Umeå University Medical Dissertations, New Series No 1406

Prostate Cancer Diagnosis Experimental and Clinical Studies With HRMAS NMR Spectroscopy

Katarina Stenman

Department of Radiation Sciences,

Diagnostic Radiology

Umeå 2011

Responsible publisher under Swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7459-154-5

ISSN: 0346-6612 1406

Omslagsbild: Katarina Stenman

Elektronisk version tillgänglig på http://umu.diva-portal.org/

Tryck/Printed by: VMC, KBC, Umeå University

Umeå, Sweden 2011

Dedicated to all men …

... and the women who care about them.

5

Table of Contents

Table of Contents 5 Abstract 7 Objective 9 Outline 10

Part I 10 Part II 10 Part III 10

Original Papers and My Contributions 11 Acknowledgements 12 Abbreviations 13 Part I 15 Chapter 1: Introduction 17 Chapter 2: The Prostate 20

2.1 What is a Prostate? Anatomy and Function 20 2.2 The Remarkable Metabolism of the Prostate 22 2.3 Man vs. Rat Prostate 26

Chapter 3: Diagnostic Methods of the Prostate 28 3.1 General Diagnostic Procedures 28 3.1.1 Digital Rectal Examination – DRE 28 3.1.2 Prostate Specific Antigen – PSA 28 3.1.3 Trans Rectal Ultra Sonography – TRUS 29 3.2 Imaging of the Prostate 29 3.2.1 Computerized Tomography - CT 30 3.2.2 Positron Emission Tomography - PET 30 3.2.3 Magnetic Resonance Imaging - MRI 31 3.2.4 Magnetic Resonance Spectroscopic Imaging – MRSI 31 3.2.5 Small Animal MRI/MRSI 32 3.3 Prostate Biopsy 33 3.4 Histopathological Interpretation of Tissue Specimens 33 3.4.1 The Gleason Score – GS 34 3.4.2 The TNM Table 35 3.4.3 Immunohistochemistry 35

Part II 37 Chapter 4: Nuclear Magnetic Resonance - NMR 39

4.1 General Introduction 39 4.2 The NMR Signal 39 4.3 The NMR Experiment 41 4.4 Chemical Shift 42 4.5 Spin-Spin Coupling 44

6

4.6 High-resolution Magic Angle Spinning (HRMAS) NMR 45 4.7 2-Dimensional NMR 46 4.7.1 TOtal Correlation SpectroscopY - TOCSY 46 4.7.2 Hetero-nuclear Single Quantum Coherence –HSQC 47

Chapter 5: Metabolomics Using HRMAS NMR 48 5.1 Concept of Metabolomics 48 5.2 Metabolomics in the Characterization of the Prostate 49 5.3 Metabolomics, Nutrition and Prostate Cancer 50

Part III 51 Chapter 6: Materials and Methods 53

6.1 Experimental Setup 53 6.1.1 Procedure of HRMAS NMR Spectroscopy 53 6.1.2 1D 1H HRMAS NMR 53 6.2 Prostate Tissue Samples 54 6.2.1 Patients 54 6.2.2 Rat 55 6.3 Interpretation of Data 55 6.3.1 Pathologic Analysis 56 6.3.2 Statistical Analysis 56

Chapter 7: Results and Comments 58 Paper I 58 Paper II 59 Paper III 61

Chapter 8: Discussion 64 Chapter 9: Suggestions for the Future 67 Chapter 10: References 68

7

Abstract

A few abnormal cells found in a small piece of prostate tissue are most consequential for a man’s future.

The prevalence of prostate cancer (PCa) is increasing globally. The main

instigating factor for this cancer is not yet known, but it appears to be the

consequence of many variables such as an increasingly older population, more

frequent PSA-testing, and factors involving lifestyle.

Prostate cancer screening, as an equivalent for breast cancer screening, has been

suggested but unfortunately there are no accurate diagnostic tools available for

this type of screening. The reason for this is simply that the prostate is one of the

most difficult organs to diagnose and, consequently, PCa screening would

generate far too many false-positive and false-negative results. The prostate is

not easily accessible as it is deeply-seated in the male pelvic area, wrapped

around the urethra and surrounded by sensitive vital organs. Furthermore, PCa

is frequently multi-focal, and the cancer cells have a tendency of assimilating

among normal cells and, thus, do not always form solid lumps. Therefore,

prostate tumors are often not felt by digital rectal examination (DRE) or

identified by imaging. The PSA-test is not reliable as it is more prostate-specific

than cancer-specific. Due to increasing prostate awareness, more early-stage

and locally confined PCas are being detected. This is saving lives, although there

is a high risk of over treatment and unnecessary side-effects. The increased

detection of PCa requires sophisticated diagnostic methods and highly skilled

clinicians who can discern between indolent and aggressive cancers. The current

“gold-standard” for PCa diagnosis is biopsy grading by pathologists using the

Gleason score system, which is a difficult task. Therefore, innovative methods to

improve the precision of prostate diagnosis, by increased biopsy sensitivity and

tumor localization, are of essence.

In light of these difficulties, the metabolomic approach using 1D and 2D high-

resolution magic angle spinning (HRMAS) NMR spectroscopy combined with

histopathology on intact prostatectomy specimens was evaluated in this research

project. The non-destructive nature of HRMAS NMR enables spectroscopic

analysis of intact tissue samples with consecutive histological examinations

under light microscope. Metabolomics aids in the unraveling and the discovery

of organ-specific endogenous metabolites that have the potential to be reliable

indicators of organ function and viability, extrinsic and intrinsic perturbations,

as well as valuable markers for treatment response. The results may, therefore,

be applied clinically to characterize an organ by utilizing bio-markers that have

the capacity to distinguish between disease and health.

8

The aim was to characterize the human and the rat prostate in terms of its

intermediary metabolism, which is shown here to differ between species and

anatomical regions. Furthermore, the aim is to seek the verification of HRMAS

NMR derived metabolites which are known to be a part of the prostate

metabolome such as, citrate, choline, and the polyamines which were performed,

but also the identification of metabolites not previously identified as part of the

local prostate metabolism, such as Omega-6, which was detected in tumors. The

extended aim was to elucidate novel bio-markers with clinical potential. In this

study, the common phyto-nutrient, inositol, which appears to possess protective

properties, was identified as being a potentially important PCa bio-marker for

the distinction between the more indolent Gleason score 6 and the more

aggressive Gleason score 7 in non-malignant prostate tissues with tumors

elsewhere in the organ. Further studies in this area of PCa research are therefore

warranted.

9

Objective

The main objective of this thesis was to investigate and evaluate the application

of HRMAS NMR on intact prostate tissues. During the course of the study, it

became clear that nothing about the prostate is easy, supporting the validity of

the statement below:

“Despite the major progress that has occurred in the biological sciences during the last 50

years, it is rather remarkable that we are about to enter the twenty-first century, and still

the specific function of the prostate gland is unknown. Indeed the prostate is the largest

organ of unknown specific function in the human body.”

Dr. John Isaacs John Hopkins School of Medicine

Therefore, as a result, the initial aims and goals have been revised several times

and finally resulted in a more comprehensive overview of the complexity of the

prostate and a presentation of HRMAS NMR spectroscopy, used in conjunction

with histopathology, as a powerful tool to gain a deeper understanding of the

intracellular local metabolism of the prostate. This method has proved to be

useful in providing valuable pieces to the, far-from-finished, puzzle of the

prostate and its diseases.

10

Outline

This thesis consists of three parts and a total of 10 chapters. The aim is to guide

the reader through the complexities that involves the prostate gland in general,

throughout the theory behind NMR spectroscopy and its utility and, finally,

towards the results and its interpretation and suggestions for future studies.

Part I

Chapter 1 presents a brief, but comprehensive, introduction about the research

topic in general; prostate cancer.

Chapter 2 provides information about the prostate such as anatomy, function,

and metabolism. It also includes an overview on the subject of experimental

rodent prostate cancer models.

Chapter 3 reviews the different diagnostic options available for prostate

diagnosis.

Part II

Chapter 4 and 5 presents the method used to generate the results in this thesis,

namely NMR, with a strong emphasis on HRMAS NMR and metabolomics.

Part III

Chapters 6-10 comprise the design of experiments, methods, results, and

discussions followed by research papers.

11

Original Papers and My Contributions

The following peer-reviewed papers are included in this doctoral dissertation

and referred to by their Roman numerals.

My contributions are shown in Table 1.

I Detection of polyunsaturated omega-6 fatty acid in human

malignant prostate tissue by 1D and 2D high-resolution magic angle spinning NMR spectroscopy. Stenman K, Hauksson JB, Gröbner G, Stattin P, Bergh A, and Riklund K. MAGMA. 2009;22(6):327-331.

II Detection of Local Prostate Metabolites by HRMAS NMR Spectroscopy: A Comparative Study of Human and Rat Prostate Tissues. Stenman K, Surowiec I, Antti H, Riklund K, Stattin P, Bergh A, and Gröbner G. Magnetic Resonance Insights. 2010;4:27-41.

III 1H HRMAS NMR Derived Bio-markers Related to Tumour Grade, Tumour Cell Fraction, and Cell Proliferation in Prostate Tissue Samples. Stenman K, Stattin P, Stenlund H, Riklund K, Gröbner G, and Bergh A. Biomarker Insights. 2011;6:1–9.

Table 1: Contributions

Main contribution=1, Contributed to high extend=2, Contributed=3

Part Paper

I Paper

II Paper

III

Idea and design of study 1 1 1

Experimental design 1 1 1

Execution of experiments 1 1 1

Data interpretation and analysis

1 1 1

Writing of manuscript 1 1 1

Note: All three papers were reprinted with kind permissions from the publishers.

12

Acknowledgements

“Haba na haba, hujaza kibaba”*

Swahili kanga writing

This study has been collaboration with Professor Anders Bergh, Pathology, the

Department of Medical Biosciences, Umeå University Hospital; Professor

Gerhard Gröbner, Physical Chemistry, the Department of Chemistry, Umeå

University; and Professor Katrine Åhlström Riklund, Diagnostic Radiology, the

Department of Radiation Sciences, Umeå University Hospital. This work has

been funded at large by the Lion’s Cancer Research Foundation and also by

generous grants from the Swedish Cancer Foundation, the JC Kempe Memorial

Foundation Academic Funds, and the Wallenberg Foundation. Thanks are due

to many people and as a precaution not to unintentionally exclude anyone, only

the people in close collaboration are therefore mentioned: Dr. Jón Haukson,

Professor Mikael Karlsson, Anna Wernblom, Christina Ericsson, Linda Kocsis,

Dept. of Radiation Sciences; Professor Pär Stattin, Dept. of Surgical and

Perioperative Sciences; Dr. Hans Stenlund, Dept. of Public Health and Clinical

Medicine; Professor Dan Johnels, Ingmar Sethson, Professor Jurgen

Schleucher, Dr. Izabella Surowiec, Dr. Henrik Antti, Dept. of Chemistry;

Birgitta Ekblom, Elisabeth Dahlberg, Pernilla Andersson, Dept. of Medical

Biosciences; Kerstin Almroth and Britt-Inger Dahlin, Dept. of Surgical and

Perioperative Sciences. Lisbeth, Erica, Eivor, Anna, Ann-Kristin, Margaret, and

Ronald, – thank you very much!! Again, special thanks to: Anders Bergh who

performed all of the, very difficult and time consuming, histopathological

evaluations of the prostate tissues, prior and subsequent HRMAS NMR analysis,

and also contributed significantly in the writing of manuscripts; Gerhard

Gröbner, who joined the group in a late stage and with his ambitious and

efficient ways proved to be the “missing piece” for the completion of this

dissertation; and my supervisor, Katrine Riklund, who administered this project.

Last but definitely not least, thanks to the patients who generously donated their

prostates for this study and thereby contributed to the highest degree to this

research and this dissertation. We wish all of you the best of health!

*“Small things, when combined together make up big things”.

As this Swahili proverb says, all things are possible if you just take it little by little, step by step.

13

Abbreviations

AA Arachidonic Acid

Acetyl-CoA Acetyl-Coenzyme A

ATP Adenosine Triphosphate

AUC Area Under the ROC Curve

BPH Benign Prostatic Hyperplasia

COX Cyclooxygenase

CPMG Carr-Purcell-Meibom-Gill

CT Computerized Tomography

D2O Deuterium Oxide

DLP Dorso-Lateral Prostate

DRE Digital Rectal Examination

EFA Essential Fatty Acid

EPA Eicosapentaenoic Acid

EtOH Ethanol/Ethyl alcohol

FID Free Induction Decay

GC-MS Gas Chromatography-Mass Spectrometry

GS Gleason Scores

HRMAS High Resolution Magic Angle Spinning

HSQC Hetero-nuclear Single Quantum Coherence

Hz Hertz

MRSI Magnetic Resonance Spectroscopic Imaging

MS Mass Spectrometry

NMR Nuclear Magnetic Resonance

NSAID Non-Steroidal Anti-Inflammatory Drug

OR Odds Ratios

PCa Prostate Cancer

PG Prostaglandin

PPM Parts Per Million

PSA Prostate Specific Antigen

PUFA Polyunsaturated Fatty Acid

RF Radio Frequent

ROC Receiver Operating Characteristic

TNM Tumor, Nodes (lymph), and Metastasis

TOCSY Total Correlation Spectroscopy

TRUS Trans Rectal Ultra Sonography

VP Ventral Prostate

14

15

Part I

16

17

Chapter 1: Introduction

Globally, prostate cancer (PCa) is rapidly becoming the most common form of

male specific cancer [Thun et al., 2010]. Consequently, a large international

research community is actively engaged in trying to learn more about this deep-

seated gland in the reproductive tract of the male body. The prostate, with its

heterogeneity and anomalous metabolism, [Costello et al., 1997, 1999, 2000] is

one of the most difficult human organs to diagnose. The PCa pathology poses a

great challenge for clinicians and scientists [Fleshner et al., 2002; Swindle et al.,

2008] and in order to understand this disorder, it is of absolute necessity to

build up a thorough comprehension of this organ and learn to interpret its

biochemical signals. The majority of PCa are slow-growing cancers and, in fact,

more men die from this disease than not. The main goal in the fight against PCa

is to find a reliable method that can differentiate between the indolent and the

aggressive types [Madu et al., 2010]. There are ongoing discussions about

diverse theories relevant to the cause and instigation of PCa. Although not

exclusively, PCa and other ailments of the prostate are by far more common in

older men [Sharma et al., 2010]. In addition to age, factors, such as, genetic

susceptibility, diet and lifestyle, infectious agents, inflammation, etc. are all

possible instigators for the development of PCa [Coffey et al., 2001; Gonzales et

al., 2010; American Cancer Society, 2010]. The incidence of PCa and other

diseases related to the prostate are expected to increase due to an ever increasing

aging male population [Thun et al., 2010].

One of the main obstacles involved in the diagnosis of the prostate is its

inaccessibility in the body. Furthermore, the prostate, especially a malignant

one, is a heterogeneous structure with different cell types and pathologies, and

cancer cells often infiltrate among normal cells [Zakian et al., 2005]. Contrary to

other solid tumors, prostate tumors do not normally develop into solid lumps.

Therefore, they are not always detectable by digital rectal examination (DRE),

which, jointly with the Prostate-Specific Antigen (PSA) test, is part of the clinical

routine examination when suspecting PCa [Kelloff et al., 2009]. Indirect

diagnosis of PCa is achieved through a simple blood test, using the PSA-test.

However, PSA is not cancer particular, but more so prostate specific even though

PSA is also expressed in other tissues and in women, and thus not a reliable

indicator. The value of PCa screening using the PSA-test is not determined given

that it generates frequent false-positive results, which involves the increased risk

of unnecessary biopsies and treatments [Wolf el al, 2010; Vickers et al., 2011].

PCa is multifocal and the cancer foci can be small, i.e., only millimeters in size,

thus making DRE and biopsy sampling a challenge for the clinician.

