barrier permeability
of the BBB and BCSFB
3.1.1 Purpose of the Study
Hydrocortisone has been shown to increase barrier tightness of the BBB (Förster et al.,
2008 ), but its effect on the BCSFB remains to be investigated. The purpose of this study was to
determine whether hydrocortisone can preserve the functional barrier integrity of the BBB and
BCSFB in the presence of tumor conditioned medium and to also determine whether barrier
properties can be influenced by different tumor environments.
3.1.2 Permeability Assay Results
To assess barrier permeability, we utilized our in vitro barrier models and pretreated with
or without 550nM HC and then treated for an additional 24 hours with either medium alone,
conditioned media (CM) A549, CM LuBM5, CM MDA-MB-231, or CM BBM 3.1. Samples that
were pretreated with 550nM HC were treated with 550nM HC again on day of treatment.
Fluorescein Permeability assay allowed us to measure the leakiness of the barriers after
treatments.
Our results showed in the absence of tumor conditioned medium, the BCSFB was
significantly (***P<0.001) more than two times permeable (344.93 ± 18.66 FL) than the BBB
(156.53 ± 28.75 FL). Treatments with hydrocortisone showed decreased leakiness in both BBB
(102.86 ± 9.28 FL) and BCSFB (167.60 ± 20.89 FL). However, in the presence of HC the
BCSFB was still shown to be significantly leakier (*P<0.05) than HC-treated BBB (Figure 1A).
19

We next determined leakiness of both barriers after treatments of tumor conditioned
medium (Figure 1B). Results show no effect of BBB permeability with tumor CM treatments.
The BCSFB becomes significantly leakier with treatments of CM MDA-MB-231 (521.47 ±
79.47 FL, **P<0.01) or CM BBM 3.1 ( 481.33 ± 64.72 FL, *P<0.05). CM A549 or CM LuBM5
do not change permeability of either barrier.
HC significantly (*P<0.05) reduced BBB permeability in the presence of CM-A549
(111.82 ± 9.84 FL), CM-LuBM5 (98.70 ± 21 FL), CM MDA-MB-231 (95.18 ± 16.15 FL), and
CM BBM 3.1 (103.66 ± 19 FL; Figure 1C). HC also significantly reduced leakiness of the
BCSFB in the presence of CM-A549 (182.35 ± 22.6 FL, **P<0.01), CM-LuBM5 (138.86 ± 5.2,
*P<0.05), CM-MDA-MB-231 (308± 85.32 FL. ***P<0.001), and CM-BBM 3.1 (229.73 ± 75.32
FL, ****P<0.0001).
Under HC treatments the BBB maintained its permeability in the presence of all tumor
conditioned medium and no significant difference was observed relative to HC only (Figure
2D). In the presence of HC the BCSFB permeability did not significantly change after treatment
with CM A549 or CM LuBM5. However, CM MDA-MB-231 did significantly increase
permeability (308.05 ± 85.32 FL, *P<0.05) relative to the HC only treatments (167.60 ± 20.89
FL). CM BBM3.1 treatment showed a similar trend of increasing HC treated BCSFB’s
permeability (229.73 ± 75 FL), but results were not significantly different (ns, P >0.05)
20

Figure 1. MDA-MB-231 and BBM 3.1 induce leakiness of the BCSFB even in the presence of
HC: Fluorescein Permeability Assay determines leakiness by measuring the amount of sodium
fluorescein that crosses the barrier. (A) comparison of the BBB and BCSFB leakiness in the
absence or presence of HC. (B) Treatments of tumor CM is compared within the BBB and
BCSFB. (C) 0 nM HC and 550 nM HC treatments are compared in the presence of tumor CM.
(D) Treatments of tumor CM is compared within the 550 nM HC treated BBB and 550 nM HC
treated BCSFB. All samples were done in triplicates. For comparisons between two groups (0
nM HC vs 550 nM HC), the unpaired Student’s t test (two tailed) was used. For multiple group
analysis (B and D), one-way analysis of variance (ANOVA) with Bonferroni tests was used
followed by statistical significance. *P < 0.05, **P < 0.01, ***P < 0.001, ****P<0.0001. ns, not
significant.

21

3.2 Breast tumors have an ability to migrate through the BCSFB rather than the
BBB
3.2.1 Purpose of the Study
The purpose of this study is to assess tumor migration across the BBB and BCSFB in
vitro models. The permeability assay suggested the BCSFB became leakier in the environment
of MDA-MB-231 or BBM 3.1. We hypothesize, MDA-MB-231 or BBM 3.1 will migrate
through the BCSFB much more efficiently than through the BBB. Also, HC treatments should
slow down all tumor migration across all barriers.
3.2.2 BBM 3.1 Tumor Migration Progression
BBM 3.1 cells showed a unique migration pattern where under certain conditions these
became stuck during migration and only their leading edge budded through the other side of the
barrier. To illustrate this we took 3D images of increasing migratory progression of BBM 3.1-
GFP cells migrating through the transwell mesh (Figure 2).
In Figure 2, at the start of migration most of the tumor cells were seen in the top view of
the transwell mesh. Migration of the tumor cells is shown in the cross section where most of the
tumor cells were seen at the top and an appendage-like structure was descending through a pore
of the transwell mesh. The bottom side view shows the leading edge of the tumor cell budding
through the opposite side of the mesh. The bottom view, showed a tiny green dot of the BBM 3.1
cells either stuck or at the beginning stage of migration.
Mid-migration in Figure 2 shows approximately half of the tumor cells in the top view.
The cross section shows active migration occurred through a transwell pore, and the bottom view
shows the other half of the tumor cell.
22

