Does Glutathione Transferase loading into

exosomes from prostate cancer cells influence

progression?

YashnaSagar

B. App. Sci. (Medical Science)

Submitted in fulfilment of the requirements for the degree of

Master of Applied Sciences (Research)

School of Biomedical Sciences

Faculty of Health

Queensland University of Technology

2016

i

Keywords

Prostate, prostate cancer, metastasis, exosomes, prostasomes, microvesicles,

glutathione transferase, chemotherapy, docetaxel.

ii

Abstract

Prostate cancer affects many older men and is the most commonly diagnosed

cancer in Australia after non-melanoma skin cancers. The need for better testing

methods for diagnosis and follow-up is necessary as the current pathology based

test is sensitive but not specific for just prostate cancer. It can also be elevated in

the normal prostate and in benign prostatic conditions. Studies have shown that the

use of circulating exosomes may be useful in detecting cancers, particularly the

exosomes in seminal fluid as they are prostate specific. Multiple studies have shown

that exosomes themselves can be manipulated into packaging and transporting

chemotherapeutic agents and particular enzymes to specific areas of the body.

Exosomes pre-packaged with chemotherapeutic drugs may lead to a more specific

treatment method. An important group of enzymes in cancer biology are the

glutathione transferases. Their presence in prostate cancer has been studied,

however the main focus has been on the methylation of GSTP1 and the

polymorphisms of GSTT1, GSTM1 and GSTM3 in several populations. Due to the

lack of functional studies regarding these enzymes, the effect that they have on

prostate cancer was examined. Prostate cancer cell lines ranging from androgen

sensitive to androgen resistant were used to first determine the GST profile. RT-

PCR and Western blot analysis was used to make expression and protein profiles of

each GST enzyme. According to the GST profile, over-expressing GSTM1 in 22Rv1

cells (androgen dependent) was an appropriate choice as GSTM1 was not present

in either 22Rv1 lysates or exosome preparations. These engineered cells and the

exosomes they shed were then used in functional studies looking at cell

proliferation. It was observed that the proliferation of GSTM1 over-expressing cells

decreased however when in the presence of docetaxel, proliferation actually

increased. Chemotherapy studies using docetaxel were performed as it is a

common drug used for treating castrate resistant prostate cancer. Exosomes shed

from the 22Rv1-M1 cells were shown to protect cells from docetaxel and increase

prostate cancer cell growth. The data from this study showed functional links

between GSTM1, exosomes and the progression of prostate cancer.

iii

Table Of Contents

Keywords .................................................................................................................. i

Abstract .................................................................................................................. ii

Table Of Contents ................................................................................................... iii

List Of Figures ........................................................................................................ v

List Of Tables ....................................................................................................... vii

List Of Abbreviations ............................................................................................. viii

Statement Of Original Authorship ........................................................................... x

Acknowledgments .................................................................................................. xi

Chapter 1: Literature Review ................................................................................ 1

1.1 Introduction ....................................................................................................... 1

1.2 Aims and significance of the study ................................................................... 2

1.3 Review of the literature .................................................................................... 3

1.3.1 Characteristics of the prostate and prostate cancer .............................. 3

1.3.2 Current treatment of localised and metastasised prostate

cancer ................................................................................................... 4

1.3.3 The production of exosomes and how to characterise them ................. 7

1.3.4 The involvement of exosomes in cell-cell communication ................... 10

1.3.5 Exosome profiling ............................................................................... 11

1.3.6 The involvement of exosomes in tumour progression ......................... 13

1.3.7 Glutathione transferases ..................................................................... 17

1.3.8 The involvement of glutathione transferases in cell signalling

pathways ............................................................................................. 18

1.3.9 The involvement of glutathione transferases in the

progression of cancer .......................................................................... 19

Chapter 2: Research Design ............................................................................ 23

2.1 Methodology and Research Design .............................................................. 23

2.2 Procedures ................................................................................................... 23

2.2.1 Cell culture ......................................................................................... 23

2.2.2 Protein analysis .................................................................................. 24

2.2.3 Exosome preparations and visualisation ............................................. 26

2.2.4 Expression of GST enzymes in prostate cancer cell lines ................... 27

2.2.5 Stable transfection of 22Rv1 cells with GSTP1 and

GSTM1 expression vectors ................................................................. 33

2.2.6 Functional studies ................................................................................ 35

2.2.7 Docetaxel treatment of prostate cancer cells ...................................... 36

iv

Chapter 3: GST enzymes are present in Prostate Cancer cells

and shed exosomes ............................................................................................ 37

3.1 Introduction ...................................................................................................... 37

3.2 Materials and Methods ..................................................................................... 39

3.3 Results............................................................................................................. 39

3.3.1 Expression of GSTP1, GSTT1, GSTM1 and GSTM3

genes in prostate cancer cell lines .......................................................... 39

3.3.2 Expression of GSTP1, GSTT1, GSTM1 and GSTM3

protein in prostate cancer cell lines ......................................................... 41

3.3.3 Prostate cancer cell lines produce exosomes and

prostasomes under normal cell culture conditions ................................... 43

3.3.4 Microvesicles released from prostate cancer cells in

culture contain GST enzymes ................................................................. 46

3.3.5 Loading of GST enzymes in prostate cancer

microvesicles .......................................................................................... 48

3.4 Discussion ....................................................................................................... 53

Chapter 4: GSTM1 over-expression affects the behaviour

of Prostate Cancer and associated stromal cells .............................................. 57

4.1 Introduction ...................................................................................................... 57

4.2 Materials and Methods ..................................................................................... 59

4.2.1 Statistical tests ....................................................................................... 59

4.3 Results............................................................................................................. 59

4.3.1 Response of Prostate Cancer cells to docetaxel ...................................... 59

4.3.2 Proliferation of GSTM1 over-expressing 22Rv1 cells ............................... 62

4.3.3 Proliferation of GSTM1 over-expressing 22Rv1 cells

treated with docetaxel .............................................................................. 63

4.3.4 Effect of exosome addition during treatment of

docetaxel-resistant cells with docetaxel ................................................... 64

4.3.5 Effect of exosome addition during treatment of

docetaxel-sensitive cells with docetaxel ................................................... 68

4.3.6 Effect of exosome addition during treatment of stromal

cells (WPMY-1) with docetaxel ............................................................... 71

4.4 Discussion ........................................................................................................ 73

Chapter 5: General Discussion ........................................................................... 75

References ........................................................................................................... 80

Appendices ........................................................................................................... 94

Appendix A: Sequencing of GSTP1 in pGEM-T vector ........................................... 94

Appendix B: Sequencing of GSTP1 after site-directed

mutagenesis .......................................................................................................... 95

v

List Of Figures

Figure 1.1 The T staging of localised and locally advanced prostate cancer.

Figure 1.2 The progression of prostate cancer and the common therapies used

at each stage.

Figure 1.3 The biogenesis of exosomes.

Figure 1.4 Overall composition of exosomes.

Figure 1.5 The promotion of exosomes in cancer metastasis.

Figure 1.6 Interactions between GSTP1 and GSTM1 isozymes and regulatory

kinases implicated in the stress-signalling pathway.

Figure 2.1 Vector map of the pIRES-neo2 vector and multiple cloning site.

Figure 3.1 Expression profile of GST enzymes in prostate cancer cell lines using

RT-PCR.

Figure 3.2 Western blot images comparing GST content in prostate cancer

whole cell lysates.

Figure 3.3 Transmission Electron Microscopy image of microvesicles isolated

from conditioned medium.

Figure 3.4 Western blot images comparing GST content in exosome

preparations shed from prostate cancer cells.

Figure 3.5 Western blot showing over-expression of GSTM1 in transfected

22Rv1 whole cell lysates.

Figure 3.6 Western blot showing the packaging of GSTM1 into shed exosomes

derived from GSTM1 transfected 22Rv1 cells.

Figure 3.7 Transmission Electron Microscopy image of exosomes isolated from

GSTM1 over-expressing 22Rv1 conditioned medium.

Figure 4.1 Dose-response curves of Docetaxel resistant prostate cancer cell

lines after treatment with Docetaxel for 24 and 72 hours.

Figure 4.2 Dose-response curves of Docetaxel sensitive prostate cancer cell

lines after treatment with Docetaxel for 24 and 72 hours.

Figure 4.3 Proliferation of 22Rv1-M1 cell line compared with 22Rv1-VO and

parental 22Rv1 (22Rv1-P) cells.

Figure 4.4 Proliferation of 22Rv1-P and 22Rv1-M1 cells compared with 22Rv1-

VO after treating with 20 nM Docetaxel for 72 h.

Figure 4.5 Proliferation of DU145 cells in the absence or presence of docetaxel

and exosomes.

Figure 4.6 Proliferation of 22Rv1 cells in the absence or presence of docetaxel

and exosomes.

vi

Figure 4.7 Proliferation of PC3 cells in the absence or presence of docetaxel

and exosomes.

Figure 4.8 Proliferation of LNCaP cells in the absence or presence of docetaxel

and exosomes.

Figure 4.9 Proliferation of WPMY-1 cells in the absence or presence of

docetaxel and exosomes.

vii

List Of Tables

Table 1.1 Main differences between extracellular vesicles, in vitro isolation

techniques, markers and size.

Table 2.1 Antibodies used for Western analysis of target protein expression in

total protein and exosome lysates.

Table 2.2 List of primers used throughout the study including cloning primers and

screening primers.

Table 2.3 Primers used for sequencing of GSTP1 and GSTM1 inserts in pGEM-T

Easy and pIRES-neo2.

Table 2.4 List of site-directed mutagenesis primers.

Table 3.1 Quantitation of microvesicles (exosomes and prostasomes) in one

TEM grid from conditioned media of prostate cancer cell lines based on

diameter.

Table 3.2 Visual comparison of GST enzymes found in whole cell lysates and

exosome preparations relative to loading controls.

Table 3.3 Quantitation of microvesicles (exosomes and prostasomes) in one

TEM grid from GSTM1 over-expressing 22Rv1 cells and 22Rv1

parental cells (from Table 3.1) based on diameter.

viii

List Of Abbreviations

°C Degrees Celsius

A Adenine

ADT Androgen Deprivation Therapy

AGRF The Australian Genome Research Facility

BCA Bicinchoninic Acid Assay

BPH Benign Prostatic Hyperplasia

BSA Bovine Serum Albumin

C Cytosine

CARF Central Analytical Research Facility

cDNA Complimentary DNA

CRPC Castrate Resistant Prostate Cancer

DMSO Dimethoxyl Sulfoxide

DNA Deoxyribonucleic Acid

Doc Docetaxel

DPBS Dulbecco’s Phosphate Buffered Saline

exo-PTX Exosomes Loaded with Paclitaxel

FCS Foetal Calf Serum

GSTs Glutathione transferases

G Guanine

HCl Hydrochloric Acid

IC50 Half maximal Inhibitory Concentration

IPTG Isopropyl β-D-1-thiogalactopyranoside

LB Luria-Bertani medium

LHRH Luteinising Hormone Releasing Hormone

lncRNA Long non-coding RNA

miRNA Micro RNA

ng Nanogram

nm Nanometre

PBS Phosphate Buffered Saline

PCa Prostate Cancer

PCR Polymerase Chain Reaction

PSA Prostate Specific Antigen

RNA Ribonucleic Acid

RPMI-1640 Roswell Park Memorial Institute-1640

T Thymine

ix

TEM Transmission Electron Microscopy

TRUS Transrectal Ultrasonography

UQ University of Queensland

X-gal 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside

µg Microgram

µL Microlitre

µM Micromolar

QUT Verified Signature

xi

Acknowledgments

Most importantly I would like to thank Sally-Anne Stephenson for her support

through the very difficult circumstances that had arisen throughout my Master’s

degree. She was possibly the best primary supervisor I could have worked with,

especially as GSTs were not part of her background. She constantly provided me

with advice and guidance that I desperately needed and encouraged me to work

harder than I could have ever imagined ensuring that I could have the best future

and I will forever be grateful for that. I would also like to thank Ricarda Thier for

taking me on as her student and regardless of the difficulties that had been endured,

I definitely gained very important skills that will come in very useful in my future. I

would also like to thank Adrian Herington for accepting me as another student and

for his support and the staff at CARF especially Jamie Riches. I would like to

acknowledge other members of the Eph receptor group, particularly Mohanan,

Jessica and Anastasiya as their support, scientific expertise and jokes made this

degree bearable. I would also like to acknowledge Patrick, Gabrielle, Ena, Michelle

and Rob as they provided me with much needed support and constantly talked me

out of making rash decisions and also gladly listened to my ridiculous stories and

hopes for the future. For that I will forever be grateful to these five friends and hope

for only the best for them. Lastly, I would like to show my sincerest thanks to my

family, especially my mother and brother. They have both seen me cry, panic and

hysterically laugh during this entire process and have still managed to support me

even through my most difficult times. Finally I would like to thank Emily, Mitchell,

Kaitlin and Matthew for their endless support and irrational belief in me. I will be

forever grateful to Emily for supporting me with creative design and Mitchell for

listening to my woes at ungodly hours. I would like to dedicate this thesis to my best

friends who have always supported me through the good times and the bad and

without their help I may not have been able to produce this document.

1

Chapter 1: Literature Review

1.1 INTRODUCTION

Prostate cancer (PCa) is the most commonly diagnosed cancer in Australian

men, excluding non-melanoma skin cancers. According to the most recent data from

the Australian Institute of Health and Welfare there were 19,993 new cases of

prostate cancer diagnosed in 2011. This number is expected to increase to around

25,000 new cases per year by 2020 due to an increase in the number of men

presenting for testing, improvements to diagnostic testing and the aging population

in Australia. Whilst the mortality rates are decreasing and are expected to continue

decreasing by 2020, in 2012, 3,070 men died from prostate cancer [AIHW, 2012].

There are many risk factors associated with the increased chance of developing

PCa. The risk of developing PCa increases with age as men under the age of 50 are

unlikely to be diagnosed with PCa however in those aged 85 and over, there is a

20% chance of diagnosis. Family history increases the risk of PCa as men with a

parent, sibling or child with PCa are twice as likely to develop PCa compared to

other men. Ethnicity is also an important risk factor as African Americans are more

likely to develop PCa than Asians. Lifestyle factors such as a high fat or high

calcium diet and environmental factors such as smoking and exposure to

carcinogenic compounds are also related to an increased risk of PCa [AIHW, 2013].

The progression of PCa involves a multistage process where a small carcinoma of

low histological grade can slowly progress towards metastatic lesions of higher

grade [Gingrich, 1997]. Often, PCa is an adenocarcinoma with well-defined

glandular patterns. It also has a clinically useful biomarker which is used for

diagnosis and during treatment – prostate-specific antigen (PSA) [Byar, 1972,

Stamey, 1987]. The PSA test is sensitive but is unable to differentiate between

benign prostatic conditions such as benign prostatic hyperplasia, indolent PCa and

aggressive PCa [Duijvesz, 2010].A considerable scientific focus is now on the

identification of biomarkers that can identify the presence of aggressive PCa that will

cause the death of a patient and should be aggressively treated, and indolent PCa

that does not require treatment and will not be terminal in the patients normal life.

Exosomes have been shown to be released from a variety of cells, especially

cancer cells and are present in all bodily fluids. Their function is yet to be fully

understood however cancer cell released exosomes appear to be involved in cancer

progression, communication and formation of the metastatic niche [Jung, 2009].

Identification of prostate cancer specific exosomes in the blood, semen and urine of

2

patients with prostate cancer, and in particular, prostate cancer specific proteins,

lipids, DNA sequences and RNA sequences carried by these exosomes, might

provide a new and relatively non-invasive option for identifying aggressive PCa.

The focus of this study was the family of detoxifying enzymes called Glutathione

S-transferases. Several members of this family have been shown to be involved in

cancer progression and in prostate cancer the cytosolic GST enzymes GSTP1,

GSTT1 and GSTM1 are particularly relevant. As cytosolic proteins are often

included as cargo in exosomes, it was hypothesised that GST enzymes over-

expressed by prostate cancer cells are loaded into exosomes and can influence the

growth of cells in the tumour microenvironment, and may contribute to the

development of resistance to chemotherapies like docetaxel.

1.2 AIMS AND SIGNIFICANCE OF THE STUDY

Exosomes have been shown to be involved in the progression of local PCa and

distant metastases and this is believed to be via cargo macromolecules contained

within. PCa cells that release exosomes may contain GST enzymes as cargo which

may contribute to the progression of the disease and could be useful biomarkers for

diagnosis and prognosis and/or potential therapeutic targets.

This project had 3 specific research aims.

1. To profile GST expression in prostate cancer cell lines and determine whether

expressed GST enzymes appear as cargo in exosomes released from these

cells.

2. To determine whether over-expression of a GST enzyme in a GST naïve cell line

results in the loading of the GST into exosomes from that cell line and whether

these GST-positive exosomes alter the in vitro growth behaviour of PCa cells and

associated stromal cells.

3. To determine if GST over-expression increases resistance to docetaxel and if

exosome delivery of GST enzymes to a GST negative line could also confer

docetaxel resistance.

3

1.3 REVIEW OF THE LITERATURE

1.3.1 CHARACTERISTICS OF THE PROSTATE AND PROSTATE CANCER

The incidence of PCa increases in men with particular risk factors which can

be characterised as either non-modifiable factors or modifiable factors. Age is the

most important non-modifiable factor as the incidence of PCa increases rapidly with

age, especially in the seventh decade of life. The relative risk of developing PCa is

also higher in men who have a first-degree relative with PCa than in men without a

first-degree relative with PCa. This risk is also higher in men younger than 65 years

with a first-degree relative than in older men with a first-degree relative with PCa,

especially if the affected relative is a brother and not a father [Kiciński, 2011].

Modifiable risks include smoking, diet, weight and physical activity and all of these

risks are associated with a moderate increase in PCa risk. Increased body-mass

index is also associated with an increase in advanced PCa [Huncharek, 2010;

Discacciati, 2014].

The definitive diagnosis of PCa depends on the presence of an

adenocarcinoma on prostate biopsy cores or operative specimens. Transrectal

ultrasonography (TRUS) guided biopsy of the prostate is the most widely accepted

method to collect histological specimens to diagnose PCa. Currently, the main

indications for prostate biopsy include an abnormal digital rectal examination (DRE)

or elevated serum PSA level. This however is problematic as many patients

undergoing prostate biopsies often show negative histology results that result in

false negative results and requires a repeat biopsy to confirm the adenocarcinoma.

It is evident that refinements in patient selection for biopsy are required. Advances in

molecular biology and radiologic imaging are generating novel data allowing

clinicians to identify patients needing a prostate biopsy with a greater degree of

accuracy [Pal, 2012].

In practice, serum PSA is highly prostate specific however it is not prostate

cancer specific. Non-malignant prostatic conditions can also result in elevated

serum PSA which reduces the PCa specificity of measuring the serum levels

[Thompson, 2004]. Furthermore, the inability of PSA testing to separate indolent

from high grade aggressive disease leads to major drawbacks in the clinical setting,

however routine PSA screening is still one of the best clinical tests available for

PCa. It reduces disease-specific mortality by detecting early prostate tumours that

are still responsive to radical treatment with curative intent [Zappa, 2014]. Advocates

of screening argue that the acceptable and inexpensive PSA test may reduce the

significant mortality caused by PCa through earlier detection and treatment of organ-

4

confined disease. As a result of the low specificity of PSA and the high number of

clinically latent PCa, those against screening argue that there would be unnecessary

biopsies, diagnosis and treatment and that healthcare systems would suffer from

spending money on unnecessary curative treatments and side effects. Despite

these drawbacks, many studies have shown that early testing of serum PSA may

determine the risk of future PCa. The odds ratio of being diagnosed with PCa was

shown to be 19 times higher in men with a PSA of 2.0-3.0 ng/mL at 50 years of age

in comparison of those individuals with a PSA of <0.5 ng/mL with the median time

from PSA testing to cancer diagnosis of 17 years [Lilja, 2007]. Those with higher

PSA values at an early age were more likely to be diagnosed with advanced cancer

[Ulmert, 2008]. Recent data confirming these findings showed that patients with PSA

>2.0 ng/mL at the age of 60 years are at an increased risk of dying from PCa than

those with PSA below this threshold who had an almost negligible risk of clinically

significant cancer [Vickers, 2010]. Although the overall risk of mortality remains low,

over 90% of PCa deaths occur in patients with PSA >2.0 ng/mL at the age of 60

[Vickers, 2010]. This data suggests that an early PSA reading in men may

determine the intensity of future PCa screening.

1.3.2 CURRENT TREATMENT OF LOCALISED AND METASTASISED

PROSTATE CANCER

PCa treatment is dependent on the life expectancy and co-morbidities of the

patient but more importantly, on patient preference for the type of procedure that

may be involved, potential long term side effects and overall quality of life. Risk

assessment depends on the grade and stage of the tumour as well as the PSA

result. The tumour is graded according to the Gleason grading system on tissue

from a core-needle biopsy. The tissue is given two grades between 1 and 5; 1

indicating well-differentiated tissue and 5 indicating poorly-differentiated tissue. The

two grades are then added together and this is known as the Gleason score

[Gleason, 1974].T staging of localised and locally advanced PCa is based on

specific criteria: T1 refers to an impalpable tumour,T2 is where the tumour is

confined to one or both lobes, T3a is where the tumour is on the outside capsule

and T3b tumours have progressed into the seminal vesicle. In Australia, risk

grouping is used where low risk PCa is T1 or T2 with a Gleason score of ≤ 6 and

PSA of ≤ 10.0 ng/mL. High risk patients have a T3 tumour, and Gleason score of 8-

10 or PSA of > 20.0 ng/mL (Figure 1.1). Patients that are classified as an

intermediate risk are those whose T score, Gleason score and PSA result falls

between the high and low risk values as described above [Duchesne, 2011].

5

Figure 1.1: The T staging of localised and locally advanced prostate cancer.

T3a* and T3b* lesions are extraprostatic and are generally managed with radiation therapy

and androgen deprivation therapy. Adapted from: Duchesne, 2011.

PCa treatments are aimed to eradicate the cancer whilst minimising

treatment-associated side effects and costs. The goal for treating localised PCa is to

target patients who are likely to benefit from intervention and minimise treatment-

related complications as well as avoiding disability or death. Treatment depends

upon patient-related factors such as any existing comorbidities, age, life expectancy

and treatment preferences and also depends on treatment-related factors like the

efficacy of the treatment, adverse effects and any potential treatment complications

[Nelson, 2013; Zeliadt, 2006]. There are many different treatment options available

for patients with early-stage PCa including radiation therapy, brachytherapy,

cryotherapy and radical prostatectomy. These therapies can be combined;

brachytherapy with radiation therapy are often considered, however the risks and

side effects vary for each treatment [Marberger, 2012]. Radiation therapy includes

multiple doses of radiation from an external source applied over several days to

weeks to eradicate PCa cells [AHRQ, 2008]. In locally advanced as well as high-risk

patients, radiation may be combined with hormonal therapy which may improve

efficacy [NIH, 2014]. Brachytherapy includes using radioactive implants placed

under radiologic guidance to emit radiation to cancer cells. Permanent (low dose

rate) or temporary (high dose rate) brachytherapy may be used. Cryotherapy

involves the destruction of cells by rapid freezing and thawing using transrectal

guided probe placement and injection of freezing/thawing cryo gases. Radical

prostatectomy is the surgical removal of the entire prostate gland with seminal

vesicles, ampulla of vas and sometimes, the pelvic lymph nodes. It includes

conventional techniques like open retropubic or open perineal, and newer

6

techniques such as laparoscopic and robotic-assisted approaches [AHRQ, 2008].

The advantages of newer laparoscopic and robotic-assisted approaches include

reduced blood loss, faster convalescence and shorter hospital stays [Allen, 2012].

Another important aspect of prostate cancer is the androgen receptor (AR).