Furthermore, there is currently no imaging method available that can be used

with high accuracy to guide biopsies towards the tumors and particularly

18

towards the most aggressive ones. Multiple biopsies are therefore taken from

different parts of the prostate and as biopsies are small, many tumors are often

missed [Andriole, 2009]. Biopsy sampling is generally performed in conjunction

with Trans Rectal Ultra Sonography (TRUS) which provides an image of the

internal organs, i.e. the prostate, and thus aids in guiding of the sampling

towards the prostate although many tumors are often not seen. Requiring a

skilled pathologist, biopsy specimens are examined and rated through the

Gleason grading system; this diagnosis will be the basis for the treatment plan

[Berney et al., 2007]. A high Gleason score indicates a poor prognosis; most men

are detected with low or intermediate grade tumors and for them prognosis is

highly variable and unpredictable. The diagnosis and prognostication of PCa are

therefore complicated but research and technological advancements of clinical

diagnostic methods and instruments are hopefully changing the situation for the

better [Kelloff et al., 2009].

The prostate is often enlarged due to Benign Prostatic Hyperplasia (BPH) in men

past the age of 50, sometimes even younger [Kirby, 2000]. BPH and its

consequences such as inflammations [De Marzo et al., 2007], does not make PCa

diagnosis easier. In fact, it is difficult to know what a completely healthy prostate

looks like in a middle-aged man [McLaughlin et al., 2005] and as the senescence

factor is a major part of PCa [Sharma et al., 2010] it is, consequently, virtually

impossible to find completely healthy, older human controls for prostate cancer

research studies. A wide variety of experimental rodent PCa models are available

for pre-clinical studies. Many of these models are successfully utilized since it is

often not possible to use human subjects due to obvious ethical, practical and

economic reasons. However, the anatomy and the metabolisms of rodents differ

from that of man. This is a major drawback [Lamb et al., 2005; Kamb, 2005].

There is currently no standardized method that can accurately diagnose, stage

and predict the outcome for PCa that affects such a large proportion of the male

population [Kelloff et al., 2009; Wolf et al., 2010]. Histological evaluation of

prostate biopsies by skilled pathologists is presently the gold standard, used in

conjunction with the PSA-test [Rajinikanth et al., 2008; Chun et al., 2010].

Novel diagnostic methods that can detect and accurately stage non-palpable and

aggressive tumors at an early stage are urgently needed [Kelloff et al., 2009]. In-

vivo Magnetic Resonance Spectroscopic Imaging (MRSI) [Giusti et al., 2010;

Klijn et al., 2011; Scheenen et al., 2011], especially when used in conjunction

with ex-vivo High Resolution Magic Angle Spinning Nuclear Magnetic

Resonance (HRMAS NMR) spectroscopy, has shown to be a promising

alternative for the future [Kurhanewicz et al., 2002; Swanson et al., 2003;

Santos et al., 2010]. Spectroscopy generates a fingerprint of the metabolism

including a visualization of its perturbations due to disease, i.e., loss of

homeostasis, which often occurs before morphological changes [van der Greef et

19

al., 2005; Bathen et al., 2010]. HRMAS NMR spectroscopy is a non-destructive

method that allows the analyzed specimen to be examined under light

microscopy post NMR [Beckonert et al., 2010; Moestue et al., 2011]. This feature

makes it possible to correlate the spectrum with the specific pathology that can

be verified with advanced immunohistochemistry.

All of the factors involved in PCa and discussed above affect the metabolism one

way or another. Therefore, the focus of this study was to learn more about the

local metabolism of the prostate, malignant as well as non-malignant/healthy in

both man and rat. HRMAS NMR spectroscopy with subsequent

histopathological analysis was used to obtain a correlated spectroscopic and

morphologic “fingerprint” of the metabolism.

20

Chapter 2: The Prostate

2.1 What is a Prostate? Anatomy and Function

The name prostate means protector in Greek. It is believed that its main

function is to protect the sensitive reproductive area and prevent urinary tract

infections. The prostate is an exocrine gland slightly larger than a walnut that fits

snugly, wrapped around the urethra, in the male pelvis. Its strategic location is

right below the bladder and in front of the rectum (Figure 1) [McNeal, 1981].

Figure 1: The male reproductive region.

The terminology of anatomy or functional anatomy of the human prostate is

complicated. This is due to the result of the anatomical changes that normally

occur with age due to hypertrophy. Thus, the prostate of a young man is not

comparable to the prostate of an older man [McLaughlin et al., 2005]. The

location of the prostate is not ideal, (Figure 1), when it comes to clinical

examinations and treatments as it is deeply seated and located close to

vulnerable organs such as the bladder, the rectum, the sphincter muscles

contracting the urethra, and the nerves regulating erection. Consequently

treatments for prostate cancer may cause problems with the nearby organs with

side-effects such as incontinence and impotence.

Four distinct regions or zones make up the prostate: the peripheral, the

transitional, the central and the anterior fibro-muscular (Figure 2). The largest

and most essential, adjacent to the rectum, the peripheral zone, contains most

glands; it is also the site where most cancers arise. The second most important

21

one, including the periurethral gland region, is the transitional zone which

directly surrounds the urethra. This is the zone where BPH originates and

explains why BPH may result in urinary obstruction. The central zone surrounds

the ejaculatory ducts and is often free of disease. The anterior fibro-muscular

zone consists mainly of connective tissues and smooth muscles. The prostate is,

at least in part, enveloped by a fibro muscular layer that is referred to as the

capsule. [McNeal, 1981; Villers et al., 1991; McLaughlin et al., 2005]

Figure 2: The zones of the prostate.

Image adapted from “De Marzo et al., 2007”.

The prostate contains two different compartments: the glands and the fibro

muscular cells. The glands contain luminal/secretory epithelial cells and basal

epithelial cells as well as a few neuroendocrine cells [Sherwood et al., 1991; Roy

Burman et al., 2004]. The stroma is contractile but it also generates many

growth factors regulating the glands. Besides PSA, the prostate glands secrete a

clear and slightly alkaline (pH 7.3) fluid [Mann, 1974], that contains compounds

such as: citrate, zinc, spermine, potassium, acid phosphates, and calcium, many

of which have protective and antibacterial effects. Including the slightly alkaline

fluid produced in the Cowper’s glands, the fluid is drained via the prostatic

ductal system into the urethra where it is combined with spermatozoa and fluid

generated in the seminal vesicle. Collectively, these compounds make up the

semen, expelled during ejaculation. The sperm is produced in seminiferous

22

tubules that are located in each testis and connected to the epididymal duce

which in turn is connected to the vas deferens (Figure 1).

2.2 The Remarkable Metabolism of the Prostate

The human body is equipped with a highly sophisticated metabolic system

committed to keeping us alive and healthy. The core of this system is the cellular

respiration which is defined as the metabolic reactions and processes that take

place in the cells in order to convert biochemical energy from nutrients into

adenosine triphosphate (ATP), with the concomitant release of waste products.

There are three main steps in this process: the Glycolysis which takes place in

the cytosol, and this is where glucose is converted into pyruvate. Pyruvate is

converted into Acetyl-Coenzyme A (acetyl-CoA) which enters the mitochondria

and the Krebs cycle if oxygen is present. The oxidative phosphorylation step in

the mitochondrial cristae comprises the electron transport chain that establishes

a chemiosmotic potential across the inner membrane. This step drives the

phosphorylation of ADP to produce ATP – the “transporter of chemical energy”

(Figure 3) [Mathews et al., 1999].

Figure 3: The cellular respiration system.

The Krebs cycle is also known as the citric acid cycle or the tricarboxylic acid

cycle (TCA cycle). It also goes by the name the Szent-Györgyi-Krebs cycle [Krebs,

1940], named after its two discoverers: Albert Szent-Györgyi (1893 – 1986) and

Hans Adolf Krebs (1900 – 1981). Krebs cycle is probably the most important

metabolic pathway in aerobic cells and organisms as it provides information

23

about the chemical intracellular conversions of carbohydrates, fats and proteins

into carbon dioxide and water to generate energy. This dynamic and amphibolic

process involves a series of enzyme catalyzed chemical reactions that occurs in

the mitochondria of the cell. The citric acid cycle begins with the high energy

compound acetyl-CoA which condenses with oxaloacetate, and catalyzed by

citric acid synthetase, to generate citrate. Citrate is oxidized further in this cycle

through a series of isomerisation-, oxidation-, and decarboxylation-steps that

finally regenerate oxaloacetate in the final step in the Krebs cycle before the cycle

is repeated (Figure 3).

The unusual, truncated Krebs cycle of the prostate (Figure 4) has been, and is

still, largely overlooked by a large section of the research community although

this is exceptionally significant in order to gain a complete understanding of this

complex organ. Contrary to the average mammalian cell, normal and BPH

prostate glandular cells accumulate high levels of both zinc and citrate [Mawson

et al., 1952; Cooper et al., 1963, Costello et al., 1997, 1998, 2005; Singh et al.,

2006].

Figure 4: The intermediary metabolism of the prostate.

Image by permission from Costello LC, Franklin RB. Mol Cancer. 2006;5:17.

The reason for this, besides the protective effects, is not fully understood and

this is not associated with other cells in the human body. High quantities of both

zinc and citrate are secreted to the prostatic fluid (Figure 4) where they are

24

believed to function as antibacterial agents while the fluid is entering and

flushing the many ductal systems in the gland [Fair et al., 1976, 1981; Kavanagh,

1985,]. One of the functions of mitochondrial zinc is to inhibit mitochondrial

aconitase (m-aconitase) which in turn prevents the oxidation of citrate in the

Krebs cycle (Figure 4). Conversely, malignant epithelial cells are metabolically

transformed to citrate-oxidizing, energy producing cells that lose the ability to

accumulate zinc (Figure 4). This feature, unique for the prostate epithelial cells,

is exploited in MRS where the spectroscopic levels of citrate are detectable

(Figure 5), [Liney et al., 1996; Kurhanewicz et al., 2002; Swanson et al., 2003;

Gillies et al., 2005]. The reduction of intracellular zinc is characteristic of PCa

and is not associated with prostatities or BPH [Zaichick et al., 1996; Costello et al

2009]

Figure 5: The characteristic double doublet of citrate seen in a

HRMAS NMR spectrum obtained from a healthy prostate tissue.

The reduction of mitochondrial zinc is, however, not due to a reduction of circulatory zinc in the body. Instead, there seems to be a hindrance in the intracellular function that regulates the uptake of zinc which is in agreement with the down-regulation of the ZIP1 transporter in cancer cells [Franklin et al., 2005]. Malignant prostate cells, just as with other malignant cells, have a higher need for energy (ATP) to handle the rapid proliferation involved. Since Otto Warburg’s pioneering studies in the 1920’s, it has been well known that most tumors have an increased glucose metabolism which has been termed the

25

Warburg effect [Warburg, 1956; Kim et al., 2006; Hsu et al., 2008]. However, the healthy prostate epithelial cell has a very slow metabolism, i.e., it is an energy inefficient cell with high aerobic glycolysis and low aerobic oxidation and its overall reaction is:

Glucose + 2 Aspartate + 2 O2 2 Citrate + 2 CO2 + 14 ATP

The malignant prostate epithelial cells transforms into energy producing cells

with a fully operational Krebs cycle with complete oxidation of glucose and with

a low glycolysis:

Glucose + 6 O2 6 CO2 + 38 ATP

Thus, the malignant prostate epithelial cells resemble typical, energy efficient

non-dividing normal mammalian cells. As previously mentioned, citrate is

produced in the mitochondria and is either oxidized by the Krebs cycle or

exported to the cytosol for the conversion to acetyl-CoA. Cytoplasmic acetyl-CoA

is essential for the production of fatty acids and sterols. In healthy prostate cells

most of the citrate is exported to the cytosol and secreted into the prostatic fluid.

In malignant prostate cells, there is an increased export of citrate from the

mitochondria into the cytosol where it is cleaved by ATP-citrate lyase into the

two compounds acetyl-CoA and OAA. Thereafter, the prostate cancer cells most

likely uses the acetyl-CoA (after carboxylation) mainly derived from citrate, and

some from acetate, for the increased requirement of the de novo fatty acid

biosynthesis [Costello et al., 2006]. This complex pathway is suggested in Figure

6.

Figure 6: Prostate cancer cell metabolism.

Image by permission from Costello LC, Franklin RB. Mol Cancer. 2006;5:17.

26

Prostate tumors (selectively) over express fatty acid synthase (FAS) which is the

enzyme responsible for the de novo synthesis of fatty acids [Swinnen et al.,

2000; Baron et al., 2004]. However, an alternative pathway to provide OAA and

acetyl-CoA for the citrate initiated lipogenesis/cholesterogenesis pathway in PCa

is suggested; namely the glutamate pathway where the two major products are

citrate and alanine. Another option is the cytosolic direct synthesis of acetyl-Coa

[Costello et al., 2006]. As acetate takes part in the cytoplasmic lipogenesis, the

PET tracer 11C-Acetate is used and studied in PCa diagnosis; however, its role as

a tracer is not determined [Jadvar, 2011]. Choline compounds are increased in

prostate tumors and this feature is exploited by MRS [Swanson et al., 2003;

McLean et al., 2010] and 11C-Choline PET [Jadvar, 2011]. As citrate alone is not

sufficient for the discrimination between PCa and BPH its resonance is usually

combined with the choline compounds, which are increased in PCa vs. BPH, in

the (total-choline + creatine)/citrate (tCho+Cre/Cit) ratio [Gillies et al., 2005].

However, increases of choline compounds may not to be correlated with

proliferation [Gillies et al., 2005]. As a matter of fact, the increased de novo

lipogenesis in tumor cells may have a more complex role than earlier believed,

possible even as a cellular protector from endogenous and exogenous insults by

promoting membrane lipid saturation [Rysman et al., 2010].

Again, this points to the complexity of the human body and, consequently,

further studies to elucidate the prostate cancer pathways are warranted in order

to assess innovative bio-markers that can be used clinically with accurate results.

2.3 Man vs. Rat Prostate

For decades, researchers have tried to find experimental models that mimic the

human prostate and its diseases; however, the perfect model has yet to be

discovered [Kamb, 2005]. The underlying reason for these difficulties is, simply,

the fact that the prostate is a very complex organ with large species differences.

Factors such as heterogeneity, anatomical, zonal and cellular differences, and

unpredictable metabolic pathways etc. needs to be taken into considerations

when choosing an appropriate model. One important aspect that, unfortunately,

is often set aside is age [Badawai et al., 2004; Jara et al., 2004] since it is often

practically difficult to age match subjects in a research study. There is an

economic factor involved in this [Hahn, 2004], namely the cost and time it

requires to breed experimental animals to equal the age of an aging man which

is, after all, the main category that acquires problems with the prostate. Thus,

there is, unfortunately, not a universal tumor metabolism that is applicable to all

tumors, species and anatomical zones. Extrapolations of experimental PCa

results into clinical studies should therefore only be made under strict caution

[Costello et al., 2005].

27

Rodent models are widely used in PCa studies, with variable success rate. Each

model possesses strengths and weaknesses and is more appropriate for some

disorders than others [Lucia et al., 1998]. The rat prostate (which is fairly similar

to that in mice) is a large gland surrounding the urethra and is, contrary to the

human prostate, composed of distinctly different lobes, the ventral, dorsal,

lateral, and anterior lobes. The anterior lobe is also called the coagulating gland.

The ventral prostate lobe has been in the focus of several studies although it has

no equivalent in the human prostate [Roy-Burman et al., 2004], one reason is

that it is easily accessible and comprises roughly half of the entire prostatic

complex. The dorsal and the lateral lobes are often combined as the dorsolateral

lobe (DLP) as the share the same ductal system and it is also practically difficult

to separate the two. The LP is the lobe of the rat prostate that is the most similar

to the human peripheral zone. The lateral lobe actually comprises two different

ductal systems: the lateral type 1, and the lateral type 2 [Hayasi et al., 1991]. It is

interesting to note that the lateral lobe of the rat prostate demonstrates similar

cellular citrate and zinc accumulating characteristics as with the human

glandular peripheral zone. The rat ventral lobe shares the same characteristics

but not the dorsal lobe [Costello et al., 1998].