End-migration in the top view of Figure 3 shows the very tail end of the tumor cells.
The cross section shows the tail end following movement of the bulk of the tumor cells to the
bottom side of the transwell. The bottom view shows tumor cells that have completed most of
their migration to the other side.
When quantifying migration results, all bottom view migratory progressions were
counted whether it was at the beginning, middle, or end stages of its migration. In some
conditions, no migration is detected not even the beginning stages which shows in figure 3.
Figure 2. 3D images of stages of tumor cell migration across transwell culture: (A) From left
to right is the beginning migration, middle of migration and the end of migration. Top view is the
top of the transwell where tumor cells initially begin their migration. Top side view shows the
sample at an angle where the top is still visible. Side view shows the sample rotated to 90%
relative to top view. Cross section shows tumor cell protrusion through transwell pores. Bottom
side view shows the underside of the sample at an angle. Bottom view shows the underside of the
sample 180 degrees relative to top view. (B) Illustration of transwell shows the locations of top
view and bottom view. Green cells represent GFP positive tumor cells

23

3.2.3 Transmigration Results
To evaluate migration across the barriers, our in vitro barrier models were pretreated with
or without HC, and A549-GFP, LuBM5-GFP, BBM 3.1-GFP, or MDA-MB-231 -GFP were then
allowed to migrate across for 24 or 48 hours. If barriers were pretreated with HC, then they were
treated with 550nM HC again during migration. After designated timepoints, the migrated cells
were quantified as mentioned in the methods sections above.
A549 cell 24 hour migration through the BBB (7 ± 4 cells/field) was not significantly
different ( P > 0.05) than through the BCSFB (7 ± 3 cells/field) in the absence of HC. In the
presence of HC, A549 cell migration was completely inhibited (0 cells per field) after 24 hours.
After 48 hours, the trend was similar and showed no difference between BBB (11 ± 6 cells/field)
and BCSFB (10 ± 4 cells/field) in the absence of HC. After 48 hours, HC significantly
(****P<0.0001) reduced migration across the BBB (1 ± 1 cells/field), and completely inhibited
migration through the BCSFB (Figure 3A)
LuBM5 cells also showed non-preferential migratory patterns similar to A549 cells.
LuBM5 cell 24 hour migration through the BBB ( 25 ± 5 cells/field) was not significantly
different than through the BCSFB (26 ± 14 cells/field). In contrast to A549 cells, HC did not
completely inhibit LuBM5 cell 24 hour migration through the BBB (18 ± 6 cells/field) and
BCSFB (14 ± 7 cells/field), and although the count averages were lower, they were not
significantly different ( P > 0.05). After 48 hours, the migratory preferences did not change and
showed no significant difference between the BBB (47 cells ± 6 cells/field) and BCSFB (41.71 ±
11 cells/field). Under HC conditions, LuBM5 cell migration remained not significantly different
after 48 hours through BBB (32 ± 10) and BCSFB (33 ± 10; Figure 3B).
24

MDA-MB-231 cells migrated significantly (****P <0.0001) more through the BCSFB (
54 ± 30 cells/field) than through the BBB (2 ± 2 cells/field) after 24 hours. HC inhibited all
MDA-MB-231 cell migration through the BBB and significantly reduced migration through the
BCSFB (5 ± 5 cells/field) after 24 hours. In the absence of HC, the trend was similar after 48
hours where migration through BCSFB (83 ± 14 cells/field) was significantly (**** P < 0.0001)
more than through BBB (1 ± 1 cell/field). In the presence of HC, the 48 hour migration was still
completely inhibited through the BBB and significantly reduced through the BCSFB (30 ± 18,
****P < 0.0001) relative to untreated BCSFB (83 ± 14 cells/field; Figure 3C).
After 24 hours, BBM 3.1 cells showed no migration through the BBB, and showed an
average of 3 cells per field (± 2 S.D) through the BCSFB which is a statistically significant
difference (* P < 0.05). In the presence of HC, BBB still showed no migration but a slight
increase in migration was seen through the BCSFB ( 5 ± 2 cells/field, *P<0.05). After 48 hours
in the absence of HC migration was still not detected through the BBB, and average migration
through the BCSFB remained in the single digits (5 ± 2 cells/field). However, 48 hour HC
treatments showed a significant increase of migration through the BCSFB (29 ± 11 cells/field).
Although BBM 3.1 cells did have an average of 1 cell per field (±1 S.D) under HC 48 h BBB
migration, the results were not significantly different to untreated 48 h BBB which was 0 cells
per field (Figure 3D).