The normal development and maintenance of the prostate is dependent on

androgen acting through the AR and expression is maintained throughout prostate

cancer progression. Mutations of the AR may contribute to the progression of

prostate cancer and the failure of endocrine therapy by allowing AR transcriptional

activation in response to anti-androgens. By inhibiting AR activity, this may delay

prostate cancer progression [Heinlein, 2004]. Hormone therapies include androgen

deprivation therapy (ADT) where injectable medications are used to lower

testosterone levels, or the surgical removal of testicles to lower or block circulating

androgens [AHRQ, 2008]. ADT is one of the most common forms of treatment for

advanced PCa. The aim of ADT is to lower the levels of testosterone and other

androgens to castrate levels, based on the fact that testosterone is a key driver of

PCa. Combination therapies with ADT and radiotherapy have shown to improve

cancer control and overall survival in early-stage PCa for intermediate- and high-risk

patients as well as in men with locally advanced PCa [Horwich, 2013; Resnick,

2012]. ADT can normalise serum PSA in over 90% of patients and results in a

sizeable tumour response in 80 to 90% of cases. This treatment occurs with either

bilateral orchiectomy (surgical castration) or medical castration (using a luteinizing

hormone releasing hormone (LHRH) agonist or antagonist). The most widely used

approach is continuous treatment with LHRH agonists in conjunction with an AR

antagonist [Resnick, 2012]. ADT can however lead to various adverse effects

including decreased libido, impotence, hot flashes, osteopenia with increased

fracture risk, metabolic alterations and changes in mood and cognition [Widmark,

2009]. Metabolic complications include obesity, insulin resistance and dyslipidaemia

and the long-term use of ADT has also shown deleterious effects on cardiovascular

health [Ahmadi, 2013]. ADT is initially successful however many men become

resistant to ADT and will relapse. This stage is known as castrate resistant prostate

cancer (CRPC).

CRPC is treated by inhibiting the androgen-receptor axis. This can be

achieved by a maximum reduction of androgen activity. The basic treatment strategy

includes complete inhibition of testosterone production in the testicles as a result of

using LHRH agonists. Simultaneously, new medications are introduced showing a

cytotoxic effect, such as docetaxel or cabazitaxel, or other drugs that block the

prostatic production of testosterone (abiraterone) which may lead to a further

7

reduction of testosterone concentration or block the androgen receptor

(bicalutamide). The use of docetaxel, made from the needles of the Pacific Yew

tree, was FDA approved in 2004 for the treatment of metastatic androgen-

independent prostate cancer and has shown to be very effective in CRPC

[Eisenberger, 2012]. Despite using these particular medications, the main source of

testosterone production, the hypothalamic-pituitary-gonadal axis, remains active.

Regardless of the failure of first-line hormone therapy, LHRH is crucial for effective

treatment. Retrospective studies evaluating the effects of continued hormone

therapy in patients with CRPC have shown longer overall survival in patients using

LHRH agonists, regardless of hormone resistance [Recine, 2015; Silberstein, 2013].

A summary of the commonly employed treatment strategies is summaries below in

Figure 1.2.

Figure 1.2: The progression of prostate cancer and the common therapies

used at each stage. The treatment of prostate cancer is highly stage dependent as

different therapies are used at the initial diagnosis, such as localised surgery and radiation,

and then in CRPC where a combination of chemotherapeutic drugs is often employed.

Adapted from: http://scotdir.com/health-and-fitness-2/prostate-cancer-treatment.

1.3.3 THE PRODUCTION OF EXOSOMES AND HOW TO CHARACTERISE THEM

Exosomes are small lipid vesicles secreted from cells that contain bioactive

molecules. Within cells, the fusion of multivesicular bodies (MVBs) with the plasma

membrane causes the release of intraluminal vesicles (ILVs) into the extracellular

8

environment. These vesicles are known as microvesicles, or exosomes [Raposo,

1996]. The sorting of extracellular vesicle (EV) cargo during the internal budding of

the membrane that leads to ILV formation is an essential step in exosome

biogenesis. The endosomal-sorting complex required for transport (ESCRT) is

responsible for the accumulation and sorting of molecules that are channelled into

the ILVs (Figure 1.2) [Morvan, 2012; Adell, 2014]. The ESCRT contains four main

complexes (ESCRT-0, -I, -II and -III) which are responsible for the final delivery of

ubiquitinated proteins to the degradation machinery. It has been shown that the

depletion of specific ESCRT-family members is able to alter the exosome protein

content and the rate of exosome release from cancer cells [Colombo, 2014].

PDCD6IP (Alix) and tumour susceptibility gene 101 (TSG101) are involved in the

formation of ILVs that are then released as exosomes [Kowal, 2014]. Alix was

recently reported to interact with syndecans by the cytosolic adaptor syntenin which

leads to exosome formation in MCF-7 and HeLa cells [Baietti, 2012]. Proteins

frequently found to be involved in exosome biogenesis in other systems, such as

TSG101, have been used as exosome markers in benign and cancer models

however reports demonstrating their functional role in the formation of exosomes are

lacking.

Exosomes are membrane-bound, 30-120 nm in diameter and contain proteins,

lipids, nucleic acids - DNA, mRNA and microRNAs. Exosomes have been isolated

and purified in various biological fluids such as seminal fluid, urine, blood, breast

milk and from the conditioned media of in vitro cultures of numerous cell types. They

have also been considered to be a potential source of biomarkers for several

diseases due to their ubiquitous presence and stability in biological fluids. Several

studies have also indicated that exosomes play a role in immune responses and are

involved in the pathogenesis of various diseases such as Parkinson’s disease,

Alzheimer’s disease, breast cancer, melanoma, pancreatic cancer and in many

other cancer types [Phuyal, 2014].

Some tumour cells spontaneously release large plasma membrane-derived

extracellular vesicles (PM-derived EVs) containing metalloproteinases with pro-

invasive properties [Di Vizio, 2012; Muralidharan-Chari, 2009]. The release of PM-

derived EVs is more commonly induced by stimuli leading to a rise in intracellular

calcium and cytoskeleton remodelling [Pasquet, 1996]. Calcium ionophores directly

trigger MV release as well as extracellular signals like formyl-Met-Leu-Phe on

neutrophils and simple feeding of tumour cells with media containing fresh FCS

[Hess, 1999; Ginestra, 1997]. Unexpected triggers also stimulate or modify exosome

9

secretion. In some studies, to avoid co-culture with FCS-derived exosomes with cell-

derived exosomes, cells are changed to FCS-free medium and the stress caused by

this abrupt starvation results in altered quantitative and/or qualitative exosome

secretion [Li, 2015]. The common differences between extracellular vesicles is

outlined in Table 1.1

Figure 1.2: The biogenesis of exosomes. The plasma membrane buds inwards

forming a membrane-bound vacuole. Exosomes are formed from MVBs, also known as late

endosomes, originally derived from lysosomes. The production of exosomes can be

triggered by many factors including extracellular stimuli like microbial attack and other

stresses. Exosomes can be released into the extracellular environment by fusion of MVBs

with the cell surface [Kourembanas, 2015].

Exosomes have successfully been purified from conditioned medium of in vitro

cell cultures. The original and most commonly used protocol for exosome

purification involves several centrifugation and ultracentrifugation steps. In some

cases, the first centrifugation steps can be replaced by a single filtration step. An

extra purification step involving sucrose gradients yields very pure exosomes.

Exosomes are also able to be trapped on beads that are covered by an antibody

specific for exosomal surface proteins [Théry, 2006; Clayton, 2001]. The

identification of vesicles as exosomes requires morphological analysis due to their

small size by visualisation with an electron microscope (EM). The purity of exosome

preparations as well as their characterisation should be determined by EM.

10

Table 1.1: Main differences between extracellular vesicles, in vitro

isolation techniques, markers and size.

Microvesicles Size Origin Features

Exosomes 30 - 120 nm

diameter

Endosome Simple phospholipid

bilayer

Markers - TSG101,

CD9, Alix, CD63,

Hsp70

Prostasomes 50 - 500 nm

diameter

Unclear - membranes that

are rich in cholesterol and

sphingolipids including the

plasma membrane, late

secretory pathway, and

endocytic compartments

Cholesterol- and

sphingomyelin-rich

lipid multilayer

PM-derived

extracellular

vesicles

Plasma membrane Contain

metalloproteins

Oncosomes 1 - 10 µm

diameter

Plasma membrane Non-apoptotic blebs

Apoptotic

bodies

1 - 5 µm

diameter

Plasma membrane Annexin V positive

Table 1.1 shows the different microvesicles that can be isolated in vitro, their origin and

features used to identify these vesicles [Borges, 2013].

1.3.4 THE INVOLVEMENT OF EXOSOMES IN CELL-CELL COMMUNICATION

Examining the interaction of single exosomes with target cells is difficult as the

resolution limit of optical microscopy is 200 nm and exosomes are smaller in size.

Various studies however have analysed the bulk interaction of exosomes with

recipient cells using fluorescence microscopy, flow cytometry or functional transfer

of surface molecules. Exosomes have been shown to serve as “communication

shuttles” between cells and can also transduce signals between cells. This

communication may be responsible for re-encoding genes of target cells and is

reported to play a part in the development of tumours, invasion and metastasis and

creation of the metastatic niche [Azmi, 2013]. Exosomes contain the same adhesion

molecules of their cell of origin andcan be captured through specific receptor-ligand

interaction by target cells that specifically recognise them rather than just by any cell

11

that is present in the microenvironment. After this interaction, the recipient cells may

behave differently [Lösche, 2004].

Cells communicate in an auto-, para- and endocrine manner by sensing

circulating endogenous bioactive compounds including proteins, lipoproteins, lipids

and steroids, simpler compounds such as eicosanoids, monoamines, endorphins or

cannabinoids and finally, extracellular miRNA molecules associated with protein

chaperones. The majority of these signalling compounds are also found in

exosomes. Microvesicles that target cells are able to produce varied biological

effects resulting from direct exosome-cell stimulation and the action of transferred

exosomal content. Exosomes release their content into an acceptor cell cytoplasm

after clathrin-mediated endocytosis, phagocytosis or micropinocytosis [Ratajczak,

2006].

Once exosomes are shed, they can cover some distance which enables

horizontal transfer of bioactive molecules and deposition of packaged bioactive

effectors at distal sites [Muralidharan-Chari, 2010]. The cargo contained within

exosomes is able to be released into a recipient cell but also into the extracellular

environment which can result in consequences for the surrounding environment. For

example, neutrophil-derived exosomes are packed with cytokines that release anti-

inflammatory molecules and then later, can function as pro-inflammatory mediators

[Koppler, 2006; Mack, 2000].

1.3.5 EXOSOME PROFILING

Exosomal content has been well-studied and includes RNA and protein

profiles. Interestingly, in some cases the RNA and protein content from exosomes

have a different profile from that of their cell of origin even though they contain the

same adhesion molecules marking their origin [Kogure, 2011; Lösche, 2004]. This

would suggest that there is selective packaging of various proteins and nucleic acid

sequences into the shed exosomes by the cell and these may be important to the

functional purpose for exosome shedding.

Exosomes are distinguished from other shed released vesicles based on their

cellular contents. For example, unlike apoptotic bodies, shed vesicles don’t contain

cytosolic organelles or nuclear fragments [Taylor, 2008]. Furthermore, microvesicles

are actin positive whereas cytoskeletal blebs are only transiently associated with the

actin cytoskeleton during retraction [Charras, 2005; Paluch, 2005]. Proteins that are

derived from the nucleus, mitochondria or endoplasmic reticulum are also excluded

from the exosomal pathway [Schneider, 2013]. Figure 1.3 shows the content of

exosomes that may not be present in all exosomes.

12

In a recent study comparing the molecular profiling of docetaxel sensitive

(Tax-sen) or docetaxel resistant (Tax-res) PCa derived exosomes (DU145 prostate

cancer cell line), it was found that the resistant cells secreted a higher number of

vesicles compared to the sensitive cells [Kharaziha, 2015]. In the same study, using

clinical samples, the median number of particles present in the serum of CRPC Tax-

res patients was higher than in CRPC Tax-sen patients. It was also found that the

exosomes secreted by the Tax-sen and Tax-res cells were not only different in

number but also in the physical characteristics of the secreted exosomes. For

example, the Tax-sen exosomes floated at a sucrose density of 1.12 to 1.19 g/mL

whereas the Tax-res exosomes were present in a more narrow, although over-

lapping, sucrose density of 1.13 to 1.18 g/mL. Cellular component and protein class

analysis further showed that the exosomes derived from these two cell lines have

subtle yet distinct differences which may regulate their biological function on

recipient cells. The Tax-res patients had a higher protein concentration of

extracellular vesicles compared to the Tax-sen patients which suggests that these

may be a source of protein biomarkers that can predict chemoresistance [Kharaziha,

2015]. More recently, studies have reported that exosomes are involved in

chemotherapy resistance [Challagundla, 2015; Bryniarski, 2015].

Small nucleic acid sequences including miRNA are also found packaged

within exosomes with the mechanism of exosome-mediated transfer of miRNAs first

verified in 2007 [Valadi, 2007]. Exosomes have the ability to envelop specific

miRNAs, protect the cargo whilst in circulation and function as a tool for invasive

access while reflecting the condition of donor cells [Palma, 2012; Ge, 2014;Kowal,

2014]. Exosomal miRNA is also a potential cancer biomarker. Interestingly,

exosomes can be transfected with free extracellular miRNA which can then be used

to target cells [Challagundla, 2015; Bryniarski, 2015].

13

Figure 1.4: Overall composition of exosomes. Schematic showing exosome

composition. Each listed component may be present in some exosomes and not others.

Abbreviations: ARF, ADP ribosylation factor; ESCRT, endosomal sorting complex required

for transport; LAMP, lysosome-associated membrane protein; MHC, major histocompatibility

complex; MFGE8, milk fat globule-epidermal growth factor-factor VIII; RAB, Ras-related

proteins in brain; TfR, transferrin receptor [Colombo, 2014].

1.3.6 THE INVOLVEMENT OF EXOSOMES INTUMOUR PROGRESSION

Tumour progression, metastasis and chemoresistance depend on the

communication between the tumour itself and its surrounding microenvironment.

Tumour cells can release exosomes in large quantities and if the number of

circulating exosomes increases in patients with cancer this correlates with poor

prognosis [Martínez, 2005]. It has been shown that the production of a metastatic

niche is probably enhanced by the release of exosomes from the primary tumour

cells as seen in Figure 1.4. A few studies have also shown that exosomes shed by

cancer cells can be taken up by stromal fibroblasts and other associated stromal

cells suggesting communication between cancer and stromal cells that may also

play a role in tumour cell spread. Rat pancreatic adenocarcinoma-derived exosomes

may help form the pre-metastatic niche leading to the development of lung

14

metastasis [Jung, 2009]. Pre-metastatic changes in vivo were shown to be the result

of the co-operation of exosomes and the soluble matrix. Exosomes may also

transfer receptors between normal cells and tumour cells. Janowska-Wieczorek,

[2001] showed the adhesion molecule CD41 can be transferred from platelets to

tumour cells by exosomes, and provides the tumour cells with pro-adhesive

properties.

Invadopodia have emerged as a docking site for exosomes promoting cancer

invasion [Hoshino, 2013]. Exosome secretion and invadopodia formation appear to

be co-dependent as the inhibition of exosome biogenesis affects invadopodia

formation and stability and in contrast, the inhibition of invadopodia formation greatly

decreases exosome secretion.

Exosomes influence the major type of stromal cells – cancer-associated

fibroblasts, vascular endothelial cells and immune cells. Fibroblasts become

“activated” as they progress through changes during the neoplastic process

including morphologic and protein expression changes. Disorganised and

uncontrolled growth as well as increased collagen production and hyaluronate

synthesis stimulation also occur [Tlsty, 2001]. These activated fibroblasts are able to

both hinder and promote tumour growth and progression, however this depends on

the molecular state of the tumour epithelial cells. They are also capable of

accelerating growth and promoting tumour cell invasion [Orimo, 2005; Camps, 1990;

Gleave, 1991; Kaneruka, 2002; Sameshima, 2000]. Cancer exosomes can trigger

fibroblast transformation through the TGF-β/Smad pathway and stimulate effects

unique from soluble TGF-β [Gu, 2012; Webber, 2010; Webber, 2015]. When tumour

fibroblasts were co-cultured with non-transformed epithelial cells, immature

pleomorphic epithelial cells with enlarged nuclei and aberrant mitosis could be seen.

These cells had an increased rate of proliferation, lost their polarity and changed cell

cycle protein expression. p53, PCNA, Ki67 and cytokeratin expression were

increased and in contrast, p21 nuclear expression and Bcl2 were decreased [Atula,

1997].

For a tumour to grow beyond a few cubic millimetres, the tumour cells require

access to the host vasculature andthe ability to divert blood to the tumour. The

creation of a tumour blood supply requires a significant increase of pro-angiogenic

factors to overcome the anti-angiogenic factors [Bergers, 2003]. Exosomes have

been shown to play a role in increasing angiogenesis. Once hypoxia is sensed

throughout the tumour, carcinoma cells secrete exosomes into the

microenvironment to start signalling and turn on the angiogenic switch to ensure

adequate oxygenation [Giordano, 2001; Park, 2010]. Hypoxia-induced proteins

15

secreted by tumour exosomes are taken up by the normal host endothelial cells

where the exosomal cargo stimulates new tubule formation which eventually leads

to a network of blood vessels to supply that area [Skog, 2008]. Endothelial cells that

have taken up hypoxic tumour exosomes start releasing growth factors and

cytokines that stimulate pericytes through the PI3K/AKT pathway [Kucharzewska,

2013].

In the early stages of cancer, the immune system may stop tumour

progression however growing tumours can activate immune-suppressive pathways

and evade immune surveillance. Exosomes produced by immune cells are able to

initiate an anti-tumour immune response. Mast cell-derived exosomes can induce

dendritic cell differentiation and activate T and B cells, whereas dendritic cell-derived

exosomes sensitise other immune cells to tumour antigens [Skokos, 2003 andThéry,

1999]. Tumour-derived exosomes are involved in immune escape by increasing

myeloid progenitor cell differentiation, decreasing T cell proliferation and effect or

functions and also by cancelling the natural cytotoxic responses controlled by

natural killer cells [Valenti, 2007; Xiang, 2009].

The metastatic potential of the parent cell has been shown to affect particular

characteristics of the exosomes that they release. In breast cancer cells, exosomes

isolated from cells with the highest metastatic potential induced the largest increase

in motility. These exosomes also induced migration to a degree that was dependent

on the metastatic potential of the parent cell type. Exosomes derived from parent

metastatic cell lines promoted cell mobility faster and/or to a further extent than

exosomes isolated from non-metastatic cell lines or control media [Harris, 2015].

16

Figure 1.5: The promotion of exosomes in cancer metastasis. Tumour-associated

exosomes influence cells and modulate the microenvironment involving key steps in the

cancer metastasis cascade. 1. In the primary site, tumour cells secrete exosomes to induce

EMT and degrade the matrix. The Wnt pathway in cancer cells is also activated by

exosomes during migration. 2. During intravasation, the endothelium is disturbed directly by

tumour-secreted exosomes and indirectly by macrophages activated by exosomes from

tumour cells. 3. Circulating tumour cells (CTCs) and tumour-activated platelets secrete

exosomes affecting the immune cells and CTCs. 4. Adhesive molecules on endothelial cells

are upregulated by exosomes from the adherent tumour cell. 5. Disseminated tumour cells

provide a micrometastasis at an appropriate niche, remoulded by exosomes from the

primary site [An, 2015].

17

In the Kharaziha et al.(2015) study comparing docetaxel sensitive and

resistant exosomes from DU145 cells, DU145 cells treated with 10 µg/mL of DU145-

Tax-res exosomes had a higher ability to degrade the extracellular matrix (ECM)

compared to DU145 cells treated with either PBS or DU145-Tax-sen exosomes. The

amount of degraded ECM, depicted as regions of matrix where fluorescein was

degraded, was quantified and found to be higher in Tax-res exosome treated DU145

cells than in the matrix of PBS or Tax-sen treated DU145 cells [Kharaziha, 2015].

The Src/PI3K/AKT and the Ras/Raf/MEK/ERK signalling cascades have been

shown to be vital for the promotion of migration, invasion and metastasis [Kinkade,

2008]. Whether or not the addition of Tax-res exosomes could modulate the active,

phosphorylated levels of AKT and ERK1/2 was also investigated. There was no

difference in AKT phosphorylation levels however there was an increase in the

ERK1/2 phosphorylation levels. This data indicates that Tax-res exosomes may

have the ability to promote migration and invasion in recipient PCa cells. It was also

shown that the isolated exosomes from PCa patient sera could induce enhanced

cell proliferation and invasion of the 22Rv1 and DU145 cell lines [Kharaziha, 2015;

Corcoran, 2012].

1.3.7 GLUTATHIONE TRANSFERASES

There are three distinct families of human glutathione transferases (GSTs)

including cytosolic, mitochondrial and membrane-bound microsomal groups. The

cytosolic GSTs are differentiated into seven classes: alpha (α, A), mu (µ, M), omega

(Ω, O), pi (π, P), sigma (σ), theta (θ, T) and zeta (Ζ) [Tew, 2011]. The different

classes of mammalian GSTs differ in the chromosomal locations of their

corresponding genes and their gene structure, resulting in distinct properties and

protein structures [Hayes, 1995].The GST genes are found in human chromosomes

at 11q for class GSTP1, 22q for GSTT1 and 1p for GSTM1. Cytosolic GSTs are

dimeric and each protein chain is identical and depicted as Arabic numbers in the

GST name i.e. GSTP1. Each subunit consists of two distinct domains which share

common folds within classes. The two subunits are catalytically independent

however the assembly of a dimeric GST is important to stabilise the enzymes. Both

subunits are responsible for binding the physiological co-substrate glutathione

(GSH) and the respective exogenous substrate at the active site [Reinemer, 1997].

The GST enzymes studied here - GSTP1, GSTT1, GSTM1 and GSTM3, are

distinguished by the different residues responsible for their catalytic activity. For

example, the pi (P) and mu (M) classes possess key tyrosine residues whereas the

theta class (T) contain serine residues [Atkinson, 2009].

18

Functional GST enzymes exist as dimers of either identical (homodimers) or

different (heterodimers) subunits, with subunit molecular weights ranging from 17 to

28 kDa. The formation of heterodimers is restricted to subunits of the same class as

monomers of different classes are unable to dimerise due to the incompatibility of

the interfacial residues. Generally, members of the same class share more than 40-

50% sequence identity but have less than 25-30% sequence identity with GSTs of

other classes [Tew, 2012; Board, 2013; Hebert, 2007].

GSTs play a role in internal cell detoxification and were found to neutralise

toxic effects caused by drugs, food additives, environmental chemicals and

carcinogens [McElwee, 2007].The natural substrate of GST is glutathione which is a

water-soluble molecule that is widely available in both the oxidised and reduced

(GSHRed) forms in eukaryotes. The binding of GSHRed with GSTs causes oxidation of

the substrate which in turn maintains the balance of redox states in cells [Liska,

1998]. Internal detoxification by GSTs occurs in the conjugation pathway (Phase II)

where GSTs bind with substrates forming conjugates, such as the GSH S-

conjugate. The conjugate is then catalysed to cysteine S-conjugate by γ-

glutamyltranspeptidase which is finally excreted in faeces and urine [Lash,

1994;Kursula, 2005].

1.3.8 THE INVOLVEMENT OF GLUTATHIONE TRANSFERASES IN CELL

SIGNALLING PATHWAYS

GSTP1 has been shown to play a role in cell signalling as it can inhibit c-Jun

N-terminal kinase 1 (JNK1) by direct protein-protein interaction as seen in Figure

1.5. JNK is a MAP kinase involved in stress response, apoptosis, inflammation and

cellular differentiation and proliferation. Low JNK activity is observed in non-stressed

cells due to the sequestration of the protein in a GSTP1:JNK complex [Adler, 1999].

The monomeric form of GSTP1 binds to the C-terminal of JNK and suppresses its

kinase activity [Wang, 2001]. The dimeric form of GSTP1 (from two monomers) is

demonstrated after the induction of apoptosis by hydrogen peroxide in

neuroblastomas showing that the suppression of JNK activity is reversed under

oxidative stress conditions. This leads to the dissociation of the GSTP1:JNK

complex. By increasing the levels of monomeric GSTP1, the rebinding of GSTP1 to

JNK occurs resulting in attenuation [Bernardini, 2002]. Mouse embryo fibroblast

cells engineered to be null for GST expression were found to contain high basal

levels of JNK activity which was reduced when these cells were transfected with

GSTP1 cDNA [Chen, 2002].