28

Chapter 3: Diagnostic Methods of the Prostate

The main clinical challenges in PCa treatment and management rely on accurate

diagnosis and initial staging. As the PSA-test has become readily available

[Wever et al., 2010] more clinically localized PCa are detected by consecutive

histopathological diagnosis and Gleason grading on biopsies. Regrettably, the

treatment for early stage disease is poorly defined [Berney et al., 2007]. Early

detection of PCa has caused a rapid increase in terms of incidence [Wever et al.,

2010]. Early detection and treatment of prostate cancer saves lives [Schröder et

al., 2009]. Unfortunately however, early detection of cancer inevitably also leads

to an increased risk of overtreatment due to the difficulties involved in

determining disease aggressiveness and the accomplishment of accurate staging

[Berney et al., 2007; Reese et al., 2011].

3.1 General Diagnostic Procedures

The general procedure for a clinical prostate examination includes: Digital

Rectal Examination (DRE) and the Prostate Specific Antigen (PSA) test. Trans

Rectal Ultra Sonography (TRUS) imaging if often used to guide biopsies [Cupp

et al., 1993]. This examination provides basic, initial information.

3.1.1 Digital Rectal Examination – DRE

This is a routine examination that allows the doctor’s finger to feel the back wall

of the prostate through the rectum. Any abnormalities detected will warrant for

additional exams and tests. The front and the middle of the prostate are not felt

during this exam. This is a subjective test and an urologist most likely has an

advantage over a general practitioner.

3.1.2 Prostate Specific Antigen – PSA

PSA is a kallekrein (an enzyme that cleaves peptide bonds in proteins), which is

produced by the prostate epithelial cells and it is a major component of the

semen, where its function is to break down coagulated semen post-ejaculation.

PSA is secreted via the prostatic ductal system and leaks into the bloodstream

where it can be detected by a simple blood test. An elevation of PSA may be a

sign of PCa since prostate cancer do not have a functioning ductal system that

drains the fluid into the urethra. The lack of free passage forces the PSA into the

29

prostate where it eventually leaks into the blood stream. However, increased

values can also be due to BPH, irritations of the prostate such as inflammation

and infection. The PSA-test is prostate specific and not cancer specific. Therefore

the optimal way to interpret a PSA-test and use it as an indicator for PCa may be

to include factors such as: age, race, free vs. bound PSA, PSA velocity (when

monitoring yearly), volume of prostate (especially the transition zone where

BPH arises), and decide individual cut-off levels based on its baseline value

[Oesterling, 1991]. However, a total PSA cutoff value of 3.0 ng/ml is often used

as an indication for biopsy although some countries prefer 2.5 ng/ml while

others prefer 4.0 ng/ml. The value of prostate cancer screening using the PSA-

test is unclear [Vickers et al., 2011], although there seems to be a reduction in

mortality albeit with a risk of overtreatment as high as 50% [Schröder et al.,

2009]. PSA expression has also been reported in other tissues and also in

females [Kraus et al., 2010]

3.1.3 Trans Rectal Ultra Sonography – TRUS

TRUS provides an ultrasound (harmless sound waves) generated image of the

prostate and surrounding tissue and allows the physician to access prostate

volume and examine the gland for abnormalities. Although this technique is able

to detect many of the palpable tumors, it is of less value for early stage

diagnostics where the tumors are non-palpable and, furthermore, some tumors

are iso-echogenic and are therefore not visualized by TRUS. TRUS is valuable to

guide biopsy sampling and to monitor treatments. Technical advancements in

this field include contrast-enhanced (microbubble contrast) agent techniques

[Aigner et al., 2010], and image fusion with TRUS and MRI [Turkbey et al.,

2011].

3.2 Imaging of the Prostate

Abnormal DRE and/or PSA-test calls for a more advanced evaluation using

image based diagnostic procedures including, anatomical, functional and

molecular imaging techniques. Imaging techniques, such as Computerized

Tomography (CT), Magnetic Resonance Imaging (MRI) (Figure 7), and Positron

Emission Tomography (PET), facilitates the evaluation of internal organs

without the need for surgery. Imaging of the prostate holds an important role in

the integrative approach for a patient with PCa where diagnosis, staging, and

monitoring in watchful waiting or the response of treatment are part of the

regimen. The field of imaging has expanded from the characterization of locally

advanced or metastatic disease to include intra and extra prostatic tumor

delineation, including morphology and zonal anatomy. Current clinical imaging

30

techniques cannot detect early disease and consequently generate limited

information to be used for accurate staging; therefore, new techniques that can

provide anatomical, functional and molecular imaging information are urgently

needed [Kelloff et al., 2009].

Figure 7: CT and MR examinations of prostate with benign hyperplasia and PCa

in the same patient. Black arrow (CT) and white arrow (MR) indicate regions

containing malignant tissue in hyperplastic prostate gland.

3.2.1 Computerized Tomography - CT

CT allows for a cross-sectional visualization of the internal organs, with or

without an intravenous contrast material (Figure 7, left). CT does not offer high

diagnostic accuracy in PCa and are, therefore, often excluded from the battery of

diagnostic tools available for the evaluation of the prostate. It is, however,

valuable for the planning of radiation therapy. CT is not indicated in early PCa

but can be used to diagnose distant metastases and sometimes lymph node

metastases.

3.2.2 Positron Emission Tomography - PET

PET is a nuclear medicine imaging technique which produces a 3D image of

functional processes in the body. PET is actually more correctly a 4D modality

where the 4th dimension is time. The system uses small amounts of radioactive

tracers that indirectly emit pairs of gamma rays which are detected on by the

PET scanner. The radiotracers are fused on with biologically active molecules,

such as glucose, and introduced into the body. The location and amount of

radiotracer absorbed is visualized on images offering details on both the

structure (from CT) and function (from PET) of organs and tissues. To get

31

optimal anatomical, functional and molecular information PET combined with

CT is used as; PET/CT [Bouchelouche et al., 2010].

18F-18-fluorodeoxyglucose is the most common PET tracer for tumors; however,

it is not useful in local PCa detection [Avril et al., 2010]. The roles of 11C/18F-

acetate or 11C/18F-choline as PCa PET tracers for primary diagnosis are not clear

[Jambor et al., 2010].

3.2.3 Magnetic Resonance Imaging - MRI

MRI, preferably with an endorectal coil, is used to improve prostate tumor

detection. MRI offers high-quality images of the internal organs (Figure 7, right).

The method is based on the magnetic behavior of hydrogen atoms as part of the

body’s inherent biomolecules and water in an outer magnetic field and their

manipulation by radiofrequency waves; an approach which generates images

with valuable information about the human anatomy and morphology.

Normally, a T2-weighted MRI obtained by with standard 1.5 Tesla scanners

equipped with endorectal coils, is used to detect and localize PCa. Tumors are

characterized by decreased signal intensity compared to the normal peripheral

zone. 3.0 Tesla scanners provides increased spatial resolution and thus

improved localization and staging. However, this MRI method does not provide

the sensitivity and specificity needed for accurate results of early stage PCa as

the cancer cell clusters often are minuscule. Additionally, there are problems

detecting tumors in zones other than the peripheral one [McMahon et al., 2009].

Multiparametric MRI combines T2-weighted imaging with one or more of

following approaches: Diffusion Weighted Imaging (DWI), Magnetic Resonance

Spectroscopy (MRS), and Dynamic Contrast Enhanced (DCE) functional

techniques. This is performed in one single study combining the different

imaging protocols [De Visschere et al., 2010; Weidner et al., 2011].

Contrast agents such as Ultra Small Supermagnetic Iron Oxide (USPIO) particles

add additional, valuable information to an MRI scan by increasing the signal of

PCa cells in lymph nodes [Thoeny et al., 2009]. However, lymph node staging is

still limited, also with MR. Currently MRI is of great value in difficult cases,

although it is not part of the clinical routine.

3.2.4 Magnetic Resonance Spectroscopic Imaging – MRSI

MRSI is a highly promising, non-invasive method, useful in the field of PCa

diagnosis. This cross-sectional technique offers an excellent overview of the

32

organ in all dimensions in-vivo, and possibilities for longitudinal assessments of

therapy, short and long-term changes related to disease and healing, volume and

density changes etc.

MRSI, preferably with an endorectal coil, provides visual anatomical and

morphological information simultaneous with spectral information about the

local metabolism and provides assessment of metabolic changes during a disease

process or treatment responses. This is of special interest since metabolic

perturbations often occur before the concomitant morphological changes. The

most commonly studied bio-markers for PCa found in the prostate metabolism

include citrate, choline compounds, and creatine. The ratio of choline

compounds (t-choline) and citrate are by far the most utilized biomarker for the

discrimination of malignant and non-malignant tissue, however, it may not be

the most optimal one [Swindle et al., 2003].

MRSI is not used routinely for prostate evaluations since it is more in the

experimental stage with room for improvements and furthermore, it is both

costly and time consuming. The present drawbacks of MRSI include difficulties

in obtaining optimal spectral resolution, poor spatial resolution, artifacts that

are, for example, caused by movements, in homogeneous external magnetic

field, metals, radiofrequent “noise” etc., but also the requirements for specialized

software and experienced personnel to acquire and interpret the data since this

technique is very much operator dependent and thus, as with most other

diagnostic tools, highly subjective [Kreis, 2004]. Furthermore, the prostate is, as

mentioned earlier, a heterogeneous organ and the tumor foci are often infiltrated

among normal tissue and, thus, will not appear clearly with MRSI. Moreover, a

more precise definition and understanding of functional anatomy of the prostate

is needed [McLaughlin et al, 2005]. However, MRSI is of immense use in

difficult cases. Technological advancements, such as functional, 2D and 3D

techniques and higher field strength, will improve tumor localization and render

MRSI a highly promising technique for the future [Thomas et al., 2008; Kelloff

et al., 2009; Larson et al., 2010].

3.2.5 Small Animal MRI/MRSI

Small animal MRI/MRSI scanners are rapidly emerging in modern research

facilities around the world. Animal models, particularly rodents, and the use of

MRI/MRSI have greatly advanced researchers’ understanding of various

pathologies and the effects of treatments etc. Since this technique is non-

invasive it has drastically reduced the number of animals needed since each

animal can be used more than once and thus followed longitudinally, i.e. each

animal can be used as its own control [Nastiuk et al., 2007]. However, given the

33

small size of rodents and the associated technical challenges, however, prostate

imaging studies in live animals only recently have become feasible, thanks to

technological improvements.

3.3 Prostate Biopsy

Taking and evaluating prostate biopsies are a team effort between the urologist

and the pathologist, and this is the most significant factor for the final diagnostic

verdict which strongly influences decisions regarding options for therapy. TRUS

guided biopsy sampling is most commonly used where the needle is inserted

through the thin rectal membrane into the prostate. Tissue samples are

extracted, with the biopsy needle, from different areas of the prostate. As tumors

often cannot be detected by imaging the prostates are systematically sampled

with 10-12 or more specimens along the sides of the peripheral zone and it is

recommended to sample a few laterally. Prostate tumors usually grow like thin

sheets laterally and consequently it is easy to misguide the biopsy needle past the

target. A larger prostate requires a more extensive sampling than a smaller size.

The biopsies in total however sample less than 1% of the prostate volume. This

means that there can be a significant amount of aggressive cancer present that

will be undetected by the biopsies. Again, imaging methods such as TRUS, CT or

MRI do not have the diagnostic accuracy to detect microscopic disease. Thus,

image guided biopsy sampling is not straight forward. PCa awareness, in the era

of PSA-testing, has increased the number of men diagnosed with low-grade/low-

risk cancers. This has led to treatment options such as focal therapy and active

surveillance to avoid unnecessary radical therapy with its many negative side-

effects. Since PCa often is multi-focal, both treatment options require improved

diagnostic methods in order to rule out aggressive tumor foci elsewhere in the

gland [Lazzeri et al., 2010; Ukimura et al., 2011]. Biopsy is an invasive

procedure with risks for complications such as: infection, bleeding, and in some

cases sepsis. Therefore, it is highly desirable to avoid unnecessary biopsies and

to discover bio-markers that possess the ability to detect microscopically small

amounts of aggressive cancer cells in the specimen and perhaps indicate that a

cancer could be present in the nearby tissue.

3.4 Histopathological Interpretation of Tissue Specimens

Prostate tissue is exceptionally heterogeneous by nature which complicates

classification and grading. Tumor grading and staging of prostate cancer is

fundamental in characterization and prognosis. The Gleason score system

provides information regarding the type of cancer cells while the TNM staging

system provides additional information concerning the extent of the cancer.

34

3.4.1 The Gleason Score – GS

The most widespread method of PCa tissue grading used today is the GS system

which is based on microscopic tumor patterns and is the single most important

prognostic factor [Rajinikanth et al., 2008]. The tissue specimen is examined

under light microscope where cells or groups of cells that are markedly different

from healthy prostate tissue are categorized. The greater the disparity between

the healthy cells and the malignant cells the higher the grade and the more likely

the tumor is aggressive and will metastasize. The GS diagram exhibit five

distinctly different tissue patterns that are technically referred to as tumor

“grades” (Figure 8). The microscopic determination of the loss of normal

glandular structure caused by the cancer is represented by a grade number

ranging from 1 to 5, with 5 being the worst grade possible (Figure 8).

Figure 8: The Gleason score system.

The final Gleason score is a sum, ranging from 2-10, of the primary grade which

represents the most common type of tumor cells (>50 %), and a secondary grade

which represents the second most common type of tumor cells (<50% but >5%),

that are found in the specimen [Gleason et al., 1974]. However, this

determination is highly subjective and thus requires a skilled pathologist. The

GS from a biopsy frequently differ from the GS of radical prostatectomy (RP). It

has been reported that the difference may be as much as 60-70 % and that under

grading is often the case in biopsy although the precision of GS has improved

over the last years [Rajinikanth et al., 2008]. A modified Gleason score is also

35

suggested, a score based on the most common and the worst grades present

[Epstein et al., 2010].

3.4.2 The TNM Table

TNM stands for Tumor, Nodes (lymph nodes), and Metastasis. This widely used

staging system gives information about the extent of the primary tumor (T

stage), the absence or presence of spread to nearby lymph nodes (N stage) and

the absence or presence of distant spread, or metastasis (M stage) (Figure 9).

Figure 9: The TNM Table.

TNM staging, just as with the Gleason score system, requires skilled interpreters.

Under and over staging is common, i.e., the clinical staging (pre-surgery) results

might often not correspond with the pathological staging (post-surgery or ex-

vivo evaluation). This is, however, not surprising since different information for

the staging are used. Correct staging is critical because treatment is directly*

related to disease stage [Hoedemaeker et al., 2000].

3.4.3 Immunohistochemistry

Interpretation of prostate tissue specimens is a challenge. The reason for this is

the heterogeneous nature of prostate tissues in general but also the new

36

demands that comes with earlier detection in the era of PSA-tests/screening

[Paner et al., 2008; Chun et al., 2010]. Differentiation of minimal cancer (<5 %

of biopsy specimen), atypical small acinar proliferation (ASAP), and high-grade

prostatic intraepithelial neoplasia (HGPIN) have increased the use of

immunohistochemical methods detection; α-methyl-acyl-coenzyme A racemase

(AMACR), nuclear p63, and the basal cell cytokeratin (34βE12). With the use of

such markers cancer can generally be distinguished from other conditions, but

as there are several mimickers of prostatic adenocarcinoma these

immunohistochemical markers are not universally applicable [Epstein, 2004].

37

Part II

38

39

Chapter 4: Nuclear Magnetic Resonance - NMR

4.1 General Introduction

In the recent decades, NMR spectroscopy [Pople et al., 1959; Ernst et al., 1987]

emerged as a powerful method to provide structural and dynamical information

of molecules at atomic level as well as unique diagnostic information in

medicine. Therefore, not surprisingly, a range of Nobel Prizes was awarded in

connection with NMR; Otto Stern (USA, 1943), Isidor I. Rabi (USA, 1944), Felix

Bloch, and Edward M. Purcell (USA, 1952), Richard R. Ernst (Switzerland, 1991),

Kurt Wüthrich (Switzerland, 2002), [The Nobelprize., 2010]. Chemist Paul C.