25

26

Figure 3. MDA-MB-231 and BBM 3.1 cells migrate more efficiently through the BCSFB than
the BBB even in the presence HC: (A-D) GFP positive tumor cell transmigration through BBB
(left 4 panels) and BCSFB (right 4 panels) + or – 550 nM Hydrocortisone for 24 or 48 hour. Top
graph is quantification for 24 hour migration and bottom graph for 48 hour migration. (A) A549-
GFP cells. (B) LuBM5-GFP cells. (C) MDA-MB-231-GFP cells (D) BBM 3.1 cells. 7 fields
were captured and cells were averaged among the fields. The unpaired Student’s t test (two
tailed) was used to detect statistical significant differences. *P < 0.05, **P < 0.01, ***P < 0.001,
****P<0.0001. ns, not significant.

BBM 3.1 cell migration in the 48 hr 550 nM HC treated samples showed larger cell size
compared to the untreated sample (Figure 3D). This was due to the BBM 3.1 not fully migrating
through the barrier. In the untreated samples, as well as in 24 hour HC treated samples, the cells
were stuck or barely beginning to bud through the barrier. To illustrate the differences between
budding cells and fully migrated cells we provide 3D images of BBM 3.1 cell migration
progression (Figure 2).

3.3 Breast tumor environments degrade BCSFB tight junctions
3.3.1 Purpose of study
The purpose of this study is to examine tight junction modulation in the presence of
breast tumor environment with or without HC. Our results from the permeability assay and
transmigration, showed BCSFB breakdown was susceptible to MDA-MB-231 cells and BBM
3.1 cells. To determine which tight junction proteins were being influenced, CPE cells were co-
cultured in transwell (CPE cells in bottom compartment and breast tumor cells in top
compartment) with MDA-MB-231 or BBM 3.1 cells for 48 hours with or without HC. The CP
cells were then collected and Claudin -1, -3, -5, ZO-1 and occludin mRNA expression was
examined by qPCR.
27

3.3.2 Tight Junction Results
Ribosomal protein large subunit P0 (RPLP0) gene expression was used to normalize TJ
gene expression and obtain delta delta CT to calculate fold change relative to untreated choroid
plexus cells.
HC increased both claudin-5 expression by 1.5 fold ( ± 0.1 S.D, **P <0.01), and ZO-1 by
2.1 fold ( ± 0.1 S.D, **P <0.05). HC downregulated expression of occludin by 30% (± 0.01 ,
P<0.001), claudin-1 by 50% (±0.1, P<0.05), and claudin-3 by 50% (±0.1, P<0.05) (Figure 4A).
Increased expression at the protein level was confirmed for claudin-5 (Figure 4B) and ZO-1
(Figure 4C). Under HC treatment, CP cells were in such close proximity that nuclei could not be
differentiated between individual cells.

28

Figure 4. HC downregulates main barrier-forming TJs in CP cells while upregulating CLDN-
5 and ZO-1 on the mRNA and protein level: (A) qPCR results. Claudin-5, ZO-1, occludin,
claudin-1, claudin-3 relative fold change to 0 nM HC (B-C) ICC of CP cells treated with 0 nM
or 550 nM HC for 48 h. (B) DAPI blue, claudin-5 (red), phalloidin (white), Merge. (C) DAPI
(blue), ZO-1 (green), phalloidin (white), merge. The unpaired Student’s t test (two tailed) was
used to detect statistical significant differences. *P < 0.05, **P < 0.01, ***P < 0.001,
****P<0.0001. ns, not significant.
Figure 4 shows within groups not treated with HC (0 nM HC), MDA-MB-231 or BBM
3.1 cell co-cultures did not downregulate expression of any of the tight junction genes. MDA-
MB-231 co-culture increased OCLN expression by 1.1-fold (±0.1 S.D , * P<0.05) and CLDN-1
by 1.2-fold (±0.1 S.D , * P<0.05). Within groups treated with HC (550 nM HC), expression did
not change across all tight junction genes.
Our data suggest that claudin-5 and ZO-1 were responsible for inducing a stronger barrier
when treated with HC. Since no difference of tight junction gene expression was observed in co-
culture samples, we hypothesize that tight junction disruption is not done at the transcription
level but at the protein level.
29

Figure 5. Breast tumor co-cultures do not decrease tight junction gene expression. Bar graphs
are fold changes relative to 0 nM HC CP only. RPLPO gene expression was used to normalize
data. For multiple group analysis, one-way analysis of variance (ANOVA) with Bonferroni tests
was used followed by statistical significance. *P < 0.05, **P < 0.01, ***P < 0.001,
****P<0.0001. ns, not significant. Significance test was relative to CP only control within 0 nM
HC and 550 nM HC