19

In a human granulosa-like tumour cell line (KGN), the depletion of GSTT1

inactivated the p38-MK2 signalling pathway and reduced apoptosis after exposure

to hydrogen peroxide. MK2 (MAPK-activated protein kinase 2) is a downstream

transporter of p38 and is involved in age-associated change in the subcellular

localisation of p38 [Ito, 2011].

GSTM1 is able to bind to apoptosis signal-reducing kinase 1 (ASK1) and

inhibit its ability to activate the JNK and p38 signalling pathways [Cho, 2001]. ASK1

is a Mitogen-activated protein kinase kinase kinase (MAPKKK) which activates both

JNK and p38. TNF-receptor-associated factor 2 (TRAF2) and other TRAF proteins

have been reported to interact with and activate ASK1 [Shiizaki, 2013]. This

mechanism is similar to the one proposed for GSTP1:JNK. In an unstressed cellular

environment, ASK1 shows low activity because of its sequestration via GSTM1.

During conditions of oxidative or chemical stress, dissociation of the GSTM1:ASK1

complex occurs, releasing GSTM1 for oligomerisation, then activating ASK1 and

subsequently inducing apoptosis [Dorion, 2002]. Ovarian GSTM1 has been shown

to be involved in negatively regulating ASK1 with the formation of protein complexes

that dissociate during stress and lead to ASK1-induced apoptosis [Bhattacharya,

2013].

20

Figure 1.6: Interactions between GSTP1 and GSTM1 isozymes and regulatory

kinases implicated in the stress-signalling pathway. GSTs interact with several

MAPKs in non-stressed conditions. Following exposure to reactive oxygen species (ROS),

drugs and/or cytokines or UV radiation, protein complexes dissociate leading to the

activation of signalling pathways. MAPK – Mitogen-activated protein kinase, JNK – c-Jun N-

terminal kinase, ASK1 – apoptosis signalling-regulating kinase 1, TRAF – TNF-receptor-

associated factor 2, EGFR – epidermal growth factor receptor, PKC – protein kinase C, PKA

– protein kinase A [Adapted from: Singh, 2015].

1.3.9 THE INVOLVEMENT OF GLUTATHIONE TRANSFERASES IN THE

PROGRESSION OF CANCER

GST over-expression has been observed in both in vitro and in vivo studies

in a range of neoplastic cells. GST over-expression was observed in carcinogenesis

and drug resistance, potentially allowing GSTs to be used as a marker protein for

cancer progression. There is a lot of emphasis on GSTP1 as it is the most widely

distributed and the most abundant isozyme. Its expression is mainly associated with

the development and resistance of tumours against commonly used anti-cancer

drugs [Oakley, 2011]. GST expression has been found to be upregulated at both the

protein and mRNA level in a panel of 60 human tumour cell lines used in the NCI

Drug Screening Program [Tew, 1996]. GSTP1 was also found to be the most

predominant GST isozyme in these cancer cells.

21

There have been many controversial studies regarding the over-expression or

down-regulation of GSTP1 and its involvement in tumour progression. GSTP1

positive cancer associated fibroblasts (CAFs) have been involved with GSTP1

negative cancer cells in primary breast cancer and axillary lymph node metastases

[Chaiwun, 2011]. In a significant number of cases, CAFs were positive for GSTP1

expression. These GSTP1-positive CAFs in the tumour microenvironment may

protect the GSTP1-negative cancer cells by catalysing GSTP1-mediated reactions

that would neutralise chemotherapeutic agents [Chaiwun, 2011]. In a study

examining the expression of drug resistance gene products in lung carcinoma

progression, GSTP1 expression was elevated in tumour tissue compared to normal

lung tissue. GSTP1 expression also increased with increasing tumour volume and

differentiating grade. The results suggest that the level of GSTP1 in lung cells

increases as cells progress from normal to a transformed state [Mattern, 2002]. In a

separate study looking at the prognostic role that GSTP1 has in invasive breast

cancer, it was concluded that GSTP1 expression could be an independent predictor

of poor prognosis in breast cancer patients. Survival rates were worse in patients

with GSTP1 positive breast tumours and the relative risk of tumour recurrence was

also higher [Huang, 2003]. Alternatively, the absence of GSTP1 expression in

human oesophageal squamous carcinoma was associated with a significantly

shorter overall survival [Wang, 2010]. This correlates with the findings of prostate

adenocarcinomas where the down-regulation of GSTP1 was observed and the loss

of GSTP1 expression was associated with malignant transformation [Okino, 2009,

Mavis, 2009].

In breast cancer cells, GSTP1 enhanced proliferation and migration.

Recombinant GSTP1 added to culture medium of human breast cancer cells bound

directly to the cell surface and increased the viability, growth rate and migration of

the cells [He, 2011].

GSTP1 may influence cancer cell resistance to doxorubicin through the

suppression of doxorubicin conversion to semiquinone free radicals and the

subsequent production of superoxide anion radicals and peroxides [Finn, 2011].

Radiation kills cells by forming highly reactive oxygen radicals that damage tumour

cell DNA and GSTP1 reduces these reactive oxygen radicals, making tumours

resistant to radiotherapy [Kanwal, 2014]. This suggests that the over-expression of

GSTP1 may be associated with the development of drug resistance in cancer cells,

where it increases detoxification of the agents and also the suppression of cellular

reactive oxygen species inducing apoptosis.

22

Extensive studies have examined the relationship between single nucleotide

polymorphisms (SNPs) of GST and null genotypes of GSTM1 and GSTT1 with

susceptibility to prostate cancer. In an Asian population, a meta-analysis showed

that GST polymorphisms may be able to predict disease susceptibility and the null

GSTM1 allele could be associated with lower risk of prostate cancer but an

increased risk of other cancers [Wei, 2012]. An opposing study has shown that

polymorphisms may actually not be associated with disease outcome and any

further possible recurrence [Wiklund, 2009] and in fact many of the studies

regarding polymorphisms have inconsistent findings [Cotignola, 2009; Nock, 2009;

Agalliu, 2006]. Epigenetic changes are changes in gene expression that aren’t

caused by alterations in the primary sequence of the nucleotides of the gene. DNA

hypermethylation is the most common epigenetic change and one of the most

common molecular alterations in human cancers [Phé, 2010]. CpG dinucleotides are

often found in clusters called CpG islands in promoter regions. CpG islands of many

genes, including tumour suppressor genes, are unmethylated in normal tissues and

often methylated to varying degrees in multiple cancer types which can cause gene

transcription silencing and the inactivation of these tumour suppressor genes [Phé,

2010; Chiam, 2014]. In prostate cancer, it has been found that the promotor regions

of several genes were hypermethylated tested by methylation-specific PCR [Liu,

2011; Ellinger, 2008; Bastian, 2005]. GSTP1 promoter hypermethylation is currently

the best readily available DNA-based biomarker for prostate cancer tissues as it is

only rarely present in benign prostate tissue [Phé, 2010]. A major issue with using

this technique as a biomarker is that GSTP1CpG island hypermethylation has also

been reported in other cancers so it is not specific to prostate cancer [Esteller,

2001]. There is also evidence that GSTP1 methylation could trigger “epigenetic

catastrophe” where tumour suppressor genes like APC and RAR-beta may also

become hypermethylated [Yegnasubramanian, 2004].

23

Chapter 2: Research Design

2.1 METHODOLOGY AND RESEARCH DESIGN

There were 3 main research aims for this project. The first was to profile GST

expression in prostate cancer cell lines using Reverse Transcriptase-PCR and

Western analysis, then determine whether GST enzymes expressed by these cells

appeared as cargo in exosomes that were released into the growth medium. The

second was to use a cell line that did not express a particular GST (as identified in

Aim 1), over-express that GST in that cell line by cloning the complete coding

sequence into the mammalian expression vector (pIRES-neo2) then determine

whether the GST was subsequently packaged into exosomes. The effect of

exogenous over-expression of the GST enzyme on in vitro cell growth

characteristics was determined using a proliferation assay and GST-enriched

exosomes were collected from medium and added to parental cells and stromal

prostatic fibroblasts to determine whether GST enzymes delivered via exosomes to

GST negative cells may alter growth of those cells. Finally, because GST enzymes

are important in detoxification of drugs, the third aim was to (1) determine whether

GST over-expression increased resistance to the common anti-prostate cancer

chemotherapy docetaxel and (2) whether exosome delivery of GST enzymes to a

GST negative line could also provide resistance to docetaxel.

2.2 PROCEDURES

2.2.1 CELL CULTURE

Cell culture of various PCa cell lines (DU145, PC3, LNCaP and 22Rv1) and

stromal prostatic fibroblast cells (WPMY-1) was vital to the study in order to collect

exosomes from cell culture medium and to examine the contents of these

exosomes. Prostate cancer cells were originally obtained from the American Type

Culture Collection. Cultured cells were also used to produce cDNA for RT-PCR

expression analysis and cloning, to make protein lysates for Western blotting, and

were used in various functional studies.

2.2.1.1 PREPARATION OF EXOSOME-DEPLETED FOETAL CALF SERUM

Equal volumes (19.25 mL) of foetal calf serum (Thermo Fisher Scientific,

Victoria, Australia) and phenol-red free RPMI-1640 medium (Thermo Fisher

Scientific) were combined and the mixture centrifuged at 101,000xg for 1 h. The

24

supernatant was filtered through a 0.22 µm filter then centrifuged again at

101,000xg for 16 h. All centrifugations were performed at 4°C with the Optima™

XPN ultracentrifuge (Beckman-Coulter, New South Wales, Australia). The

supernatant was collected and stored at -30°C until used to make 5% exosome-

depleted FCS RPMI medium for culturing cells.

2.2.1.2 CELL CULTURE OF PROSTATE CANCER CELL LINES

Prostate cancer cell lines including DU145, PC3, LNCaP and 22Rv1 were

maintained in 5% exosome depleted FCS+RPMI. Cells were maintained in an

incubator at 37°C, 5% CO2 in a humid environment. Cells were grown and

maintained in either T-75 or T-175 flasks (Thermo Fisher Scientific). After reaching

80-90% confluency, conditioned media was collected for exosome isolation and the

cells washed carefully with phosphate buffered saline (PBS) for passaging. The PBS

was then aspirated and cells detached from the bottom of the flask with 0.25%

Trypsin (Thermo Fisher Scientific) with incubation at 37°C for 5 min. The trypsin was

then inhibited by the addition of fresh exosome-depleted media, the cells removed

from the flask and collected by centrifugation at 100 x g for 5 min. The supernatant

was aspirated and the cell pellet resuspended in fresh medium. Cells were then split

1:8 and seeded into new T-75 or T-175 flasks and placed back into the incubator for

continued cell passaging. Culture media was aspirated and replaced with fresh

media 3 days after passaging and the cells grown for a further 24-48 h before the

conditioned media was collected and cells passaged again. All conditioned media

was stored at 4°C for 1 week until exosome isolations were performed. For all

experiments described here, only cells passaged less than 40 times were used.

2.2.2 PROTEIN ANALYSIS

The GST profile in a range of PCa cell lines was vital to the study to see if

specific GSTs were present in more aggressive PCa cells or in less aggressive PCa

cells or in a variety of PCa cells. The presence or absence of GSTs in exosomes

was also important to see whether or not the GST expression profile followed that of

its parent cells or if it was inconsistent.

2.2.2.1 PREPARATION OF TOTAL PROTEIN LYSATES

Prostate cancer cell lines were grown to >90% confluence before total

protein lysates were made using M-PER mammalian protein extraction reagent lysis

buffer (Thermo Fisher Scientific) supplemented with Protease Inhibitor mini tablets

(Pierce, Thermo Fisher Scientific) and Phosphatase Inhibitor mini tablets (Pierce,

25

Thermo Fisher Scientific). Briefly, adherent cell monolayers were washed with ice-

cold Phosphate Buffered Saline (1X PBS is 8 g/L NaCl, 0.2 g/L KCl, 1.44 g/L

Na2HPO4, 0.24 g/L KH2PO4 adjusted to pH 7.4 with HCl in dH2O). The PBS was

removed before the addition of 1 mL M-PER buffer and the flask placed on a rocking

platform for 30 min at 4°C. The bottom of the flask was then scraped with a rubber

cell scraper and the lysate transferred to a 1.6 mL Eppendorf tube. Lysate was

passed through a 26 gauge needle then centrifuged at 21,130xg for 1 h at 4°C to

remove insoluble proteins and any remaining cell debris. After centrifugation, the

supernatant was carefully removed without disturbing the pellet and placed in a

sterile 1.6 mL Eppendorf tube. Soluble protein concentration was determined using

the BCA assay (Pierce, Thermo, Rockford, IL) with bovine serum albumin (BSA)

diluted from 0.1 mg/mL to 3 mg/mL used to determine the standard curve. Lysates

were then used immediately or stored at -80°C until required for future experiments.

2.2.2.2 PREPARATION OF EXOSOME PROTEIN LYSATES

Isolated exosome samples were lysed directly from the PBS suspension (see

Section 2.2.3.1 Exosome preparations from cell culture medium).Soluble protein

concentration was determined using the BCA assay (Pierce, Thermo, Rockford, IL)

with bovine serum albumin (BSA) diluted from 0.01 mg/mL to 1 mg/mL used to

determine the standard curve. Lysates were then stored at -80°C until required for

future experiments.

2.2.2.3 WESTERN ANALYSIS

Protein lysate samples (40 μg) and exosome samples (5 μg)for Western

analysis were resuspended in1X SDS Loading buffer (4X SDS-Loading Dye is 0.25

M Tris-HCl pH 6.8, 40% glycerol, 8% SDS, 0.01% Bromophenol Blue) with fresh β-

mercaptoethanol added to 5%, and the sample heated to 95°C for 5 min. Proteins

were separated by electrophoresis using SDS-PAGE gels (4% stacking and 10%

separating) at 150 V for 1 h and then transferred at 90 V for 90 min onto BioTrace™

NT nitrocellulose membrane (Pall, Pensacola, FL) using SDS-Transfer buffer (25

mMTris, 190 mM glycine, 20% methanol, pH 8.5). The membrane was washed once

in Tris buffered saline with Tween-20 (TBS-T is 8 g/L NaCl, 3 g/L Tris, 0.2 g/L KCl,

0.1% Tween-20 pH 7.4) then blocked with 5% skim milk / TBST for 1 h at room

temperature before being incubated with the primary antibody diluted in 5% skim

milk / TBST on a rocker at 4°Covernight (Table 2.1). Membranes were washed 3

times for 10 min each before bound primary antibodies were detected using

horseradish peroxidase-labelled secondary antibodies diluted in 5% skim milk / 1X

26

TBST with incubation for 1 h at room temperature. Blots were washed again 3 times

for 10 min each before detection of bound secondary antibody using LumiGLO®

chemiluminescent substrate (Cell Signaling Technologies, Danvers, MA) with

exposure to Hyperfilm ECL X-ray film (Amersham, Parramatta) for 2 to 15 min.

Table 2.1: Antibodies used for Western analysis of target protein expression

in total protein and exosome lysates.

Target protein Species and type of Ab Antibody company Dilution

TSG101 mouse monoclonal Abcam, Australia 1:1,000

GSTP1 rabbit polyclonal Proteintech, Chicago, USA 1:1,000

GSTT1 rabbit polyclonal Proteintech, Chicago, USA 1:1,000

GSTM1 rabbit polyclonal Proteintech, Chicago, USA 1:2,000

GSTM3 rabbit polyclonal Proteintech, Chicago, USA 1:5,000

GAPDH mouse monoclonal Abcam, Australia 1:10,000

anti-rabbit IgG goat anti-rabbit IgG,

HRP-conjugated

Proteintech, Chicago, USA 1:5,000

anti-mouse

IgG

horse anti-mouse IgG,

HRP- conjugated

Cell Signaling, Beverly, USA 1:5,000

2.2.3 EXOSOME PREPARATIONS AND VISUALISATION

Exosome preparations were vital to the study to examine the contents of the

exosomes. The isolated exosomes required visual confirmation to ensure that the

majority of the exosome preparations were in fact exosomes and not cellular debris.

2.2.3.1 EXOSOME PREPARATIONS FROM CELL CULTURE MEDIUM

Conditioned medium was first centrifuged at 300xg for 10 min to pellet and

remove viable cells then centrifuged again at 2,000xg for 20 min to remove dead

cells. The supernatant was then collected and centrifuged at 10,000xg for 30 min to

remove cell debris, filtered through a 0.22 µM filter to remove any large

microvesicles, then the supernatant centrifuged at 101,000xg in an Optima™ XPN

ultracentrifuge for 90 min to pellet the exosomes. The supernatant was discarded

and the exosome pellet washed in 800 µL sterile Dulbecco’s Phosphate Buffered

Saline (DPBS - Thermo Fisher Scientific) and centrifuged at 101,000xg for 2 h in an

Optima MAX-XP centrifuge (Beckman-Coulter). All centrifugation steps were

performed at 4°C.The wash solution was discarded and the exosome pellet

resuspended in 100 µL of sterile DPBS. An aliquot of 5 µL was stored at -80°C for

27

transmission electron microscopy (TEM) and the remaining 95 µL either used

immediately for quantitation and Western analysis or stored at -80°C for use in

future experiments.

2.2.3.2 VISUALISATION OF EXOSOMES USING TRANSMISSION ELECTRON

MICROSCOPY

The 5 µL exosome aliquots were thawed and applied to copper-coated

Formvar/Carbon grids (200 mesh) and allowed to incubate for 5 min at room

temperature before negatively staining with 5 µL uranylacetate for a further 5 min.

The grids were then examined with a JEOL 1400 electron microscope at 80 kV at a

range of magnifications from 12,000 X to 15,000 X magnification (Central Analytical

Research Facility, QUT).

2.2.4 EXPRESSION OF GST ENZYMES IN PROSTATE CANCER CELL LINES

To determine whether the over-expression of GSTs in PCa cells could also

increase the level of GSTs in exosomes shed by those cells, the full length coding

sequences of GSTP1 and GSTM1 were cloned, initially into pGEM-T Easy vector

(Promega, Madison, WI), then subsequently into the mammalian expression vector

pIRES-neo2 for constitutive expression in prostate cancer cells. Protein expression

of the transfected cell lines is driven by the human cytomegalovirus immediate-early

enhancer/promoter (CMV).

2.2.4.1 PRIMER DESIGN

Primers were designed using the NCBI database for the GSTP1 and

GSTM1genes for RT-PCR expression analysis and cloning.GSTT1 and GSTM3

primers were also designed for RT-PCR expression analysis only. Primers to

amplify the complete coding sequence of both GSTP1 (Genbank sequence

NM_000852.3) and GSTM1 (Genbank sequence NM_000561.3) were designed to

include restriction enzyme sites for directional cloning into the mammalian

expression vector pIRES-neo2 (Clontech, Mountain View, CA). Restriction enzyme

NheI was included on the forward primer that spanned the start codon and BamHI

was included on the reverse primer that spanned the stop codon (Table 2.2).Primers

were manufactured by Sigma-Aldrich (Sydney, Australia) and restriction enzymes

purchased from New England Biolabs (NEB, Ipswich, MA).

28

Table 2.2: List of primers used throughout the study including cloning primers

and screening primers.

Name Primer Sequence (5’ – 3’) Product Tm oC

Primers for RT-PCR expression analysis only

GSTT1 F CATCGTGGATCTGATTAAAGG

403bp

60°C GSTT1 R AGCTCCGTGATGGCTATGAGG

GSTM3 F GCTTTCATGTGCCGTTTTGAG

184 bp

65°C GSTM3 R CAGTTGTCCAGGTAGCTCTTCC

PBGD F CTTTCCAAGCGGAGCCATGTCTGG

377bp

60°C PBGD R CATGAGGGTTTTCCCGCTTGCAGA

Primers for both RT-PCR expression analysis and

cloning

GSTP1 F

GTCAGGCTAGCGCCACCATGCGGCCCTACACCGTGGT

C

632bp

60°C

GSTP1 R ACGTCGGATCCCTAACCCTCACTGTTTCCCGTTGC

GSTM1 F

GTCAGGCTAGCGCCACCATGCCCATGATACTGGGGTAC

TGG

656bp

72°C

GSTM1 R ACGTCGGATCCCTACTTGTTGCCCCAGACAGCCAT

Green: NheI; pink: Kozak sequence; blue: BamHI; underlined: start codon. Primers were

obtained from Sigma-Aldrich.

2.2.4.2 RNA EXTRACTION

RNA was isolated from DU145, PC3, LNCaP and 22Rv1 cells with1 mL TRIzol

Reagent (Thermo Fisher Scientific). Cells were scraped off the tissue culture flask

with a rubber cell scraper and moved to 1.6 mL Eppendorf tubes then incubated at

room temperature for 5 min. A 200 µL aliquot of chloroform was then added to each

tube and the tubes shaken vigorously for 15 sec, incubated for 2-3 min at room

temperature then centrifuged at 12,000xg for 15 min at 4°C. The solution separated

into three layers and the top layer (aqueous phase) containing the RNA was moved

to a new tube without disturbing the other layers. RNA was then precipitated using

500 µL of isopropanol added to the aqueous phase then incubated for 10 min at

room temperature and collected by centrifugation at 12,000xg for 10 min at 4°C. The

supernatant was removed and 1 mL of 70% ethanol added and centrifuged at

7,500xg for 5 min at 4°C. The supernatant was removed completely and the RNA

pellet was dried before being resuspended in 30 µL of UltrapureTMDNase/RNase

free distilled water (Invitrogen, Waltham, MA). The RNA concentration and purity

29

was determined using a Nanodrop1000 Spectrophotometer (Thermo Fisher

Scientific).

2.2.4.3 cDNA PREPARATION

Total RNA (2 μg) from each cell line was used for cDNA synthesis using

Superscript III (Thermo Fisher Scientific).To this, 1 µL of random primers, 2 µg RNA,

1 µL 10 mM dNTP mix was added to a 0.2 mL PCR tube and made up to 13 µL with

Ultrapure™ water. The tube was then incubated at 65°C for 5 min and placed on ice

for 1 min. The tube was briefly centrifuged and 4 µL 5x First-Strand Buffer, 1 µL 0.1

M DTT, 1 µL RNaseOUT Recombinant RNase Inhibitor and 1 µL of SuperScript™ III

RT was added. The reaction was mixed by gently pipetting up and down and

incubated at 25°C for 5 min. The reaction was then incubated at 55oC for 30-60 min

and then heat-inactivated at 70°C for 15 min. Three samples were made, 1 test and

2 negative controls. The negative control samples consisted of a no enzyme control

where the SuperScript™ III enzyme wasn’t added to the reaction to identify any

genomic DNA contamination. The second control contained reverse transcription

reagents only (no RNA control).

2.2.4.4 POLYMERASE CHAIN REACTION AMPLIFICATION OF GSTs

The complete coding sequences of GSTP1 and GSTM1were amplified using

the pair of primers specific to each gene (Table 2.2) in polymerase chain reactions

with 1 μL of the cDNA as the template and amplification using MyFi™ DNA

Polymerase (Bioline, Alexandria, NSW) in a 25 µL reaction following the

manufacturer’s recommendations. For semi-quantitative RT-PCR of other GST

genes, primers were also designed to GSTT1 and GSTM3 (Table 2.2). Primers for

specific amplification of the housekeeping gene porphobilinogen deaminase (PBGD)

were also used to confirm integrity of each cDNA sample. Cycling conditions

included an initial denaturation cycle of 95°C for 2 min followed by 30-40 cycles of

94°C for 30 sec, 60°C (GSTP1), 60°C (GSTT1), 72°C (GSTM1),65°C (GSTM3) or

60°C (PBGD)for 30 sec, 72°C for 30 sec, with a final extension cycle of 74°C for 7

min and were performed in a Mastercycler Pro (Eppendorf, Hamburg, Germany).

PCR products were separated on 1% tris-borate agarose gels containing a

1:40, 000 dilution of Ethidium Bromideat 115 V for 20 min(Thermo Fisher Scientific)

and photographed using a Gel Doc system (Syngene, MD, USA). The Bioline

HyperLadder 50 bp DNA molecular weight marker (Marker) was separated in the

first lane of each gel and used to confirm approximate PCR product sizes

30

2.2.4.5 LIGATION OF GSTP1 AND GSTM1 INTO THE pGEMT-Easy VECTOR

Following electrophoresis, the amplified PCR products corresponding to the

GSTP1 and GSTM1 coding sequences were excised from the gel and purified using

the Wizard® SV Gel and PCR Clean-Up System (Promega). The PCR products

were then ligated into the pGEMT-Easy vector (Promega) following the

manufacturer’s protocol. Reactions contained 1 µL vector, 1 µL T4 ligase, 1 µL 10 x

ligase buffer and 7 µL of the insert and were incubated at 14°C overnight.