Lauterbur, USA and physicist Peter Mansfield, United Kingdom shared the

Nobel Prize in Medicine 2003 for the development of magnetic resonance

imaging (MRI). MRI, which has become an invaluable diagnostic method in

medicine, is directly derived from NMR, a fact that is often neglected in the

medicinal community [Fry et al., 2004].

The technique relies on the property of various nuclei, to possess a nuclear spin,

whose size is defined by the quantum number I. The nuclear spin is connected

with specific nuclear magnetic properties whose discovery by Rabi was awarded

the Nobel Prize in 1944. These magnetic properties, mainly the magnetic nuclear

moment, enabled researchers to study these nuclei in magnetic fields for

obtaining unique structural, chemical, physical and electronic information about

molecules since 1958.

4.2 The NMR Signal

The most common nuclei used in NMR in life science are 1H, 13C, 19F and 31P. In

general, these NMR active nuclei possess a property called spin with a value of

1/2. Insertion of a sample into a permanent outer magnetic field (Bo) causes the

nuclear magnetic moment of these nuclei to adopt different positions with

respect to this magnetic field, accompanied by a splitting up of their energy into

discrete energy levels. A nucleus with a spin of ½ adopts two possible magnetic

quantum states; m=1/2 which is the lower energy state also called α state or E1,

and m=-1/2 which is the higher energy state also called the β state or E2 (Figure

10).

40

Figure 10: Energy levels

This process is accompanied by a precession of the magnetic moments around

the Bo like spinning tops. The frequency of this precession is called Larmor

frequency and corresponds to:

E= νh

v = Larmor frequency

h = Planck’s constant: 6.63x10-34 J s

The energy difference (between two adjacent energy levels) of the nuclear spins

in the outer magnetic field and it is dependent on the gyromagnetic ratio γ of the

nucleus, and the outer magnetic field strength, B0.

ν/2=-γBo

γ = gyromagnetic ratio

B0 = magnetic field strength

However, due to the energy difference between both states (in the case of I=1/2),

there will be a small excess of spins aligned in the direction of B0, at the lower

energy state E1. This population difference will result in a macroscopic

magnetization of the sample parallel to the outer magnetic field, represented as

41

equilibrium magnetization vector, M0. This resulting vector Mo equals MZ since it

is aligned along the Z-axes. MZ is referred to as the longitudinal magnetization

(Figure 11).

Figure 11: NMR coordinate system.

The size of MZ is dependent on the Boltzmann distribution, as it is determined

by the following equation;

NE2/NE1 = e-∆E/kT

NE2 = higher energy state

NE1 = lower energy state

∆E = energy difference between the spin states

k = Boltzmann's constant: 1.3805 x 10-23 J/Kelvin

T = temperature in Kelvin.

4.3 The NMR Experiment

By applying in an NMR experiment radio frequency pulses at a frequency

which corresponds to the energy difference E between both states, it is possible

to flip the nuclear spins from their respective to the states via absorption of

this energy. An absorption of energy occurs when the energy E of the radio

frequent pulse meets the resonance conditions, namely the Larmor frequency

corresponding to the energy difference ∆E between the two spin states E2 and E1

which is described in the formula;

E =∆E=ħγB0

42

The radio frequent (RF) pulse applied along the X-axis creates a secondary

magnetic field that disturbs the alignment of the magnetic moments and causes

them to flip away from the equilibrium state. The required frequency is

dependent upon the static magnetic field, B0, and the nuclei observed. A

perpendicular flip of the initial magnetization Mz from the Z-axis to the XY-

plane where the transverse magnetization MY can be detected, requires a 900

pulse along the X-axes. This net magnetization may be described as an

oscillating magnetic field in excess that induces an alternating voltage in the

detection coil. This is the NMR signal also called free induction decay (FID). In

general, this signal diminishes over time due to the decay of the transverse

magnetization MXY in the XY-plane, characterized by the spin-spin relaxation

time T2 according to:

MXY =MXYoe-t/T2

In addition, there is a process, called spin lattice relaxation, which describes the

return of the longitudinal MZ magnetization towards its equilibrium (as prior to

the application of the NMR pulse). This process is characterized by a sample

specific spin lattice relaxation time, T1, and is described in the equation:

Mz = Mo

(1 - e-t/T1)

In an NMR experiment, the FID is time-dependent and can be converted into

the frequency domain by Fourier transformation [Jacobsen, 2007]. This will

provide the typical NMR spectrum, where frequency differences (magnetic field

differences) are plotted as a function versus intensity on a graph (spectrum).

This allows the visual observation of specific quantum mechanical magnetic

properties of the selected atomic nuclei. The most common type of nuclei

frequently analyzed by NMR is hydrogen (1H) which is, besides being the

lightest, by far the most abundant atom in the universe. NMR is nowadays

frequently utilized in the medical fields in the form of MRSI.

4.4 Chemical Shift

The chemical shift is a function of the nucleus and its local chemical

environment, thereby giving detailed information about the part of the molecule

where the nucleus is residing. By understanding different chemical

environments it is possible to obtain structural information about the molecule

in a sample [Chamberlain, 1959]. This is because the electronic environment,

that shields the nucleus from the magnetic field, is distinctive of the particular

nucleus and, thus, provides the exact resonance frequency of the nucleus. The

difference between the applied magnetic field and the field at the nucleus is

43

termed the nuclear shielding. The differences in the resonance frequency are

referred to as the chemical shift and it provides the information needed to

identify the compounds visible by NMR spectroscopy. The frequency shifts are

extremely small in comparison to the fundamental NMR frequency. In order to

detect the minuscule frequency differences the applied magnetic field must be

kept constant throughout the whole sample volume. This is done by using shim

coils. Since B0 differs to some extent between spectrometers the frequency

measurement in Hertz (Hz) was converted to a standardized scale that measures

in parts per million (ppm). This was done in a way that the NMR chemical shift

can be universally compared in ppm regardless of magnetic field strength. For 1H-NMR the chemical shift range is just over 10 ppm. The chemical shift is

calibrated using an internal standard as a reference; such as the commonly used

(tetramethylsilane) TMS [Van Dyke Tiers, 1958]. The chemical shift is obtained

by using the following calculations:

δ = (ν - ν0)/ν0

δ = chemical shift

ν = resonant frequency

ν0 = standard frequency

Frequently, a secondary reference of known chemical shift can be used instead of

the common standards such as TMS. A chemical reference standard of known

concentration is added to the sample and the concentration of the solute can

then be determined by comparing the integrals of the signals. The secondary

references are often the residual solvent signals or the unsuppressed H2O signal

[Cheng et al., 1998].

The secondary references are often used in quantitative NMR. However, there

are challenges involved in this, especially when quantitating HRMAS spectral

data. The standard to be used must be compatible with the sample, be

chemically inert, have a longitudinal relaxation time (T1) that, preferably, is

shorter than that of the sample, and furthermore, not generate resonances that

cause overlaps [Akoka et al., 1999]. A semi-quantitative approach is to compare

spectral ratios within each sample which does not require the need for suitable

internal standards. However, this method does not allow for separate resonances

to be studied independent from each other [Swanson et al., 2006; Sitter et al.,

2009]. Technological advancements have contributed to the development of new

types of internal standards such as the Electronic REference To access In vivo

Concentrations (ERECTIC). As the ERETIC signal is generated by an electronic

device it is therefore not dependent on the sample. This electronic reference

signal aids in the determination of absolute concentrations [Akoka et al., 1999;

44

Albers et al., 2009]. Modification and improvements of the ERETIC reference

include, as an example, the Amplitude-corrected Referencing Through Signal

Injection or ARTSI [Mehr et al., 2008].

4.5 Spin-Spin Coupling

Spin-spin coupling is an essential attribute of NMR as it provides detailed

structural and conformational information of the molecule. This spin-coupling

describes the phenomena that one nucleus feels the presence and spin state of

another nucleus nearby via connecting bonds (transferred via the magnetic

properties of the bond electrons). Spin-spin coupling can cause splitting of the

signal for each type of nucleus into two or more lines depending on the number

of chemically bonded neighboring nuclei. Thus, coupling to n equivalent (spin

½) nuclei will split the signal into an n+1 multiplet with intensity ratios

according to Pascal's triangle, as can be seen in Figure 12. Pascal's triangle [Lash

et al., 1987], named after the French mathematician Blaise Pascal, entails a set of

numbers aligned in the shape of a pyramid. These numbers represent the

binomial coefficients. Nuclei that are chemically equivalent (isochronous) do not

exhibit spin-splitting between each other and thus have no effect on the NMR

spectra.

Figure 12: Splitting patterns.

The size of the splitting; i.e., the coupling constant which is annotated J, is

independent of the magnetic field and is therefore measured as an absolute

frequency, usually in Hertz.

45

4.6 High-resolution Magic Angle Spinning (HRMAS) NMR

The line width of an NMR resonance depends on orientation-dependent factors,

such as chemical shift, chemical shielding anisotropy, and magnetic

susceptibility of the sample and susceptibility differences within the sample. The

rapid isotropic motion of the molecules in liquid state NMR averages the

anisotropic interactions and reduces broadenings due to dipolar couplings

resulting in narrow line widths. Furthermore, by choosing the correct sample

geometry, the broadenings, due to magnetic susceptibilities, are minimized

[Brown et al., 2001].

Solid or semi-solid samples are more difficult to analyze by NMR due to the lack

of molecular mobility and therefore the anisotropic nature of the corresponding

NMR spectra, which are usually wideline spectra without any significant

resolution. Due to this anisotropic nature, the NMR signal depends on the

orientation of the individual molecules with respect to the outer magnetic field;

an orientation dependence which for most NMR relevant interactions can be

described (in the case of spin I=1/2 systems) by the 2nd Legendre Polynom

P2(cosθ), with P2(cosθ)=(3cos2θ-1)/2. This term becomes zero when:

(3cos2θ-1)/2 = 0

cos2θ = 1/3

θ = 54.74º

This 54.74º angle is called the “Magic Angle” [Andrew et al., 1959]. By spinning

the sample rapidly at the magic angle, 54.74º with respect to B0 the orientation

dependent P2 term will be time averaged to zero. This will reduce anisotropic

interactions to zero, or in the case of chemical shift anisotropy to their isotropic

value and will break up the wideline NMR spectra into a set of individual narrow

lines which can be assigned as in the case of liquid NMR spectroscopy (Figure

13), [Maas et al., 1996].

46

Figure 13: HRMAS NMR spectra obtained from non-malignant

prostate tissue (left) and malignant prostate tissue (right).

High Resolution Magic Angle Spinning (HRMAS) NMR thus enables the analysis

of intact tissue samples, where the MAS approach even minimizes the

susceptibility broadening of the NMR lines due to the heterogeneous nature of

the samples (Figure 13). The non-destructive nature of HRMAS NMR allows for

consecutive histological examinations under light microscope [Moestue et al.,

2011].

4.7 2-Dimensional NMR

Contrary to 1D NMR, 2D NMR will plot intensity as a function of two

frequencies; F1 and F2 [Martin et al., 1988]. This type of spectrum is often made

as a topographical map where each peak is plotted as a contour with the co-

ordinates corresponding to frequencies at F1 and F2. F2 is the equivalent to the

x-axes in a 1D spectrum. 2D NMR is necessary when trying to elucidate and

identify resonances, i.e. confirming the compounds found in the sample.

4.7.1 TOtal Correlation SpectroscopY - TOCSY

2D TOCSY gives information about spin systems within a molecule. This is

especially useful when assigning compounds with similar chemical shifts and

multiplet overlap. TOCSY yields through bond correlation. The TOCSY

experiment correlates all protons of a spin system via spin-spin couplings

[Braunschweiler et al., 1983]. A typical TOCSY spectrum has a mixing period

during which only scalar coupling is acting while chemical shift is suppressed.

The basic process is transfer of magnetization between pairs of coupled spins.

47

4.7.2 Hetero-nuclear Single Quantum Coherence –HSQC

HSQC has been used since its invention in 1980 [Bodenhausen et al., 1980]. This

type of 2D spectrum displays the chemical shift of 1H on the x-axes and the

indirect chemical shift of a hetero-nucleus such as 13C or 15N on the y-axes. The

spectrum show the cross peaks for each unique proton attached to the hetero-

nucleus. This provides the chemical shift of both the proton and the hetero-

nucleus coupled to the same proton, and thus, important qualitative information

for the identification of a compound [Soubias et al., 2003].

48

Chapter 5: Metabolomics Using HRMAS NMR

5.1 Concept of Metabolomics

Metabolomics is part of the omics sciences, such as genomics, proteomics,

transcriptomics, in systems biology. Although the term “metabolomics” is rather

novel this type of approach to elucidate metabolic perturbation in biological

systems due to disease is not new. The notion that changes in tissues and

biological fluids, such as urine, are indicative of disease were explored in the

ancient Greece and China [van der Greef et al., 2005; Nicholson et al., 2008],

and Dalgliesh et al investigated metabolic patterns in urine and tissue by using

GC-MS in 1966 [Nordström et al., 2010].

There has been some confusion regarding the definitions of the terms

“metabolomics” and “metabonomics”. Nicholson et al introduced the expression

“metabonomics” in 1999 [Nicholson et al., 1999], while Fiehn et al launched the

term “metabolomics” just a year later [Fiehn et al., 2000]. Although

metabolomics is the more common term in use today, there are differences

between the two slightly similar terms: “metabolomics” is usually referred to as

the more comprehensive method covering a global range of metabolites typically

analyzed by mass spectrometry (MS). “Metabonomics” is often referred to as

being based on the use of NMR and thus not as comprehensive as MS. This is

due to the fact that MS detects metabolites that are present at concentrations

orders of magnitude lower than what can be detected by NMR [German et al.,

2006]. The term metabolomics, though NMR based, will be used in this thesis.

The metabolome is defined as the complete set of small molecule metabolites,

endogenous as well as exogenous, found in a biological sample, i.e., specific cell,

organ or organism [Wishart et al., 2007]. The exploration of the metabolome

aids in the search for unique fingerprints responsible for different phenotypes

[Koulman et al., 2008]. The distinction between phenotype and genotype was

proposed in 1911 by Wilhelm Johannsen with the aim of clarifying the difference

between an organism's heredity and what that heredity produces [Johannsen,

1911]. Phenotypes are the outcome of genetic expressions as well as the influence

of environmental factors, and the interactions between the two; and this may be

detected in the metabolome.

Both extrinsic and intrinsic factors have input on the metabolome. Extrinsic

factors may include lifestyle choices, nutrients and non-nutrients, physical

activity, stress, drugs, and colonic micro flora (micro biome). Intrinsic factors

may comprise genotype, age, body composition, tissue turnover, reproductive

49

status, and diurnal cycle [Koulman et al., 2008, Goodacre, 2007]. The

metabolites present in a metabolome include metabolic intermediates, amino

acids, fatty acids, etc. Metabolomics aids in the unraveling and the discovery of

organ-specific endogenous metabolites that has the potential to be reliable

indicators of organ function and viability as well as valuable markers for

treatment response [Conti et al., 2006]. The metabolites may be viewed

individually but in a more holistic context or as a group in, what is termed,

metabolic profiling [Beckonert et al., 2007]. Metabolomics is an excellent tool in

translational research as it provides a reflection of, for example, the unique

metabolism of experimental models where the results ultimately are to be

applicable clinically. This is of special interest in the emerging field of molecular

imaging where diagnostic technologies such as MRSI or PET are utilized

[Spratlin et al., 2009]. The comprehensive investigation of the metabolome will

provide a more holistic picture of the biochemical perturbations which are

expressed subsequent gene expressions and protein activities [Bathen et al.,

2010]. The results may therefore be applied clinically in-vivo to characterize an

organ by utilizing bio-markers that have the capacity to discriminate between

disease and health. A powerful application of metabolomics is in the

differentiation of malignancy vs. non-malignancy. However, despite this, the

approach is not commonly used in the clinics for oncology [Spratling et al.,

2009].