To determine whether tight junctions were disrupted at the protein level, an ICC was
performed on CP cells co-cultured with MDA-MB-231 or BBM 3.1 cells treated with or without
HC for 48 hours. CP cells were stained for claudin-5 or ZO-1.
The claudin -5 and ZO-1 signal did not change in co-cultures among groups not treated
with HC (Figure 6A,B). However, among the groups treated with HC (Figure 6C) claudin -5
did show a decrease in signal in both MDA-MB-231 or BBM 3.1 cell co-cultures relative to CP
only. ZO-1 (Figure 6D) maintained a strong signal across both co-cultures but no obvious
decrease in ZO-1 signal was observed.
30

Figure 6. Beast tumor co-cultures decrease CP claudin-5 protein in the presence of
hydrocortisone: Fluorescent intensity threshold was determined by adjusting the fluorescent
intensity until no signal was detected on negative primary antibody controls. These settings were
used for imaging corresponding positive antibody samples (claudin-5 or ZO-1). (A-D) Images
include CP cells. Conditions include CP only, CP co-cultured with MDA-MB-231, and CP co-
cultured with BBM 3.1. 48 hour treatments of 0 nM HC (A-B) or 550 nM HC (C-D). ICC stains
included DAPI (blue) claudin-5 (red), ZO-1 (green), and Phalloidin (white).
3.4 Effects of hydrocortisone on tumor cell migration

3.4.1 Purpose of the study
The purpose of this study is to determine whether HC has any effect on tumor migration
independent of barrier resistance. BBM 3.1 was found to be more invasive during transmigration
through the BCSFB under HC treatments. To observe migration without barrier resistance, tumor
cells were pretreated with or without 550nM HC for 24 hours and seeded onto transwells and
allowed to migrate to the other side of the transwell for 12 hours. We hypothesize that HC will
induce tumor cell migration.
3.4.2 Tumor transmigration independent of barrier resistance results
MDA-MB-231 cells showed no significant difference in migration rates between
untreated (50 ± 10 cells/field) and HC treated samples (57 ± 17 cells/field) (Figure 7A). HC
significantly (****P<0.0001) increased BBM 3.1 cell migration from 4 cells/field (±2 S.D) to 17
cells/field (±6 S.D) (Figure 7B). There was no significant difference in A549 cell migration
between untreated (3 ± 2 cells/field) and HC treated samples (2 ± 1 cells/field) (Figure 7C). HC
increased LuBM5 cell migration significantly ( *** P<0.001) from 9 cells/field ( ±5 S.D) to 19
cells/field (± 5 S.D) (Figure 7D).
31

Figure 7 HC increases migration independent of barrier resistance in LuBM5 and BBM3.1.
(A-D) GFP positive tumor transmigration through blank transwells - or + 550 nM
Hydrocortisone for 12 hours. Graphs are quantifications of 0 nM HC vs 550 nM HC migrations
(A) MDA-MB-231-GFP (B) BBM 3.1, DAPI stain is used because GFP signal is very low the
first 12 hours of initial BBM 3.1 migration (C) A549-GFP (D) LuBM5-GFP. (A-D) 7 fields were
captured and cells were averaged among the fields. The unpaired Student’s t test (two tailed) was
used to detect statistical significant differences. *P < 0.05, **P < 0.01, ***P < 0.001,
****P<0.0001. ns, not significant.
32

Chapter 4 – Discussion

The stress hormone cortisol has been shown to be increased in patients with breast
cancer, and patients with breast metastases showed even higher levels. (G. van der Pompe 1996).
Until recently it has been shown that increased levels of cortisol actually promote metastases in
breast cancer (Obradovic et al., 2019). In this study we examined whether cortisol’s role in tumor
invasiveness outweighs cortisol’s supporting role in strengthening the CNS barriers.
Since the BCSFB was a possible point of entry, we had to determine whether cortisol had
the same effect on the BCSFB as it does on BBB. The BBB has been shown to respond to
cortisol by increasing tight junction protein expression and barrier strength (Förster et al., 2008 ).
We were able to confirm that both BBB and BCSFB did become tighter in the presence of
cortisol. Interestingly, claudin-5 was upregulated upon HC treatment in the BCSFB while
claudin-1 and -3 were down regulated. These findings were not expected because claudin-1 and
claudin-3 are considered the main barrier-forming tight junctions of the BCSFB while claudin-5
is known to be exclusively found in the BBB (Goncalves et al., 2013, Steinemann et al., 2016,
Kratzer et al., 2012). We confirmed claudin-5 protein was being expressed in the CP cells by
ICC, but strong claudin-5 signal was only observed in the presence of HC (Figure 4B).
These findings have guided us to form the hypothesis that exposure to high levels of
cortisol is inducing the BCSFB to become more BBB-like. Although the BCSFB becomes
tighter, it is still leakier than the in vitro BBB model. The BCSFB has two main functions which
are producing CSF and maintaining barrier function. The BCSFB contains tight junction proteins
such as claudin -2 and -11 which act only as pore-forming junctions and allow for the exchange
of water and solutes from the blood and CSF. (Goncalves et al., 2013,). This makes the BCSFB
33