2.2.4.6 HEAT SHOCK TRANSFORMATION INTO E. COLI AND SELECTION OF

TRANSORMED COLONIES

Aliquots of chemically competent TOP10 E. coli was available in our

laboratory. A 100 µL aliquot of cells was thawed on ice then combined with 3 µL of

the ligation reaction. After a 20 min incubation on ice, the cells were heated to 42°C

for 45 sec and then returned to ice for a further 2 min. The cells were then

resuspended in 400 µL of Luria-Bertani (LB) medium (1% Bacto-tryptone, 1% NaCl,

0.5% Bacto-yeast, pH 7.5) and incubated at 37°C for 1 h in a 200 rpm shaking

incubator. The transformed cells were then plated on LB Agar plates (LB medium

with 1.5% Bacto-agar) supplemented with 1 mM X-gal (5-bromo-4-chloro-3-indolyl-

β-D-galactopyranoside), 1 mM IPTG (Isopropyl β-D-1-thiogalactopyranoside) and 1

mM Ampicillin and grown overnight at 37°C.

2.2.4.7 PREPARATION OF PLASMID DNA

Six white colonies were selected and individually inoculated in 5 mL LB

medium supplemented with 1 mM Ampicillin and incubated overnight at 37°C with

shaking at 200 rpm. Plasmids were purified from bacterial cultures using the

Wizard®Plus SV Minipreps DNA Purification System (Promega) and the

manufacturer’s protocol was followed. The purified plasmid DNA was then quantified

using a NanoDrop and stored at 4°C. A double digest using restriction enzymes

NheI and BamH Iwas performed to confirm the presence of the GSTP1 or GSTM1

inserts prior to sequencing. Reactions were performed containing 1µg of plasmid

DNA, 1 µL of 10 x Buffer, 1 µL of NheI, 1µL of BamHI and was made to 10 µL with

DNase/RNase free water. This reaction was incubated for 1h at 37°C and the

fragments separated by electrophoresis through a 1.0% agarose gel at 110V for 20

min and photographed using the Gel Doc system (Syngene).

31

2.2.4.8 SEQUENCING OF CONSTRUCTS

pGEM-T Easy plasmids containing an insert of the appropriate size were

chosen for sequencing of the insert. Inserts were sequenced fully on both strands.

For each reaction, 800 ng of purified plasmid DNA was used with 6 pmol primer

(either T7 or SP6 - Table 2.3) and the reaction volume adjusted to 12 µL with

DNase/RNase free water. Sequencing was performed by the Australian Genome

Research Facility (AGRF) at the University of Queensland, Brisbane, Australia.

Identity of the sequences as GSTP1 or GSTM1 was confirmed by comparison to the

reference sequence in the BlastN database (NM_000852.3 and NM_000561.3,

respectively) and through alignment with this sequence using the Sequencher4.7

Demo software package (Gene Codes, Ann Arbor, MI).

Table 2.3: Primers used for sequencing of GSTP1 and GSTM1 inserts in

pGEM-T Easy and pIRES-neo2.

Primers were obtained from Sigma-Aldrich.

2.2.4.9 PRODUCTION OF THE pIRES-neo2 VECTOR

A glycerol stock of the pIRES-neo2 vector in E. coli JM109 was used to

inoculate 8 mL of LB medium supplemented with 100 µg/mL Ampicillin and this was

cultured overnight at 37°C in a shaking incubator at 200 rpm. Plasmid was isolated

from this bacterial culture using the Wizard® Plus SV Minipreps DNA Purification

System (Promega) and the manufacturer’s instructions and stored at -20°C until

required.

2.2.4.10 CLONING INTO THE pIRES-neo2 VECTOR

GSTP1 and GSTM1 coding sequences were removed from the pGEM-T

vector via a double digest of 1 µg of each construct with the restriction enzymes

NheI and BamHI as previously detailed. Similarly, 2 µg of the mammalian

expression vector pIRES-neo2(Figure 2.1) was also digested with NheI and BamHI.

The restriction digests were then electrophoresed into a 1.0% agarose gel at 115V

for 20 min to separate the insert sequence from the pGEM-T plasmid sequence and

Primer Name Vector Primer Sequence (5’ – 3’)

T7 pGEM-T Easy TAATACGACTCACTATAGGG

SP6 pGEM-T Easy ATTTAGGTGACACTATAG

CMV F pIRES-neo2 CGCAAATGGGCGGTAGGCGTG

32

the small fragment of the pIRES-neo2 multiple cloning site from the rest of the

vector. The bands corresponding to the GSTP1 and GSTM1 coding sequences and

pIRES-neo2 were then excised from the gel using a sterile scalpel blade and the

DNA purified from the gel slice using the Wizard® SV Gel and PCR Clean-Up

System (Promega) following the manufacturer’s instructions. The GSTP1 and

GSTM1 coding sequences were then ligated into the pIRES-neo2 vector using T4

DNA ligase (Promega). In each 10 µL reaction, 1 µL pIRES-neo2 vector was ligated

with 7 µL of GST insert in the presence of 1 µL T4 ligase and 1 µL 10x ligase buffer.

After an overnight incubation at 14°C, 3 µL of the ligation was used to transform E.

coli TOP10 cells by heat shock which were then spread onto LB agar plates

supplemented with 100 µg/mL Ampicillin for growth and selection overnight at 37°C.

As before, several individual colonies were then selected for plasmid isolation using

the Wizard Plus SV Minipreps DNA Purification System (Promega).

Figure 2.1: Vector map of the pIRES-neo2 vector and multiple cloning site.

GSTP1 and GSTM1 inserts were cut out of the pGEM-T Easy vector and subsequently

cloned into the pIRES-neo2 vector in the multiple cloning site (MCS).

2.2.4.11 SEQUENCING OF CONSTRUCTS

The plasmid DNA samples containing the appropriate sized insert were

selected after restriction analysis and 800 ng used in a sequencing reaction with 1

µL of the CMV forward primer (Table 2.3). As before, sequencing was performed by

33

AGRF (UQ) and the identity of the sequences confirmed by BlastN and Sequencher

4.7 alignments to the reference sequence of each gene.

2.2.4.12 SITE-DIRECTED MUTAGENESIS

A site-directed mutagenesis (SDM) protocol was used to change an

incorrect base in the GSTP1 sequence. Two overlapping SDM primers were

designed (Table 2.4), containing the correct base at nucleotide +5 of the coding

sequence, and these were used in a PCR reaction with the GSTP1 pIRES-neo2

vector (pIRES-P1) as the template. Amplifications were again performed using

MyFi™ DNA Polymerase in a 25 µL reaction that included 30 ng of purified DNA, 5

µL of 5 x Buffer, 1 µL forward SDM primer, 1µL reverse SDM primer, 1 µL MyFi™

DNA Polymerase, adjusted to 25 µL with Ultrapure™ distilled water. Cycling

conditions included an initial denaturation at 95oC for 2 min, then 19 cycles of 95oC

for 30 sec, 65oC for 30 sec and 72oC for 6 min (1 min per kb), with a final extension

step at 74oC for 10 min.

Table 2.4: List of site-directed mutagenesis primers.

Site-Directed

Mutagenesis

Primers

Primer Sequence (5’ – 3’)

GSTP1 - SDM F GTCAGGCTAGCGCCACCATGCCGCCCTACACCGTGGTCTATT

GSTP1 - SDM R AATAGACCACGGTGTAGGGCGGATGGTGGCGCTAGCCTCAG

Primers were obtained from Sigma-Aldrich.

Template plasmid in the PCR reaction product was then linearised by the direct

addition of 1 µL of methylation sensitive restriction enzyme DpnI and incubation at

37°C for 1 h. E. coli TOP10 cells were then transformed with 10µL of the DpnI-

digested PCR reaction using the heat shock protocol and plated on LB Agar-

Ampicillin plates. Plasmids were isolated from several individual colonies, the

presence of the insert confirmed by restriction digestion and the insert then

sequenced at AGRF to confirm the base change.

2.2.5 STABLE TRANSFECTION OF 22Rv1 CELLS WITH GSTP1 AND

GSTM1EXPRESSION VECTORS

Expression constructs containing GSTP1 and GSTM1 coding sequences

(pIRES-P1 and pIRES-M1, respectively) and the empty vector containing no insert

34

(VO) were used to transfect prostate cancer cell line 22Rv1 to create a model for

studying GST loading into exosomes.

2.2.5.1 LINEARISATION OF GST EXPRESSION CONSTRUCTS AND CO-

PRECIPITATION

The pIRES-P1, pIRES-M1 and pIRES-neo2 vectors (6 µg) were linearised

by digestion with NruI at 37°C for 1 h. The linearised plasmid was then precipitated

by adding 1/10th of the reaction volume of 3M sodium acetate, pH 5.2, and 3

volumes of ice-cold 100% ethanol. Precipitation was optimised by placing the

samples at -80°C for 1 hour before collecting DNA using centrifugation at 15,000

rpm for 30 min at 4°C. The supernatant was removed and the DNA pellet washed

using 200µL of ice-cold 70% ethanol with a second centrifugation at 15,000 rpm for

5 min at 4°C. The 70% ethanol wash solution was then removed and the pellet

dried thoroughly before being resuspended in Opti-MEM® medium (Thermo Fisher

Scientific) for transfection.

2.2.5.2 TRANSFECTION OF pIRES-neo2 CONSTRUCTS

Two six well plates were seeded with 22Rv1 cells to 60-80% confluency

and the cells allowed to attach overnight under normal cell culture growth conditions.

Cells were then transfected with either the linearisedpIRES-P1, linearised pIRES-M1

or linearised pIRES-neo2 (vector only, VO) using Lipofectamine® 2000 (Thermo

Fisher Scientific). Briefly, for each reaction, 6µL of Lipofectamine® 2000 reagent

was diluted in 150 µL Opti-MEM® Medium in a sterile Eppendorf tube and

separately, 2.5 µg of DNA was resuspended in 150 µL of Opti-MEM® Medium in a

second tube. Diluted DNA was then added to the tube containing the diluted

Lipofectamine® 2000 reagent slowly and gentle mixing by rotating the tube. After 5

min incubation at room temperature, 100 µL of the DNA-lipid complex was added to

the cell growth medium in two duplicate wells and the medium swirled gently to mix.

As a control, a mock transfection was performed using a sample to which no DNA

was added. Transfected cells were then incubated under normal growth conditions

for 4 h before changing medium to RPMI with 5% exosome-depleted FCS. Cells

were grown for 24-48 h before starting selection of stably transfected cells.

2.2.5.3 SELECTION OF TRANSFECTED CELLS

Stably transfected 22Rv1 cells were selected by resistance to G418. G418

was added to exosome-depleted medium at a final concentration of 1.5 mg/mL and

this was used for culture of transfected 22Rv1 cells with regular changes every 3-4

35

days for 2 weeks until the mock transfected cells had all died and colonies were

seen growing in the wells containing cells transfected with the expression constructs

pIRES-M1 and VO. Cells transfected with pIRES-P1 could not be generated despite

several attempts.

2.2.5.4 MAINTENANCE AND CRYOPRESERVATION OF TRANSFECTED CELLS

After selection, the transfected cells were expanded and samples

cryopreserved. Cells were removed from the flask by trypsin digestion and

passaged at a 1:6 split ratio into new flasks for continued cell culture. For long term

storage, samples of 700 µL of cells suspended in 5% exosome-depleted FCS

growth medium were combined with 200 µL of fresh exosome-depleted FCS and

100 µL of 100% dimethyl sulphoxide (DMSO) in 1.2 mL cryovials and slowly cooled

to -80°C in a Mr Frosty Freezing container (Thermo Fisher Scientific). Cells were

then stored in liquid nitrogen vapour until required.

2.2.6 FUNCTIONAL STUDIES

A variety of functional studies were performed to determine the effect of

GSTM1 expression on the growth characteristics of the transfected 22Rv1 cells.

Exosomes released from 22Rv1 transfected cells were examined for GSTM1

loading. The effect of GSTM1-loaded exosomes on the growth of normal prostatic

fibroblast cell line, WPMY-1 was also tested.

2.2.6.1 CELL PROLIFERATION ASSAY

The proliferation of transfected cells (pIRES-M1 and VO) was compared to

parental 22Rv1 cells using the colorimetric CellTiter 96® Aqueous One Solution Cell

Proliferation Assay (Promega). Cells were lifted from the tissue culture flask using

trypsin, counted with a cell counter (BioRad, Gladesville, NSW), and diluted to 2 x

105 cells/mL before 2000 cells were seeded into individual wells of 96-well cell

culture plates in 100 µL of RPMI containing 5% exosome-depleted FCS. Cells

were grown under normal cell culture conditions for 4, 24 and 72 h after seeding. At

each time point, 20 µL of Aqueous One Solution was added directly to the wells and

the reaction incubated at 37°C for 1 h prior quantitation of the formazan product by

measuring the absorbance at 490 nm on a Multiskan™ GO Microplate

Spectrophotometer (Thermo Fisher Scientific). Proliferation of 22Rv1 exosome-

andM1 exosome-treated WPMY-1 cells was also examined using the same method

and compared to proliferation of untreated WPMY-1 cells.

36

2.2.7 DOCETAXEL TREATMENT OF PROSTATE CANCER CELL LINES

GSTs are involved in detoxifying chemotherapeutic agents and treatments

with such agents may affect the level of GSTs that are present in PCa cells and the

exosomes these treated cells shed. Dose-response curves were made to determine

the IC50 of each cell line so a clinically appropriate concentration of Docetaxel would

be used in functional studies that examine whether or not GSTM1 loaded exosomes

could have a positive or negative effect on prostate cancer and stromal prostatic

fibroblast cell lines.

2.2.7.1 DOSE-RESPONSE CURVES OF PROSTATE CANCER CELL LINES

DU145, PC3, LNCaP and 22Rv1 cells were seeded into 96 well plates at

different seeding densities for the cell line - DU145 at 5,000 cells/well, PC3 at 8,000

cells/well, LNCaP at 10,000 cells/well and 22Rv1 cells at 5,000 cells/well. The

densities were based on how confluent the monolayer was after a 24 h incubation.

Cells were treated with Docetaxel diluted to various concentrations - 0.1, 1, 5, 10, 20

and 40nM in 5% exosome-depleted FCS RPMI growth medium. Cell proliferation

was quantitated at two different time points, 24 and 72 h using the CellTiter 96®

Aqueous One Solution Cell Proliferation Assay (Promega). Wells containing LNCaP

cells were pre-coated in poly-L-lysine to ensure cell adherence.

2.2.7.2 MEASURING PROLIFERATION OF PROSTATE CANCER CELLS AFTER

CO-CULTURING WITH DOCETAXEL AND GSTM1 EXOSOMES

DU145, PC3, LNCaP, 22Rv1 and WPMY-1 cells were seeded into 96 well

plates at various seeding densities - DU145 at 5,000 cells/well, PC3 at 8,000

cells/well, LNCaP at 10,000 cells/well, 22Rv1 cells at 5,000 cells/well and WPMY-1

at 5, 000 cells/well. A range of controls were used including: a vehicle control

(0.01% DMSO), cells treated with 100and 200 µg/mL 22Rv1-VO exosomes and

cells treated with 20 nM Doc only. The test cells were treated with both 20 nM Doc

and 100 or 200µg/mL 22Rv1-VO exosomes as well as treating with 20 nM Doc and

100 or 200µg/mL 22Rv1-M1 exosomes. Cell proliferation was quantitated at 3

different time points, 0, 24 and 72 h using the CellTiter 96® Aqueous One Solution

Cell Proliferation Assay (Promega). Wells containing LNCaP cells were pre-coated

in poly-L-lysine to ensure cell adherence.

37

Chapter 3: GST enzymes are present in Prostate

Cancer cells and shed exosomes

3.1 INTRODUCTION

In recent years, considerable scientific focus has turned to the study of

exosomes, endosomal-derived microvesicles with diameters of 20-120 nm, with

potential roles in communication, modulation of the immune system and preparation

of the tumour microvenvironment prior to dissemination of metastatic cells

[Mathivanan, 2010; Colombo, 2014; Greening, 2015; Webber, 2015]. Exosomes are

released from many different cells in the body, including both normal and tumour

cells, but their cargo of proteins, RNA sequences (miRNA, lncRNA) and genomic

DNA often reflects their cell of origin, and they are therefore considered a potential

source of biomarkers for diagnosis and prognosis of diseases including prostate

cancer [Giusti and Dolo, 2014]. In patients with prostate cancer, exosomes can be

isolated from bodily fluids including plasma, seminal fluid, saliva and urine using

centrifugation and antibody-enrichment methods [Bjinsdorp, 2013; Duijvesz, 2015;

Zheng, 2014; Jia, 2014; Taylor and Shah, 2015].

Extracellular vesicles released from the normal prostate also include

prostasomes, reported to differ from exosomes in their biogenesis, size and their

membrane structure. They range from 50 to 500 nm diameter, have a cholesterol-

and sphingomyelin-rich lipid multilayer in comparison to the more simple lipid bilayer

of exosomes [Poliakov, 2009]. Prostasomes also have a demonstrated function in

normal processes of human reproduction, releasing ATPase which increases the

motility of spermatozoa and delaying the acrosomal reaction to progesterone via the

membrane cholesterol [Stegmayr and Ronquist, 1982; Cross and Mahasreshti,

1997]. They also appear to protect sperm from phagocytosis by the immune cells of

the female reproductive system through an immunosuppressive activity [Kelly, 1991;

Poliakov, 2009].

The largest cancer-derived extracellular vesicles are the oncosomes, atypical

vesicles of 1-10 μm in diameter that form from non-apoptotic blebbing of the plasma

membrane [Al-Nedawi, 2008]. Di Vizio et al. (2012) reported an increase in the

number of large oncosomes present in the circulation of patients with metastatic

prostate cancer.

Glutathione-S-transferases (GSTs) are a family of related proteins important for

internal cell detoxification of toxic secondary metabolites formed during oxidative

stress agents via the enzymatic conjugation of glutathione (GSH) [Hayes, 2005;

38

McIlwain, 2006]. Conjugation of GSH with target compounds mostly results in the

formation of inactive products that are pumped from the cell. In this way, GST

enzymes expressed by cancer cells can neutralise the toxic effects of electrophilic

chemotherapeutic drugs including taxanes that generate reactive oxygen species in

cancer cells, and facilitate drug elimination through the ABC transporter MRP1

[Reinemer, 1997; McElwee, 2007]. Resistance to the key chemotherapeutic agent

Docetaxel (Taxotere™, Sanofi–Aventis, Paris, France), a taxoid derivative used

against many solid tumours, including advanced castrate-resistant and/or metastatic

prostate cancers, correlates with increased expression and catalytic activity of

GSTP1 [Park, 2002]. In rare cases however, the GSH-conjugate formed may even

be more reactive than the parent compound contributing to development of cancer

through cell and DNA damage. GSTT1, expressed at high levels in the prostate,

has been linked to the formation of genotoxic intermediates that increase prostate

cancer risk in this way [Sherratt, 1997]. Similarly, GSTM1 is also expressed in the

prostate but its potential contribution to prostate cancer has not been explored in

detail.

The hypothesis of this study is that detection of these four cytosolic GST

enzymes, GSTP1, GSTT1, GSTM1 and GSTM3 within microvesicles shed from

prostate cancer cells may 1) assist in screening of patient fluid samples for the

diagnosis of prostate cancer and 2) identify those cancers with increased risk of

developing resistance to chemotherapeutic agents like docetaxel. To explore this

hypothesis, GSTP1, GSTT1, GSTM1 and GSTM3 expression profiles were firstly

examined in prostate cancer cell lines using RT-PCR and their protein profiles

correlated with gene expression using Western blot analysis. Microvesicles were

then prepared from the medium of cultured prostate cancer cell lines and their

isolation verified with Western analysis and transmission electron microscopy.

Which GST proteins were present within microvesicles released from different

prostate cancer cell line cells was then determined using Western blot analysis.

Finally, an over-expression approach was used to determine whether exogenously

over-expressed GST enzymes could be loaded into microvesicles produced by

GST-naïve prostate cancer cells.

39

3.2 MATERIALS AND METHODS

The common materials and methods relevant to this chapter have been outlined

in Chapter 2.

3.3 RESULTS

3.3.1 EXPRESSION OF GSTP1, GSTT1, GSTM1 AND GSTM3 GENES IN

PROSTATE CANCER CELL LINES

Reverse transcriptase polymerase chain reaction (RT-PCR) was used to

determine the expression profile of GST enzymes in prostate cancer cell lines to

select model cell lines for use in the study. RNA was extracted from >90% confluent

monolayers of prostate cancer cell lines DU145, PC3, LNCaP and 22Rv1 cells then

reverse transcribed into cDNA. This cDNA was then used as the template in PCR

reactions with primer pairs designed to specifically amplify a PCR product

corresponding to the individual GST enzymes GSTP1, GSTT1, GSTM1 and GSTM3

and PBGD as a housekeeping control. PCR products were separated by agarose

gel electrophoresis, visualised by staining with Ethidium Bromide and the size of the

product estimated by comparison to a DNA molecular weight marker (Figure 3.1).

Amplification of the housekeeping control PBGD was used to confirm the integrity of

the cDNA from each cell line with an amplified product seen in the appropriate

lanes. All negative controls contained no amplified product as expected. A strongly

staining band corresponding to GSTP1 was present in the lanes for cell lines

DU145, PC3, LNCaP but this band stained only faintly in the lane for 22Rv1 cells.

This suggests that GSTP1 is robustly expressed in DU145, PC3 and LNCaP but is

only poorly expressed in 22Rv1. The GSTT1 gene was expressed by all cancer cell

lines but as this band stained only very faintly this result would suggest that the

gene is not highly expressed in any of these lines (Figure 3.1). The GSTM1 gene

was expressed in DU145, PC3 and LNCaP cells and did not appear to be expressed

by the 22Rv1 cell line. GSTM3 appeared to be highly expressed by all of the cell

lines.

40

Figure 3.1: Expression profile of GST enzymes in prostate cancer cell lines

using RT-PCR. A PCR product corresponding to the GSTP1 cDNA sequence was present

in samples from DU145, PC3, LNCaP and faintly in 22Rv1 cell lines at the expected size of

632 bp.GSTT1 was seen in DU145, PC3, LNCaP and 22Rv1 cells with an expected size of

403 bp.GSTM1 was present in DU145, PC3 and LNCaP cell lines at 656 bp and GSTM3 in

all prostate cancer cell lines at 184bp. Various controls including a reaction with no reverse

transcriptase (RT -ve), another with no RNA (RNA -ve) and a non-template control (PCR -ve)

were included with each PCR. PBGD was used as a housekeeping gene which was present

in samples from all of the prostate cancer cell lines at 377 bp. The Bioline HyperLadder 50

bp DNA molecular weight marker (Marker) was separated in the first lane of each gel and

used to confirm approximate PCR product sizes.

41

3.3.2 EXPRESSION OF GSTP1, GSTT1, GSTM1 AND GSTM3 PROTEIN IN

PROSTATE CANCER CELL LINES

To determine whether protein expression levels correlated with gene

expression, total protein lysates were prepared using the same prostate cancer cell

lines and separated using reducing SDS-PAGE then transferred to nitrocellulose for

Western blot analysis using antibodies specific to each of the four selected GST

enzymes. A strong immunoreactive band corresponding to GSTP1 at the predicted

molecular weight of 23 kDa was seen in the lanes corresponding to DU145 and PC3

but was only very faint in LNCaP and absent in 22Rv1 (Figure 3.2A). GSTT1 was

identified at the monomeric size of 27 kDa in all prostate cancer lysates (Figure

3.2B). GSTM1 was identified as a band at 55 kDa in the LNCaP protein sample with

no band consistent with the monomer of 27 kDa seen in any of the samples (Figure

3.2C and data not shown). Given that samples were denatured by boiling and

reduced with -mercaptoethanol, it is likely that this 55 kDa band is dimeric GSTM1

and the two monomers in this are covalently linked through interactions other than

disulphide bridges. The expression pattern of GSTM3 was similar to that of GSTT1,

with more protein detected in the DU145 and PC3 cell lines when compared with

LNCaP and 22Rv1, although GSTM3 was also detected in these cell lines as a band

of much lower intensity when related to GAPDH (Figure 3.2D). An antibody to the

housekeeping control GAPDH confirmed even protein loading in Westerns (Figure

3.2 A, B, C) with a clear immunoreactive band identified at 37 kDa in each lane and

of a similar intensity.

42

Figure 3.2: Western blot images comparing GST content in prostate cancer

whole cell lysates. The proteins of interest, GSTP1, GSTT1, GSTM1 and GSTM3 were

examined in total protein lysates from four prostate cancer cell lines as indicated. Detection

of GAPDH using a specific antibody was used as a control for each lysate and even loading

of protein is indicated by consistent immunoreactivity seen at 37 kDa in the lower panels. (A)

A single band was detected using an antibody specific to GSTP1 at the monomeric size of

23 kDa and was present in the DU145, PC3 and faintly in the LNCaP lysates. (B) A single

band was detected using an antibody specific to GSTT1 in all of the prostate cancer lysates

at the monomeric size of 27 kDa. (C) A single band was detected using an antibody specific

to GSTM1 only in LNCaP lysates and as a dimeric protein at 55 kDa. (D) A single band was

detected using an antibody specific to GSTM3 and was present in all of the lysates at the

monomeric size of 27 kDa.

37 kDa

DU

145

GSTP1 23 kDa

GAPDH 37 kDa

PC

3

LN

Ca

P

22R

v1

A B

DU

145

PC

3

LN

Ca

P

22R

v1

GSTT1 27 kDa

GAPDH 37 kDa

D

GSTM3

PC

3

LN

Ca

P

22R

v1

DU

145

GAPDH

27 kDa

DU

145

LN

Ca

P

22R

v1

55 kDa GSTM1 (dimer)

C

PC

3

GAPDH 37 kDa

43

3.3.3 PROSTATE CANCER CELL LINES PRODUCE EXOSOMES AND

PROSTASOMES UNDER NORMAL CELL CULTURE CONDITIONS

Prostate cancer cells release microvesicles that range in size from 30 nm to

10 µm and for this reason ultracentrifugation was chosen as the method of

microvesicle isolation for this study, despite the increased risk of collection of

contaminating cellular debris. Prostate cancer cell lines were grown in medium

supplemented with foetal calf serum that had been depleted of any bovine

microvesicles using a validated ultra-centrifugation protocol [Théry, 2006]. Medium

was removed from the cell culture flasks in which various prostate cancer cell lines

were growing and microvesicles isolated from this using the method described in

Chapter 2 (Method 2.2.3 – Exosome preparations and visualisation). Microvesicles

are 30 – 120 nm in diameter and in each preparation, vesicles with diameters that

fall within this range can be clearly identified but these microvesicles do vary in size

and interestingly, there also appears to be two clear subsets of exosomes that can

be separated depending on their size – small exosomes with an average diameter of

40 nm (range 20 nm to 80 nm) and large exosomes with an average diameter of 100

nm (range 80 nm to 120 nm). In addition, several much larger microvesicles can

also be seen and because these larger vesicles have an average diameter of 200

nm (range >120 nm to 250 nm) these are most likely prostasomes (reported

diameters between 50 and 500 nm). No microvesicles larger than this (eg.

oncosomes of 1 – 10 µm) were identified but due to their much larger size these

may have been removed from the exosome purification process during removal of

cellular debris. In all microvesicle preparations, vesicles of different sizes can be

seen, some of which have the cup-shaped morphology characteristic of

microvesicles prepared for TEM. Figure 3.3 shows vesicles that were shed from

each prostate cancer cell line and the summary of these results is shown in Table

3.1.DU145, PC3 and LNCaP cells shed exosomes (averaging 30-120 nm in

diameter) whereas 22Rv1 cells shed larger vesicles, the prostasomes (averaging

150 nm in diameter).

44

Figure 3.3: Transmission Electron Microscopy image of microvesicles isolated

from conditioned medium. Conditioned medium was collected from prostate cancer

cells after 24 h of incubation and differentially centrifuged to remove cellular debris. The

microvesicles were then resuspended in DPBS and analysed by TEM using a negative

staining method with 1% uranyl acetate with the microscope operating at 80 kV. The

microvesicles appear spherical and cup shaped ranging from 20-250 nm. (A) DU145 derived

microvesicles ranged from 20-200 nm at 12,000X magnification. (B) PC3 derived

microvesicles ranged from 40-230 nm at 12,000X magnification. (C) LNCaP derived

microvesicles ranged from 40-230 nm at 15,000X magnification. (D) 22Rv1 derived

microvesicles ranged from 40-150 nm at 12,000X magnification. A scale bar indicating 500

nm is shown for each image.

A B

C D

DU145 PC3

LNCaP 22Rv1

45

Table 3.1: Quantitation of microvesicles (exosomes and prostasomes) in one TEM grid from conditioned media of prostate cancer cell

lines based on diameter.

Table 3.1 summaries the different types and sizes of vesicles that were found from each prostate cancer cell line. DU145 and LNCaP cells

mostly shed small exosomes whereas PC3 cells shed mostly large exosomes. 22Rv1 cells released an average number of small and large

exosomes however they also shed more prostasomes than the other cell lines.

Cell line Total Number of Small

Exosomes

Average diameter

and range

% of total

Number of Large

Exosomes

Average diameter and

range

% of total

Number of Prostasomes

Average diameter and

range

% of total

DU145 44 23 30 nm 20 - 80nm

52% 17 120 nm 80 - 120 nm

39% 4 200 nm 120 - 250 nm

9%

PC3 23 7 40 nm 20 - 80 nm

30% 12 110 nm 80 – 120 nm

52% 4 230 nm 120 – 250 nm

17%

LNCaP 23 15 40 nm 20 - 80 nm

65% 4 100 nm 80 – 120 nm

17% 4 230 nm 120 – 250 nm

17%

22Rv1 47 17 40 nm 20 - 80 nm

36% 18 120 nm 80 – 120 nm

38% 12 150 nm 120 – 250 nm

25%

46

3.3.4 MICROVESICLES RELEASED FROM PROSTATE CANCER CELLS IN

CULTURE CONTAIN GST ENZYMES

The TEM results indicated the presence of microvesicles and this was further

verified with Western blot analysis to detect the exosome associated protein

TSG101. Prepared microvesicles resuspended in DPBS were lysed using M-PER

mammalian lysis buffer and the proteins separated by SDS-PAGE before being

transferred onto nitrocellulose membranes. The membranes were then

immunoblotted using antibodies specific to TSG101 and a single immunoreactive

band was identified in all exosome samples at the predicted molecular weight of 48

kDa (bottom panels in Figure 3.4). TSG101 often appears highly enriched in 22Rv1

preparations most likely due to the differences in exosome biogenesis for the

different cell lines which was also seen in another study [Yoshioka, 2013].

To determine whether microvesicles released from prostate cancer cells

contain GST enzymes, exosome preparations were also examined by Western blot

analysis using antibodies specific to each of the four chosen GST enzymes (Figure

3.4). Expression of TSG101 was used as a positive control for exosomes.

Consistent with the expression of GSTP1 protein in prostate cancer cell lysates, a

single band at 23 kDa, the predicted size of monomeric GSTP1, was detected in

microvesicle preparations from DU145 and PC3 but not in LNCaP or 22Rv1

preparations (Figure 3.4A). Although the GSTT1 protein was consistently present in

all cell lysates at the monomeric size of 27 kDa, no GSTT1 protein was detected in

exosomes from any of the four prostate cancer cell lines (Figure 3.4B).GSTM1 was

detected in microvesicle preparations from LNCaP consistent with the high

expression seen in the Western analysis of LNCaP total protein lysate (compare

Figure 3.2C with Figure 3.4C). The size of the protein bands detected in these

samples were however different. GSTM1 in the LNCaP total protein lysate was 55

kDa and is most likely dimeric protein but the GSTM1 protein in the LNCaP

exosomes was present at 27 kDa and is probably the monomeric form. The

Western analysis of LNCaP exosomes also detected a band at 75 kDa but the

identity of this band is unknown and this was not detected in the LNCaP total protein

lysate. Similarly, DU145 exosomes also appear to contain monomeric GSTM1

protein despite having undetectable expression in the total protein lysate sample

tested previously (compare Figure 3.2C with Figure 3.4C). The higher molecular

weight band at 75 kDa is also detected in the DU145 exosome sample. GSTM1

expression was not seen in PC3 and 22Rv1 samples, and this is consistent withthe

result gained using total protein lysates. GSTM3 was present in all of the lysates at

the monomeric size of 27 kDa, however similarly to GSTM1, it was only present in

47

the DU145 and LNCaP exosome preparations at 27 kDa (Figure 3.4D). The

comparison between lysate and exosome content is highlighted in Table 3.2.

Figure 3.4: Western blot images comparing GST content in exosome

preparations shed from prostate cancer cells. The GST proteins of interest, GSTP1,

GSTT1, GSTM1 and GSTM3 were examined in exosome preparations. Exosome marker

TSG101 was used as an exosome marker for each sample and a single band of the correct

molecular weight (48 kDa) was present in all exosome preparations - seen in the lower

panels. (A) A single band corresponding to GSTP1 was present in the DU145 and PC3

preparations at the monomeric size of 23 kDa. (B) No bands corresponding to GSTT1

(predicted molecular weight of 27 kDa) were present in any of the exosome preparations

from the prostate cancer cells. (C) Two immunoreactive bands were seen in the Western bot

analysis using a GSTM1 specific antibody but only in the DU145 and LNCaP exosome

preparations. One was at the monomeric size of 27 kDa but a second was identified at 75

kDa and the identity of this band is unknown (?). (D) An immunoreactive band corresponding

to GSTM3 was present in only the DU145 and LNCaP exosome preparations and was at the

predicted monomeric size of 27 kDa.

D

DU

145

GSTP1 23 kDa

TSG101 48 kDa

PC

3

LN

Ca

P

22R

v1

A

DU

145

LN

Ca

P

22

Rv1

27 kDa GSTT1

B

PC

3

TSG101 48 kDa

C

DU

145

PC

3

LN

Ca

P

22R

v1

GSTM1 (dimer)

75 kDa

TSG101 48 kDa

GSTM1

55 kDa

27 kDa

? GSTM3

PC

3

LN

Ca

P

22R

v1

DU

145

TSG101 48 kDa

27 kDa

48

Table 3.2: Visual comparison of GST enzymes found in whole cell lysates and

exosome preparations relative to loading controls.

Cell lines

GST enzyme

Level of expression in total cell lysate

Level of expression in exosome lysate

DU145 GSTP1 +++ +++

GSTT1 ++ ND

GSTM1 ND ++ 75 kDa; + 27 kDa

GSTM3 ++ +++

PC3 GSTP1 +++ +++

GSTT1 ++ ND

GSTM1 ND ND

GSTM3 ++ ND

LNCaP GSTP1 + ND

GSTT1 + ND

GSTM1 +++ 55 kDa ++ 75 kDa; +++ 27 kDa

GSTM3 + +++

22Rv1 GSTP1 ND ND

GSTT1 + ND

GSTM1 ND ND

GSTM3 + ND +++ - strongly expressed, ++ - expressed, + - weakly expressed, ND – not detected.

Table 3.2 summarises the different levels of GST expression in lysates compared to

exosomes. This summary was performed visually based off a number of different

western blotting results for both the lysates and exosome preparations.

3.3.5 LOADING OF GST ENZYMES INTO PROSTATE CANCER

MICROVESICLES

Comparison of GST enzyme levels in total protein lysates and microvesicles

(Table 3.2) suggested that in some cases GSTs may be preferentially loaded into

exosomes (eg. GSTM1 and DU145 - low lysate level but high exosome level). To

determine whether increasing GST expression in a naïve cell also increases the

GST enzyme in shed exosomes, the complete coding sequences of GSTP1 and

GSTM1 were cloned into the mammalian expression vector pIRES-neo2 to create

expression constructs suitable for the stable transfection of the prostate cancer cell

line 22Rv1 that does not express either of these two GST enzymes. Integrity of the

coding sequences was confirmed by sequencing fully on both strands and in the

course of this a single nucleotide error was noticed in the sequence of GSTP1. As

this incorrect nucleotide caused an amino acid substitution from proline at amino

acid position 2 (P2) to arginine and is a change that might affect protein folding and

therefore enzyme function, this sequence error was corrected using site directed

49

mutagenesis prior to transfection (Appendix A and B). The GSTM1 sequence was

consistent with the reference sequence from the Genbank database

(NM_000561.3).

Despite a total of four individual attempts, 22Rv1 cells stably expressing

GSTP1 could not be selected. In each experiment, a G418-resistant population

containing the empty vector remained after the two week selection period

suggesting that expression of GSTP1 did not allow continued growth of the 22Rv1

cells. Four individual heterogenous populations of stable, G418-resistant 22Rv1

cells were however obtained from the cells transfected with the GSTM1 expression

vector and expression of the GSTM1 protein was confirmed using Western analysis

(Figure 3.5). Robust expression was seen in each of the four replicate populations

as evidenced by the strongly immunoreactive bands in the appropriate lysate lanes

when compared with the parental 22Rv1 cells and 22Rv1 cells transfected with the

empty vector (22Rv1-VO). In all cases the protein was migrating at 27 kDa

consistent with monomeric GSTM1 protein.

To determine if GSTM1 protein exogenously over-expressed by the 22Rv1

cells was loaded into microvesicles, microvesicles were prepared from the medium

and analysed using Western analysis. GSTM1 was consistently detected, albeit at

low levels, in the microvesicle preparations derived from the GSTM1-transfected

22Rv1 cells at the monomeric size of 27 kDa, with comparison to both the parental

22Rv1 and 22Rv1-VO cells. There was no band present at 75 kDa that had

previously been seen in DU145 and LNCaP derived exosomes. TSG101 was again

used as an exosome marker and loading control that showed even loading of the

microvesicle preparations at 48 kDa (Figure 3.6). Microvesicles were also imaged

with TEM (Figure 3.7) to observe any morphological changes that may have

occurred after transfection and as a consequence of exogenous GSTM1 expression.

The GSTM1 over-expressing cells mostly shed large exosomes which is different to

the 22Rv1 parental cells that shed a majority of prostasomes. This change in

vesicles may be due to G418 selection pressure.

50

Figure 3.5: Western blot showing over-expression of GSTM1 in transfected

22Rv1 whole cell lysates. The protein of interest, GSTM1 was present at the monomeric

size of 27 kDa in all transfected 22Rv1 replicates (22Rv1-M1). It was not present in the

vector only 22Rv1 (22Rv1-VO) cells or the 22Rv1-Parental cells as expected. GSTM1 was

present as a monomeric protein. GAPDH was used as a loading control.

Figure 3.6: Western blot showing the packaging of GSTM1 into shed

exosomes derived from GSTM1 transfected 22Rv1 cells.GSTM1 was detected in

all exosomes preparations from 22Rv1-M1 cells and not expressed in 22Rv1-P or 22Rv1-VO

cells as expected. GSTM1 was present only as a monomeric protein. TSG101 was used as

an exosome marker and loading control and indicated even loading at 48 kDa.

GSTM1 27 kDa

22R

v1-M

1 #

1

22R

v1-M

1 #

2

22R

v1-M

1 #

3

22R

v1-M

1 #

4

GAPDH 37 kDa

22R

v1-V

O

22R

v1-P

GSTM1 27 kDa

TSG101 48 kDa

22R

v1

-M1 #

4

22R

v1

-M1 #

1

22R

v1

-M1 #

2

22R

v1

-M1 #

3

22R

v1

-P

22R

v1

-VO

51

Figure 3.7: Transmission Electron Microscopy image of exosomes isolated

from GSTM1 over-expressing 22Rv1 conditioned medium. Conditioned medium

was collected from 22Rv1-M1 cells after 24 h of incubation and differentially centrifuged to

remove cellular debris. The exosome preparations were then resuspended in DPBS and

analysed with TEM by negatively staining with 1% uranyl acetate. The exosomes appeared

spherical and cup shaped at an approximate size of 30-200 nm. The image is at 15,000X

magnification and the microscope was operating at 80 kV.

52

Table 3.3: Quantitation of microvesicles (exosomes and prostasomes) in one TEM grid from GSTM1 over-expressing 22Rv1 cells and

22Rv1 parental cells (from Table 3.1) based on diameter.

Table 3.3 compares the different vesicles shed from the 22Rv1 parental and the GSTM1 over-expressing cell line. The engineered cells shed

more large exosomes than the parental cells and conversely shed a smaller amount of prostasomes than the parental cells.

Cell line Total Number of Small Exosomes

Average diameter and range

% of total

Number of Large Exosomes

Average diameter and range

% of total

Number of Prostasomes

Average diameter and range

% of total

22Rv1 47 17 40 nm 20 - 80 nm

36% 18 120 nm 80 – 120 nm

38% 12 150 nm 120 – 250 nm

25%

22Rv1-M1 42 6 30 nm 20 - 80 nm

14% 33 100 nm 80 – 120 nm

79% 3 200 nm 120 – 250 nm

7%

53

3.4 DISCUSSION

GST enzymes are involved in detoxification and if we are able to identify GST

positive enzymes in the blood of patients with prostate cancer and in the exosomes,

we may be able to predict patients who may not respond favourably to treatment

with chemotherapeutics. This would significantly decrease the amount of money

spent on treatment and more importantly, provide disease specific patient care with

fewer side effects. The detection of four particular cytosolic GST enzymes - GSTP1,

GSTT1, GSTM1 and GSTM3 - were studied where GSTP1 and GSTT1 have

previously been linked to prostate cancer however there is little information

regarding GSTM1 and GSTM3.

A variety of PCa cells were used throughout this study, representing the change

from androgen dependent to androgen independent prostate cancer. 22Rv1 cells

represented androgen responsive cancer, LNCaP cells represented androgen

dependent cancer and DU145 and PC3 cells represented androgen independent

cancer [Chlenski, 2001; Marchiani, 2010]. The GST profiles were conducted in each

of these cell lines to see if certain GSTs were expressed in different cancer stages.

GSTP1 was present at the monomeric size in DU145 and PC3 cell lysates and

faintly in LNCaP and 22Rv1 lysates. This indicates that GSTP1 may only be

expressed in androgen independent cases, such as during castrate resistant

prostate cancer. GSTT1 was present as both a monomer and dimer in all cell

lysates showing that it is present throughout the progression of prostate cancer.

GSTM1 was only present in LNCaP lysates as a putative dimer and not in any other

lysates suggesting that it may only be present in androgen dependent prostate

cancer. Also the transition from a homodimer, to a dimer to then a monomer is

protein concentration dependent. If the first transition is bimolecular, subunit

dissociation occurs and produces a monomer however if the second transition is

protein concentration dependent, then the subunits dissociate after the formation of

a dimer, requiring an extra dissociation step. The molecular origin for the ability of

dissociated mu class subunits to exist as stable structures may involve interactions

that are specific for GSTM1 like the presence of the mu loop [Hornby, 2000]. The

extra step required for dissociation of the dimer may explain why GSTM1 was only

consistently present as a dimer. The specificity of the GSTM1 antibody could also

be tested by knocking down GSTM1 and seeing if the bands remained. Like GSTT1,

GSTM3 was present in all of the lysates as a monomeric protein throughout the

progression of prostate cancer. It was unusual that GSTM3 didn’t follow the same

lysate expression profile as GSTM1 however this may be due to the functionality of

54

the enzymes themselves. Differences in the gene expression levels and protein

levels may indicate different levels of regulation for these GST enzymes. For

example, the GSTM1 gene was expressed in DU145, PC3 and LNCaP cells yet

protein was only detected in the sample from the LNCaP line. This may indicate a

further mechanism for regulation of this enzyme occurring after transcription in the

DU145 and PC3 cells or it might show that the GSTM1 protein, if made, is rapidly

degraded by the cell. Understanding all mechanisms involved in GST regulation is

important. To further understand why the extra bands were seen, GSTM1 could be

knocked down using siRNA to see if the band is no longer present which would also

validate the results.

Recent studies of exosomes from various cancer cell lines recommended

isolating these small vesicles by several different methods including

ultracentrifugation, sucrose or OptiPrep™ density-based separation and

immunoaffinity capture with anti-EpCAM-coated magnetic beads [Tauro, 2012].

These methods all have their own limitations as density-based methods exclude

vesicles based on size and buoyancy and immunoaffinity methods require

microvesicles to have EpCAM presented on the membrane. Ultracentrifugation was

used in this study as it captures a more representative pool of microvesicles. By

using ultracentrifugation, microvesicle preparations were likely to include both

exosomes and prostasomes with an overlap in sizes and differences in the

membrane composition as exosomes have a simple lipid bilayer and prostasomes

have multilayers rich in cholesterol and sphingomyelin. Prostasomes can be

differentiated from exosomes by size and also by analysing cholesterol and

sphingomyelin content. Prostasome multilayers are unable to be visualised with

TEM however they can be identified by their size as they are generally much larger

than exosomes [Chiasserini, 2015].

During this study, the prostasomes and exosomes were not differentiated as

exosome preparations yielded small and large exosomes and prostasomes so this

collective group of vesicles were referred to as microvesicles. The microvesicles

were directly loaded and probed with the appropriate GST antibodies. GSTP1 was

present in the DU145 and PC3 microvesicles as a monomer, identical to the protein

lysate profile. GSTT1 did not appear to be packaged into any of the shed

microvesicles which was unusual as it was present in all of the lysates. GSTM1 was

present in LNCaP microvesicles as a monomer and surprisingly, in DU145

microvesicles. This suggests that GSTM1 may in fact be preferentially loaded into

DU145 microvesicles. Similarly to the GSTM1 profile, GSTM3 was also present only

in DU145 and LNCaP microvesicles as a monomer. Since it was present in all of the

55

prostate cancer lysates and only in the DU145 and LNCaP microvesicles, this

shows that similar to GSTM1, GSTM3 is also preferentially loaded into particular

microvesicles. Given that GST enzyme function relies on formation of a stable

dimer, the observation that GSTM1 is present as a monomer in microvesicles might

suggest that the enzyme is in an inactive state. Furthermore, although the identity

of the 75 kDa band seen on Western blot analysis using protein lysates from LNCaP

and DU145 microvesicles is unknown, its immunoreactivity with the GSTM1-specific

antibody may have identified this as a complex of GSTM1 protein and an as yet

unidentified binding partner. This band was present in sample that had been both

denatured and reduced which would suggest a strong covalent interaction if this is

the case. An immunoprecipitation procedure coupled with mass spectrometry

analysis of this complex might clarify this.

To determine the best candidate for the over-expression studies, the protein

and microvesicle profiles were studied. It was apparent that over-expressing GSTP1

and GSTM1 in 22Rv1 cells was the most appropriate choice. This was due to the

fact that neither of these enzymes were present in the protein lysates or

microvesicles. GSTT1 and GSTM3 were not chosen as GSTT1 was not present in

the microvesicles and the biological activity of GSTM3 is very similar to the activity

of GSTM1. It was important to see if over-expressing GSTs could potentially lead to

the GSTs being loaded into the microvesicles. After performing site-directed

mutagenesis on GSTP1 to correct the incorrect base, it was transfected into 22Rv1

cells with Lipofectamine. This was attempted four times and each time the 22Rv1-

VO cells remained while the transfected 22Rv1-P1 cells died. This suggests that

since GSTP1 was unable to be expressed in 22Rv1 cells, it could be tumour

suppressive or downregulated by DNA CpG methylation, however further research

would need to be performed to confirm this [Gray, 2012]. There is much controversy

about its expression in relation to prostate cancer where the progression of cancer is

related to silencing GSTP1 by methylation and conversely, GSTP1 over-expression

being linked to tumour progression. A study by Lee et al. showed that GSTP1 was

downregulated in prostate tumour samples compared to the normal tissue [Lee,

1994]. The results from this study show that GSTP1 probably contributes to prostate

cancer progression through the production of GST intermediates which cause DNA

damage. GSTP1 is most likely cell line dependent as it was highly expressed in

DU145 and PC3 mRNA and protein as well as LNCaP mRNA. GSTP1 was only

faintly present in 22Rv1 mRNA and not at all in protein lysates suggesting that these

cells don’t require GSTP1. GSTM1 was able to be stably expressed in 22Rv1 cells.

56

This would suggest that the targets of these two GST enzymes were different in the

22Rv1 cells.

Future studies could include screening patient samples to see if patient derived

exosomes contain GST enzymes. Blood samples and seminal fluid appear to be the

most appropriate sample types. Seminal fluid is a more appropriate fluid to examine

as about 30% of seminal fluid comes from the prostate and naturally contain more

prostasomes [Tavoosidana, 2011; Aalberts, 2013]. GSTP1 content in exosomes in

particular may be a good candidate for monitoring prostate cancer progression as

methylated GSTP1 is currently being used in some prognostic studies. Early

disease states may be detected using seminal fluid as the duct structures would still

be intact and once tissue architecture has become disrupted by the tumour itself,

there would most likely be more exosomes released into the bloodstream.

57

Chapter 4: GSTM1 over-expression affects the

behaviour of Prostate Cancer and associated stromal

cells

4.1 INTRODUCTION

Glutathione-S-transferase enzymes are a superfamily of enzymes that, through

their ability to catalyse the nucleophilic addition of the glutathione tripeptide (GSH -

-Glu-Cys-Gly) to various reactive electrophiles, play a key role in cellular

detoxification and protection from carcinogens, reactive oxygen species and

chemotherapeutic agents. The expression of several GST enzymes is commonly

dysregulated in cancers including prostate cancer through a variety of different

cellular mechanisms including gene amplification, transcriptional activation, protein

stabilisation and genetic polymorphisms [Matthias, 1998; Di Pietro, 2010].

Dysregulation can be due to reduced or loss of expression or conversely, over-

expression and both have been linked to cancer progression. Reduced expression

of GST mu family enzymes is predicted to contribute to cancer initiation through

uncontrolled and increased oxidative stress and resultant DNA damage in the

prostate gland [Frohlich, 2008]. A meta-analysis combining data from 29 different

studies (4564 cases and 5464 controls) showed that individuals with a GSTM1 null

genotype have an increased risk of developing prostate cancer [Mo, 2009]. Over-

expression of GST enzymes however, has been linked by several studies to

resistance to chemotherapeutic drugs. For example, GSTP1 is often more highly

expressed in breast cancer cells of patients with poor clinical outcome and low

overall survival [Miyake, 2012; Saxena, 2012]. GSTP1 can detoxify

chemotherapeutic drugs inside neoplastic cells, making these cells resistant to

chemotherapy [Jardim, 2012]. In lung cancer cells, GSTP1 expression is

consistently up-regulated compared to normal lung tissue and appears to be directly

related to the development of chemotherapeutic resistance [Mattern, 2002].

Alternatively, over-expression and increased activity of other GST enzymes has

been linked to the production of genotoxic intermediates, as is the case for GSTT1

which is often highly expressed in the prostate where it correlates with PCa risk

[Rebbeck, 1999].

Although many different studies have linked various GSTM1 gene

polymorphisms to prostate cancer risk [Mo, 2009; reviewed in Di Pietro, 2010], only

a few have explored the expression and function of the protein itself in prostate

58

cancer cells. High expression levels of GSTM1 have previously been reported in

several drug-resistant tumour cell lines and malignant drug-resistant melanoma

metastases [Peters, 1994]. In transitional cell cancer (TCC) of the human bladder,

GSTM1 was significantly increased (average 2.8-fold) in TCC specimens compared

to levels in normal adjacent mucosa [Berendsen, 1997].

In the research described in Chapter 3, expression of GSTM1 was identified by

RT-PCR in the DU145, PC3 and LNCaP prostate cancer cell lines. Dimeric GSTM1

protein was detected by Western analysis in total protein lysates from LNCaP cells

but was undetectable in the DU145 and PC3 samples. Interestingly, GSTM1

proteins were detected in exosome preparations from both LNCaP and DU145

suggesting that in the DU145 cells, GSTM1 may be preferentially loaded into

exosomes for an extracellular function. In both LNCaP and DU145 exosomes,

GSTM1 is present in the monomeric form, with no dimeric protein identified, and a

higher molecular weight band at approximately 75 kDa may identify an exosome-

related protein complex containing GSTM1.

Little has been reported about the potential role for GSTM1 in prostate cancer

and the initial aim of the research described in this Chapter was to determine the

effects of over-expression of GSTM1 on the proliferation of GSTM1-null prostate

cancer cells. Then, because GSTs are able to protect tumour cells from the

cytotoxicity of anti-cancer drugs, prostate cancer cell lines were treated with

docetaxel to determine if over-expression of GSTM1 increased resistance to this

chemotherapeutic drug. Loading of GSTM1 into exosomes was explored by

comparing the GSTM1 levels in exosomes prepared from over-expressing, vector

only and parental cells. Finally, exosomes loaded with GSTM1 were added to the

medium of prostate cancer cells and stromal prostatic cells (WPMY-1 cell line) to

determine whether GSTM1 loaded into exosomes could protect these cells from

treatment with docetaxel, a commonly used anti-mitotic chemotherapeutic agent

[Eisenberger, 2012].

59

4.2 MATERIALS AND METHODS

The common materials and methods relevant to this chapter have been outlined

in Chapter 2.

4.2.1 STATISTICAL TESTS

Statistical analyses comparing results from three independent experiments

were completed using a two-tailed Student's t-test and either GraphPad Prism (La

Jolla, CA, USA) or Microsoft Excel Software. Data is represented as mean ±

standard error of the mean (SEM) and statistical significance was set at p≤ 0.05.

4.3 RESULTS

4.3.1 RESPONSE OF PROSTATE CANCER CELL LINES TO DOCETAXEL

DU145 and 22Rv1 have been reported to show some resistance to

docetaxel and resistant populations have been established from both lines by others

[Corcoran, 2012]. To confirm that the DU145 and 22Rv1 cells available in our

laboratory respond to docetaxel as reported, cells were treated for 24 h and 72 h

with docetaxel at a range of increasing concentrations of docetaxel from 0 nM

(vehicle only – 0.01% DMSO) to 40 nM. This range of doses was chosen according

to current literature [Hao, 2010; Toner, 2013]. Cell viability was measured using a

colorimetric MTS assay (Promega) where the MTS tetrazolium compound (Owen’s

reagent) is reduced by cells into a soluble coloured formazan product which is then

measured by determining the absorbance at 490nm. Production of the coloured

formazan product requires NADPH or NADH produced by dehydrogenase enzymes

in metabolically active cells and is therefore proportional to the number of viable

cells. Viability of DU145 cells started to decrease when the cells had been treated

for 24 h with a docetaxel concentration of 5 nM and a further reduction in viability

was seen with increasing docetaxel concentration (Figure 4.1). Cells treated for 72

h also responded to increased docetaxel concentration with less viability seen,

reaching the IC50 at approximately 12 nM. The 22Rv1 cell line population seemed to

be more resistant to docetaxel at all concentrations after 24 h but the cells did begin

to respond on further treatment for 72 h. The IC50 for 22Rv1 cells was not reached

in this experiment (even with a docetaxel concentration of 40 nM).

60

Figure 4.1: Dose-response curves of Docetaxel resistant prostate cancer cell

lines after treatment with Docetaxel for 24 and 72 hours. Cells were treated with a

range of Docetaxel (Vehicle, 0.1, 1, 5, 10, 20 and 40 nM) for 24 hours. IC50 represented by

dotted lines. Error bars = SEM. n = 3 for each cell line at each time-point.

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

40 n

M

0

5 0

1 0 0

1 5 0

D U 1 4 5 - 7 2 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

40 n

M

0

5 0

1 0 0

1 5 0

D U 1 4 5 - 2 4 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

40 n

M

0

5 0

1 0 0

1 5 0

2 0 0

2 2 R v 1 - 2 4 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

40 n

M

0

5 0

1 0 0

1 5 0

2 0 0

2 5 0

2 2 R v 1 - 7 2 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

61

To confirm that the docetaxel treatment regime was appropriate to determine

resistance, the sensitive cell lines LNCaP and PC3 were also treated with docetaxel

under the same conditions (Figure 4.2). Treatment with 10 nM of docetaxel for 24 h

was sufficient to reach the IC50of the LNCaP cells at 72 h of treatment approximately

2 nM was enough confirming that LNCaP is a docetaxel sensitive cell line and that

the docetaxel treatment was working. PC3 cells appeared less sensitive to

docetaxel than LNCaP, and the IC50 was not achieved with any concentration by 24

h. At 72 h however, 50% of the cells were no longer viable when treated with 5 nM

docetaxel.

Figure 4.2: Dose-response curves of Docetaxel sensitive prostate cancer cell

lines after treatment with Docetaxel for 24 and 72 hours. Cell lines were treated

with a range of Docetaxel (Vehicle, 0.1, 1, 5, 10, 20 (LNCaP) and 40 nM (PC3)) for 24 and

72 hours. IC50 represented by dotted lines. Error bars = SEM. n = 3 for each cell line at each

time-point.

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

40 n

M

0

5 0

1 0 0

1 5 0

P C 3 - 2 4 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

40 n

M

0

5 0

1 0 0

1 5 0

P C 3 - 7 2 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

0

5 0

1 0 0

1 5 0

L N C a P - 2 4 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

Veh

icle

0.1

nM

1 n

M

5 n

M

10 n

M

20 n

M

0

5 0

1 0 0

1 5 0

L N C a P - 7 2 h

D o c e ta x e l (n M )

Ce

ll v

iab

ilit

y (

%)

62

4.3.2 PROLIFERATION OF GSTM1OVER-EXPRESSING 22Rv1 CELLS

In Chapter 3, an experiment designed to test whether exogenous over-

expression of GSTM1 in GSTM1-naïve 22Rv1 cells would result in loading of

GSTM1 into exosomes released by these cells was described. The 22Rv1 cell line

was used as a model for this experiment as the cells produce no detectable GSTP1

or GSTM1 and only little GSTM3 and GSTT1. To our knowledge the contribution of

GSTM1 enzyme activity to docetaxel resistance has not been reported in the

scientific literature and the availability of the 22Rv1-M1 lines allowed testing of this.

Firstly, a proliferation assay was performed to compare growth of the parental

22Rv1-P, vector only 22Rv1-VO and the 22Rv1-M1 expressing populations (for

22Rv1-M1 cells, data from 4 individual heterogenous polyclonal populations were

combined). Proliferation was measured at three different time points after seeding -

at 4 h to normalise the attached cell number of each population after seeding and

then at 24 h and 72 h to compare proliferation of the different populations (Figure

4.3). Parental cells appeared to proliferate the fastest and the significant reduction in

proliferation seen in the 22Rv1-VO cells may be due to the continual selective

pressure of culturing in G418 containing medium. In comparison to the 22Rv1-VO

cells, the 22Rv1-M1 cells also showed a significant decrease in proliferation

suggesting that over-expression of GSTM1 puts further growth pressure on these

cells.

63

4 h 2 4 h 7 2 h

0

1

2

3

4

P ro life ra t io n o f 2 2 R v 1 -M 1 d e r iv a t iv e c e ll l in e

c o m p a re d w ith 2 2 R v 1 -V O

P ro life ra tio n

Fo

ld c

ha

ng

e i

n p

ro

life

ra

tio

n

2 2 R v 1 -P a re n ta l

2 2 R v1 -V O

2 2 R v1 -M 1

****

***

****

***

Figure 4.3: Proliferation of 22Rv1-M1 cell line compared with 22Rv1-VO and

parental 22Rv1 (22Rv1-P) cells. There was a significant decrease in proliferation of the

GSTM1 over-expressing cell lines compared to the vector control at 72 h. n = 3. Error bars =

SEM.*** p≤ 0.001, **** p≤ 0.0001 according to Student’s t test.

4.3.3 PROLIFERATION OF GSTM1 OVER-EXPRESSING 22Rv1 CELLS

TREATED WITH DOCETAXEL

Although 22Rv1 does appear to demonstrate some resistance to docetaxel, a

proliferation assay was used to determine whether over-expression of GSTM1 might

increase the resistance of these cells when challenged with docetaxel. Parental

22Rv1-P, vector only 22Rv1-VO and the 22Rv1-M1 populations were treated with

20 nM docetaxel for 72 h and viability measured using the MTS assay. Consistent

with the proliferation assay described above, treatment with 20 nM docetaxel

correlated with a reduction in cell viability for the cells that were transfected with the

empty vector when compared with the parental 22Rv1 cells (Figure 4.4).

Expression of GSTM1 however resulted in a significant increase in viability of cells

treated under the same conditions suggesting that increased expression of GSTM1

provides a mechanism for detoxification of docetaxel. Given the decrease in

64

proliferation seen when these cells are grown without docetaxel (Figure 4.3) this

increase in viability may be even greater than measured in this experiment.

2 2 R v 1 -P a r e n ta l 2 2 R v 1 -V O 2 2 R v 1 -M 1

0

2 0

4 0

6 0

8 0

1 0 0

D o c e ta x e l t re a tm e n t - 2 0 n M fo r 7 2 h

C e ll lin es

Ce

ll v

iab

ilit

y (

%)

****

Figure 4.4: Proliferation of 22Rv1-P and 22Rv1-M1 cells compared with 22Rv1-

VO after treating with 20 nM Docetaxel for 72 h. There was a significant increase in

cell proliferation of 22Rv1-M1 cells compared to the vector control suggesting increased

resistance to docetaxel. n = 3. Error bars = SEM. **** p≤ 0.0001 according to Student’s t test.

4.3.4EFFECT OF EXOSOME ADDITION DURING TREATMENT OF DOCETAXEL-

RESISTANT CELLS WITH DOCETAXEL

Published experimental data suggest that exosomes produced by cancer

cells can influence the behaviour of cells within the tumour microenvironment and

even possibly help to determine the pre-metastatic niche [Jung, 2009]. To determine

whether addition of exosomes would also increase the proliferation of cell lines that

show some resistance to docetaxel, exosomes prepared from 22Rv1-VO cells and

22Rv1-M1 cells were added to the medium for treatment of DU145 and parental

22Rv1 cells, again in the presence or absence of docetaxel (Figures 4.5 and 4.6).

Cells were incubated for up to 72 h and proliferation measured at both 24 h and

72 h. As seen in Figure 4.5, DU145 continued to proliferate in the presence of

docetaxel which was expected for a resistant cell line, although why this proliferation

was approximately 1.5 fold more than the vehicle-treated cells is not clear. Addition

of 100 μg/mL 22Rv1-VO and 22Rv1-M1 exosomes resulted in an increase in cell

proliferation in normal cell culture and in the presence of 20 nM docetaxel. A further

65

increase in proliferation was seen when twice as many exosomes (200 μg/mL) were

added.

Exosomes were also added to the medium of 22Rv1 parental cells (Figure

4.6). An increase in proliferation was again seen in the 22Rv1 cells treated with 100

μg/mL of the 22Rv1-VO exosomes but this proliferation response was lost when this

was increased to 200 μg/mL suggesting that too many exosomes may be inhibitory

to cell growth. This may be the case if the exosomes produced by these cells are

being used to remove unwanted cellular contents. Interestingly however, treatment

with100 μg/mL 22Rv1-M1 exosomes correlated with an increase in proliferation by

approximately 3 fold and this increased to almost 9 fold when 200 μg/mL exosomes

were added for 72 h. This would suggest that loading of GSTM1 into exosomes

may provide more resistance to docetaxel even in the parental cells.

66

Figure 4.5: Proliferation of DU145 cells in the absence or presence of docetaxel and exosomes. (A) Addition of 100 µg/mL exosomes. (B)

Addition of 200 µg/mL exosomes. Cells were untreated (Vehicle – DMSO), treated with 20 nM docetaxel or treated with docetaxel and exosomes. n = 3. Error

bars = SEM. * p≤ 0.05, ** p≤ 0.01, *** p≤ 0.001, **** p≤ 0.0001 according to Student’s t test.

Ve

hic

le

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5

D U 1 4 5

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

ro

life

ra

tio

n

2 0 n M D o c e ta x e lN o D o c e ta x e l

* ****

****0 h o u rs

2 4 h o u rs

7 2 h o u rs

Ve

hic

le

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

1

2

3

D U 1 4 5

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

roli

fera

tio

n

0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

***

**

A B

67

Figure 4.6: Proliferation of 22Rv1 cells in the absence or presence of docetaxel and exosomes. (A) Addition of 100 µg/mL exosomes. (B)

Addition of 200 µg/mL exosomes. Cells were untreated (Vehicle – DMSO), treated with 20 nM docetaxel or treated with docetaxel and exosomes. n = 3. Error

bars = SEM.*** p≤ 0.001, **** p≤ 0.0001 according to Student’s t test.

Ve

hic

le

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

2

4

6

8

1 0

2 2 R v 1

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

roli

fera

tio

n

0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

***

****

****

Ve

hic

le

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

2

4

6

8

1 0

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

roli

fera

tio

n0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

********

****

2 2 R v 1

A B

68

4.3.5EFFECT OF EXOSOME ADDITION DURING TREATMENT OF DOCETAXEL-

SENSITIVE CELLS WITH DOCETAXEL

Similar to 4.3.4, the effect of GSTM1 loaded exosomes shed by 22Rv1-M1

cells was studied to observe whether or not the exosomes provide protection from

chemotherapeutic agents. 100 μg/mL or 200 μg/mL solution of exosomes in growth

medium was added to cultures of the docetaxel sensitive cell lines PC3 and LNCaP,

with and without 20 nM docetaxel. Cells were incubated for up to 72 h and

proliferation measured at 24 and 72 h. The addition of both 22Rv1-M1 and 22Rv1-

VO exosomes to PC3 cells increased cell proliferation in both the absence and

presence of docetaxel. This was particularly interesting as the IC50 for PC3 cells at

72 h was 5 nM. The results, as seen in Figure 4.7, showed that even the addition of

exosomes themselves increased proliferation in cultures without docetaxel.

Interestingly, the proliferation also increased when cultured with docetaxel as well

and this shows that exosomes containing GSTM1 and exosomes that don’t contain

GSTM1 also provide tumour protective properties. Similar results are seen when

both concentrations of exosomes are used.

The proliferation of LNCaP cells increased dramatically when cultured with

docetaxel and exosomes. This was particularly interesting as the IC50 of LNCaP

cells was 2 nM at 72 h and this proliferation increase shows that exosomes with the

aid of GSTM1 play a definite protective role in LNCaP cells. Treatment with 100

μg/mL 22Rv1-VO exosomes resulted in a significant increase in proliferation in the

docetaxel untreated cells (from 1.8 fold to almost 5 fold). An increase in proliferation

was also seen in the cells treated with 22Rv1-M1 exosomes although this was less

than the VO cells (1.8 fold to 3.3 fold), perhaps consistent with the proliferation data

that suggests over-expression of GSTM1 has a slight negative effect on the cells.

Cells co-treated with both exosomes and docetaxel also had significantly higher

proliferation than cells treated with docetaxel alone and in this experiment there was

no significant difference in the proliferation of cells treated with exosomes from

either the 22Rv1-VO cells or the 22Rv1-M1 cells. Similar results are also seen when

200 µg/mL exosomes are used.

.

69

Figure 4.7: Proliferation of PC3 cells in the absence or presence of docetaxel and exosomes. (A) Addition of 100 µg/mL exosomes. (B)

Addition of 200 µg/mL exosomes. Cells were untreated (Vehicle – DMSO), treated with 20 nM docetaxel or treated with docetaxel and exosomes. n = 3. Error

bars = SEM. * p≤ 0.05, ** p≤ 0.01, **** p≤ 0.0001 according to Student’s t test.

Ve

hic

le

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

1

2

3

4

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

ro

life

ra

tio

n

2 0 n M D o c e ta x e lN o D o c e ta x e l

P C 3

*****

****0 h o u rs

2 4 h o u rs

7 2 h o u rs

Ve

hic

le

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

1

2

3

4

5

P C 3

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

roli

fera

tio

n

0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

*

*

**

A B

70

Figure 4.8: Proliferation of LNCaP cells in the absence or presence of docetaxel and exosomes. (A) Addition of 100 µg/mL exosomes. (B)

Addition of 200 µg/mL exosomes. Cells were untreated (Vehicle – DMSO), treated with 20 nM docetaxel or treated with docetaxel and exosomes. n = 3. Error

bars = SEM.**** p≤ 0.0001 according to Student’s t test.

Ve

hic

le

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

2

4

6

L N C a P

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

ro

life

ra

tio

n0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

****

****

****

Ve

hic

le

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

2

4

6

L N C a P

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

roli

fera

tio

n

0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

****

****

****

A B

71

4.3.6EFFECT OF EXOSOME ADDITION DURING TREATMENT OF STROMAL

CELLS (WPMY-1) WITH DOCETAXEL

To determine whether exosomes produced by cancer cells can influence the

behaviour of stromal cells found within the tumour microenvironment, cells of the

stromal prostatic fibroblast cell line WPMY-1 were then treated with medium

containing 100 or 200 µg/mL exosomes from 22Rv1-VO and 22Rv1-M1 cells in the

presence or absence of docetaxel where proliferation was measured at 4, 24 and 72

hours (Figure 4.9). Again, the effect of adding concentrations of 100 μg/mL or 200

μg/mL of exosomes to the proliferation of WPMY-1 cells was compared at 24 h and

72 h. In the vehicle only treated culture, WPMY-1 cells proliferated 1.8 fold over the

72 h time period of this experiment. As expected, docetaxel treatment caused a

decrease in proliferation of WPMY-1 cells with only a third of the untreated cell

number (0.6 fold) remaining at 72 h. Addition of 100 μg/mL exosomes, either 22Rv1-

VO or 22Rv1-M1, had little effect on the proliferation of these cells in the absence of

docetaxel suggesting that the growth advantage seen for the prostate cancer cells

might be cell context dependent. Interestingly, addition of exosomes from either

22Rv1-VO or 22Rv1-M1 cells rescued the WPMY-1 cells from docetaxel treatment,

restoring proliferation to that of the vehicle treated cells suggesting that these

exosomes both carry proteins that off-set the negative growth effect of the

docetaxel. Little benefit was seen from the addition of twice as many exosomes

when the cells were in normal cell culture but a significant increase in proliferation

was seen when 200 μg/mL exosomes were added to the WPMY-1 cells treated with

docetaxel and this may be related to the increased levels of GSTM1 in these

exosomes.

72

Figure 4.9: Proliferation of WPMY-1 cells in the absence or presence of docetaxel and exosomes. (A) Addition of 100 µg/mL exosomes. (B)

Addition of 200 µg/mL exosomes. Cells were untreated (Vehicle – DMSO), treated with 20 nM docetaxel or treated with docetaxel and exosomes. n = 3. Error

bars = SEM. **** p≤ 0.0001 according to Student’s t test.

Ve

hic

le

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

10

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

10

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

roli

fera

tio

n

0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

W P M Y -1

****

****

Ve

hic

le

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

Do

ce

tax

el

on

ly

20

0 µ

g/m

L 2

2R

v1

-VO

ex

os

om

es

20

0 µ

g/m

L 2

2R

v1

-M1

ex

os

om

es

0

1

2

3

4

W P M Y -1

T re a tm e n ts

Fo

ld c

ha

ng

e i

n p

roli

fera

tio

n

0 h o u rs

2 4 h o u rs

7 2 h o u rs

2 0 n M D o c e ta x e lN o D o c e ta x e l

****

****

A B

73

4.4 DISCUSSION

There are various studies suggesting that the expression of GST enzymes can

either be tumour promoting or suppressive. This study in particular showed that

exosomes carrying GSTM1 was tumour promoting. The 22Rv1-M1 cells significantly

decreased in proliferation when compared to both the 22Rv1-P and 22Rv1-VO cells

however in the presence of 20 nM docetaxel, cell viability of the 22Rv1-M1 cells

increased compared to the parental and vector only cells.

To determine whether exosomes carrying GST enzymes might confer

resistance to chemotherapeutic agents, cells were treated with a combination of

exosomes and docetaxel. Docetaxel was chosen for this study because it is the

most commonly used anti-mitotic agent for treatment of advanced and castrate

resistant prostate cancer. Published data suggests that GSTP1 is able to detoxify

docetaxel [Arai, 2008] however, to our knowledge, the ability of GSTM1 to detoxify

docetaxel, or other chemotherapeutic drugs, has not been determined. A panel of

prostate cancer cell lines that were variably resistant (DU145 and 22Rv1) and

sensitive (PC3 and LNCaP) to docetaxel allowed the testing of the hypothesis that

GSTM1 could be carried in exosomes and this might provide increased resistance to

docetaxel treatment. In the first part of the study reported in this Chapter, docetaxel-

sensitive 22Rv1 cells were transfected with an expression construct containing the

complete coding sequence of GSTM1under control of the constitutive CMV-IE

promoter. Although the proliferation of 22Rv1-M1 cells was lower than that of the

parental and vector only cells, proliferation was higher in 22Rv1-M1 cells than the

22Rv1-VO cells when these were treated with docetaxel for 72 h. This suggests that

although GSTM1 expression in 22Rv1 may be detrimental during conditions of

normal growth, it does provide a growth advantage when cells are challenged with

docetaxel. This result also supported the use of docetaxel in further experiments.

Exosomes are thought to be an important means of cell to cell communication

between cells within the immediate tumour microenvironment or at more distant

sites in the body. To determine whether exosomes might have local effects on other

cells including other tumour cells and stromal cells, exosome preparations were

added to various cell populations and the effect on cell proliferation determined. In

the first experiments the interaction of 22Rv1-M1 exosomes with cancer cells

including docetaxel resistant DU145 and 22Rv1 parental cells and docetaxel

sensitive LNCaP and PC3 cells, was determined. In keeping with similar studies of

gastric and breast cancers [Qu, 2009; Koga, 2005], addition of exosomes to cell

culture medium stimulated proliferation of cells, regardless of loading with GSTM1,

74

which would suggest that exosomes from prostate cancer cells carry growth

promoting macromolecules. A clear growth benefit was seen when exosomes

carrying GSTM1 were added to cells treated with docetaxel supporting the

hypothesis that GSTM1 enzymes can be carried in exosomes to other cells and

thereby assist in cancer cell resistance to docetaxel. It is perhaps not surprising that

GSTM1 exosomes showed a more beneficial effect on cells that were sensitive to

Docetaxel than those that were inherently more resistant. Reconstruction of

autocrine stimulation by treating 22Rv1 cells with exosomes shed from 22Rv1 cells

also resulted in increased proliferation of cells. Parental 22Rv1 cells appeared to be

much more responsive to what may be a self-perpetuating signal.

Exosomes released from tumour cells may also influence growth and behaviour

of stromal cells within the tumour microenvironment. In a study regarding breast

cancer, it was shown that prostatic stromal fibroblasts promoted metastasis in breast

cancer. Metastasis promotion was influenced by TGF-β1 secreted by fibroblasts in

vitro [Stuelten, 2010]. Other fibroblasts such as cancer associated fibroblasts are

able to induce chemotherapy resistance during breast cancer treatment [Martinez-

Outschoorn, 2011]. This is due to the collagen type I secreted by the CAFs as it

decreases chemotherapeutic drug uptake in tumours and plays a significant role in

regulating tumour sensitivity to a variety of chemotherapies [Loeffler, 2006]. In the

experiment reported in this Chapter, when WPMY-1 cells were treated with 22Rv1-

M1 exosomes an increase in proliferation was also seen, even in the presence of

docetaxel.

When considered as a whole, these experiments show that exosomes can

stimulate proliferative responses in a paracrine and autocrine manner and that GST

enzymes loaded as cargo into exosomes may contribute to resistance to docetaxel,

an important prostate cancer chemotherapeutic. Novel anti-cancer strategies may

therefore include those that can block exosome GST loading, inhibit GST activity or

even remove exosomes from circulation.

The results presented in this chapter have shown that GSTM1 over-expression

in prostate cancer cells can be loaded into shed exosomes. A recent study has

shown the benefits of treating Parkinson’s disease with exosomes containing

specific Parkinson’s disease treatments [Haney, 2015] and it may be possible to

express anti-cancer proteins and peptides in cells that then package these into

exosomes for transfer to tumour cells [Kim, 2015]. Exosome packaging could

potentially be a new and valuable option for delivery of anti-tumour therapies but this

can only be realised with further research.

75

Chapter 5: General Discussion

GST expression has been a widely researched and often controversial topic

in cancer. Studies suggest that the GSTP1 protein promotes tumour growth and

progression [He, 2011] and others believe that it may be tumour suppressive [Gray,

2012]. Methylated GSTP1 has been suggested as a potential prognostic marker for

prostate cancer and has been the focus of many clinical trials [Mahon, 2014].

Despite several attempts throughout the study, it did not seem possible to transfect

22Rv1 cells with an expression construct to exogenously over-express GSTP1.

This would suggest that in this particular cell line, GSTP1 may function as a tumour

suppressor however might be cell context dependent and should be tested in other

cell lines using an inducible expression system.

The functional and clinical effects in prostate cancer of many of the other

classes, and particularly GSTT1, GSTM1 and GSTM3, have not yet been studied in

detail because to date the focus has been the relevance of polymorphisms found in

several different populations [Matthias, 1998; Di Pietro, 2010; Mo, 2009; reviewed in

Di Pietro, 2010]. The data presented here suggests that GSTM1 may contribute to

resistance to docetaxel in sensitive cells and that GSTM1 can be loaded into

exosomes and delivered to cells to increase proliferation and docetaxel resistance,

two key findings that add to the knowledge of both GSTM1 enzyme and exosome

functions. This study examining the effect that GSTs and exosomes have on the

proliferation of prostate cancer cells is the first functional study of GSTM1’s cellular

effect. To the best of my knowledge, there are no published studies about GST

loaded exosomes affecting cancer progression in both untreated and treated cells.

Protein profiles of the various GSTs were conducted to see if the inactive

monomeric or active dimeric form of GSTs were present in a range of prostate

cancer cells and their shed exosomes. Proteomic profiling has shown that proteins

packaged into shed exosomes are different from the proteins present in the parent

cells [Théry, 2006; Duijvesz, 2013; Hosseini-Beheshti, 2012]. GSTT1 was present in

all of the prostate cancer lysates as a monomer but not in any of the shed exosomes

as either a monomer or dimer. GSTM1 was present in only the LNCaP lysates as a

dimer and packaged as a monomer into shed exosomes as well as being packaged

into DU145 exosomes, both as a monomer and as a complex. GSTM3 was present

in all of the lysates as a monomer and also followed the exosome profile of GSTM1

where it was present monomerically in only LNCaP and DU145 exosomes. This

suggests that some GSTs and possibly other proteins are preferentially packaged

76

into exosomes. This packaging may be a way to remove unnecessary products from

the cell or the exosomes themselves may be used as a means of transportation for

particular enzymes to other cells. The presence of dimeric GSTM1 is interesting as

the western blotting conditions (denaturing and reducing) should only yield

monomeric protein. Exosome function has been linked to tumour progression as

they act as mediators between the localised tumour and distant metastases [Quail,

2013]. After being released, the proteins within these exosomes may affect tumour

growth and this probably is the case for GSTM1. Unlike the other GSTs that were

studied, GSTP1 was present in both the DU145 and PC3 lysates and exosomes and

also only contained only monomeric GSTP1. The lysate profile was expected as the

results correlated with a previous study [Hokaiwado, 2008].

GSTM1 was stably over-expressed in 22Rv1 cells and subsequently

packaged into shed exosomes that were shown to serve a biological purpose in

which the proliferation of prostate cancer cells and stromal prostatic fibroblasts

increased as 22Rv1-M1 exosomes were added both in the presence or absence of

docetaxel. The addition of exosomes themselves (22Rv1-VO) was shown to

increase proliferation in DU145, PC3 and LNCaP at both exosome concentrations.

The increased proliferation of DU145 cells both in the absence and presence of

docetaxel was very similar. This was expected as DU145 cells are docetaxel

resistant, so cell growth was not hindered with 20 nM docetaxel. The proliferation of

22Rv1 cells when treated with 22Rv1-VO exosomes did not increase. This was also

expected since the exosomes were originally derived from 22Rv1 cells and would

contain the same proteins and miRNA. Proliferation increased significantly when

200 µg/mL 22Rv1-M1 exosomes were added to 22Rv1 cells and this shows that

loading GSTM1 into exosomes helps cell growth when culturing with and without

docetaxel. When 22Rv1-VO and 22Rv1-M1 exosomes were added to the docetaxel

sensitive cell lines PC3 and LNCaP, the proliferation of these tumour cells increased

significantly. The LNCaP cells that were cultured with docetaxel showed particularly

interesting results as tumour growth was increased with a very high concentration of

docetaxel. The IC50 for LNCaP cells was 2 nM at 72 hours (Figure 4.2) however

when vector and over-expressing exosomes were added to the culture, the

proliferation increased significantly. This shows that exosomes themselves and

GSTM1 over-expressing exosomes played a protective role in LNCaP cells from

docetaxel.

There are many possible future directions that have stemmed from this

study. The over-expression of GSTM1 could be performed in other prostate cancer

cell lines particularly LNCaP cells where it was already strongly expressed. GSTM3

77

could perhaps be over-expressed in 22Rv1 cells and the exosomes collected from

this over-expressing cell line to see if there is also a functional effect as it has not

yet been studied. These shed exosomes could be used in functional studies such as

the same proliferation studies that were performed during this study. The effect of

these exosomes on cell migration could also be tested using a transwell migration

assay. This would include using a Boyden chamber to observe the migration of cells

from the top layer to the bottom to see if the addition of exosomes increases or

decreases the rate of migration [Lee, 2013]. The exosomes from DU145 and PC3

cells that already express GSTP1 could also be used in proliferation and migration

studies. As there was no GSTT1 in any shed exosomes, this lack of GSTT1 can be

studied with these exosomes. GSTP1, GSTM1 and GSTM3could be over-expressed

in normal prostate cells such as RWPE-1 or prostatic stromal cells like WPMY-1 and

exosomes shed from these manipulated cells could then be used in similar

functional studies to see if over-expression resulted in an increase or decrease in

proliferation and if these transfected cells were less or more sensitive to

chemotherapeutic drugs.

Conversely, GST expression could also be silenced using short interfering

RNA (siRNA). Lack of exosomal GSTs could be examined to see if unlike the over-

expression, silencing decreases docetaxel detoxification. Proliferation of these GST

free cancer cells in the presence and absence of docetaxel can also be examined.

The function of GSTs and exosomes can also be examined with using other

chemotherapeutic drugs like doxorubicin and cisplatin alone or in conjunction with

docetaxel [Tsakalozou, 2012]. These are also common drugs used in prostate

cancer treatments that have different modes of action [McKay, 2015; Tsakalozou,

2012].

Furthermore, experiments using clinical samples rather than cell lines could

also be used to further enhance the study. Samples from healthy patients, to

patients with BPH and patients with low or high Gleason score prostate cancer could

be used. Exosomes would be isolated from seminal fluid, which is rich in

prostasomes and exosomes, and the GST content examined. The process to isolate

exosomes from clinical samples is outlined in Théry, 2006 and follows similar

procedures as isolating exosomes from cell culture media. Following exosome

characterisation, proliferation and migration studies could be performed both in the

absence and presence of docetaxel or another chemotherapeutic drug, doxorubicin.

The importance of exosomes in delivery of agents to inhibit tumour spread

has previously been explored. A study by Kim et al. (2015) reported loading

paclitaxel (PTX) within macrophage-released exosomes and the ability of these to

78

overcome multidrug resistance of cancer cells. In this study, the most effective

method of loading PTX into exosomes was to sonicate the exosomes in the

presence of PTX at 20% amplitude, 6 cycles for 30 seconds on/off for three minutes

with a two minute cooling period between each cycle. After sonication, the exo-PTX

solution was incubated at 37°C for 60 minutes to allow for the recovery of the

exosomal membrane. By examining the levels of Alix, TSG101 and Flotillin in

exosomes before and after sonication with Western blotting, it was apparent that

sonication did not affect the protein content of the exosomes. There were also no

major alterations of the lipid content of exosomal membranes. The final incubation

hour resulted in a complete restoration of membrane microviscosity. The cup-

shaped morphology of exosomes also remained the same [Kim, 2015].

A significant inhibition of lung metastasis growth by exo-PTX was seen in

mice after IV injecting with exo-PTX compared to treatments with saline or Taxol

[Kim, 2015]. Few cancer cells were seen in the lungs of exo-PTX treated animals

and sonicated empty exosomes showed no significant inhibition on metastasis

growth. When the intraluminal cargo is released into the cytosol of a target cells, any

drug that is also in the inner bilayer of exosomes may also be used. Airway

delivered exosomes showed almost complete co-localisation with cancer

metastases [Kim, 2015]. It was speculated that macrophage-released exosomes

were likely to have specific proteins on the surface allowing preferential

accumulation in cancer cells [Kim, 2015]. The research described previously shows

that an over-expressed protein, GSTM1, can be loaded into exosomes and this

might be an approach that can be used to load anti-tumour molecules.

Exosome-mediated cell to cell communication is important in the interactions

of cancer cells with the immune system [Finn, 2012]. Parolini et al. (2009) showed

that under acidic conditions, there was a more efficient fusion of exosomes with

target cells suggesting that tumours in an acidic microenvironment may

preferentially take up exosomes compared to the surrounding healthy tissue which

may not be as acidic [Parolini, 2009]. Therapeutically, exosomes provide a good

drug delivery system as they are able to be lyophilised and reconstituted and still

retain their morphology and other characteristics meaning that exosomes containing

drugs may be prepared and stored prior to treatment [Haney, 2015].

In summary, the progression of local PCa and any distant metastases

includes the involvement of exosomes, particularly the macromolecules they

contain. These macromolecules could have a positive or negative effect. This study

shows that GST enzymes that are loaded as cargo in exosomes released from PCa

cells may also effect the proliferation of these cells in vitro and be involved in cancer

79

progression in the presence and absence of chemotherapeutic drugs. This study

provides a basis for further investigations into whether GSTs contained in exosomes

can be used as useful biomarkers for diagnosis and prognosis of PCa or used as

potential therapeutic targets.

80

References

1. Aalberts M et al. (2013). Prostasomes: extracellular vesicles from the prostate.

Reproduction. 147(1):R1-14.

2. Adell MA et al. (2014). Coordinated binding of Vps4 to ESCRT-III drives

membrane neck constriction during MVB vesicle formation. J Cell Biol.

205(1):33-49.

3. Adler V et al. (1999). Regulation of JNK signaling by GSTp. EMBO J.

18(5):1321-34.

4. Agalliu I et al. (2006). Polymorphisms in the glutathione S-transferase M1, T1,

and P1 genes and prostate cancer prognosis. Prostate. 66:1535-1541.

5. Ahmadi H and Daneshmand S. (2013). Androgen deprivation therapy:

evidence-based management of side effects. BJU Int. 111(4):543-8.

6. AIHW. (2013). Prostate cancer in Australia. Cancer series 79. Cat. no. CAN 76.

Canberra: AIHW.

7. Allen AM et al. (2012). An evidence based review of proton beam therapy: the

report of ASTRO’s emerging technology committee. Radiother Oncol. 103(1):8-

11.

8. Al-Nedawi K et al. (2008). Intercellular transfer of the oncogenic receptor

EGFRvIII by microvesicles derived from tumour cells. Nat Cell Biol. 10(5):619-

24.

9. An Tet al. (2015). Exosomes serve as tumour markers for personalized

diagnostics owing to their important role in cancer metastasis. J Extracell

Vesicles. 4:27522.

10. Arai T et al. (2008). Association of GSTP1 expression with resistance to

docetaxel and paclitaxel in human breast cancers. Eur J Surg Oncol.

34(7):734-8.

11. Atkinson HJ and Babbitt PC. (2009). Glutathione Transferases Are Structural

and Functional Outliers in the Thioredoxin Fold. Biochemistry. 48(46):11108-

16.

12. Atula S et al. (1997). Fibroblasts can modulate the phenotype of malignant

epithelial cells in vitro. Exp Cell Res. 235(1):180-7.

13. Azmi AS et al. (2013). Exosomes in cancer development, metastasis, and drug

resistance: a comprehensive review.

14. Baietti MF et al. (2012). Syndecan-syntenin-ALIX regulates the biogenesis of

exosomes. Nat Cell Biol. 14(7):677-85.

81

15. Bastian PJ et al. (2005). Preoperative serum DNA GSTP1 CpG island

hypermethylation and the risk of early prostate-specific antigen recurrence

following radical prostatectomy. Clin Cancer Res. 11(11):4037-43.

16. Berendsen CL et al. (1997). Glutathione S-transferase activity and subunit

composition in transitional cell cancer and mucosa of the human bladder.

Urology. 49(4):644-51.

17. Bergers G and Benjamin LE. (2003). Tumorigenesis and the angiogenic switch.

Nat Rev Cancer. 3(6):401-10.

18. Bernardini S et al. (2002). Role of GSTP1-1 in mediating the effect of etoposide

on human neuroblastoma cell line Sh-Sy5y. J Cell Biochem. 86(2):340-7.

19. Bhattacharya P et al. (2013). Glutathione S-transferase class µ regulation of

apoptosis signal-regulating kinase 1 protein during VCD-induced ovotoxicity in

neonatal rat ovaries. Toxicol Appl Pharmacol. 267(1):49-56.

20. Bjinsdorp IV et al. (2013). Exosomal ITGA3 interferes with non-cancerous

prostate cell functions and is increased in urine exosomes of metastatic

prostate cancer patients. J Extracell Vesicles. 2. doi: 10.3402/jev.v2i0.22097.

21. Board PG and Menon D. (2013). Glutathione transferases, regulators of cellular

metabolism and physiology. Biochim Biophys Acta. 1830(5):3267-88.

22. Borges FT et al. (2013). Extracellular vesicles: structure, function, and potential

clinical uses in renal diseases. Braz J Med Biol Res. 46(10):824-30.

23. Bryniarski K et al. (2015). Free Extracellular miRNA Functionally Targets Cells

by Transfected Exosomes from Their Companion Cells. PLoS One.

10(4):e0122991.

24. Byar D and Mostofi F. (1972). Carcinoma of the prostate: Prognostic evaluation

of certain pathologic features in 208 radical prostatectomies. Cancer. 30(1):5-

13.

25. Camps JL et al. (1990). Fibroblast-mediated acceleration of human epithelial

tumor growth in vivo. Proc Natl Acad Sci USA. 87(1):75-79.

26. Cancer Staging. National Cancer Institute website. Retrieved June 5, 2015,

from http://www.cancer.gov/cancertopics/factsheet/detection/staging.

27. Chaiwun B et al. (2011). GSTPi-positive tumour microenvironment-associated

fibroblasts are significantly associated with GSTPi-negative cancer cells in

paired cases of primary invasive breast cancer and axillary lymph node

metastases. Br J Cancer. 105(8):1224-9.

28. Challagundla KB et al. (2015). Exosome-Mediated Transfer of microRNAs

Within the Tumor Microenvironment and Neuroblastoma Resistance to

82

Chemotherapy. J Natl Cancer Inst. 107(7).

29. Charras GT et al. (2005). Non-equilibration of hydrostatic pressure in blebbing

cells. Nature. 435(7040):365-9.

30. Chen N et al. (2002). Differential gene expression profiles of Jnk1- and Jnk2-

deficient murine fibroblast cells. Cancer Res. 62(5):1300-4.

31. Chiam K et al. (2014). Epigenetic biomarkers in prostate cancer: Current and

future uses. Cancer Lett. 342(2):248-56.

32. Chiasserini D et al. (2015). Identification and Partial Characterization of Two

Populations of Prostasomes by a Combination of Dynamic Light Scattering and

Proteomic Analysis. J Membr Biol. 248(6):991-1004.

33. Chlenski A et al. (2001). Androgen receptor expression in androgen-

independent prostate cancer cell lines. Prostate. 47(1):66-75.

34. Cho SG et al. (2001). Glutathione S-transferase mu modulates the stress-

activated signals by suppressing apoptosis signal-regulating kinase 1. J Biol

Chem. 276(16):12749-55.

35. Clayton A et al. (2001). Analysis of antigen presenting cell derived exosomes,

based on immune-magnetic isolation and flow cytometry. J Immunol Methods.

247(1-2):163-74.

36. Colombo M et al. (2014). Biogenesis, secretion, and intercellular interactions of

exosomes and other extracellular vesicles. Annu Rev Cell Dev Biol. 30:255-89.

37. Comparative effectiveness of therapies for clinically localized prostate cancer:

an update of a 2008 comparative effectiveness review. Agency for Healthcare

Research and Quality. Retrieved June 5, 2015, from:

http://effectivehealthcare.ahrq.gov/ehc/products/521/1434/localized-prostate-

cancer-therapy-protocol-130329.pdf.

38. Corcoran C et al. (2012). Docetaxel-resistance in prostate cancer: evaluating

associated phenotypic changes and potential for resistance transfer via

exosomes. PLoS One. 7(12):e50999.

39. Cotignola J et al. (2013). Glutathione-S-Transferase (GST) polymorphisms are

associated with relapse after radical prostatectomy. Prostate Cancer Prostatic

Dis. 16(1):28-34.

40. Cross NL and Mahasreshti P. (1997). Prostasome fraction of human seminal

plasma prevents sperm from becoming acrosomally responsive to the agonist

progesterone. Arch Androl. 39(1):39-44.

41. Di Pietro G et al.(2010). Glutathione S-transferases: an overview in cancer

research. Expert Opin Drug Metab Toxicol. 6(2):153-70.

83

42. Di Vizio D et al. (2012). Large oncosomes in human prostate cancer tissues

and in the circulation of mice with metastatic disease. Am J Pathol.

181(5):1573-84.

43. Discacciati A and Wolk A. (2014). Lifestyle and dietary factors in prostate

cancer prevention. Recent Results Cancer Res. 202:27-37.

44. Dorion S et al. (2002). Activation of the p38 signaling pathway by heat shock

involves the dissociation of glutathione S-transferase Mu from Ask1. J Biol

Chem. 277(34):30792-7.

45. Duchesne G. (2011). Localised prostate cancer - current treatment options.

Aust Fam Physician. 40(10):768-71.

46. Duijvesz D et al. (2010). Exosomes as biomarker treasure chests for prostate

cancer. Eur Urol. 59(5):823-831.

47. Duijvesz D et al. (2013). Proteomic profiling of exosomes leads to the

identification of novel biomarkers for prostate cancer. PLoS One. 8(12):e82589.

48. Duijvesz D et al. (2015). Immuno-based detection of extracellular vesicles in

urine as diagnostic marker for prostate cancer. Int J Cancer. 137(12):2869-78.

49. Eisenberger MA and Antonarakis ES. (2012). The Experience with Cytotoxic

Chemotherapy in Metastatic Castration-Resistant Prostate Cancer. UrolClin

North Am. 39(4):573-581.

50. Ellinger J et al. (2008). CpG island hypermethylation in cell-free serum DNA

identifies patients with localized prostate cancer. Prostate. 68(1):42-9.

51. Esteller M et al. (2001). A gene hypermethylation profile of human cancer.

Cancer Res. 61(8):3225-9.

52. Finn NA et al. (2011). A switching mechanism in doxorubicin bioactivation can

be exploited to control doxorubicin toxicity. PLoS Comput Biol. 7(9):e1002151.

53. Finn OJ et al. (2012). Immuno-oncology: understanding the function and

dysfunction of the immune system in cancer. Ann Oncol. 23 Suppl 8:viii6-9.

54. Frohlich DA et al. The role of Nrf2 in increased reactive oxygen species and

DNA damage in prostate tumorigenesis. Oncogene. 27(31):4353-62.

55. Ge Q et al. (2014). miRNA in plasma exosome is stable under different storage

conditions. Molecules. 19(2):1568-75.

56. Ginestra A et al. (1997). Urokinase plasminogen activator and gelatinases are

associated with membrane vesicles shed by human HT1080 fibrosarcoma

cells. J Biol Chem. 272(27):17216-22.

57. Gingrich JR et al. (1997). Androgen-independent Prostate Cancer Progression

in the TRAMP Model. Cancer Res. 57(21):4687-91.

84

58. Giordano FJ and Johnson RS. (2001). Angiogenesis: the role of the

microenvironment in flipping the switch. Curr Opin Genet Dev. 11(1):35-40.

59. Giusti I and Dolo V. (2014). Extracellular vesicles in prostate cancer: new future

clinical strategies? Biomed Res Int. 2014:561571.

60. Gleason D. (1974). Prediction of prognosis for prostatic adenocarcinoma by

combined histological grading and clinical staging. J Urol. 111(1):58-64.

61. Gleave M et al. (1991). Acceleration of human prostate cancer growth in vivo

by factors produced by prostate and bone fibroblasts. Cancer Res.

51(14):3753-61.

62. Gray SG et al. (2012). Gemcitabine reactivates epigenetically silenced genes

and functions as a DNA methyltransferase inhibitor. Int J Mol Med. 30(6):1505-

11.

63. Greening DW et al. (2015). Exosomes and their roles in immune regulation and

cancer. Semin Cell Dev Biol. 40:72-81.

64. Gu J et al. (2012). Gastric cancer exosomes trigger differentiation of umbilical

cord derived mesenchymal stem cells to carcinoma-associated fibroblasts

through TGF-β/Smad pathway. PLoS One. 7(12):e52465.

65. Haney MJ et al. (2015). Exosomes as drug delivery vehicles for Parkinson’s

disease therapy. J Control Release. 207:18-30.

66. Hao J et al. (2010). Co-expression of CD147 (EMMPRIN), CD44v3-10, MDR1

and monocarboxylate transporters is associated with prostate cancer drug

resistance and progression. Br J Cancer. 103(7):1008-18.

67. Harris DA et al. (2015). Exosomes released from breast cancer carcinomas

stimulate cell movement. PLoS One. 10(3):e0117495.

68. Hayes J and Pulford D. (1995). The Glutathione S-Transferase Supergene

Family: Regulation of GST* and the Contribution of the Isoenzymes to Cancer

Chemoprotection and Drug Resistance. Crit Rev Biochem Mol Biol. 30(6):445-

600.

69. He S et al. (2011). Glutathione-S-transferase enhances proliferation-migration

and protects against shikonin-induced cell death in breast cancer cells.

Kaohsiung J Med Sci. 27(11):477-84.

70. Hebert H and Jegerschöld C. (2007). The structure of membrane associated

proteins in eicosanoid and glutathione metabolism as determined by electron

crystallography. Curr Opin Struct Biol. 17(4):396-404.

71. Heinlein CA and Chang C. (2004). Androgen receptor in prostate cancer.

Endocr Rev. 25(2):276-308.

85

72. Hess C et al. (1999). Ectosomes released by human neutrophils are

specialized functional units. J Immmunol. 163(8):4564-73.

73. Hokaiwado N et al. (2008). Glutathione S-transferase Pi mediates proliferation

of androgen-independent prostate cancer cells. Carcinogenesis. 29(6):1134-

1138.

74. Hornby JA et al. (2000). Equilibrium folding of dimeric class mu glutathione

transferases involves a stable monomeric intermediate. Biochemistry.

39(40):12336-44.

75. Horwich A et al. (2013). Prostate cancer: ESMO Clinical Practice Guidelines for

diagnosis, treatment and follow-up. Ann Oncol. 24(suppl 6):vi106-vi114.

76. Hoshino D et al. (2013). Exosome secretion is enhanced by invadopodia and

drives invasive behavior. Cell Rep. 5(5):1159-68.

77. Hosseini-Beheshti E et al. (2012). Exosomes as Biomarker Enriched

Microvesicles: Characterization of Exosomal Proteins Derived from a Panel of

Prostate Cell Lines with Distinct AR Phenotypes. Mol Cell Proteomics.

11(10):863-885.

78. Huang J et al. (2003). Prognostic significance of glutathione S-transferase-pi in

invasive breast cancer. Mod Pathol. 16(6):558-65.

79. Huncharek M et al. (2010). Smoking as a risk factor for prostate cancer: a

meta-analysis of 24 prospective cohort studies. Am J Public Health.

100(4):693-701.

80. Ito M et al. (2011). GSTT1 is upregulated by oxidative stress through p38-MK2

signaling pathway in human granulosa cells: possible association with

mitochondrial activity. Aging (Albany NY). 3(12):1213-23.

81. Janowska-Wieczorek A et al. (2001). Platelet-derived microparticles bind to

hematopoietic stem/progenitor cells and enhance their engraftment. Blood.

98(10):3143-9.

82. Jardim BV et al. (2012). Glutathione transferase pi (GSTpi) expression in

breast cancer: An immunohistochemical and molecular study. Acta Histochem.

114(5):510-7.

83. Jia S et al. (2014). Emerging technologies in extracellular vesicle-based

molecular diagnostics. Expert Rev Mol Diagn. 14(3):307-21.

84. Jung T et al. (2009). CD44v6 dependence of premetastatic niche preparation

by exosomes. Neoplasia. 11(10):1093-105.

85. Kanekura T et al. (2002). Basigin (CD147) is expressed on melanoma cells and

induces tumor cell invasion by stimulating production of matrix

86

metalloproteinases by fibroblasts. Int J Cancer. 99(4):520-8.

86. Kanwal R et al. (2014). Protection against oxidative DNA damage and stress in

human prostate by glutathione S-transferase P1. Mol Carcinog. 53(1):8-18.

87. Kelly RW et al. (1991). Extracellular organelles (prostasomes) are

immunosuppressive components of human semen. Clin Exp Immunol.

86(3):550-6.

88. Kharaziha P et al. (2015). Molecular profiling of prostate cancer derived

exosomes may reveal a predictive signature for response to docetaxel.

Oncotarget. [Epub ahead of print].

89. Kiciński M et al. (2011). An epidemiological reappraisal of the familial

aggregation of prostate cancer: a meta-analysis. PLoS One. 6(10):e27130.

90. Kim MS et al. (2015). Development of exosome-encapsulated paclitaxel to

overcome MDR in cancer cells. Nanomedicine. pii: S1549-9634(15)00202-6.

91. Kinkade CW et al. (2008). Targeting AKT/mTOR and ERK MAPK signaling

inhibits hormone-refractory prostate cancer in a preclinical mouse model. J Clin

Invest. 118(9):3051-64.

92. Koga K et al. (2005). Purification, characterization and biological significance of

tumor-derived exosomes. Anticancer res. 25(6A):3703-7.

93. Kogure T et al. (2011). Intercellular nanovesicle-mediated microRNA transfer: a

mechanism of environmental modulation of hepatocellular cancer cell growth.

Hepatology. 54(4):1237-48.

94. Köppler B et al. (2006). Differential mechanisms of microparticle transfer to B

cells and monocytes: anti-inflammatory properties of microparticles. Eur J

Immunol. 36(3):648-60.

95. Kourembanas S. (2015). Exosomes: vehicles of intercellular signaling,

biomarkers, and vectors of cell therapy. Annu Rev Physiol. 77:13-27.

96. Kowal J et al. (2014). Biogenesis and secretion of exosomes. Curr Opin Cell

Biol. 29:116-25.

97. Kucharzewska P et al. (2013). Exosomes reflect the hypoxic status of glioma

cells and mediate hypoxia-dependent activation of vascular cells during tumor

development. Proc Natl Acad Sci USA. 110(18):7312-7.

98. Kursula I et al. (2005). Crystal structure of non-fused glutathione S-transferase

from Schistosoma japonicum in complex with glutathione. Protein Pept Lett.

12(7):709-12.

99. Lash LH. (1994). Role of renal metabolism in risk to toxic chemicals. Environ

Health Perspect. 102 Suppl 11:75-9.

87

100. Lee JK et al. (2013). Exosomes derived from mesenchymal stem cells

suppress angiogenesis by down-regulating VEGF expression in breast cancer

cells. PLoS One. 8(12):e84256.

101. Lee WH et al. (1994). Cytidine methylation of regulatory sequences near the pi-

class glutathione S-transferase gene accompanies human prostatic

carcinogenesis. Proc Natl Acad Sci U S A. 91(24):11733-7.

102. Li J et al. (2015). Serum-free culture alters the quantity and protein composition

of neuroblastoma-derived extracellular vesicles. J Extracell Vesicles. 4:26883.

103. Lilja H et al. (2007). Long-term prediction of prostate cancer up to 25 years

before diagnosis of prostate cancer using prostate kallikreins measured at age

44 to 50 years. J Clin Oncol. 25(4):431-6.

104. Liska DJ. (1998). The detoxification enzyme systems. Altern Med Rev.

3(3):187-98.

105. Liu L et al. (2011). Association of tissue promoter methylation levels of APC,

TGFβ2, HOXD3 and RASSF1A with prostate cancer progression. Int J Cancer.

129(10).2454-62.

106. Loeffler M et al. (2006). Targeting tumor-associated fibroblasts improves

cancer chemotherapy by increasing intratumoral drug uptake. J Clin Invest.

116(7):1955-62.

107. Lösche W et al. (2004). Platelet-derived microvesicles transfer tissue factor to

monocytes but not to neutrophils. Platelets. 15(2):109-15.

108. Mack M et al. (2000). Transfer of the chemokine receptor CCR5 between cells

by membrane-derived microparticles: a mechanism for cellular human

immunodeficiency virus 1 infection. Nat Med. 6(7):769-75.

109. Mahon KL et al. (2014). Methylated Glutathione S-transferase 1 (mGSTP1) is a

potential plasma free DNA epigenetic marker of prognosis and response to

chemotherapy in castrate-resistant prostate cancer. Br J Cancer. 111(9):1802-

9.

110. Marberger M et al. (2012). Novel approaches to improve prostate cancer

diagnosis and management in early-stage disease. BJU Int. 109(suppl 2):1-7.

111. Marchiani S et al. (2010). Androgen-responsive and -unresponsive prostate

cancer cell lines respond differently to stimuli inducing neuroendocrine

differentiation. Int J Androl. 33(6):784-93.

112. Martínez MC et al. (2005). Shed membrane microparticles from circulating and

vascular cells in regulating vascular function. Am J Physiol Heart Circ Physiol.

288(3):H1004-9.

88

113. Martinez-Outschoorn UE et al. (2011). Anti-estrogen resistance in breast

cancer is induced by the tumor microenvironment and can be overcome by

inhibiting mitochondrial function in epithelial cancer cells. Cancer Biol Ther.

12(10):924-38.

114. Mathivanan S et al. (2010). Exosomes: extracellular organelles important in

intercellular communication. J Proteomics. 73(10):1907-20.

115. Mattern J et al. (2002). Expression of drug resistance gene products during

progression of lung carcinomas. Oncol Rep. 9(6):1811-4.

116. Matthias C et al. (1998). The glutathione-S-transferase GSTP1 polymorphism:

effects on susceptibility to oral/pharyngeal and laryngeal carcinomas.

Pharmacogenetics. 8(1):1-6.

117. Mavis CK et al. (2009). Expression level and DNA methylation status of

glutathione-S-transferase genes in normal murine prostate and TRAMP

tumors. Prostate. 69(12):1312-24.

118. McElwee JJ et al. (2007). Evolutionary conservation of regulated longevity

assurance mechanisms. Genome Biol. 8(7):R132.

119. McIlwain CC et al. (2006). Glutathione S-transferase polymorphisms: cancer

incidence and therapy. Oncogene. 25(11):1639-48.

120. McKay RR et al. (2015). Docetaxel, bevacizumab, and androgen deprivation

therapy for biochemical relapse after definitive local therapy for prostate

cancer. Cancer. 121(15):2603-2611.

121. Miyake T et al. (2012). GSTP1 expression predicts poor pathological complete

response to neoadjuvant chemotherapy in ER-negative breast cancer. Cancer

Sci. 103(5):913-20.

122. Mo Z et al. (2009). An updating meta-analysis of the GSTM1, GSTT1, and

GSTP1 polymorphisms and prostate cancer: a HuGE review. Prostate.

69(6):662-88.

123. Morvan J et al. (2012). Pkh1/2-dependent phosphorylation of Vps27 regulates

ESCRT-I recruitment to endosomes. Mol Biol Cell. 23(20):4054-64.

124. Muralidharan-Chari V et al. (2009). ARF6-regulated shedding of tumor cell-

derived plasma membrane microvesicles. Curr Biol. 19(22):1875-85.

125. Muralidharan-Chari V et al. (2010). Microvesicles: mediators of extracellular

communication during cancer progression. J Cell Sci. 123(Pt 10):1603-11.

126. Nelson WG et al. (2013). Prostate cancer. In: Niederhuber JE, Armitage JO,

Doroshow JH, et al, eds. Abeloff’s Clinical Oncology. 5th ed. Philadelphia, PA:

Elsevier Churchill Livingstone; 1463-96.

89

127. Nock NL et al. (2009). Polymorphisms in glutathione S-transferase genes

increase risk of prostate cancer biochemical recurrence differentially by

ethnicity and disease severity. Cancer Causes Control. 20(10):1915-1926.

128. Oakley A. (2011). Glutathione transferases: a structural perspective. Drug

Metab Rev. 43(2):138-51.

129. Okino ST et al. (2009). Chromatin changes on the GSTP1 promoter associated

with its inactivation in prostate cancer. Mol Carcinog. 46(10)839-46.

130. Orimo A et al. (2005). Stromal fibroblasts present in invasive human breast

carcinomas promote tumor growth and angiogenesis through elevated SDF-

1/CXCL12 secretion. Cell. 121(3):335-48.

131. Pal RP et al. (2013). Defining prostate cancer risk before prostate biopsy.

UrolOncol. 31(8):1408-18.

132. Palma J et al. (2012). MicroRNAs are exported from malignant cells in

customized particles. Nucleic Acids Res. 40(18):9125-38.

133. Paluch E et al. (2005). Corticol actomyosin breakage triggers shape oscillations

in cells and cell fragments. Biophys J. 89(1):724-33.

134. Park JE et al. (2010). Hypoxic tumor cell modulates its microenvironment to

enhance angiogenic and metastatic potential by secretion of proteins and

exosomes. Mol Cell Proteomics. 9(6):1085-99.

135. Park JS et al. (2002). Cellular sensitivity determinants to docetaxel in human

gastrointestinal cancers. Int J Oncol. 20(2):333-8.

136. Parolini I et al. (2009). Microenvironmental pH is a key factor for exosome

traffic in tumor cells. J Biol Chem. 284(49):34211-22.

137. Pasquet JM et al. (1996). Calcium influx is a determining factor of calpain

activation and microparticle formation in platelets. Eur J Biochem. 239(3):647-

54.

138. Peters W et al. (1994). Glutathione, glutathione-s-transferase and p-170

glycoprotein in metastases of malignant melanomas. Int J Oncol. 4(6):1323-7.

139. Phé V et al. (2010). Methylated genes as potential biomarkers in prostate

cancer. BJU Int. 105(10):1364-70.

140. Phuyal S et al.(2014). Regulation of exosome release by glycophingolipids and

flotillins. FEBS J. 281(9):2214-27.

141. Poliakov A et al. (2009). Structural heterogeneity and protein composition of

exosome-like vesicles (prostasomes) in human semen. Prostate. 69(2):159-67.

142. Qu JL et al. (2009). Gastric cancer exosomes promote tumour cell proliferation

through PI3K/Akt and MAPK/ERK activation. Dig Liver Dis. 41(12):875-80.

90

143. Quail DF and Joyce JA. (2013). Microenvironmental regulation of tumor

progression and metastasis. Nat Med. 19(11):1423-37.

144. Raposo G et al. (1996). B lymphocytes secrete antigen-presenting vesicles. J

Exp Med. 183(3):1161-72.

145. Ratajczak J et al. (2006). Membrane-derived microvesicles: important and

underappreciated mediators of cell-to-cell communication. Leukemia.

20(9):1487-95.

146. Rebbeck TR et al. (1999). Glutathione S-transferase-mu (GSTM1) and –theta

(GSTT1) genotypes in the etiology of prostate cancer. Cancer Epidemiol

Biomarkers Prev. 8(4 Pt 1):283-7.

147. Recine F and Sternberg CN. (2015). Hormonal therapy and chemotherapy in

hormone-naïve and castration resistant prostate cancer. Transl Androl Urol.

4(3):355-64.

148. Reinemer P et al. (1997). The three-dimensional structure of class pi

glutathione S-transferase in complex with glutathione sulfonate at 2.3 A

resolution. EMBO J. 10(8): 1997-2005.

149. Resnick MJ and Pension DF. (2012). Quality of life with advanced metastatic

prostate cancer. Urol Clin North Am. 39(4):505-15.

150. Sameshima T et al. (2000). Glioma cell extracellular matrix metalloproteinase

inducer (EMMPRIN) (CD147) stimulates production of membrane-type matrix

metalloproteinases and activated gelatinase A in co-cultures with brain-derived

fibroblasts. Cancer Lett. 157(2):177-84.

151. Saxena A et al. (2012). GSTP1 methylation and polymorphism increase the

risk of breast cancer and the effects of diet and lifestyle in breast cancer

patients. Exp Ther Med. 4(6):1097-1103.

152. Schneider A and Simons M. (2013). Exosomes: vesicular carriers for

intercellular communication in neurodegenerative disorders. Cell Tissue Res.

352(1):33-47.

153. Prostate cancer treatment. ScotDir.com. Retrieved July 17, 2016, from

http://scotdir.com/health-and-fitness-2/prostate-cancer-treatment

154. Sherratt PJ et al. (1997). Evidence that human class Theta glutathione S-

transferase T1-1 can catalyse the activation of dichloromethane, a liver and

lung carcinogen in the mouse. Comparison of the tissue distribution of GST T1-

1 with that of classes Alpha, Mu and Pi GST in human. Biochem J. 326 (Pt

3):837-46.

155. Shiizaki S et al. (2013). Activation mechanisms of ASK1 in response to various

91

stresses and its significance in intracellular signaling. Adv Biol Regul.

53(1):135-44.

156. Silberstein JL et al. (2013). Current clinical challenges in prostate cancer.

Transl Androl Urol. 2(3):122-36.

157. Singh S. (2015). Cytoprotective and regulatory functions of glutathione S-

transferases in cancer cell proliferation and cell death. Cancer Chemother

Pharmacol. 75(1):1-15.

158. Skog J et al. (2008). Glioblastoma microvesicles transport RNA and protein

that promote tumor growth and provide diagnostic biomarkers. Nat Cell Biol.

10(12):1470-1476.

159. Skokos D et al. (2003). Mast cell-derived exosomes induce phenotypic and

functional maturation of dendritic cells and elicit specific immune responses in

vivo. J Immunol. 170(6):3037-45.

160. Stamey T et al. (1987). Prostate-Specific Antigen as a Serum Marker for

Adenocarcinoma of the Prostate. N Engl J Med. 317(15):909-916.

161. Stegmayr B and Ronquist G. (1982). Promotive effect on human sperm

progressive motility by prostasomes. Urol Res. 10(5):253-7.

162. Stuelten CH et al. (2010). Transient tumor-fibroblast interactions increase

tumor cell malignancy by a TGF-Beta mediated mechanism in a mouse

xenograph model of breast cancer. PLoS One. 5(3):e9832.

163. Tauro BJ et al. (2012). Comparison of ultracentrifugation, density gradient

separation, and immunoaffinity capture methods for isolating human colon

cancer cell line LIM1863-derived exosomes. Methods. 56(2):293-304.

164. Tavoosidana G et al. (2011). Multiple recognition assay reveals prostasomes

as promising plasma biomarkers for prostate cancer. Proc Natl Acad Sci U S A.

108(21):8809-14.

165. Taylor DD and Shah S. (2015). Methods of isolating extracellular vesicles

impact down-stream analyses of their cargoes. Methods. pii: S1046-

2023(15)00092-4.

166. Taylor RC et al. (2008). Apoptosis: controlled demolition at the cellular level.

Nat Rev Mol Cell Biol. 9(3):231-41.

167. Tew K and Townsend D. (2012). Glutathione-S-Transferases As Determinants

of Cell Survival and Death. Antioxid Redox Signal. 17(12):1729-1737

168. Tew KD and Townsend DM. (2011). Redox platforms in cancer drug discovery

and development. Curr Opin Chem Biol. 15(1):156-61.

169. Tew KD et al. (1996). Glutathione-associated enzymes in the human cell lines

92

of the National Cancer Institute Drug Screening Program. Mol Pharmacol.

50(1):149-59.

170. Théry C et al. (1999). Molecular characterization of dendritic cell-derived

exosomes. Selective accumulation of the heat shock protein hsc73. J Cell Biol.

147(3):599-610.

171. Théry C et al. (2006). Curr Protoc Cell Biol. Wiley Online Library, Chapter

3:Unit 3.22.

172. Thompson IM et al. (2004). Prevalence of prostate cancer among men with a

prostate-specific antigen level < or =4.0 ng per milliliter. N Engl J Med.

350(22):2239-46.

173. Tlsty TD and Hein PW. (2001). Know thy neighbor: stromal cells can contribute

oncogenic signals. Curr Opin Genet Dev. 11(1):54-9.

174. Toner AP et al. (2013). The novel toluidine sulphonamide EL 102 shows pre-

clinical in vitro and in vivo activity against prostate cancer and circumvents

MDR1 resistance. Br J Cancer. 109(8):2131-41.

175. Tsakalozou E et al. (2012). Combination effects of docetaxel and Doxorubicin

in hormone-refractory prostate cancer cells. Biochem Res Int. 2012:832059.

176. Ulmert D et al. (2008). Prostate-specific antigen at or before age 50 as a

predictor of advanced prostate cancer diagnosed up to 25 years later: a case-

control study. BMC Med. 6:6.

177. Valadi H et al. (2007). Exosome-mediated transfer of mRNAs and microRNAs

is a novel mechanism of genetic exchange between cells. Nat Cell Biol.

9(6):654-59.

178. Valenti R et al. (2007). Tumor-released microvesicles as vehicles of

immunosuppression. Cancer Res. 67(7):2912-5.

179. Vickers AJ et al. (2010). Prostate specific antigen concentration at age 60 and

death or metastasis from prostate cancer: case-control study. BMJ. 341:c4521.

180. Wang T et al. (2001). Glutathione S-transferase P1-1 (GSTP1-1) inhibits c-Jun

N-terminal kinase (JNK1) signaling through interaction with the C terminus. J

Biol Chem. 276(24):20999-1003.

181. Wang Z et al. (2010). Decreased expression of GST pi is correlated with a poor

prognosis in human esophageal squamous carcinoma. BMC Cancer. 10:352.

182. Webber J et al. (2010). Cancer exosomes trigger fibroblast to myofibroblast

differentiation. Cancer Res. 70(23):9621-30.

183. Webber JP et al. (2015). Differentiation of tumour-promoting stromal

myofibroblasts by cancer exosomes. Oncogene. 34(3):290-302.

93

184. Wei B et al. (2012). Association of GSTM1 null allele with prostate cancer risk:

evidence from 36 case-control studies. PLoS One. 7(10):e46982.

185. Wickmark A et al. (2009). Endocrine treatment, with or without therapy, in

locally advanced prostate cancer (SPCG-7/SFUO-3): an open randomised

phase III trial. Lancet. 373(9660):301-8.

186. Wiklund FE et al. (2009). Established prostate cancer susceptibility variants are

not associated with disease outcome. Cancer Epidemiol Biomarkers Prev.

18(5):1659-1662.

187. Xiang X et al. (2009). Induction of myeloid-derived suppressor cells by tumor

exosomes. Int J Cancer. 124(11):2621-33.

188. Yegnasubramanian S et al. (2004). Hypermethylation of CpG islands in primary

and metastatic human prostate cancer. Cancer Res. 64(6):1975-86.

189. Yoshioka Y et al. (2013). Comparative marker analysis of extracellular vesicles

in different human cancer types. J Extracell Vesicles. 2:10.3402.

190. Zappa M et al. (2014). A different method of evaluation of the ERSPC trial

confirms that prostate-specific antigen testing has a significant impact on

prostate cancer mortality. Eur Urol. 66(3):401-3.

191. Zeliadt SB et al. Why do men choose one treatment over another? A review of

patient decision making for localized prostate cancer. Cancer. 106(9):1865-74.

192. Zheng X et al. (2014). Exosome analysis: a promising biomarker system with

special attention to saliva. J Membr Biol. 247(11):1129-36.

94

Appendices

Appendix A: Sequencing of GSTP1 in pGEM-T easy vector

Sequencing of GSTP1 in the pGEM-T easy vector showed a base pair change at

nucleotide +5 changing the corresponding amino acid from P to R (Proline to

Arginine).

95

Appendix B: Sequencing of GSTP1 after site-directed mutagenesis

Site-directed mutagenesis of the GSTP1 sequence showing the base change at

nucleotide +5 from G to C.

C C G