5.2 Metabolomics in the Characterization of the Prostate

Successful therapy for PCa depends on early diagnosis and accurate evaluation

of progression and outcome. Bio-markers aimed for the characterization of the

prostate are especially difficult due to the heterogeneity of the prostate,

especially when malignant, and the fact that it is difficult to differentiate

between the aggressive type vs. the indolent type of PCa [Madu et al., 2010], and

in addition, there is a variability between patients which presents a challenge

using the diagnostic tools available [Abate-Shen et al., 2009]. Thus, clinical

indicators in the form of robust bio-markers are important in modern medicine.

The human prostate metabolome, visible by HRMAS NMR spectroscopy,

consists of small, freely rotating organic molecules [Swanson et al., 2003, 2006;

Tessem et al., 2008]. The Krebs cycle intermediate citrate is one of the most

important metabolites found in the prostate metabolism and has been

researched thoroughly by different means. By assessing the biochemistry of, for

example, the Krebs cycle intermediates using the metabolomic approach where

extrinsic and intrinsic factors are included, a more comprehensive observation

will be provided [Griffin et al., 2004].

50

5.3 Metabolomics, Nutrition and Prostate Cancer

The prevention of the development of disease and prostate cancer is even more

important than the search for bio-markers [Silberstein et al., 2010]. There is a

growing body of evidence suggesting a relationship between nutrition, health,

and PCa risk, and where there seems to be a particularly strong link between

obesity and PCa [Coffey, 2001; Ornish et al., 2009; Gonzales et al., 2010; Kahn

et al., 2010; Thun et al., 2010). Furthermore, the roles of certain food

components as protectors against disease are documented [Davis et al., 2004;

Afman et al., 2006]. By defining an individual’s healthy metabolome at baseline

under specific nutritional conditions and lifestyle factors may ultimately lead to

personalized medicine. The baseline metabolome would aid in the interpretation

of disease and a comprehension of optimal nutritional intake and the

maintaining of homeostasis on an individual basis [van der Greef 2004; German

et al., 2005; Gibney et al., 2005; Trujillo et al., 2006; Goodacre, 2007; Holmes et

al 2008]. Essential nutrients are especially interesting since they must be

obtained by dietary means. These nutrients: vitamins, dietary minerals, essential

fatty acids, and essential amino acids, are required for normal body functioning.

The essential polyunsaturated fatty acids (PUFAs) omega-6 and omega-3 has

received much attention. Omega-3 is believed to possess protective properties

while omega-6 has been known as an instigator for PCa. These PUFAs, along

with numerous other nutrients, are detectable by HRMAS NMR.

51

Part III

52

53

Chapter 6: Materials and Methods

“Mti hawendi ila kwa nyenzo”*

Swahili kanga writing

6.1 Experimental Setup

6.1.1 Procedure of HRMAS NMR Spectroscopy

All spectroscopic data were acquired at 11.7 T on a Bruker AMX2 NMR

spectrometer (Bruker Biospin GMBH, Karlsruhe, Germany), operating at a 1H

NMR frequency of 500.13 MHz and equipped with a 4 mm dual band, 1H/13C,

HRMAS probe head from Bruker. NMR spectra were acquired at room

temperature (20◦C). The magic angle spinning rate was set to 5 kHz for all

experiments except for the 2D rotor-synchronized TOCSY experiments were a

41.8 ms mixing time at 2.25 kHz sampling was used. The 2D 1H-13C GE-HSQC

(phase-sensitive) was performed at 5 kHz. The MAS spinning rate suppresses

chemical shift anisotropy and the line broadening effects due to dipolar

couplings.

Prostate tissues were cut with a surgical steel blade into smaller specimens

weighing ~15+5 mg so as to fit the MAS rotor with insert. 4 mm cylindrical ZrO2

rotors with spherical inserts were utilized. Kel-F caps were used to secure the

rotors. D2O were used for 2H field locking and ten micro liters of D2O were

added to each specimen and allowed to soak in for a few minutes. Prostate tissue

specimens were packed in rotors and where either immediately placed in the

HRMAS probe to be analyzed or temporarily stored at -20◦ C. In order to avoid

possible errors all samples were treated as identically as possible.

6.1.2 1D 1H HRMAS NMR

For the 1D HRMAS NMR spectroscopy a rotor-synchronized Carr-Purcell-

Meibom-Gill (CPMG) [Meiboom et al., 1958], pulse sequence (90o-(τ-180o- τ)n-

acquisition) with a 5 kHz spinning rate was used. A T2 filter of 20 ms duration

*You need to have proper tools to carry out any task.

54

was used to suppress broad signal contributions from macromolecular

components. The T2=20 ms was chosen based on T2 time studies performed in-

house. The excitation band width was set to 10309 Hz and each spectrum was

recorded with 256 transients, using 16K complex time domain data points and

with a repetition time of 5.0 seconds. To suppress the resonances arising from

H2O a presaturation pulse during the relaxation delay prior to the CPMG pulse

sequence. All 1D spectra were generated, from start to finish, at a maximum time

of 30 minutes.

Spectral 1D data were analyzed off-line using Mestre-C 4.9.9.6 and/or MNova

6.0.4. (Mestrelab Research, A Coruña, Spain). The free induction decays (FID’s)

of the CPMG edited 1H HRMAS spectra were subjected to a 1.5 Hz exponential

multiplication, zero-filled to 60k complex points preceding Fourier

transformation. The resulting spectra were manually phased and baseline

corrected using a multi-point polynomial function. The spectra were referenced

to the residual water peak by setting the chemical shift of the lactate doublet

equal to exactly 1.33 ppm. A selection of metabolites, that were relatively clear of

spectral overlaps, was subjected to a line-fitting procedure/deconvolution so as

to obtain metabolite peak areas.

6.2 Prostate Tissue Samples

Excised tissues are no longer under enzymatic control outside of its natural

environment, which is the body. Therefore, all biological specimens: tissues, and

fluids, require strict handling to avoid unnecessary oxidation and degradation

[Sitter et al., 2009]. It is not practically feasible to analyze fresh tissues,

therefore all excised tissue samples were immediately snap-frozen in N2(l) and

thereafter stored in a freezer with a temperature of -70C◦. Thawing does have an

effect on the specimens [Middleton et al., 1998; Wu et al., 2003] which needs to

be taken into consideration when interpreting the results.

6.2.1 Patients

Informed consent was obtained from the patients and the studies were approved

by the Research Ethics Committee of Umeå University Hospital, Umeå, Sweden.

Forty patients with PCa were included in this study. All of them underwent

prostatectomy and agreed to donate the surgical specimen by a form of written

consent. Study participants, aged between 41 and 71 years, were drawn from the

northern part of Sweden. The tumor stage was in the range of 1 to 3, and the

Gleason scores were GS 6 (3+3) and GS 7 (3+4). Serum PSA value was 12.41 +

55

13.02 (range 3.42 – 50.03). None of the patients had undergone any form of

treatment before surgery.

Specimens were chilled immediately after surgery. Prostate surfaces were inked

for anatomical identification purposes and thereafter the gland was cut into

horizontal 1 cm thick sections. Between 2 and 20 tissue punches, approximately

0.5 cm in diameter, were obtained from the peripheral zone of each section. All

specimens were snap-frozen in N2(l) within 30 minutes after surgery and

thereafter stored in -70 C freezers. In order to identify the punches sampled

each prostate slice were fixed using formalin and embedded in paraffin before

being cut into 5-µm thick sections. Each section was stained with haematoxylin

and eosin, thus enabling tumor foci identification using microscopy. The foci

were thoroughly marked and the frozen punches were preliminary classified as

malignant or non-malignant by studying the tissue surrounding the holes from

where the punches originated from. The frozen tissue punches were randomly

cut into smaller specimen and analyzed by HRMAS NMR.

6.2.2 Rat

The study protocol for the animal experiments was approved by the Animal

Research Ethics Committee of Umeå, Umeå, Sweden.

Nine adult male Sprague-Dawley rats (B&K, Stockholm, Sweden) were included

in this study. The Sprague Dawley strain is an outbred multipurpose model. The

rats were anaesthetized according to the previously described method [Lissbrant

et al., 2004], and their ventral prostate (VP) and dorso-lateral (DLP) lobes were

removed and immediately frozen in liquid nitrogen.

Six adult male Copenhagen x Fisher F1 rats (ALAB, Uppsala, Sweden) were

transplanted subcutaneously with pieces of highly differentiated Dunning R3327

PAP rat prostatic adenocarcinoma in their right flank as earlier described

[Häggström et al., 1998]. Upon growth to 1-2 cm in diameter each tumor was

removed and immediately frozen in liquid nitrogen.

6.3 Interpretation of Data

The interpretation of NMR data requires, besides obligatory spectroscopic skills,

a thorough comprehension of the particular species analyzed, and general

biochemical knowledge. Assignments of the spectral resonances in this study

were made by consulting published data, and by 2D TOCSY, and 2D GE-HSQC

experiments. The 2D NMR experiments are valuable when assigning overlapping

56

resonances which are common in HRMAS spectra. It is not uncomplicated to

quantify this type of spectral data due to the overlapping resonances, as already

mentioned, furthermore, there are challenges involved in weighing the samples

correctly due to the natural variation in tissue composition of intact biological

specimens. In this study the basic deconvolution method to quantify spectral

data was used: linefitting of resonances and thereafter using the spectral ratios

within each sample [Inouye et al., 2010].

6.3.1 Pathologic Analysis

Formalin fixed, paraffin embedded specimens were cut in 5 micron thick

sections and stained with hematoxylin-eosin (H&E). Frozen samples were cut

unfixed and then stained either with H&E or immunostained for high molecular

weight cytokeratin (cytokeratin-hmw, DAKO, Stockholm, Sweden) detecting

basal epithelial, or the cell proliferation marker Ki67 (DAKO).

The slides were reviewed by a highly experienced uro-pathologist who

determined the percentage of non-malignant glandular, non-malignant stroma,

and the percentage and grade of prostate cancer. Cancer-containing prostate

tissues (glands lacking cytokeratin-hmw positive basal epithelial cells) were

compared to non-malignant tissues (glands with an intact basal epithelial cell

layer) regardless of relative glandular and stromal composition. The fraction of

malignant vs. non-malignant prostate tissue was established using a light

microscope with a square-lattice mounted in the eye-piece to count the number

of grid-intersections falling on each tissue compartment. The distance between

the remaining holes from the punches and the closest tumour was measured on

the whole mount sections. As “completely healthy,” i.e. benign, prostate tissues

were lacking non-malignant were used as such.

6.3.2 Statistical Analysis

Upon the completion of the NMR measurements and post NMR processing, the

results, in the form of integral areas, were brought into Excel (Microsoft Office

Excel 2003) for the generation of a matrix. The results from the

histopathological analysis and pathological staging were added.

The data matrix was calculated with: SPSS (v. 16.0, SPSS, Chicago, IL, USA)

where basic descriptive statistics such as means and SD was generated, and

STATA (v. 10, StataCorp LP, College Station, TX, USA), where the data was

standard error-adjusted in order to take clustering (dependencies between

samples obtained from the same patient) into considerations. The following

57

statistical methods was used; Multivariate Linear Regression, Binary Logistic

Regression, Receiver Operating Characteristic (ROC) and Area Under the ROC

Curve (AUC). Odds Ratios (ORs) and 95% confidence intervals (CIs) were

calculated. P-values <0.05 were considered to be statistically significant.

58

Chapter 7: Results and Comments

Various types of intact prostate tissues, obtained from both human and rats

have been studied using HRMAS NMR spectroscopy. The results from these

studies are reported below.

Paper I

The aim of this study was, firstly, to verify metabolites known to be part of

the local prostate metabolism according to previous publications. Secondly,

our extended aim was to assess and elucidate additional compounds,

associated with the prostate but not previously published.

The result from the 1D 1H HRMAS NMR analysis confirmed the compounds

related to the prostate metabolism but also revealed a spectral resonance at

2.8 ppm that had previously not been documented in spectroscopic studies

of the prostate. The 2.8 ppm resonance in the 1D spectrum, which was found

in prostate cancer specimens only, was indicative of a polyunsaturated fatty

acid (PUFA). There are two PUFAs that are considered essential, namely the

omega-3: α-linolenic acid (18:3 ω3) and the omega-6: linoleic acid (18:2 ω6),

which are structurally similar as both are 18 carbon fatty acids; however,

omega-3 has three double bonds while omega-6 has two (Figure 14). The fact

that omega-3 has its first double bond at the third carbon counting from the

methyl end while omega-6 first double bond is connected to the sixth carbon

from the methyl end, provides the information needed to distinguish these

two compounds from each other (Figure 14) by using 2D 1H,13C-HSQC NMR.

The PUFA in our study was identified as an omega-6 specie. This finding was

not surprising as linoleic-acid omega-6 is by far the most common PUFA in

western diet [Simopoulos, 2010].

Figure 14: The two essential, poly-unsaturated fatty acids;

alpha-linolenic acid/omega-3, and linoleic acid/omega-6.

59

Linoleic acid is the omega-6 compound which is considered to be the

“parent” of several n-6 compounds varying in length from 14 to 40 carbons

and thus would be the essential fatty acid (EFA) [Cunnane, 2003]. Several

studies have indicated an excess of omega-6 and its derivatives as instigators

of inflammation and also PCa [Kelavkar et al., 2006, 2009; Ukoli et al.,

2010]. One of the more known n-6 EFAs arachidonic acid (AA); 20:4 ω6, is a

precursor in the production of eicosanoids which are compounds involved in

the regulation of inflammation and immunity [Funk, 2001]. The n-3

eicosapentaenoic acid (EPA) competes with AA for prostaglandin and

leukotriene synthesis at the cyclooxygenase (COX) and lipoxygenase (LOX)

level [McEntee et al., 2010]. However, the term EFA is ambiguous and the

metabolism of essential and non-essential fatty acids is very complex and

more interconnected than previously recognized. The requirements and

health effects of EFAs are individual and dependent on variables such as: age

and tissue type, but also on genetic differences in metabolism [Simopoulos,

2010]. α-linolenic acid and linoleic acid cannot be synthesized endogenously

“de novo”, but it seems that this is possible with the 16 carbon precursors;

16:3 ω3, and 16:3 ω6 [Cunnane, 2003]. Both ω6 and ω3 are readily β-

oxidized and recycled into cholesterol and palmitic acid, which is the most

common saturated fatty acid in mammals [Beare-Rogers et al., 2001], and,

less easily replaced in tissue lipids than the non-essential common fatty acids

such as palmitic and stearic acid [Cunnane et al., 2003].

HRMAS NMR detects freely rotating compounds in the semi-solid and intact

tissue specimens investigated. Furthermore, a filter to suppress signals from

macro-molecules was used therefore limiting the detection of compounds

tightly bound to membranes. As arachidonic acid is the most biologically

important PUFA of cell membranes and, when released by phospholipase,

directly affect cell signaling pathways and serve as substrate in the COX and

LOX pathways, a hypothesis is therefore that this n-6 PUFA was released

from membrane phospholipids due to inflammatory activation. Other

options include: the conversion of PUFAs for the, apparently, protective

properties of membrane lipid saturation [Rysman et al., 2010], or simply

from a PUFA buildup needed for the generation of unsaturated fatty acids or

cholesterol to be used for the increased synthesis of cellular membranes due

to increased proliferation.

Paper II

This study was designed to be a comparative investigation where both

human and rat prostate tissues were scrutinized. The human specimens

originated from the peripheral zone only, and the rat specimens were

60

obtained from three different lobes, whereas two of them were combined

into one entity. In addition, a subcutaneous experimental rat PCa tumor was

also included in the design. The HRMAS NMR spectrum provides an image

depicting the local metabolism found in the specimen analyzed, hence,

comparison of tissues obtained from different species is straightforward. Our

result revealed obvious spectral differences between the diverse prostate

tissue types, indicating metabolic distinctions.