inherently more permeable than the BBB (Johanson et al., 2011) Although it has been reported
the main barrier-forming tight junction proteins of the BCSFB are claudin-1, and -3 (Dias et al.,
2019), our data suggest that these tight junction proteins are also appropriate to aid the
production of BCSFB. The BBB’s main TJ protein is claudin-5 and its main function is to create
a barrier. Since our results show that upregulated claudin-5 correlated with a tighter BCSFB, we
hypothesize the BCSFB is becoming more BBB-like to protect the brain from peripheral
invaders. To further support this evidence, future studies to knocking down claudin-5 in CP cells
would allow us to determine whether it is a key mediator in HC-induced barrier strengthening.
Moreover since our results show cortisol did induce stronger barriers across both the
BBB and BCSFB, we further elucidated the barrier properties in the presence of tumor cells. We
used lung and breast cancer cell lines as a model because these are the most common in brain
metastases. Our results revealed that the BBB was able to maintain its barrier tightness in the
environment of all tumor lines. However, the BCSFB became compromised in environments of
MDA-MB-231 and BBM 3.1 cells, but it maintained its barrier integrity in both lung tumor
environments. This suggested that breast tumors release components into their environment that
are detrimental specifically to the BCSFB. Even under barrier strengthening conditions (HC) the
BCSFB still becomes compromised in breast tumor environments.
Under HC treatments, we were able to confirm CP claudin-5 disruption under BBM 3.1
and MDA-MB-231 cell environments (Figure 6). Since there was no change in mRNA tight
junction expression among co-cultures, we concluded the decrease of claudin-5 was done at the
post-transcriptional level or protein level. We also determined that tight junction disruption
primarily occurred in the extracellular domain since ZO-1 remained unaffected in co-cultures. In
the absence of HC, changes of claudin-5 were not observable. This was possibly due to changes
34

only at the post-transcriptional level when treated with HC. Independent of HC, we did observe
an increase in BCSFB permeability under breast tumor conditioned medium treatments.
Therefore, we hypothesize that tight junction disruption is not specific to claudin-5. If we want to
confirm this, we would have to investigate tight junction proteins that are highly expressed
independently of HC such as claudin-1 and -3.
Our transmigration results (Figure 3) confirmed that breast tumors had specificity to the
BCSFB. It is possible that whatever is released into the breast tumor environment is what allows
the breakdown and migration across the BCSFB. The BBB results (Figure 3C, 3D) support this
idea, because the breast tumor cell environment had no influence on BBB integrity and therefore
showed little to no migration through the BBB. Lung tumor cell lines A549 and LuBM5 showed
no migration specificity for BBB or BCSFB and migrated across both at equal rates. Although
lung tumor cells had no effect on barrier strength (Figure 1B-D) , they still managed to cross the
barriers (Figure 3A,B)).
Transmigration under HC treatments provided more evidence of breast to BCSFB
preference. HC stopped breast tumor migration across the BBB, but migration was still
observable through the BCSFB. If high levels of cortisol completely prevent breast tumor entry
via BBB, then circulating breast tumor cells would continue to circulate through the BBB’s
vascular system until these reached a point a vulnerability, which in this case is the BCSFB. In
vivo metastases experiments involving hydrocortisone treatments and breast to brain localization
would be needed to support this claim.
Interestingly, MDA-MB-231 and BBM 3.1 cells showed different migration patterns in
the presence of HC. HC was able to slow down MDA-MB-231 cell migration, but BBM 3.1 cell
migration actually increased with HC. These results showed cortisol was having a direct effect
35

on BBM 3.1 making it more invasive. Although both MDA-MB-231 and BBM 3.1 are triple
negative breast tumors, they both vary in terms of cell behavior. First, BBM 3.1 cells are tumor
cells that have already metastasized to the brain. Therefore, these cells went through several
changes to develop the necessary mechanisms which allowed them to leave the primary tumor
site, travel through the circulatory system, break through the CNS barrier and successfully
implant and grow inside the brain parenchyma. All of these necessary changes BBM 3.1 cells
underwent make them distinguishable from those of the primary tumor tissue site.
One of the more important components developed by BBM 3.1 cell we discovered was
its ability to respond to cortisol. We showed BBM 3.1 cell migration rates, independent of barrier
resistance, were increased with HC treatments but MDA-MB-231 cells were unaffected. A study
showed breast metastases occur in mice when cortisol levels were high (Obradovic et al., 2019).
Thus, we postulate that high cortisol levels are perhaps the typical conditions for BBM 3.1. To
confirm this, we would need to compare the level of glucocorticoid receptors across primary
breast and secondary breast-to-brain tumors. Moreover, we hypothesize that secondary breast-to-
brain tumors have increased glucocorticoid receptors when compared to primary breast tumors.
This increased GR expression may be the key to its breast-to-brain metastases evolution.
Another finding which was not suspected was HC also increased migration rates,
independent of barrier resistance, in LuBM5 but not A549 cells. A cortisol dependent migration
mechanism may also be playing a role in lung-to-brain metastases. This becomes apparent when
observing the migration of LuBM5 cells across the barriers treated with HC. HC does not
significantly slow down migration of LuBM5 cells, but it is able to completely stop A549 cell
migration across both barriers. The correlation between cortisol and lung tumor metastases has
yet to be investigated.
36