Rat, mouse, and man share approximately 99 % of the same genes [Gibbs et

al., 2004], which clarifies the popularity of using rodents as experimental

PCa models [Shirai et al., 2000]. However, a similar genome does not,

unfortunately, provide the key to perform realistic comparisons [Buttyan et

al., 1993; Lamb et al., 2005]. The rat lateral prostate is the closest analogue

to the human peripheral zone where most cancers occur. Furthermore, the

LP, but not the DP, contains citrate producing cells. As it is difficult to

separate the LP from the DP these lobes are usually combined into the DLP

entity, which is what we used in our study. One of the finding was the

presence of a 2.8 ppm resonances stemming from omega-6 fatty-acids, in all

of the Dunning tumors specimens and in some of the DLP specimens. This is

in line with our findings in Paper I, i.e., this 2.8 ppm resonance was found in

some PCa specimens. One of the questions here is to why the completely

healthy DLP show similar characteristics as the Dunning tumors and human

PCa tissues. One answer is suggested: the Dunning tumor is believed to

originate from the DP and consequently both should therefore share similar

metabolic characteristics. Furthermore, the DP is the lobe in the DLP entity

that lacks citrate accumulating cells. When comparing the spectra obtained

from the DLP and the Dunning tumor it is evident that they share similar

metabolic characteristics as compared with the VP which expresses a vastly

different metabolism. The Dunning tumor spectra most closely resemble the

spectra obtained from human PCa tissue. The one important feature that

indicates cancer in both human PCa and the Dunning tumor is the increased

resonance of (GPCho+PCho) over the free choline resonance. However, this

sign is also evident in completely healthy DLP. Therefore, it would be of

great value to study the metabolism of the DP and the LP separately. Another

important reason for this is to further study and to clarify the functions of

these two lobes in regards to hormonally induced PCa and senescence.

Obviously, it is not optimal to compare a prostate tissue sampled from a 65

years old man with one obtained from a 12 wks old rat. This is of special

interest in this study as age-dependent overgrowth and pathological changes

that leads to cancer usually arises in the DLP but not in the VP. Interestingly,

the COX-2 pathway expression is both higher and dependent on the ageing

process in the DLP as it increases with age, but seems to be unaffected in the

VP. [Badawi et al., 2004]. It is logical that inflammation increases with age.

61

As already discussed above in Paper I, the COX enzymes convert the omega-

6 fatty acid arachidonic acid into the pro-inflammatory, and also pain

producer, prostaglandin E2 (PGE2). There exist two forms of the COX

enzyme; COX-1, which is expressed in most tissues and organs, and the

inducible enzyme COX-2 which is primarily localized in inflammatory cells.

COX-2 is therefore considered to be the key enzyme in fatty acid metabolism

and inflammation. To inhibit the conversion of AA to PGE non-steroidal

anti-inflammatory drugs (NSAIDs), including COX-2 inhibitors, are

frequently used, in spite of significant negative side effects [Seibert et al.,

1994]. Omega-3 is believed to function as a natural inhibitor of COX-2 and

by reducing inflammation indirectly decreases the risk of PCa. Importantly,

however, this potential effect appears to be modified by genetic variation in

COX-2 [Fradet et al., 2009; Simopoulos, 2010]. Furthermore, a recent study

showed that dietary modification that increased tumor EPA and LA (n-3)

content enhanced responses to hormone ablation therapy. Conversely, a

relapse to androgen independent growth (measured by PSA and tumor

volume) positively correlated with tumor AA (n-6) content [McEntee et al.,

2008]. As mentioned earlier, the optimal study would be to separately

scrutinize the DP and the LP obtained from senescent rats and compare with

human PCa tissue samples. In addition, a thorough dietary investigation is

recommended to be included in the study to evaluate the therapeutic effects

of PUFAs.

Paper III

The third study entailed a more comprehensive metabolomic investigation,

based upon the knowledge generated from study I and II. The aim was to

discover spectral bio-markers that correlate/relate to the various dependents

obtained from the clinical examinations of human prostate tissues. In this

study, bio-markers known to be part of the prostate metabolism were

investigated by exploiting peak ratios and their variation due to metabolic

perturbations, using the basic deconvolution method. This approach is

termed as “the use of biochemical prior knowledge” and is used to decrease

uncertainties with overlapping resonances [Inouye et al., 2010]. This

method, using ratios with arbitrary units, circumvent the problems involved

in finding the optimal standard reference and leak-proofing the rotors, and

the need for T2 value adjustments, all in order to make accurate quantitative

calculations [Swanson et al., 2006; Albers et al., 2009].

Histological evaluation of needle biopsy cores from the prostate is the initial

step in diagnosis and treatment decision making. A representative biopsy

may provide an indication as to the nature of the cancer in the prostate gland

62

based on the Gleason score and the volume of disease. Cores containing

sufficient amount of cancer are essential for the evaluation. However, there

is no imaging method available that can guide towards the most malignant

part of the prostate. Because more men are now being diagnosed with local

PCa, due to increased PSA-testing, it is necessary to evaluate and improve

diagnostic options with the aim of minimizing under grading of biopsies. As

many prostate tumors are latent and will not pose a threat, it is desirable to

find a biomarker than is able to differentiate between the indolent and

aggressive types in order to avoid overtreatment. Therefore, the aim of this

study was to identify potential bio-markers for the discrimination of

aggressive vs. the more latent type of cancer and also to estimate the fraction

of tumor cells in any given prostate tissue specimen.

As most PCa studies have focused on the tumor, and as it is not possible to

accurately detect tumors in-vivo by the imaging methods currently available,

we used an opposite approach, i.e., by searching for bio-markers in the non-

malignant tissue. The metabolomic bio-marker that was prominent in our

study was the naturally occurring nutrient inositol which apparently has the

potential of sensing tumors and their corresponding grades elsewhere in the

organ. Inositol can be phosphorylated in seven variations to form

phosphoinositides that act as cellular mediators of signal transduction,

metabolic regulation, and overall growth [Mathews et al., 2000].

By exploiting the ratio of myo-inositol/scyllo-inositol discrimination

between GS 6 and GS 7 was achieved. This is potentially a very important

finding as GS 6 tumors usually have a better prognosis than GS 7. There are

also different treatment approaches for these two different types of prostate

cancer. This novel finding needs to be explored further as bio-markers that

have the power to detect tumors and also estimate their Gleason score in

non-malignant tissue is highly desired. Such feature would minimize the

number of biopsy cores sampled as it would therefore not be necessary to

aim for the tumor itself. This type of bio-marker is currently lacking and is

essential for successful focal therapy and/or active surveillance [Lazzeri et

al., 2010]. This finding suggests protective properties of inositol which is in

line with numerous other studies [Vucenic et al., 2003, 2006; Fardet, 2010].

The detection of loss/failure of cellular protection at an early stage has the

potential of being the basis for preventative medicine.

Curiously, we did not find an equivalent bio-marker in malignant prostate

tissue. The ratio of citrate/creatine came close to statistic significance when

not adjusting for clustering (uneven sampling among the patients in our

study). The focal point here is the complexity of biochemical pathways in all

living organisms and the importance of viewing each patient as an individual

63

rather than a gross average, as there are many factors that have an effect on

the metabolome. A tumor marker was identified in a previous study (Paper

II); (GPCho+PCho)/Cre. An increased choline metabolism is a well known

marker for cancer and is verified in other PCa studies. Our study investigated

the correlative strength of this bio-marker ratio; (GPCho+PCho)/Cre, and

the fraction of tumor cells in any given prostate tissue sample. Indeed, we

found statistically significant correlation. We also found a significant,

although very weak, correlation with the proliferation marker Ki67 and

(GPCho+PCho)/Cre. This raises questions about the underlying reasons for

the increased choline metabolism, which in fact, is poorly understood

[Moestue et al., 2010], and as it might not to be correlated with proliferation

[Gillies et al., 2005], our observation with Ki67 and (GPCho+PCho)/Cre

might be a coincidence. Our study showed large variations in the

concentration of choline compounds, especially in prostate specimens

containing >50 % cancer cells. A recent study where choline metabolic

profiles in breast cancer xenograft models was investigated found

associations with differences in gene expression for basal-like vs. luminal-

like breast cancer [Moestue et al., 2010]. However, there are no such types of

prostate cancer.

Despite uncertainties regarding the exact function of choline metabolites its

increase is still a marker for tumors. This feature is valuable for tumor

verification and localization and may be used parallel with other bio-markers

such as inositol.

64

Chapter 8: Discussion

A few abnormal cells found in a small piece of prostate tissue are most

consequential for a man’s future.

As thoroughly discussed in this dissertation, the prostate is a very difficult

organ to diagnose [Chun et al., 2010; Colleselli et al., 2011] thus the concerns

about incorrect diagnosis and treatments are most valid. Prostate health

awareness, involving PSA-testing and biopsy, has increased the number of

early stage and localized PCa [Wolf et al., 2010; Wever et al., 2010].

Treatment options for this diagnosis are radical therapy such as surgery

and/or radiotherapy and brachytherapy where there are side effects such as

urinary and sexual dysfunctions. The alternative option is active surveillance

with the risk of under treatment but also the psychological burden involved.

Prostate sparing focal therapy is an alternative treatment option although its

effects are not determined yet [Lazzeri et al., 2010]. As PCa more often than

not is multifocal, focal therapy has a higher demand for increased biopsy

sensitivity and precise tumor localization [Meiers et al., 2007; Scheenen et

al., 2011]. Bio-markers with the power of accurately discriminating between

aggressive and indolent prostate tumors are most important as this diagnosis

will be the basis for therapy choice [Xu et al., 2010]. Therefore, innovative

methods to increase the precision of prostate diagnostic are of essence in the

field of prostate research. The ex-vivo and non-destructive method of

HRMAS NMR spectroscopy, in combination with histopathology, shows

promise in this area as it provides metabolomic information, both spectral

and morphologic, with the potential of being used as pre-clinical information

for the development of clinical diagnostic imaging methods for MRSI and

PET [Spratlin et al., 2009]. HRMAS NMR also holds the potential of being

used as a complement to clinical histopatological evaluations, although this

has to be investigated further [van Asten et al., 2008; Santos et al., 2010].

There are many positive aspects and strengths in this approach using

HRMAS NMR in combination with histopathological analysis as is presented

here. Firstly, this is a non-destructive method, thus detecting both lipophilic

and hydrophilic compound in their present form simultaneously and the

tissue is preserved for post-NMR analysis and verification under light

microscope. This strength is exploited in Paper I: “Detection of

polyunsaturated omega-6 fatty acid in human malignant prostate tissue by

1D and 2D high-resolution magic angle spinning NMR spectroscopy”,

where both fatty acids and small organic, water soluble compounds such as

citrate are discussed. Furthermore, Paper I presents the versatility of

HRMAS NMR where unexpected resonances in the spectra arising from

65

unknown compounds may be identified by various 2D methods. The

spectrum depicts the metabolism as it is and this is correlated with results

from histopathology, and thus metabolic profiling is possible. This offers the

characterization and visualization of the diverse metabolism in different

species, organs, anatomical regions, and different types of cells. This type of

characterization is presented in Paper II: “Detection of Local Prostate

Metabolites by HRMAS NMR Spectroscopy: A Comparative Study of

Human and Rat Prostate Tissues”. Such information is of key value when

designing experiments [Costello et al., 2006], especially in translational

research [Ruttenberg et al., 2007]. Paper II further discusses spectrally

derived bio-markers which are applicable on humans may not be suited for

experimental rodent models. HRMAS NMR detects cellular perturbations

due to disease vs. health, in other words; the dynamics of cellular

homeostasis. The loss of homeostasis, detected at an early stage, is an

important key in the understanding of disease [van der Greef, et al 2005].

This type of perturbation/loss is presented in Paper III: “1H HRMAS NMR

Derived Bio-markers Related to Tumour Grade, Tumour Cell Fraction, and

Cell Proliferation in Prostate Tissue Samples”, where it appears as the

homeostasis of non-malignant cells in the prostate, with tumors elsewhere in

the organ, is being disturbed. The key metabolite here is the common phyto-

nutrient inositol which apparently possesses protective properties [Vucenik

et al., 2003, 2006; Roy et al., 2009; Bacić et al., 2010] and is presently under

clinical trials investigations [www.ClinicalTrials.gov]. Thus, jointly with the

finding in Paper I regarding essential PUFA omega-6, HRMAS NMR

provides a window into the area of nutrition. This is of particular interest as

lifestyle choices, especially diet, have impact on prostate health and disease

[Moyad et al., 2008; Gonzales et al., 2010]. Thus, chemoprevention of

prostate cancer/disease using dietary means, and other preventative

measures, is a promising area in prostate research [Silberstein et al., 2010].

This approach, using HRMAS NMR in combination with subsequent

histopathology, is an excellent way of detecting the metabolome, however, it

cannot be stressed enough that this method, like with any other study

involving ex-vivo biological tissues, requires strict handling and a robust

study protocol as there are many possible short-comings. Biological tissues

without the enzymatic control and exposed to oxygen are at high risk of

oxidation and degradation. The time from the freezer to post-NMR handling

must be as short and efficient as possible to avoid metabolic break down and

steady, low temperature is of essence [Beckonert et al., 2010]. The shimming

of the NMR probe to generate a homogeneous magnetic field is equally

important as this affects the line-width and thus resolution which is critical

for line-fitting. Tuning and matching for the correct nucleus is equally

important factor to increase the signal-to-noise ratio [Kreis, 2004]. As both

66

shimming and tuning are manually adjusted it is important to decrease

operator dependent errors. This may be done by using the same operator for

all the experiments as we did in our study. Operator dependent errors should

also be considered when processing the spectra.

The results presented here are merely a fraction of what can be examined

using the metabolomic approach of HRMAS NMR spectroscopy combined

with histopathology.

67

Chapter 9: Suggestions for the Future

“Akufukuzae hakwambii toka”*

Swahili kanga writing

When reviewing my years involved in studying the prostate, I realize that the

most valuable knowledge has been gained simply by trials and errors. I have

also come to the conclusion, like many others, that this is a very complex,

highly multi-facetted, organ to study. It requires skills in many, widely,

separate areas such as, urology, pathology, chemistry, biology, radiology,

physics, statistics, demography, sociology, politics etc., and most

importantly, a thorough knowledge about the anatomy and function of both

human and experimental prostate models. This realization has made me

question the current approach of performing prostate cancer research (and

research in general), i.e., by separate entities of research groups specializing

in proprietary areas with very little, if any, intra-communication. This is

unfortunate since it is, simply, necessary in order to gain complete

understanding of this complex organ.

Interdisciplinary research is, however, becoming more popular and should

be highly encouraged. My one suggestion for the future is, therefore,

consequently, to encourage separate research groups to communicate and

share and discuss findings and not view conflicting results as intimidating

but as valuable information to be used for collective improvements that will

ultimately benefit the single most important factor – the patient.

*Inner feelings are mostly communicated through attitude and behavior and much less

through words.

68

Chapter 10: References

Abate-Shen C, Shen MM. Diagnostics: The prostate-cancer metabolome. Nature. 2009;457:799-800.

Afman L, Muller M. Nutrigenomics: From molecular nutrition to prevention of disease. J Am Diet Assoc. 2006;106:569–576.

Aigner F, Mitterberger M, Rehder P, Pallwein L, Junker D, Horninger W, Frauscher F. Status of transrectal ultrasound imaging of the prostate. J Endourol. 2010;24(5):685-691.

Akoka S, Barantin L, Trierweiler M. Concentration measurement by proton NMR using the ERETIC method. Anal Chem. 1999;71(13):2554–2557.

Albers MJ, Butler TN, Rahwa I, Bao N, Keshari KR, Swanson MG, Kurhanewicz J. Evaluation of the ERETIC method as an improved quantitative reference for 1H HR-MAS spectroscopy of prostate tissue. Magn Reson Med. 2009;61(3):525–532.

American Cancer Society. Cancer Prevention & Early Detection Facts & Figures 2010. Atlanta: American Cancer Society; 2010.

Andrew ER, A. Bradbury A, Eades RG. Removal of dipolar broadening of nuclear magnetic resonance spectra of solids by specimen rotation. Nature.1959;183:1802-1803.

Andriole GL. The lottery of conventional prostate biopsy. Nat Rev Urol. 2009;6:188-189.

Avril N, Dambha F, Murray I, Shamash J, Powles T, Sahdev A. The clinical advances of fluorine-2-D-deoxyglucose--positron emission tomography/computed tomography in urological cancers. Int J Urol. 2010;17(6):501-11.

Bacić I, Druzijanić N, Karlo R, Skifić I, Jagić S. Efficacy of IP6 + inositol in the treatment of breast cancer patients receiving chemotherapy: prospective, randomized, pilot clinical study. J Exp Clin Cancer Res. 2010;29:12.

Badawi AF, Liu Y, Eldeen MB, Morrow W, Razak ZR, Maradeo M, Badr MZ. Age-associated changes in the expression pattern of cyclooxygenase-2 and related apoptotic markers in the cancer susceptible region of rat prostate. Carcinogenesis. 2004;25(9):1681-1688.

Baron A, Migita T, Tang D, Loda M. Fatty acid synthase: a metabolic oncogene in prostate cancer? J Cell Biochem. 2004;91:47–53.

69

Bathen TF, Sitter B, Sjøbakk TE, Tessem MB, Gribbestad IS. Magnetic resonance metabolomics of intact tissue: a biotechnological tool in cancer diagnostics and treatment evaluation. Cancer Res. 2010;70(17):6692-6696.

Beare-Rogers J, Dieffenbacher A, Holm JV. Lexicon of lipid nutrition (IUPAC Technical Report). Pure Appl Chem. 2001;73(4): 685-744.

Beckonert B, Coen M, Keun HC, Wang Y, Ebbels T, Holmes E, Lindon JC, Nicholson JK. High-resolution magic-angle-spinning NMR spectroscopy for metabolic profiling of intact tissues. Nature Protocols. 2010;5:1019-1032.

Beckonert O, Keun HC, Ebbels TMD, Bundy J, Holmes E, Lindon JC, Nicholson JK. Metabolic profiling, metabolomic and metabonomic procedures for NMR spectroscopy of urine, plasma, serum and tissue extracts. Nat Protocols. 2007;2(11):2692–2703.

Berney DM, Fisher G, Kattan MW, Oliver TD, Scardino P, Foster CS, et al. (Trans-Atlantic Prostate Group). Major shifts in the treatment and prognosis of prostate cancer due to changes in pathological diagnosis and grading. BJU Int. 2007;100:1240-1244.

Bodenhausen G, Ruben DJ. Natural abundance nitrogen-15 NMR by enhanced heteronuclear spectroscopy. Chem Phys Lett. 1980;69:185-188.

Bouchelouche K, Turkbey B, Choyke P, Capala J. Imaging prostate cancer: an update on positron emission tomography and magnetic resonance imaging. Curr Urol Rep. 2010;11(3):180-190.

Braunschweiler L, Ernst RR. Coherence transfer by isotropic mixing: Application to proton correlation spectroscopy. J Magn Reson. 1983;53:521-528.

Brown SP, Spiess HW. Advanced solid-state NMR methods for the elucidation of structure and dynamics of molecular, macromolecular, and supramolecular systems. Chem Rev. 2001;101(12):4125-4156.

Buttyan R, Slawin K. Rodent models for targeted oncogenesis of the prostate gland. Cancer Metastasis Rev. 1993;12(1):11-19.

Chamberlain NF. Determining molecular structure by nuclear magnetic resonance of hydrogen. Anal Chem. 1959;31:56-77.

Cheng LL, Chang IW, Smith BL, Gonzalez RG. Evaluating human breast ductal carcinomas with high-resolution magic-angle spinning proton magnetic resonance spectroscopy. J Magn Reson. 1998;135(1):194-202.

Chun FK, Epstein JI, Shariat SF, Schröder FH, Scattoni V, et al. Optimizing performance and interpretation of prostate biopsy: a critical analysis of the literature. Eur Urol. 2010;58(6):851-864.

70

Coffey DS. Similarities of prostate and breast cancer: Evolution, diet, and estrogens. Urology. 2001;57(4,S1):31-38.

Colleselli D, Hennenlotter J, Schilling D, Krueger SA, Stenzl A, Schwentner C, et al. Impact of clinical parameters on the diagnostic accuracy of endorectal coil MRI for the detection of prostate cancer. Urol Int. 2011 (In press).

Conti F, Manganaro M, Miccheli A. Metabolomics and medical practice. Clin Ter. 2006;157(6):549-552.

Cooper JE, Farid I. The role of citric acid in the physiology of the prostate. A chromatographic study of citric acid cycle intermediates in benign and malignant prostatic tissue. J Surg Res. 1963;3:112–121.

Costello LC, Franklin RB. Citrate metabolism of normal and malignant prostate epithelial cells. Urology. 1997;50:3–12.

Costello LC, Franklin RB. Novel role of zinc in the regulation of prostate citrate metabolism and its implications in prostate cancer. Prostate. 1998;35(4):285-296.

Costello LC, Franklin RB, Narayan P. Citrate in the diagnosis of prostate cancer. Prostate. 1999;38(3)237-245.

Costello LC, Franklin RB. The intermediary metabolism of the prostate: a key to understanding the pathogenesis and progression of prostate malignancy. Oncology. 2000;59:269-282.

Costello LC, Franklin RB, Feng P. Mitochondrial function, zinc, and intermediary metabolism relationships in normal prostate and prostate cancer. Mitochondrion. 2005;5(3):143-153.

Costello LC, Franklin RB. ’Why do tumour cells glycolyse?’: From glycolysis through citrate to lipogenesis. Mol Cell Biochem. 2005;280(1–2):1–8.

Costello LC, Franklin RB. The clinical relevance of the metabolism of prostate cancer; zinc and tumor suppression: connecting the dots. Mol Cancer. 2006;5:17.

Costello LC, Franklin RB. Prostatic fluid electrolyte composition for the screening of prostate cancer: a potential solution to a major problem. Prostate Cancer Prostatic Dis. 2009;12:17–24.

Cunnane SC, Ryan MA, Nadeau CR, Bazinet RP, Musa-Veloso K, McCloy U. Why is carbon from some polyunsaturates extensively recycled into lipid synthesis? Lipids. 2003;38(4):477.

Cunnane SC. Problems with essential fatty acids: time for a new paradigm? Prog Lipid Res. 2003;42(6):544-568.

71

Cupp MR, Oesterling JE. Prostate-specific antigen, digital rectal examination, and transrectal ultrasonography: their roles in diagnosing early prostate cancer. Mayo Clin Proc. 1993;68(3):297-306.

Davis CD, Milner J. Frontiers in nutrigenomics, proteomics, metabolomics and cancer prevention. Mutat Res. 2004;551:51–64.

De Marzo AM, Platz EA, Sutcliffe S, Xu J, Grönberg H, Drake CG, Nakai Y, Isaacs WB, Nelson WG. Inflammation in prostate carcinogenesis. Nat Rev Cancer. 2007;7(4):256-269.

De Visschere P, Oosterlinck W, De Meerleer G, Villeirs G. Clinical and imaging tools in the early diagnosis of prostate cancer, a review. JBR-BTR. 2010;93(2):62-70.

Epstein JI. Diagnosis and reporting of limited adenocarcinoma of the prostate on needle biopsy. Modern Pathology. 2004;17:307–315.

Epstein JI. An update of the Gleason grading system. J Urol. 2010;183(2):433-440.

Ernst RR, Bodenhausen G, Wokaun A. Principles of Nuclear Magnetic Resonance in One and Two Dimensions. Oxford University Press, NY, USA. 1987. ISBN-0-19-855647-0.

Fair WR, Couch J, Wehner N. Prostatic antibacterial factor. Identity and significance. Urology. 1976;7(2):169-177.

Fair WR, Parrish RF. Antibacterial substances in prostatic fluid. Prog Clin Biol Res. 1981;75A:247-264.

Fardet A. New hypotheses for the health-protective mechanisms of whole-grain cereals: what is beyond fibre? Nutr Res Rev. 2010;23(1):65-134.

Fiehn O, Kopka J, Dormann P, Altmann T, Trethewey RN, Willmitzer L. Metabolite profiling for plant functional genomics. Nature Biotechnology. 2000;18(11):1157-1161.

Fleshner N, Klotz L. Role of ”saturation biopsy” in the detection of prostate cancer among difficult diagnostic cases. Urology. 2002;60:93-97.

Fradet V, Cheng I, Casey G, Witte JS. Dietary omega-3 fatty acids, cyclooxygenase-2 genetic variation, and aggressive prostate cancer risk. Clin Cancer Res. 2009;15(7):2559-2566.

Franklin RB, Feng P, Milon B, et al. hZIP1 zinc uptake transporter down regulation and zinc depletion in prostate cancer. Mol Cancer 2005;4:32.

Fry CG. The Nobel Prize in Medicine for Magnetic Resonance Imaging. J Chem Educ. 2004;81(7)922.

72

Funk CD. Prostaglandins and leukotrienes: advances in eicosanoid biology. Science. 2001;294(5548):1871-1875.

German JB, Hammock BD, Watkins SM. Metabolomics: building on a century of biochemistry to guide human health. Metabolomics. 2005;1:3–9.

Gibbs RA, Weinstock GM, Metzker ML, Muzny DM et al. Genome sequence of the Brown Norway rat yields insights into mammalian evolution. Nature. 2004;428(6982):493-521.

Gibney MJ, Walsh M, Brennan L, Roche HM, German B, van Ommen B. Metabolomics in human nutrition: opportunities and challenges. Am J Clin Nutr. 2005;82:497–503.

Gillies RJ, Morse DL. In vivo magnetic resonance spectroscopy in cancer. Annu Rev Biomed Eng. 2005;7:287-326.

Giusti S, Caramella D, Fruzzetti E, Lazzereschi M, Tognetti A, Bartolozzi C. Peripheral zone prostate cancer. Pre-treatment evaluation with MR and 3D (1)H MR spectroscopic imaging: correlation with pathologic findings. Abdom Imaging. 2010;35(6):757-763.

Gleason DF, Mellinger GT. Prediction of prognosis for prostatic adenocarcinoma by combined histological grading and clinical staging. J Urol. 1974;111:58–64.

Gonzales CA, Riboli E. Diet and cancer prevention: Contributions from the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Eur J Cancer. 2010;46(14):2555-2562.

Goodacre R. Metabolomics of a superorganism 1–3. J Nutr. 2007;137:259S–266S.

Griffin JL, Shockcor JP. Metabolic profiles of cancer cells. Nat Rev Cancer. 2004;4(7):551–561.

Häggström S, Wikström P, Bergh A, Damber JE. Expression of vascular endothelial growth factor and its receptors in the rat ventral prostate and the Dunning R3327-PAP adenocarcinoma before and after castration. Prostate. 1998;36:71-79.

Hahn WC. Experimental models of human cancer. Cell Cycle. 2004;3(5):604-605.

Hayashi N, Sugimura Y, Kawamura J, Donjacour AA, Cunha GR. Morphological and functional heterogeneity in the rat prostatic gland. Biol Reprod. 1991;45(2):308-321.

Hoedemaeker RF, Vis AN, Van Der Kwast TH. Staging prostate cancer. Microsc Res Tech. 2000;51(5):423-429.

73

Holmes E, Loo RL, Stamler J, Bictash M, Yap IKS, Chan Q, Ebbels T, Nicholson JK, Elliott P. Human metabolic phenotype diversity and its association with diet and blood pressure. Nature. 2008;453:396-400.

Hsu PP, Sabatini DM. Cancer cell metabolism: Warburg and beyond. Cell. 2008;134:703-707.

Inouye M, Kettunen J, Soininen P, Silander K, Ripatti S, Kumpula LS, Peltonen L. Metabonomic, transcriptomic, and genomic variation of a population cohort. Molecular Systems Biology. 2010;6:441.

Jacobsen NE. NMR Spectroscopy Explained: Simplified Theory, Applications and Examples for Organic Chemistry and Structural Biology. 2007. John Wiley & Sons, Inc. Hoboken, NJ, USA.

Jadvar H. Prostate Cancer: PET with 18F-FDG, 18F- or 11C-Acetate, and 18F or 11C-Choline. J Nucl Med. 2011;52(1):81–89.

Jambor I, Borra R, Kemppainen J, Lepomäki V, Aronen HJ, Minn H, et al. Functional imaging of localized prostate cancer aggressiveness using 11C-acetate PET/CT and 1H-MR spectroscopy. J Nucl Med. 2010;51(11):1676-1683.

Jara M, Carballada R, Esponda P. Age-induced apoptosis in the male genital tract of the mouse. Reproduction. 2004;127:359-366.

Johannsen W. The genotype conception of heredity. American Naturalist. 1911;45:129-159.

Kamb A. What's wrong with our cancer models? Nat Rev Drug Discov. 2005;4(2):161-165.

Kavanagh JP. Sodium, potassium, calcium, magnesium, zinc, citrate and chloride content of human prostatic and seminal fluid. J Reprod Fert. 1985;75:35-41.

Kelavkar U, Lin Y, Landsittel D, Chandran U, Dhir R. The yin and yang of 15-lipoxygenase-1 and delta-desaturases: Dietary omega-6 linoleic acid metabolic pathway in prostate. J Carcinog. 2006;5:9.

Kelavkar UP, Hutzley J, McHugh K, Allen KG, Parwani A. Prostate tumor growth can be modulated by dietarily targeting the 15-lipoxygenase-1 and cyclooxygenase-2 enzymes. Neoplasia. 2009;11(7):692-699.

Kelloff GJ, Choyke P, Coffey DS. Challenges in clinical prostate cancer: role of imaging. AJR. 2009;192:1455-1470.

Khan N, Afaq F, Mukhtar H. Lifestyle as risk factor for cancer: Evidence from human studies. Cancer Lett. 2010;293(2):133-143.

74

Kim J, Dang CV. Cancer’s molecular sweet tooth and the Warburg effect. Cancer Res. 2006;66(18)8927-8930.

Kirby RS. The natural history of benign prostatic hyperplasia: what have we learned in the last decade? Urology. 2000;56(5,S1):3-6.

Klijn S, De Visschere PJ, De Meerleer GO, Villeirs GM. Comparison of qualitative and quantitative approach to prostate MR spectroscopy in peripheral zone cancer detection. Eur J Radiol 2011 (in print).

Koulman A, Volmer DA. Perspectives for metabolomics in human nutrition: an overview. Nutrition Bulletin. 2008;33(4):324–330.

Kraus TS, Cohen C, Siddiqui MT. Prostate-specific antigen and hormone receptor expression in male and female breast carcinoma. Diagn Pathol. 2010;5:63.

Krebs HA. The citric acid cycle and the Szent-Györgyi cycle in pigeon breast muscle. Biochem J. 1940;34(5):775-779.

Kreis R. Issues of spectral quality in clinical 1H-magnetic resonance spectroscopy and a gallery of artifacts. NMR Biomed. 2004;17:361-381.

Kurhanewicz J, Swanson MG, Nelson SJ, Vigneron DB. Combined magnetic resonance imaging and spectroscopic imaging approach to molecular imaging of prostate cancer. J Magn Reson Imaging. 2002;16:451-463.

Lamb DJ, Zhang L. Challenges in prostate cancer research: animal models for nutritional studies of chemoprevention and disease progression. J Nutr. 2005;135(12):3009S-3015S.

Larson PEZ, Bok R, Kerr AB, Lustig M, Kurhanewicz J, Vigneron DB, et al. Investigation of tumor hyperpolarized [1-13C]-pyruvate dynamics using time-resolved multiband RF excitation echo-planar MRSI. Magn Reson Med. 2010;63(3):582–591.

Lash TD, Shirkey Lash S. The use of Pascal-like triangles in describing first order NMR coupling patterns. J Chem Educ. 1987;64(4):315.

Lazzeri M, Guazzoni G. Focal therapy meets prostate cancer. Lancet. 2010;376(9746):1036-1037.

Liney GP, Lowry M, Turnbull LW, Manton DJ, Knowles AJ, Blackband SJ, Horsman A. Proton MR T2 maps correlate with the citrate concentration in the prostate. NMR Biomed. 1996;9(2):59-64.

Lissbrant IF, Hammarsten P, Lissbrant E, Ferrara N, Häggström Rudolfsson S, Bergh A. Neutralizing VEGF bioactivity with a soluble chimeric VEGF-receptor protein flt(1-3)IgG inhibits testosterone-stimulated prostate growth in castrated mice. Prostate. 2004;58:57–65.