We initially hypothesized elevated levels of cortisol would decrease breast-to-brain
metastasis by reinforcing the CNS barrier properties. Although HC decreased migration through
the BBB, breast tumor cell migration through the BCSFB was not decreased. In conclusion, we
found that cortisol facilitates breast-to-brain metastasis by inducing a more invasive tumor
phenotype and breaches the cortisol strengthened BCSFB. Because cortisol is commonly used in
the clinic to treat cancer patients, it is very important to continue studies on cortisol and tumor
progression. Since our study was only able to focus on the end points of breast-to-brain
metastasis, it would be ideal to do a long-term tracking study where patient samples would be
continually harvested and analyzed as the tumor progressed. With this data we could ultimately
pinpoint the essential changes involved in breast-to-brain metastasis and develop therapies to
prevent its initial metastatic colonization.

37

References
[1] N.J. Abbott, L. Ronnback, E. Hansson, Astrocyte-endothelial interactions at the blood-brain
barrier, Nat Rev Neurosci 7(1) (2006) 41-53.
[2] A.S. Achrol, R.C. Rennert, C. Anders, R. Soffietti, M.S. Ahluwalia, L. Nayak, S. Peters, N.D.
Arvold, G.R. Harsh, P.S. Steeg, S.D. Chang, Brain metastases, Nat Rev Dis Primers 5(1) (2019)
5.
[3] H.C. Bauer, I.A. Krizbai, H. Bauer, A. Traweger, "You Shall Not Pass"-tight junctions of the
blood brain barrier, Front Neurosci 8 (2014) 392.
[4] M. Castro Dias, J.A. Mapunda, M. Vladymyrov, B. Engelhardt, Structure and Junctional
Complexes of Endothelial, Epithelial and Glial Brain Barriers, Int J Mol Sci 20(21) (2019).
[5] R. Daneman, A. Prat, The blood-brain barrier, Cold Spring Harb Perspect Biol 7(1) (2015)
a020412.
[6] G. Delitala, P. Tomasi, R. Virdis, Prolactin, growth hormone and thyrotropin-thyroid
hormone secretion during stress states in man, Baillieres Clin Endocrinol Metab 1(2) (1987) 391-
414.
[7] A. Denton, J. Shaw, Corticosteroids in palliative care - perspectives of clinicians involved in
prescribing: a qualitative study, BMC Palliat Care 13(1) (2014) 50-50.
[8] B. Engelhardt, L. Sorokin, The blood-brain and the blood-cerebrospinal fluid barriers:
function and dysfunction, Semin Immunopathol 31(4) (2009) 497-511.
[9] C. Förster, Tight junctions and the modulation of barrier function in disease, Histochem Cell
Biol 130(1) (2008) 55-70.
[10] C. Forster, M. Burek, I.A. Romero, B. Weksler, P.O. Couraud, D. Drenckhahn, Differential
effects of hydrocortisone and TNFalpha on tight junction proteins in an in vitro model of the
human blood-brain barrier, J Physiol 586(7) (2008) 1937-49.
[11] A. Gonçalves, A.F. Ambrósio, R. Fernandes, Regulation of claudins in blood-tissue barriers
under physiological and pathological states, Tissue barriers 1(3) (2013) e24782-e24782.
[12] C. Greene, M. Campbell, Tight junction modulation of the blood brain barrier: CNS
delivery of small molecules, Tissue Barriers 4(1) (2016) e1138017.
[13] C. Greene, N. Hanley, M. Campbell, Claudin-5: gatekeeper of neurological function, Fluids
and Barriers of the CNS 16(1) (2019) 3.
[14] W.A. Hall, H.R. Djalilian, E.S. Nussbaum, K.H. Cho, Long-term survival with metastatic
cancer to the brain, Med Oncol 17(4) (2000) 279-86.
38