75

Lucia SM, Bostwick DG, Bosland M, Cockett ATK, Knapp DW, Watkins BA, et al. Workgroup 1: Rodent models of prostate cancer. Prostate. 1998;36:49-55.

Maas WE, Laukien FH, Cory DG. Gradient, High Resolution, Magic Angle Sample Spinning NMR. J Am Chem Soc.1996;118(51):13085–13086.

Madu CO, Lu Y. Novel diagnostic bio-markers for prostate cancer. J Cancer. 2010;1:150-177.

Mann T. Secretory function of the prostate, seminal vesicle and other male accessory organs of reproduction. J Reprod Fert. 1974;37:179-188.

Martin GE, Zekter AS. Two-Dimensional NMR Methods for Establishing Molecular Connectivity. 1988. John Wiley & Sons, Inc. NY, USA.

Mathews CK, van Holde KE, G. Ahern KG. Biochemistry. 3rd Edition, 2000. Addison Wesley Longman, Inc. San Francisco, Ca, USA.

Mawson CA, Fischer MI. The occurrence of zinc in the human prostate gland. Can J Med Sci. 1952;30:336–339.

McEntee MF, Carol Ziegler C, Reel D, Tomer K, Ahmed Shoieb A, Lih FB, O’Rourke D, Whela J. Dietary n-3 polyunsaturated fatty acids enhance hormone ablation therapy in androgen-dependent prostate cancer. Am J Pathol. 2008;173:229–241.

McLaughlin PW, Troyer S, Berri S, Narayana V, Meirowitz A, Roberson PL, Montie J. Functional anatomy of the prostate: implications for treatment planning. Int J Radiat Oncol Biol Phys. 2005;63(2):479-491.

McLean MA, Barrett T, Gnanapragasam VJ, Priest AN, Joubert I, Lomas DJ, Neal DE, Griffiths JR, Sala E. Prostate cancer metabolite quantification relative to water in 1H-MRSI in vivo at 3 Tesla. Magn Reson Med. 2010 (In print).

McMahon CJ, Bloch BN, Lenkinski RE, Rofsky NM. Dynamic contrast-enhanced MR imaging in the evaluation of patients with prostate cancer. Magn Reson Imaging Clin N Am. 2009;17(2):363-383.

McNeal JE. The zonal anatomy of the prostate. Prostate. 1981;2(1):35-49.

Mehr K, John B, Russell D Avizonis D. Electronic referencing techniques for quantitative NMR: pitfalls and how to avoid them using amplitude-corrected referencing through signal injection. Anal Chem. 2008;80(21):8320–8323.

Meiboom S, Gill D. Modified spin-echo method for measuring nuclear relaxation times. Rev. Sci. Instrum. 1958;29:688-691.

76

Meiers I, Waters DJ, Bostwick DG. Preoperative prediction of multifocal prostate cancer and application of focal therapy: review 2007. Urology. 2007;70(6S):3-8.

Middleton DA, Bradley DP, Connor SC, Mullins PG, Reid DG. The effect of sample freezing on proton magic-angle spinning NMR spectra of biological tissue. Magn Reson Med. 1998;40(1):166–169.

Moestue S, Sitter B, Bathen TF, Tessem MB, Gribbestad IS. HR MAS MR spectroscopy in metabolic characterization of cancer. Curr Top Med Chem. 2011;11(1):2-26.

Moestue SA, Sitter B, Børresen-Dale AL, Engebraaten O, Gribbestad IS, et al. Distinct choline metabolic profiles are associated with differences in gene expression for basal-like and luminal-like breast cancer xenograft models. BMC Cancer. 2010;10:433.

Moyad MA, Lowe FC. Educating patients about lifestyle modifications for prostate health. Am J Med. 2008;121(8,S2):S34-42.

Nastiuk KL, Liu H, Mark Hamamura M, Muftuler LT, Orhan Nalcioglu O, and John J Krolewski JJ. In vivo MRI volumetric measurement of prostate regression and growth in mice. BMC Urology. 2007;7:12.

Nicholson JK, Lindon JC, Holmes E. 'Metabonomics': understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica. 1999;29(11):1181-1189.

Nicholson JK, Lindon JC. Systems biology: Metabonomics. Nature. 2008;455(7216):1054-1056.

Nordström A, Lewensohn R. Metabolomics: moving to the clinic. J Neuroimmune Pharmacol. 2010;5(1):4-17.

Oesterling JE. Prostate specific antigen: a critical assessment of the most useful tumor marker for adenocarcinoma of the prostate. J Urol. 1991;145(5):907-923.

Ornish D, Magbanua MJM, Shinohara K, Haqq CM, Carroll PR, et al. Changes in prostate gene expression in men undergoing an intensive nutrition and lifestyle intervention. Proc Natl Acad Sci USA. 2008;105(24):8369–8374.

Paner GP, Luthringer Dj, Amin MB. Best practice in diagnostic immunohistochemistry. Prostate carcinoma and its mimics in needle core biopsies. Arch Pathol Lab Med. 2008;132:1388-1396.

Pople JA, Schneider WG, Bernstein HJ. High-Resolution Nuclear Magnetic Resonance. 1959. McGraw-Hill, NY, USA.

77

Rajinikanth A, Manoharan M, Soloway CT, Civantos FJ, Soloway MS. Trends in Gleason score: concordance between biopsy and prostatectomy over 15 years. Urology. 2008;72(1):177-182.

Reese AC, Sadetsky N, Carroll PR, Cooperberg MR. Inaccuracies in assignment of clinical stage for localized prostate cancer. Cancer. 2011;15(117):283-289.

Roy S, Gu M, Ramasamy K, Singh RP, Agarwal C, Siriwardana S, Sclafani RA, Agarwal R. p21/Cip1 and p27/Kip1 Are essential molecular targets of inositol hexaphosphate for its antitumor efficacy against prostate cancer. Cancer Res. 2009;69(3):1166-73.

Roy-Burman P, Wu H, Powell WC, Hagenkord J, Cohen MB. Genetically defined mouse models that mimic natural aspects of human prostate cancer development. Endocr Relat Cancer. 2004;11:225–254.

Ruttenberg A, Clark T, Bug W, Samwald M, Bodenreider O, Herman I, Cheung KH, et al. Advancing translational research with the Semantic Web. BMC Bioinformatics. 2007;8(S3):S2.

Rysman E, Brusselmans K, Vanderhoydonc F, Smans K, Waelkens E, Verhoeven G, Swinnen JV, et al. De novo lipogenesis protects cancer cells from free radicals and chemotherapeutics by promoting membrane lipid saturation. Cancer Res. 2010;70(20):8117-8126.

Santos CF, Kurhanewicz J, Tabatabai ZL, Carroll PR, Ha CM, Swanson MG, et al. Metabolic, pathologic, and genetic analysis of prostate tissues: quantitative evaluation of histopathologic and mRNA integrity after HR-MAS spectroscopy. NMR Biomed. 2010;23(4):391-398.

Scheenen TW, Fütterer J, Weiland E, van Hecke P, Roell S, Heerschap A, et al. Discriminating cancer from noncancer tissue in the prostate by 3-Dimensional proton magnetic resonance spectroscopic imaging. A prospective multicenter validation study. Invest Radiol. 2011;46(1):25-33.

Schröder FH, Hugosson J, Roobol MJ, Tammela TLJ, Ciatto S, Nelen V, Auvinen A. Screening and prostate-cancer mortality in a randomized european study. N Engl J Med. 2009;360:1320-1328.

Seibert K, Zhang Y, Leahy K, Hauser S, Masferrer J, Perkins W, Lee L, Isakson P. Pharmacological and biochemical demonstration of the role of cyclooxygenase 2 in inflammation and pain. Proc Natl Acad Sci USA. 1994;91(25):12013-12017.

Sharma S, Shin JS, Grimshaw M, Clarke RA, Lee CS. The senescence pathway in prostatic carcinogenesis. Pathology. 2010;42(6):507-511.

78

Shirai T, Takahashi S, Cui L, Futakuchi M, Kato K, Tamano S, Imaida K. Experimental prostate carcinogenesis - rodent models. Mutat Res. 2000;462(2-3):219-226.

Silberstein JL, Parsons JK. Prostate cancer prevention: concepts and clinical recommendations. Prostate Cancer Prostatic Dis. 2010;13(4):300-306.

Simopoulos AP. Genetic variants in the metabolism of omega-6 and omega-3 fatty acids: their role in the determination of nutritional requirements and chronic disease risk. Exp Biol Med (Maywood). 2010;235(7):785-795.

Singh KK, Desouki MM, Franklin RB, Costello LC. Mitochondrial aconitase and citrate metabolism in malignant and nonmalignant human prostate tissues. Mol Cancer. 2006;5:14.

Sitter B, Bathen TF, Tessem MB, Gribbestad IS. High-resolution magic angle spinning (HR MAS) MR spectroscopy in metabolic characterization of human cancer. Prog Nucl Mag Res Sp. 2009;54(3-4):239-254.

Soubias O, Piottob M, Saurela O, Assematb O, Réata V. Milon A. Detection of natural abundance 1H–13C correlations of cholesterol in its membrane environment using a gradient enhanced HSQC experiment under high resolution magic angle spinning. J Magn Reson. 2003;165(2):303-308.

Spratlin JL, Serkova NJ, Eckhardt GS. Clinical applications of metabolomics in oncology: a review. Clin Canc Res. 2009;15(2):431-440.

Swanson MG, Vigneron DB, Tabatabai ZL, Males RG, Schmitt L, Carroll PR, James JK, Hurd RE, Kurhanewicz J. Proton HR-MAS spectroscopy and quantitative pathologic analysis of MRI/3D-MRSI targeted post-surgical prostate tissues. Magn Reson Med. 2003;50(5):944–954.

Swanson MG, Zektzer AS, Tabatabai ZL, Simko J, Jarso S, Keshari KR, Schmitt L, Carroll PR, Shinohara K, Vigneron DB, Kurhanewicz J. Quantitative analysis of prostate metabolites using 1H HR-MAS spectroscopy. Magn Reson Med. 2006;55(6):1257–1264.

Swindle P, McCredie S, Russell P, Himmelreich U, Khadra M, Lean C, Mountford C. Pathologic characterization of human prostate tissue with proton MR spectroscopy. Radiology. 2003;228:144-151.

Swindle P, Ramadan S, Stanwell P, McCredie S, Russell P, Mountford C Proton magnetic resonance spectroscopy of the central, transition and peripheral zones of the prostate: assignments and correlation with histopathology. Magn Reson Mater Phy. 2008;21:423-434.

Swinnen JV, Vanderhoydonc F, Heyns W, Verhoeven G, et al. Selective activation of the fatty acid synthesis pathway in human prostate cancer. Int J Cancer. 2000;88(2):176-179.

79

Tessem MG, Swanson MG, Nelson SJ, Vigneron DB, Gribbestad IS, Kurhanewicz J, et al. Evaluation of lactate and alanine as metabolic bio-markers of prostate cancer using 1H HR-MAS spectroscopy of biopsy tissues. Magn Reson Med. 2008;60(3):510–516.

The Nobel Prizes. Nobelprize.org.2010. http://nobelprize.org/nobel_prizes/

Thoeny HC, Triantafyllou M, Vermathen P, Studer UE, et al. Combined ultrasmall superparamagnetic particles of iron oxide–enhanced and diffusion-weighted magnetic resonance imaging reliably detect pelvic lymph node metastases in normal-sized nodes of bladder and prostate cancer patients. Eur Urol. 2009;55(4):761-769.

Thomas AM, Lange T, Velan SS, Nagarajan R, Raman S, Boesiger p, et al. Two-dimensional MR spectroscopy of healthy and cancerous prostate in vivo. MAGMA. 2008;21(6):443-458.

Thun MJ, DeLancey JO, Center MM, Jemal A, Ward EM. The global burden of cancer: priorities for prevention. Carcinogenesis. 2010;31(1):100-110.

Trujillo E, Davis C, Milner J. Nutrigenomics, proteomics, metabolomics,and the practice of dietetics. J Am Diet Assoc. 2006;106:403–413.

Turkbey B, Xu S, Kruecker J, Locklin J, Pinto PA. Documenting the location of prostate biopsies with image fusion. BJU International. 2011;107(1):53–57.

Ukimura O, Hung AJ, Gill IS. Innovations in prostate biopsy strategies for active surveillance and focal therapy. Curr Opin Urol. 2011;21(2):115-120.

Ukoli FA, Fowke JH, Akumabor P, Oguike T, Taher KA, et al. The association of plasma fatty acids with prostate cancer risk in African Americans and Africans. J Health Care Poor Underserved. 2010;21(1S):127-147.

van Asten J, Cuijpers V, Hulsbergen-van de Kaa C, Soede-Huijbregts C, Witjes JA, Verhofstad A, Heerschap A. High resolution magic angle spinning NMR spectroscopy for metabolic assessment of cancer presence and Gleason score in human prostate needle biopsies. Magn Reson Mater Phy. 2008;21:435-442.

van der Greef J, Smilde AK. Symbiosis of chemometrics and metabolomics: past, present, and future. J Chemometrics. 2005;19:376–386.

Van Dyke Tiers G. Reliable proton nuclear resonance shielding values by

“Internal Referencing” with Tetramethyl-silane. J PhysChem.

1958;62(9):1151–1152.

Vickers AJ, Till C, Tangen CM, Lilja H, Thompson IM. An empirical evaluation of guidelines on prostate-specific antigen velocity in prostate cancer detection. J Natl Cancer Inst. 2011;103:1–8.

80

Villers A, Steg A, Boccon-Gibod L. Anatomy of the prostate: review of the different models. Eur Urol.1991;20(4):261-268.

Vucenik I, Shamsuddin AM. Cancer inhibition by inositol hexaphosphate (IP6) and inositol: from laboratory to clinic. J Nutr. 2003;133(11,S1):3778S-3784S.

Vucenik I, Shamsuddin AM. Protection against cancer by dietary IP6 and inositol. Nutr Cancer. 2006;55(2):109-125.

Warburg O. On the origin of cancer cells. Science. 1956;123:309-314.

Weidner AM, Michaely HJ, Lemke A, Breitinger L, Wenz F, Marx A, Schoenberg SO, Dinter DJ. Value of multiparametric prostate MRI of the peripheral zone. Z Med Phys. 2011 (in print).

Wever EM, Draisma G, Heijnsdijk EA, Roobol MJ, Boer R, Otto SJ, de Koning HJ. Prostate-specific antigen screening in the United States vs in the European Randomized Study of Screening for Prostate Cancer-Rotterdam. J Natl Cancer Inst. 2010;102(5):352-355.

Wishart DS, Tzur D, Knox C, Eisner R, Guo AC, Young N, et al. HMDB: the Human Metabolome Database. Nucl. Acids Res. 2007;35(S1):D521-D526.

Wolf AMD, Wender RC, Etzioni RB, Thompson IM, D’Amico AV, et al. American Cancer Society guideline for the early detection of prostate cancer. CA Cancer J Clin. 2010;60:70–98.

Wu CL, Taylor JL, He WL, Zepeda AG, Halpern EF, Bielecki A, Gonzalez RG, Cheng LL. Proton high-resolution magic angle spinning NMR analysis of fresh and previously frozen tissue of human prostate. Magn Reson Med. 2003;50(6):1307–1311.

Xu J, Zheng SL, Isaacs SD, Wiley KE, Wiklund F, Grönberg H, Isaacs WB. Inherited genetic variant predisposes to aggressive but not indolent prostate cancer. Proc Natl Acad Sci U S A. 2010;107(5):2136-2140.

Zaichick VY, Sviridova TV, Zaichick SV. Zinc concentration in human prostatic fluid: normal, chronic prostatitis, adenoma and cancer. Int Urol Nephrol. 1996;28:687–694.

Zakian KL, Sircar K, Hricak H, Chen H-N, Shukla-Dave A, Eberhardt S. Correlation of proton MR spectroscopic imaging with Gleason score based on step-section pathologic analysis after radical prostatectomy. Radiology. 2005;234:804-814.