[15] C.E. Johanson, E.G. Stopa, P.N. McMillan, The blood-cerebrospinal fluid barrier: structure
and functional significance, Methods Mol Biol 686 (2011) 101-31.
[16] J. Klas, H. Wolburg, T. Terasaki, G. Fricker, V. Reichel, Characterization of immortalized
choroid plexus epithelial cell lines for studies of transport processes across the blood-
cerebrospinal fluid barrier, Cerebrospinal Fluid Res 7 (2010) 11.
[17] I. Kratzer, A. Vasiljevic, C. Rey, M. Fevre-Montange, N. Saunders, N. Strazielle, J.F.
Ghersi-Egea, Complexity and developmental changes in the expression pattern of claudins at the
blood-CSF barrier, Histochem Cell Biol 138(6) (2012) 861-79.
[18] B. Lee, K.M. Moon, C.Y. Kim, Tight Junction in the Intestinal Epithelium: Its Association
with Diseases and Regulation by Phytochemicals, J Immunol Res 2018 (2018) 2645465.
[19] C. Menard, M.L. Pfau, G.E. Hodes, V. Kana, V.X. Wang, S. Bouchard, A. Takahashi, M.E.
Flanigan, H. Aleyasin, K.B. LeClair, W.G. Janssen, B. Labonte, E.M. Parise, Z.S. Lorsch, S.A.
Golden, M. Heshmati, C. Tamminga, G. Turecki, M. Campbell, Z.A. Fayad, C.Y. Tang, M.
Merad, S.J. Russo, Social stress induces neurovascular pathology promoting depression, Nat
Neurosci 20(12) (2017) 1752-1760.
[20] F. Metzger, D. Mischek, F. Stoffers, The Connected Steady State Model and the
Interdependence of the CSF Proteome and CSF Flow Characteristics, Front Neurosci 11 (2017)
241.
[21] A. Nicolas-Robin, J.D. Barouk, E. Darnal, B. Riou, O. Langeron, Free cortisol and accuracy
of total cortisol measurements in the diagnosis of adrenal insufficiency in brain-dead patients,
Anesthesiology 115(3) (2011) 568-74.
[22] R.H. Oakley, J.A. Cidlowski, The biology of the glucocorticoid receptor: new signaling
mechanisms in health and disease, J Allergy Clin Immunol 132(5) (2013) 1033-44.
[23] M.M.S. Obradović, B. Hamelin, N. Manevski, J.P. Couto, A. Sethi, M.-M. Coissieux, S.
Münst, R. Okamoto, H. Kohler, A. Schmidt, M. Bentires-Alj, Glucocorticoids promote breast
cancer metastasis, Nature 567(7749) (2019) 540-544.
[24] I. Perogamvros, D.W. Ray, P.J. Trainer, Regulation of cortisol bioavailability--effects on
hormone measurement and action, Nat Rev Endocrinol 8(12) (2012) 717-27.
[25] M.A. Pufall, Glucocorticoids and Cancer, Adv Exp Med Biol 872 (2015) 315-33.
[26] S. Ranabir, K. Reetu, Stress and hormones, Indian J Endocrinol Metab 15(1) (2011) 18-22.
[27] Z. Redzic, Molecular biology of the blood-brain and the blood-cerebrospinal fluid barriers:
similarities and differences, Fluids Barriers CNS 8(1) (2011) 3.
[28] H.M. Reichardt, R. Gold, F. Luhder, Glucocorticoids in multiple sclerosis and experimental
autoimmune encephalomyelitis, Expert Rev Neurother 6(11) (2006) 1657-70.
39

[29] T. Rhen, J.A. Cidlowski, Antiinflammatory action of glucocorticoids--new mechanisms for
old drugs, N Engl J Med 353(16) (2005) 1711-23.
[30] K.E. Sandoval, K.A. Witt, Blood-brain barrier tight junction permeability and ischemic
stroke, Neurobiol Dis 32(2) (2008) 200-19.
[31] A. Shih, K.C. Jackson, 2nd, Role of corticosteroids in palliative care, J Pain Palliat Care
Pharmacother 21(4) (2007) 69-76.
[32] D. Skuse, A. Albanese, R. Stanhope, J. Gilmour, L. Voss, A new stress-related syndrome of
growth failure and hyperphagia in children, associated with reversibility of growth-hormone
insufficiency, Lancet 348(9024) (1996) 353-8.
[33] A. Steinemann, I. Galm, S. Chip, C. Nitsch, I.P. Maly, Claudin-1, -2 and -3 Are Selectively
Expressed in the Epithelia of the Choroid Plexus of the Mouse from Early Development and into
Adulthood While Claudin-5 is Restricted to Endothelial Cells, Front Neuroanat 10 (2016) 16.
[34] G. van der Pompe, M.H. Antoni, C.J. Heijnen, Elevated basal cortisol levels and attenuated
ACTH and cortisol responses to a behavioral challenge in women with metastatic breast cancer,
Psychoneuroendocrinology 21(4) (1996) 361-74.
[35] H. Wang, C. Zhang, J. Zhang, L. Kong, H. Zhu, J. Yu, The prognosis analysis of different
metastasis pattern in patients with different breast cancer subtypes: a SEER based study,
Oncotarget 8(16) (2017) 26368-26379.
[36] C.R. Weber, Dynamic properties of the tight junction barrier, Ann N Y Acad Sci 1257
(2012) 77-84.
[37] S. Whirledge, J.A. Cidlowski, Glucocorticoids, stress, and fertility, Minerva Endocrinol
35(2) (2010) 109-25.

Abstract (if available)

Abstract More than 90% of brain tumors are the result of metastasis. Lung cancer is the most common source of brain metastasis for men and for women it’s breast cancer. For brain metastasis to occur a tumor needs to successfully breach the central nervous system (CNS) barriers, which are made of the blood-brain-barrier (BBB) and blood-cerebrospinal fluid barrier (BCSFB). ❧ Studies have shown that women with breast cancer have elevated basal cortisol levels when compared to healthy women. Cortisol has two counterintuitive properties with respect to its potential effects of cortisol on the evolution of breast-to-brain metastases. First, cortisol has pro-metastatic properties in breast cancer. Second, cortisol strengthens the BBB thus protecting the brain from microbes and peripheral immune cells. In this study, we determine whether cortisol’s role in tumor invasiveness outweighs cortisol’s supporting role in strengthening the CNS barriers. We expanded our study to include the BCSFB, which has been an underestimated site of tumor entry. We provide evidence that hydrocortisone (HC), a cortisol analog, increased the tightness of the BCSFB similar to BBB by upregulating a tight junction protein reported being exclusive to the BBB. ❧ To determine whether extracellular components released from lung or breast tumor cells influence barrier properties, we treated our in vitro barriers with tumor conditioned medium (CM). Our results showed that under HC treatments, lung tumor CM had no effect on either barrier, but breast tumor CM was able to break down the BCSFB. We hypothesize extracellular components released from breast tumors influence breast-to-brain metastasis through the BCSFB. In vitro transmigration results confirmed this hypothesis by showing breast tumor migration occurred significantly more through the BCSFB than the BBB, and lung tumors showed equal migration across both barriers. This preference became more apparent under HC treatments and showed no breast tumor migration through the BBB but still able to cross the HC-treated BCSFB. We also showed cortisol actually increased breast tumor cell migration through the BCSFB. We conclude increased cortisol levels facilitate breast-to-brain metastasis through the BCSFB. This suggest that cortisol plays a pro-metastatic role in breast-to-brain metastasis and caution is needed when using glucocorticoids to treat breast cancer patients.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The opening of the blood brain barrier by homogenized perillyl alcohol

PDF

Circulating tumor cells as biomarkers in lung cancer

PDF

Neuronal master regulator SRRM4 in breast cancer cells facilitates CNS-acclimation and colonization leading to brain metastasis

PDF

Design, synthesis and validation of Axl-targeted monoclonal antibody probe for microPET imaging of human lung cancer

PDF

Molecular targets for treatment of glioblastoma multiforme

PDF

Anti-cancer effects of novel glidobactin type proteasome inhibitors

PDF

Decreased levels of expression of transmembrane protein 56 (TMEM56) in breast cancer tissues

PDF

Interleukin-11: a study of its effects in glioblastoma multiforme

PDF

Molecular mechanisms of chemoresistance in breast cancer

PDF

Development of a temozolomide-perillyl alcohol conjugate, NEO212, for the treatment of hematologic malignancies

PDF

The role of endoplasmic reticulum chaperone protein GRP78 in breast cancer

PDF

Radiosensitization of nasopharyngal carcinoma cells by NEO212, a perillyl alcohol-temozolomide conjugate drug

PDF

Role of a novel transmembrane protein, TMEM56, in tumorigenic growth of human PC3 prostate cancer cell line

PDF

Role of a novel transmembrane protein, MTTS1 in mitochondrial regulation and tumor suppression

PDF

Understanding the role of APP and DYRK1A in human brain pericytes

PDF

Treatment of HPV driven tumors with SH5-07, a small molecule inhibitor targeting STAT3

PDF

Reactivation of the Epstein Barr virus lytic cycle in nasopharyngeal carcinoma by NEO212, a novel temozolomide-perillyl alcohol conjugate

PDF

Effect of NEO212, a novel perillyl alcohol-temozolomide conjugate, on mycosis fungoides and Sézary syndrome

PDF

Cytotoxic effect of NEO212, a novel perillyl alcohol-temozolomide conjugate, on canine lymphoma

PDF

Mechanism of secretion and function of heat shock protein-90 (Hsp90) family genes

Asset Metadata

Creator Herrera, Robert Augustine (author)

Core Title Corisol's role in breast-to-brain metastasis

School Keck School of Medicine

Degree Master of Science

Degree Program Molecular Microbiology and Immunology

Publication Date 07/02/2020

Defense Date 03/05/2020

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

Tag BBB,BCSFB,breast-to-brain metastasis,OAI-PMH Harvest,permeability,transmigration

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Schonthal, Axel (committee chair), Hofman, Florence (committee member), Neman-Ebrahim, Josh (committee member), Tahara, Stanley (committee member)

Creator Email herrerra@usc.edu,rh.rober7@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-323700

Unique identifier UC11673203

Identifier etd-HerreraRob-8631.pdf (filename),usctheses-c89-323700 (legacy record id)

Legacy Identifier etd-HerreraRob-8631.pdf

Dmrecord 323700

Document Type Thesis

Format application/pdf (imt)

Rights Herrera, Robert Augustine

Type texts

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

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

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA