retinoic acid receptor expression and polymorphisms …

RETINOIC ACID RECEPTOR

EXPRESSION AND

POLYMORPHISMS IN HUMAN

BREAST CANCER

by

Danny Mok BSc(Hons)

This thesis is presented for the degree of

Doctor of Philosophy

of The University of Western Australia

2002

Department of Clinical Biochemistry, PathCentre

Department of Pathology

The University of Western Australia

Sir Charles Gairdner Hospital

Nedlands, Western Australia

Acknowledgments

ACKNOWLEDGMENTS

I would firstly like to thank Dr. John Beilby for his valuable supervision of my

laboratory work, preparation of m y thesis, as well as his tremendous support in times of

trouble and when things were going great.

I gratefully thank Professor John Papadimitriou for his enormous contributions in

reviewing m y thesis, as well as manuscripts for publication and his support of m y PhD

project.

I would also like to thank Professor Roland Hahnel and his wife, Erika, for without

them, this project would not have been possible.

Thanks also to Drs. Peter Garcia-Webb and Chootoo Bhagat for allowing me to use the

facilities within the Department of Clinical Biochemistry, PathCentre. Many thanks to

Dr. David Ingram for allowing m e access to his breast tissue samples.

I wish to thank Associate Professor Theo Gotjamanos for all the support and advice he

has given m e in helping to complete m y PhD and for his mediation of all

correspondence with the University with regards to m y degree.

I am also grateful of the help and advice Associate Professor Ralph Martins and

Guiseppe Verdile offered in the Western blots.

To all my colleagues in the lab: Christine Chin, Clive Hunt, Jodi Burley, Caroline

Chapman, Janeen Bennett, John Prior and Kim Jury, for their great friendship, gossip

sessions and support over the years.

I would like to thank all the staff in the Departments of Clinical Biochemistry,

PathCentre, and Pathology, U W A , for their friendship throughout m y PhD.

ii

Acknowledgments

To all m y postgraduate student buddies: Anthea Downs, Linda Cunningham, Michael

Swarbrick, Conchita Kuek, David Dunn, Vincent Williams, Sherine Chan, Tane

McNiven, Carlo Pirn, Amerigo Carrello and Peter Mark, for sharing many stories of

failed experiments, disasters in the lab, as well as understanding how difficult a PhD

student's life can be.

Big thanks to all my friends who kept asking the question all PhD students hate hearing:

"how's the PhD going?" Thanks Richie, Tajinder, Pete, Michelle, Marguerite, Dave,

Tracey, Scott, Suzie, Julian, Rebecca, Geoff, Carrie, Angie, Adam, Hassan, Mark, Cath,

Jo and Jurgen.

Very big thanks to my love, Elysia Hollams, for her unending support and

encouragement during the hard times in the laboratory, and for her invaluable effort in

helping m e organise m y thesis writing. Good luck with your PhD, Smurfette!

Finally, I wish to thank my parents, Caroline and David, who always worked hard to

support an education for m e that unfortunately were not available to them. For that, I

dedicate this thesis to m u m and dad. Thanks also to m y brother and sister, Darren and

Davena, just for being there when I needed them.

iii

Thesis Abstract

THESIS ABSTRACT

Retinoic acid is a natural compound belonging to a group of molecules called

retinoids, which include vitamin A and its biologically active derivatives. Retinoic acid

has been shown to induce cell differentiation both in vivo and in vitro and to inhibit

growth of a wide range of neoplastic cells, including those derived from mammary

carcinomas. The effects of retinoic acid are mediated by nuclear receptors known as the

retinoic acid receptors (RARs) and three isotypes have been identified: RARcc, P and y.

The RARs dimerize with a related nuclear receptor, RXR, and function as ligand-

activated transcription factors by binding to retinoic acid response elements found in the

promoter region of target genes. Loss of expression of RARs has been documented in

various diseases, including malignant cancers, although little information is available

about the status of RARs in breast biopsies. This thesis hypothesised that decreased

expression of RARs and the presence of polymorphisms in their mRNA are involved in

the development and/or progression of human mammary carcinoma. Using PCR, all

three RAR isotypes were detected in normal breast and mammary carcinoma biopsies,

although there was a small loss of RARP expression and a significant loss of RARy

expression observed in mammary carcinoma than in normal breast tissue. This may

suggest that the absence of RARs in breast tissue could lead to malignancy. Further

analysis of RAR mRNA levels in breast tissue using competitive RT-PCR showed

significant decreases in RARP and y expression in mammary carcinoma compared to

normal breast. Semi-quantitative analysis of protein levels also revealed a significant

decrease in the expression of RARP and y in mammary carcinoma compared with

patient-matched normal breast tissue. Statistically, there was no relationship between

Thesis Abstract

RARcc and oestrogen receptor status, in contrast to previous reports in the literature.

However, a positive correlation was observed between ER status and RARy expression,

which has not been previously documented and may have important ramifications on

the use of RARy-selective retinoids in the treatment of breast cancer. No relationship

was observed between RAR protein levels and tumour grade, suggesting that decrease

in RAR expression may be an early event in mammary carcinogenesis. Finally, SSCP

studies to detect polymorphisms in RARs identified a serine to leucine base change at

amino acid 427 in RARy. This polymorphism was observed in 3/72 normal specimens

compared to 7/72 found in mammary carcinoma. Although this was not a significant

difference, analysis of a three-generation family with the S427L polymorphism revealed

it to be heritable. Follow-up studies of the normal specimens could not be performed to

determine if they had developed breast cancer, therefore it is unknown if this

polymorphism predisposes to disease. Whether the S427L polymorphism of the RARy

gene has any role in the development of breast cancer remains to be elucidated. The

results of this thesis show that reduced levels of RARs, as well as genetic

polymorphisms in their exons, may be associated with the initiation or progression of

mammary carcinoma due to insufficient signalling of the normal retinoid response

through its receptors. Clinical trials are currently under way in which breast cancer

patients are given retinoids as treatment of this disease. It is therefore of great

importance that reliable methods are developed for the detection and quantitation of

RARs in breast cancer patients, to determine whether there are sufficient receptors

present in the breast tumour tissue to transduce an effective retinoid response. This

thesis describes such a method, using competitive RT-PCR.

v

Statement

STATEMENT

This statement declares that all the work presented in this thesis was performed solely

by the candidate, unless otherwise stated within the thesis.

Danny M o k (Candidate)

Table of Contents

TABLE OF CONTENTS

ACKNOWLEDGMENTS ii

THESIS ABSTRACT iv

STATEMENT vi

TABLE OF CONTENTS vii

LIST OF FIGURES xiii

LIST OF TABLES xviii

ABBREVIATIONS xix

CHAPTER ONE Literature Review 1

1.1 Historical Overview 2

1.2 Metabolism of Retinoic Acid 4

1.2.1 Uptake of Dietary Retinoids 4

1.2.2 Release of Retinol into Cells 6

1.3 Molecular Biology of Retinoid Receptors 7

1.3.1. Types of Retinoid Receptors 10

1.3.2. Structure of RARs 11

1.3.3 Mechanism of RAR-gene targeting 14

1.3.3.1 Up-Regulation of Target Genes 14

1.3.3.2 Down-Regulation of Target Genes 17

1.4 Regulation of Retinoic Acid Receptor Signalling 20

1.5 Retinoids, RARs and Disease 20

1.6 Retinoids, RARs and Mammary Carcinoma 26

1.7 Research Proposal, Aims and Hypothesis of this Study 32

vii

Table of Contents

CHAPTER TWO Materials and Methods 37

2.1 Specimens 38

2.1.1 Specimens used for PCR Analysis 3 8

2.1.2 Specimens used for Protein Analysis 38

2.1.3 Specimens used for SSCP Analysis 38

2.2 Extraction of Total RNA from Human Breast Tissue 39

2.2.1 Reagents 39

2.2.2 Procedure 40

2.3 Estimation and Integrity Check of Total RNA 41

2.3.1 Reagents 41

2.3.2 Procedure 41

2.4 Synthesis ofcDNA from RNA Samples 42

2.4.1 Reagents 42

2.4.2 Procedure 42

2.5 Extraction of DNA from Whole Blood 43

2.6 Extraction of DNA from Human Serum 43

2.7 Extraction of Protein from Human Breast Tissues 43

2.7.1 Reagents 43

2.7.2 Procedure 44

2.8 Estimation of Total Cell Protein from Breast Tissues 44

2.8.1 Reagents 44

2.8.2 Procedure 44

2.9 Preparation of RARa DNA Competitor for Semi-Quantitative

Competitive Polymerase Chain Reaction (PCR) 45

viii

Table of Contents

2.10 Preparation of Competitor R N A for Competitive RT-PCR 47

2.10.1 Generation of Deletion RAR DNA Competitor Fragments 47

2.10.2 Cloning of Deletion RAR DNA Competitor Fragments 49

2.10.3 Construction of RAR RNA Competitors 49

2.11 PCR Detection of RARs in Human Breast Tissue cDNA 50

2.11.1 Reagents and Equipment 50

2.11.2 PCR Detection of RARs in Human Breast Tissues 51

2.11.3 Semi-Quantitative Competitive PCR Detection of RARa

in Human Breast Cancer Tissue 51

2.11.4 Competitive PCR Detection of RARs in Human Breast

Tissue 53

2.11.5 PCR Detection of RAR Polymorphisms in Human Breast

Tissue 55

2.11.6 PCR Detection of RARy Polymorphism in Serum and

Buffy Coats 59

2.12 Agarose Gel Electrophoresis of PCR Products 59

2.12.1 Reagents 59

2.12.2 Procedure 59

2.13 Preparation of Polyacrylamide Gels for SSCP Analysis 60

2.13.1 Reagents 60

2.13.2 Procedure 60

2.14 SSCP Screening and Sequence Analysis of Polymorphic RARs 61

2.15 Protein Separation on SDS-Polyacrylamide Gels (SDS-PAGE) 62

2.15.1 Reagents 62

Table of Contents

2.15.2 Procedure 63

2.16 Electroblot Transfer of Proteins onto Membranes 63

2.16.1 Reagents 63

2.16.2 Procedure 64

2.17 Immunoblotting (Western Blot) of Breast Tissue Protein for

RARs 64

2.17.1 Reagents 64

2.17.2 Procedure 65

2.18 Quantitation of RAR Competitive PCR Fragments and Protein

Bands 65

CHAPTER THREE PCR Detection of RARs in Breast Tissue 67

3.0 Introduction 68

3.1 PCR Detection of RARs in Breast Tissue 68

3.1.1 Detection of RARs in Normal and Malignant Mammary

Tissue

3.1.2 RAR Status and Tumour Grading in Mammary Carcinoma 69

3.2 Quantitation of RARs using RT-PCR 74

3.2.1 Evaluation of the Competitive RT-PCR Method 74

3.2.1.1 Exponential Amplification of RARa, p and y

mRNA 74

3.2.1.2 Linearity of Competitive RT-PCR 81

3.2.2 Comparison between Semi-Quantitative and Competitive

RT-PCR 81

Table of Contents

3.2.3 Quantitation of RARs in Normal and Malignant Breast

Tissues using Competitive RT-PCR 86

3.2.4 Comparison between RAR mRNA Levels and ER Status

in Mammary Carcinoma 92

3.3 Discussion 96

3.4 Conclusions 104

CHAPTER FOUR Analysis of Retinoic Acid Receptor Protein Levels in

Breast Tissues 106

4.0 Introduction 107

4.1 Distribution of RAR Protein Levels in Breast Tissues 107

4.2 Comparison of RAR and Oestrogen Receptor Protein Levels 111

4.3 Comparison of RAR Protein Levels and Grading of Mammary

Carcinoma Tissue 116

4.4 Discussion 118

4.5 Conclusions 121

CHAPTER FIVE Identification of Retinoic Acid Receptor Polymorphisms 122

5.0 Introduction 123

5.1 Absence of Polymorphisms in RARa and P mRNA in

Normal and Malignant Breast Tissue 124

5.2 Detection of Polymorphisms in RARy mRNA in Normal and

Malignant Breast Tissue 124

Table of Contents

5.3 Detection of RARy Polymorphism in Serum from Females with

Mammary Carcinoma 141

5.4 Detection of RARy Polymorphism in Serum and Whole Blood

from Normal Females 141

5.5 HeritabilityofRARyS427L Polymorphism 142

5.6 Discussion 144

5.7 Conclusions 151

CHAPTER SIX General Discussion 152

6.0 General Discussion 153

6.1 Future Work 159

REFERENCES 161

xii

List of Figures

LIST OF FIGURES

Figure 1.1.1 Structures of common natural retinoids 3

Figure 1.2.2.1 Metabolism of dietary retinoids 8

Figure 1.3.2.1 Basic structure of retinoic acid receptors 12

Figure 1.7.1 Schematic representation of competitive RT-PCR 35

Figure 2.9.1 Pvu II restriction map of the RARa cDNA region amplified

as described in Section 2.10.2 46

Figure 2.10.1 Restriction maps of the RAR cDNA regions amplified as

described in Section 2.11.2 48

Figure 3.2.1.1.1 Exponential amplification range for RARa using 1 OOng

total sample RNA and 40 pg competitor RNA 75

Figure 3.2.1.1.2 Exponential amplification range for RARP using 1 OOng


Figure 3.2.1.1.3 Exponential amplification range for RARy using 1 OOng


Figure 3.2.1.1.4 Stability of PCR amplification over a number of cycles for

RARa using 100 ng total sample RNA and 40 pg competitor

RNA 78

Figure 3.2.1.1.5 Stability of PCR amplification over a number of cycles for

RARP using 100 ng total sample RNA and 2 pg competitor

RNA 79

xiii

List of Figures

Figure 3.2.1.1.6 Stability of P C R amplification over a number of cycles for

RARy using 100 ng total sample RNA and 40 pg competitor

RNA 80

Figure 3.2.1.2.1 Linearity assay for RARa competitive RT-PCR 82

Figure 3.2.1.2.2 Linearity assay for RARp competitive RT-PCR 83

Figure 3.2.1.2.3 Linearity assay for RARy competitive RT-PCR 84

Figure 3.2.2.1 Semi-quantitative competitive RT-PCR of RARa using a

DNA competitor 85

Figure 3.2.2.2 Comparison between semi-quantitative and competitive

RT-PCR 87

Figure 3.2.3.1 Detection of RARa mRNA levels by competitive RT-PCR 88

Figure 3.2.3.2 Detection of RARp mRNA levels by competitive RT-PCR 89

Figure 3.2.3.3 Detection of RARy mRNA levels by competitive RT-PCR 90

Figure 3.2.3.4 Quantitation of mRNA levels by competitive RT-PCR in

normal breast and primary mammary carcinoma tissue 91

Figure 3.2.4.1 Relationship between ER status and RARa mRNA levels

in mammary carcinoma tissues 93

Figure 3.2.4.2 Relationship between ER status and RARp mRNA levels


Figure 3.2.4.3 Relationship between ER status and RARy mRNA levels


Figure 4.1.1 Detection of RARa protein in normal breast and mammary

carcinoma tissue 108

xiv

List of Figures

Figure 4.1.2 Detection of R A R p protein in normal breast and mammary


Figure 4.1.3 Detection of R A R y protein in normal breast and mammary


Figure 4.1.4 Comparison of R A R protein levels in normal breast and

mammary carcinoma tissues matched for the same patient 112

Figure 4.2.1 Relationship between R A R a protein levels and E R protein

levels in mammary carcinoma tissues 113

Figure 4.2.2 Relationship between R A R p protein levels and E R protein


Figure 4.2.3 Relationship between R A R y protein levels and E R protein


Figure 5.1.1 Representative S SCP band patterns observed for R A R a

breast tumour c D N A samples using Alphal primer set 125

Figure 5.1.2 Representative SSCP band patterns observed for R A R a

breast tumour c D N A samples using Alpha2 primer set 126





Figure 5.1.5 Representative S SCP band patterns observed for R A R a


Figure 5.1.6 Representative SSCP band patterns observed for R A R p

breast tumour c D N A samples using Betal primer set 130

' ^^ = xv

List of Figures

Figure 5.1.7 Representative SSCP band patterns observed for R A R p

breast tumour cDNA samples using Beta2 primer set 131

Figure 5.1.8 Representative SSCP band patterns observed for RARp






Figure 5.2.1 Representative SSCP band patterns observed for RARy

breast tumour cDNA samples using Gammal primer set 135


breast tumour cDNA samples using Gamma2 primer set 136







Figure 5.2.6 Band shift patterns seen for RARy in mammary carcinoma

tissues 140

Figure 5.5.1 Detection of RARy S427L polymorphism in a three-

generation family 143

Figure 5.6.1 Agarose gel showing PCR products of RARy 146

xvi

List of Figures

Figure 5.6.2 Agarose gel of RARy PCR amplification comparing

extraction methods 147

Figure 6.0.1 Model of mammary carcinogenesis due to disruptions

in RAR signalling 160

xvii

List of Tables

LIST OF TABLES

Table 1.3.3.1.1

Table 2.11.2.1

Table 2.11.4.1

Table 2.11.5.1a

Table 2.11.5.1b

Table 2.11.5.2a

Table 2.11.5.2b

Table 2.11.5.3a

Table 2.11.5.3b

Table 3.1.1.1

Table 3.1.1.2

Table 3.1.2.1

Table 4.3.1

Genes known to contain R A R E s in their promoter

regions 16

Sequences of oligonucleotide primers 52

Sequences of oligonucleotide primers for competitive

RT-PCR 54

Sequences of oligonucleotide primers for RARa 56

PCR conditions for amplification of RARa cDNA 56

Sequences of oligonucleotide primers for RARp 57

PCR conditions for amplification of RARp cDNA 57

Sequences of oligonucleotide primers for RARy 58

PCR conditions for amplification of RARy cDNA 58

Detection of RARs in normal human breast tissues 70

Detection of R A R s in primary mammary carcinomas 71

RAR status and tumour grading of mammary carcinoma

biopsies 73

Analysis of RAR protein levels and tumour grading of

mammary carcinoma biopsies 117

xviii

Abbreviations

ABBREVIATIONS

9cRA

AEV

ANOVA

APL

ARAT

ATF

atRA

CBP

CRABP

CRBP

CREB

DCIS

DEPC

DMBA

dNTP

DR

ECL

ER

FABP

GR

GAPDH

IFN

9-cw-retinoid acid

avian erythroblastosis virus

analysis of variance

acute promyelocytic leukemia

acyl coA:retinol acyltransferase

activating transcription factor

all-fran^-retinoic acid

C R E B binding protein

cellular retinoic acid binding protein

cellular retinol binding protein

cyclic A M P response element binding protein

ductal carcinoma in situ

diethylpyrocarbonate

dimethylbenz (a) anthracene

deoxyribonucleotide triphosphate

direct repeat

enhanced chemiluminescence

oestrogen receptor

fatty acid binding protein

glucocorticoid receptor

glyceraldehyde-3-phosphate dehydrogenase

interferon

xix

Abbreviations

LDL

LRAT

NADP+

NCoR

NPM

NuMA

PCAF

p/CIP

PCR

PLZF

PML

PPAR

PR

PVDF

RAR

RARE

RBP

RT-PCR

RXR

RXRE

RZR

SDS

S.E.M.

SMRT

TBE

low density lipoprotein

lecithimretinol acytransferase

oxidised nicotinamide adenine dinucleotide

nuclear corepressor protein

nucleophosmin

nuclear mitotic apparatus protein

p300/CBP associated factor

p300/CBP interacting protein

polymerase chain reaction

promyelocytic leukaemia zinc finger

promyelocytic leukaemia gene

peroxisome proliferator-activated receptor

progesterone receptor

polyvinylidene difluoride

retinoic acid receptor

retinoic acid receptor response element

retinol binding protein

reverse transcription-polymerase chain reaction

retinoid-X-receptor

retinoid-X-receptor response element

retinoid-Z-receptor

sodium dodecyl-sulphate

standard error of the mean

silencing mediator for retinoid and thyroid hormone receptors

Tris/Borate/EDTA

Abbreviations

TBST

TGF

TIF

TPA

TR

TRE

Tris

Tth

UV

VDR

Tris-buffered saline plus Tween 20

transforming growth factor

transcription intermediary factor

12-o-tetradecanoyl-phorbol-13 -acetate

thyroid hormone receptor

T P A responsive element

Tris(hydroxymethyl)methylamine

Thermus thermophilus

ultra-violet

vitamin D hormone receptor

CHAPTER ONE

LITERATURE

REVIEW

1

Chapter 1 Literature Review

1.1 HISTORICAL OVERVIEW

Retinoids are a family of biologically active molecules that consists of both

natural and synthetic analogues of vitamin A (retinol, see Figure 1.1.1a). The profound

effects of vitamin A on cellular differentiation and proliferation were discovered more

than 70 years ago. Wolbach and Howe (1925) demonstrated that rats fed a vitamin A-

deficient diet developed squamous cell metaplasia in several tissues of epithelial origin,

including epithelial cells of the respiratory, gastrointestinal and genitourinary tracts,

cornea, skin, salivary glands and prostate gland. Not only was hyperproliferation found

in some epithelia, hyperkeratinization was also observed in many tissues. However,

when these rats were fed a normal diet containing vitamin A, the squamous metaplasia

completely reverted to columnar, ciliated and mucous-secreting cells (Wolbach and

Howe, 1933).

The discovery of naturally occurring, biologically active derivatives of vitamin A added

further complexity to our understanding of the retinoid response in the body. It was

found that 11-cw-retinal (Figure 1.1.1b) was the light absorbing chromophore found in

the visual pigment, rhodopsin (Wald, 1968). Furthermore, all-fraH.s-retinoic acid (atRA)

(Figure 1.1.1c) was shown to support growth in vitamin A-deficient rats (Arens, 1946).

Although this experiment used synthetically prepared atRA, it was later discovered to

form naturally from retinol in the rat intestine and liver (Emerick, et al, 1967). Aside

from its effects on vision and growth and differentiation of a variety of cell types,

vitamin A and its derivatives are associated in a wide spectrum of biological activities,

including reproduction, metabolism, as well as serving as morphogenic agents during

embryonic development and formation of limbs (Eichele, 1989;


C H 2 O H

a. alWraws-retinol (vitamin A )

CHO b. 11-cw-retinal

COOH

c. all-fraw.s-retinoic acid

COOH

d. 9-cw-retinoic acid

Figure 1.1.1 Structures of common natural retinoids


Brockes, 1989; D e Luca, 1991; Mangelsdorf, etal, 1994). It appears that 11-ds-retinal

is uniquely involved in vision, whereas all-fraws-retinoic acid is recognised as the key

retinoid in essentially all other biological functions of vitamin A.

Although the anti-proliferative and differentiation effects of retinoids in cells

had long been known, the mechanism of action had eluded scientists until recently.

Following studies of other hydrophobic hormones in the 1970s (Evans, 1988), it was

postulated that retinoids act similarly to steroid and thyroid hormones via interaction of

specific protein receptors and segments of DNA, and regulate gene transcription.

Retinoid receptors were discovered a little over 10 years ago and the discovery of these

receptors have provided researchers a greater insight into the signalling effects of

retinoids in biological systems. Of particular interest are studies involving retinoid

receptors and their involvement in human disease, investigations which may help in

identifying causes of disease and perhaps a method of treatment and cure.

1.2 METABOLISM OF RETINOIC ACID

1.2.1 Uptake of Dietary Retinoids

Retinoids cannot be synthesised de novo in the body and are thus required as a

nutrient. They are present in the diet as either provitamin A carotenoids (e.g., P-

carotene) or preformed retinoids. There are over 600 naturally occurring retinoids, of

which approximately 50 are converted to vitamin A (Blaner and Olson, 1994). Upon

uptake of dietary retinoids in the intestine they are hydrolysed and reduced to all-trans-

retinol (vitamin A, "retinol"), presumably by pancreatic carboxyl (cholesteryl) ester


hydrolase and/or brush border-associated retinyl ester hydrolase (Rigtrup and Ong,

1992).

Most of the retinol is eventually transported from the small intestine into the

liver for storage, which is done in several ways. The majority of the retinol is esterified

in the gut wall with palmitic or stearic acid, incorporated into chylomicrons and secreted

into the lymphatic system (see Figure 1.2.2.1). Upon catabolism of chylomicrons by

lipoprotein lipase into chylomicron remnants, retinyl esters are first taken up by liver

parenchymal cells where they are further hydrolysed to retinol and re-esterified before

being stored in the stellate cells of the liver (Goodman, et al, 1965; Blaner, et al,

1987). Re-esterification occurs by two pathways. One involves the transesterification

of a fatty acyl group from the Ai-position of lecithin to retinol, which is catalysed by

lecithimretinol acyltransferase (LRAT) (MacDonald and Ong, 1988). The second

pathway involves an acyl-coenzyme A dependent esterification of retinol that is

catalysed by acyl-coA:retinol acyltransferase (ARAT) (Ross, 1982).

Retinol can also be transported in the plasma by transthyretin complexed with

retinol binding protein (RBP), a 21 kDa protein containing a single binding site

specifically for retinol (Blaner, 1989). Alternatively, p-carotene can be transported in

plasma by low density lipoprotein (LDL), and retinoic acid, which is present in very low

concentrations in blood and presumably taken up by diffusion in the intestine, can be

transported to the liver for storage by association with albumin (Ross, 1993a). Other

organs also able to store retinoids include adipose tissue, kidney, testis, lung, bone

marrow and the eye (reviewed in Blaner and Olson, 1994).

5


1.2.2 Release of Retinol into Cells

Retinol is released from storage organs and transported to target cells by binding

to the RBP-transthyretin complex in the plasma. The molecular mechanisms whereby

target cells take up retinol are unknown but presumably involve membrane receptors

specific for RBP (Bavik, et al, 1991). Retinol is a hydrophobic molecule and is present

in the cell bound to a small binding protein called cellular retinol binding protein

(CRBP). There are two types of CRBPs: CRBP-I and CRBP-II, each with a molecular

weight of approximately 15kDa but different tissue distribution (Bashor, et al, 1973;

Ong, 1984). CRBP-I appears to be widely distributed, with highest expression in liver,

kidney and proximal epididymis, while CRBP-II is found only in the small intestine

(Levin, et al, 1987). The esterification of retinol as described previously also occurs in

the presence of CRBP (MacDonald and Ong, 1988). Retinol bound to CRBP can also

undergo reversible oxidation to retinal in the presence of NADP+-dependent retinol

dehydrogenase (Posch, etal, 1991). Retinal can then be used in the visual pathway, or,

still bound to CRBP, be further oxidised to all-fran.s-retinoic acid in the presence of

retinal dehydrogenase (Posch, et al, 1992). The synthesis of retinoic acid releases

CRBP, upon which another binding protein, termed cellular retinoic acid binding

protein (CRABP), binds to retinoic acid. Retinoic acid can thus enter the nucleus and

interact with receptors to initiate transcription of target genes (discussed in next

section). Alternatively, CRABP bound to retinoic acid can be further catabolised into

more polar metabolites, which may or may not be biologically active, via the

cytochrome P450 system. Such polar metabolites include 4-hydroxy- and 4-oxoretinoic

acid, 13-cw-retinoic acid and 3,4-didehydroretinoic acid, all presumably formed by a

class of mixed-function oxidases that contain cytochrome P450 (Skare, et al, 1982;

Leo, et al, 1984; Thaller and Eichele, 1990; Eckhoff, et al, 1991). Another

6


biologically important metabolite of all-trans-retinoic acid is 9-cis-retinoic acid, which

has been shown to form by isomerization in bovine liver microsomes (Urbach and

Rando, 1994).

Similar to CRBPs, there are also two types of CRABPs, type I and II, each of

approximately 15kDa, which also differ in their tissue distribution (Astrom, et al,

1991). Both types of CRBPs and CRABPs belong to a larger gene superfamily of fatty

acid binding proteins (FABP) that encode for small proteins which function in binding,

transport and metabolism of hydrophobic ligands (Gordon, et al, 1991). As described

above, the primary function of CRBPs and CRABPs is to direct specific enzymatic

reactions involved in retinoid metabolism. Another role of these binding proteins has

been suggested in that they act as cytoplasmic buffers, by sequestering and limiting the

availability of free retinoids in cells (Ross, 1993b). A summary of the metabolism of

retinoids to the common biologically active derivative, all-^raws-retinoic acid, is

depicted in Figure 1.2.2.1.

1.3 MOLECULAR BIOLOGY OF RETINOID RECEPTORS

With the identification and characterisation of retinoid binding proteins in the 1970's, it

was suggested that CRABP-I was the mediator of the biological effects of atRA (Chytil

and Ong, 1987). It was shown that F9 embryonal carcinoma cells, which had detectable

CRABP-I, differentiated upon addition of atRA to the medium. Furthermore, mutant

embryonal carcinoma cells selected for their failure to differentiate in response to atRA

were shown to be lacking in CRABP-I protein (Jetten and Jetten, 1979; Schindler, et al,

1981). However, this theory was later refuted when it was shown that cells of the

human myelocytic leukaemia cell line, HL-60, differentiated into

7


SMALL INTESTINE Dietary retinyl esters Dietary carotenoids

PLASMA

RBP-ROH

RBP

LIVER

TARGET CELL

RBP-ROH

PLASMA

Figure 1.2.2.1. Metabolism of dietary retinoids. Abbreviations: R O H , retinol; RE, retinyl ester; C M ,

chylomicron; LPL, lipoprotein lipase; CRBP, cellular retinol binding protein; LRAT, lecithin:retinol

acyltransferase; ARAT, acyl-coA:retinol acyltransferase; RBP, retinol binding protein; ROH DHase,

retinol dehydrogenase; RAL, retinal; RAL DHase, retinal dehydrogenase; atRA, all-jVa/w-retinoic acid;

CRABP, cellular retinoic acid binding protein; RAR, retinoic acid receptor


mature granulocytes in the presence of atRA, yet did not have any detectable CRABP-I

(Breitman, et al, 1982). Moreover, stably transfected F9 teratocarcinoma stem cells

overexpressing CRABP-I protein showed an 80-90% reduction in differentiation-

specific genes upon atRA treatment (Boylan and Gudas, 1991).

Prior to the discovery of retinoic acid receptors, cloning of cDNA sequences for other

nuclear steroid hormone receptors had been accomplished. These included the

oestrogen (ER) (Green, et al, 1986), progesterone (PR) (Misrahi, et al, 1987) and

glucocorticoid receptors (GR) (Weinberger, et al, 1985a), all of which have structural

and sequence similarity with the avian erythroblastosis virus (AEV) v-erbA oncogene

(Weinberger, et al, 1985b; Krust, et al, 1986). Interestingly, the mammalian

counterpart to this oncogene is the non-steroidal thyroid hormone nuclear receptor

(Weinberger, et al, 1986). All these nuclear proteins belong to the steroid/thyroid

hormone superfamily of receptors that function as ligand-activated transcription factors,

modulating gene expression through DNA-specific binding to regulatory regions of

target genes (Evans, 1988). These receptors have a modular structure and share highly

conserved DNA-binding and ligand-binding domains. It was assumed that the effects of

retinoic acid would be mediated via nuclear receptors similar to these steroid/thyroid

hormone receptors. A screening technique was therefore devised on the basis that the

DNA-binding domain of the putative receptor, or 'orphan receptor', would be

homologous to those of the nuclear receptor superfamily.

9


1.3.1 Types of Retinoid Receptors

In 1987, two independent research groups discovered an orphan receptor that

bound with high affinity to atRA (Giguere, et al, 1987; Petkovich, et al, 1987). This

receptor, termed retinoic acid receptor alpha, or RARa, shared sequence homology in

its DNA-binding domain and ligand-binding domain with other steroid and thyroid

hormone nuclear receptors. This pioneering work provided a cornerstone for

unravelling the mechanisms of action of retinoic acid and other retinoids in cells.

Subsequently, two more nuclear receptors, RARP and RARy, were identified that bound

to atRA with high affinity. All three types of RARs are found in mammals, amphibians

and birds (de The, et al, 1987; Benbrook, et al, 1988; Brand, et al, 1988; Krust, et al,

1989; Mangelsdorf, et al, 1994). The genes for these three RARs map to different

chromosomes, and in humans localise on 17q21.1, 3p24 and 12pl3 for RARa, RARp

and RARy, respectively (Mattei, et al, 1988a, 1988b; Ishikawa, et al, 1990).

Furthermore, various isoforms exist for each type of RAR, which differ structurally only

in their amino-terminal regions. There are two isoforms for RARa (ai, ai), four for

RARP (Pi, P2, P3, P4) and two for RARy (yi, ^2). These isoforms arise by different

mechanisms: differential usage of two promoters in each RAR gene and/or alternative

splicing of the amino-terminal region of each receptor (reviewed in Leid, et al, 1992a).

The ongoing search for orphan receptors that bound to specific ligands led to the

discovery of another retinoid signalling pathway. This receptor, termed retinoid X

receptor, or RXR, was originally found to be activated by atRA (Mangelsdorf, et al,

1990). However, sequence homology revealed that its ligand-binding domain shared

less than 27% identity with that of RARa, suggesting that another retinoid must activate

this receptor. This was later shown to be the case when a natural derivative of all-trans-


retinoic acid, 9-cw-retinoic acid (9cRA) (Figure Id), was found to bind with high

affinity to RXR (Heyman, et al, 1992; Levin, et al, 1992). As for RARs, three types of

human RXRs have been identified; RXRa, RXRp and RXRy (Mangelsdorf, et al,

1990; Fleischhauer, et al, 1992; Leid, et al, 1992b) which are also found in mammals,

birds and amphibians (Mangelsdorf, et al, 1994).

Other orphan receptors have recently been discovered that are similar to RARs

and RXRs. A receptor termed LXRa was shown to activate gene expression in the

presence of 9cRA (Willy, et al, 1995). It is specifically found in kidney, intestine, liver

and spleen, which thereby restrict its function to certain tissues. This receptor was also

found to form heterodimers (discussed in Section 1.3.3) with RXRa and peroxisome

proliferator-activated receptor a (Miyata, et al, 1996). Another receptor, called RZR

(retinoid-Z receptor), was shown to be structurally similar to RARs and consist of two

types, a and p (Becker-Andre, et al, 1993; Carlberg, et al, 1994). However, retinoids

are not their cognate ligands, nor are they functionally related to RARs.

1.3.2 Structure of RARs

Similar to steroid and thyroid hormone receptors, RARs are modular in structure

and contain six domains, A-F (Figure 1.3.2.1), based on sequence homology studies

(Green and Chambon, 1988). The amino-terminal A domain is highly variable and

gives rise to the type of RAR as well as the isoform. The A/B region contains a ligand-

independent *ra«.s-activation function, termed AF-1 or x\. Domain C contains the

cysteine-rich DNA binding domain and possesses two zinc finger motifs that allow the

11


c o

• t-H

N

s • 1—1

Q

-0 p TJ

P

1 •i-H

Q

K O O u _L

o

5 \ CO

a> PH

P4

13 •»_>

57

X

B If

.1 N

tin

W

u

PQ

fe •

V?

•1

•3 CA

CO

<D CO

•c

I O

o u

CO

a • ^ H

13 f-H

P TJ o S CO

•HH

<4H

o <u

§ -a 8 fe CO

B

u U l- i

T3 p e<3 a 'W*

Q p • H H

CJ <4-l

o o

o

1 CO O

• H H

CO

PQ .

_—1

tN

c. H—1

a> _i j—j.

.2<

^ - N

a '3

a o Tj xi

93 <W

o p o "-P PH

•c O CO ta T3 T3 o •a •K

TJ (SJ IH

<s m H—1

p o •HH H->

o <u CO o -*-» I-I .u <+H <L>

fe

12


receptor to recognise specific D N A sequences. This 66 amino acid region is highly

conserved (>95% amino acid identity) among the three types of RARs. Conversely, the

central D domain is a hinge region that bears no homology between the various RARs,

and a function is yet to be determined. However, by analogy with other nuclear steroid

hormone receptors, this region may include a nuclear localization signal (Giguere,

1994). The ligand-binding domain is found in domain E and is also highly conserved,

with at least 75% amino acid homology among the RARs. This region is the largest

domain on the RAR and also contains a ligand-dependent trans-activation function,

termed AF-2 or T4 (Mangelsdorf and Evans, 1995). Furthermore, the E domain contains

a heptad repeat motif consisting of a leucine-rich region that is involved in dimerization

of these receptors (Forman, et al, 1989). The final domain, F, is at the carboxyl-

terminal end of the receptor that has an undefined function and bears little homology

with other nuclear receptors.

It is interesting to note that the amino acid sequence homology of any one type

of RAR in different species is far greater than the similarity found between each RAR

type in a given species. This, coupled with the presence of isoforms, suggests that

RARs were derived from a common ancestral receptor and that each RAR performs

separate functions (Chambon, 1994). This is further supported by observations of

differential expression patterns of RARs in various organs as well as during embryonic

development. The RARa receptor appears to be ubiquitously expressed. In the mouse

it has been found in the brain, skin, muscle, kidney, heart, lung, liver and intestine.

Furthermore, the levels of the two RARa isoforms appear to differ between these

organs (Leroy, et al, 1991a). Similarly, murine RARpi and p3 isoforms appear to be

prevalent in the lung, skin and brain, while p2 is found in kidney, liver, lung, heart,

13


muscle and brain, but at varying levels (Zelent, et al, 1991). The R A R y receptor

appears to have a restricted pattern of expression, being primarily found in the lung and

skin (Zelent, et al, 1989; Kastner, et al, 1990). However, despite these receptors being

differentially expressed, in single knockout studies of each RAR type, the defects

observed are milder than expected, suggesting that there may be some functional

redundancy of transcriptional control between RAR types and/or isoforms (reviewed in

Kastner, et al, 1995).

1.3.3 Mechanism of RAR-gene targeting

1.3.3.1 Up-regulation of Target Genes

As RARs are ligand-inducible transcription factors, they mediate their effects on gene

activation beginning with binding of atRA to the E domain of the receptor. It was

originally thought that RARs, as well as the thyroid hormone (TR) and vitamin D

(VDR) receptors, functioned as homodimers in a similar fashion to their steroid

hormone receptor counterparts. Although they are able to function as homodimers, it

was apparent that an accessory factor was required for high affinity binding of RARs,

TR and VDR to their target genes (Glass, 1994). This accessory factor was later to be

RXR, which binds heterodimerically with other nuclear receptors (RARs, TR, VDR,

peroxisome proliferator-activated receptors (PPAR)) as well as homodimerically with

itself to activate or repress target genes (Lehmann, et al, 1993), (MacDonald, et al,

1993; Mangelsdorf and Evans, 1995). Hence, RXR is commonly referred to as a master

regulator in nuclear receptor signalling pathways. However, it has been shown that

RARa can form heterodimers with VDR to mediate gene activation in response to atRA

(Schrader, etal, 1993).

14


Once the R A R / R X R heterodimer forms, it directly binds to specific D N A

sequences usually located in the promoter region of target genes. These sequences are

known as retinoic acid response elements (RAREs) for RAR/RXR complex binding,

and 9cRA response elements (RXREs) for RXR homodimer binding. Analysis of the

response element shows a direct repeat of a consensus hexad sequence of the

configuration PuGGTCA(X)„PuGGTCA, where X is a nucleotide spacer and n refers to

the number of nucleotides. This hexad repeat appears to be a common motif for other

non-steroid nuclear receptors. The number of spacer molecules is usually between 1 to

5 nucleotides and this is known as the "l-to-5" rule for non-steroid receptors of the

superfamily (Giguere, 1994). Different nuclear receptors recognise direct repeats of

specific spacing. For example, RAR/RXR heterodimers usually bind to RAREs with a

spacing of 2 or 5 bases (termed DR-2 and DR-5, respectively), while RXR homodimers

bind to RXREs that have half sites separated by one nucleotide (DR-1). This rule

however is not strict, as RAREs have been encountered with DR-1 as well as DR>5,

although they are less frequently found (Mangelsdorf, et al, 1994; Kato, et al, 1995).

For RAR/RXR heterodimers, it appears that RXR occupies the 5'-up-stream half site,

while RAR binds to the 3'-down-stream half site in DR-2 and DR-5 RAREs (Predki, et

al, 1994).

As RARs act as transcription factors for other genes, it is not surprising that

research has been done in identifying such target genes and is still an ongoing task.

There are many genes that have been discovered in humans and other animals so far that

contain RAREs in their promoter regions, although the pattern of the direct repeat

sequence may not be perfect. However, they are similar to the consensus motif. A list

of genes known to contain RAREs in their promoter region is shown in Table 1.3.3.1.1.

15


Table 1.3.3.1.1. Genes known to contain RAREs in their promoter regions

Gene

a-fetoprotein Alcohol dehydrogenase (ADH3) Alu elements Apolipoprotein(a) Apolipoprotein A-I Apolipoprotein CIII Cathepsin D

Cholesterol 7-a-hydroxylase CRABP-II CRBP-I Cytochrome P4501A1 Fructose-1,6-6r's-phosphatase H I 0 histone HoxAA Insulin-linked polymorphic region Keratin Lactoferrin LamininBl Mitochondrial uncoupling protein Oxytocin Phosphoeno/pyruvate carboxykinase Placental lactogen

R A R p Thyroid-stimulating hormone Tissue transglutaminase type II Tissue-type plasminogen activator

Reference

(Liu, etal, 199A) (Duester, etal, 1991) (Vansant and Reynolds, 1995) (Ramharack, etal, 1998) (Rottman, etal, 1991) (Lavrentiadou, etal, 1999) (Sheikh, et al, 1996)

(Crestani, etal, 1998) (Durand, etal, 1992) (Smith, etal, 1991) (Vecchini, etal, 1994) (Fujisawa, et al, 2000) (Bouterfa^a/., 1995) (Doerksen, etal, 1996) (Clark, et al, 1995) (Tomie-Canie, etal, 1996) (Lee, etal, 1995) (Vasios, etal, 1989) (Alvarez, etal, 1995) (Richard and Zingg, 1991) (Lucas, etal, 1991) (Stephanou and Handwerger, 1995)

(deThe, etal, 1990b) (Breen, etal, 1995) (Nagy, etal, 1996) (Bulens, etal, 1995)


1.3.3.2 Down-regulation of Target Genes

In addition to their effects on up-regulating target genes, RARs are also capable

of down-regulating genes through interactions with other factors. One such factor is

AP-1 (activating protein-1), which is a complex consisting of the protein products from

the protooncogenes c-jun and c-fos and activating transcription factor (ATF), in which

the Jun protein can bind as a homodimer or as a Jun/Fos heterodimer (Angel and Karin,

1991). Activation of AP-1 can be mediated with the tumour promoter TPA (12-o-

tetradecanoyl-phorbol-13-acetate) and acts as a transcription factor by binding to

specific response elements called TREs (TPA-responsive elements) in promoter regions

of target genes, in much the same way as RARs (Angel and Karin, 1991). A mutual

relationship exists between RARs and AP-1 in that treatment of cells with atRA can

inhibit the action of AP-1, as demonstrated by studies on the human osteocalcin gene

(Schule, et al, 1990). Likewise, activation of AP-1 synthesis by the phorbol ester TPA

effectively blocks the action of atRA. Closer examination of the mechanism behind this

revealed a "cross-coupling" between the two transcription factors where the promoter of

the osteocalcin gene contained an overlapping TRE/RARE site (Schule, et al, 1990).

Thus, simply occupying the response element with either AP-1 or RAR can regulate

each other's action, which is vital in preventing tumourigenesis and invasion. Indeed,

regulation of collagenase and of the metalloproteinase stromerysin, both of which

degrade many components of the extracellular matrix, occur through AP-1/RAR cross-

coupling and if left unbalanced, this may result in tumour invasion (Nicholson, et al,

1990; Schule, et al, 1991). Recently, an alternative mechanism hass been proposed,

where ligand-bound RARa disrupts Jun homodimerization or Jun/Fos

heterodimerizafion, thereby preventing the formation of AP-1 complexes. However,

17


whether this disruption occurs through direct protein-protein interaction (ie, AP-1-RAR)

or via recruitment of an accessory factor is still unclear (Zhou, et al, 1999).

Recent findings have broadened and refined our understanding of the molecular

mechanisms of RAR signalling. Currently, it is thought that RAR-RXR heterodimers

act via two opposing pathways: activation of target genes in the presence of ligand and

repression in its absence. These transcriptional responses are mediated through

coregulator proteins, which bind with the RAR-RXR heterodimer complex. In the

absence of atRA, a group of cofactors known as corepressors bind to the silencer core of

RARs, which corresponds to the hinge region (domain D) and part of the ligand binding

region (domain E) (Baniahmad, et al, 1992). Such corepressors include the silencing

mediator for retinoid and thyroid hormone receptors (SMRT), nuclear receptor

corepressor (N-CoR) and transcription intermediary factor-1 a and p (TIFla and P)

proteins (Chen and Evans, 1995; Horlein, et al, 1995; Le Douarin, et al, 1995, 1996).

These corepressors function by stabilising the chromatin structure through recruitment

of histone deacetylase into close proximity to the RAR-RXR heterodimer

(Collingwood, et al, 1999). Moreover, these corepressors form part of a multiprotein

complex which mediates gene promoter silencing by RARs (McKenna, et al, 1999).

Conversely, upon binding of atRA to RAR, other cofactors bind to the nuclear

receptor to augment transcription. Several types of coactivators have been identified

which bind to RARs and include p300/cyclic AMP response element (CREB) binding

protein (CBP), p300/CBP associated factor (PCAF) and steroid receptor-coactivator-3,

otherwise known as p/CIP (p300/CBP interacting protein) (Hanstein, et al, 1996;

Kamei, etal, 1996; Torchia, etal, 1997; Blanco, etal, 1998; McKenna, etal, 1999).

All these coactivators belong to a family of molecules related to pi60 (Collingwood, et

18


al, 1999). W h e n atRA binds to R A R , coactivators bind to the receptor, which in turn

releases corepressors from the RAR-RXR heterodimeric complex, thereby activating

gene transcription via AF-2 control (Minucci and Pelicci, 1999). Although AF-2 is

required for transcriptional activation, it is not required for dissociation of corepressors

from the silencer core (Baniahmad, et al, 1997). Unlike corepressors, these

coactivators modify the chromatin structure by recruiting histone acetyltransferase, once

they are bound to RAR (Torchia, et al, 1998). Interestingly, the pi60 family of

coactivators share a common structural feature, which is the alpha-helical LXXLL

amino acid motif and its integrity is required to mediate binding of coactivators to

ligand-bound RAR-RXR heterodimers (Heery, et al, 1997).

Although the discovery of corepressors and coactivators has defined a dual role

for RAR-RXR heterodimers, which also serves as a useful model for transcriptional

regulation, there are still questions to be answered to validate this model. For example,

it is unclear as to whether there is constitutive binding of RAR-RXR heterodimers to

their cognate RAREs in target genes, which would appear to be a logical requirement

for a dual regulatory role of these receptors involving an active or repressed state.

19


1.4 REGULATION OF RETINOIC ACID RECEPTOR SIGNALLING

As shown in Table 1.3.3.1.1, there are a large number of genes that can be

regulated by RAR/RXR heterodimers. Similar to other hormone systems, there appear

to be positive and negative feedback autoregulation mechanisms in the retinoid

signalling pathway. The genes encoding RARa, RARp and RARy are known to have

RAREs in one of their promoters (de The, et al, 1990b; Leroy, et al, 1991b; Lehmann,

et al, 1992). This implies that autoinducing the expression of RARs by retinoic acid

could lead to an amplification of the retinoid signal, which has been shown by de The

and others for the RARp gene (de The, et al, 1990b).

In addition to autoregulation of retinoid signalling at the RAR transcript level,

feedback mechanisms are also present in the atRA synthesis pathway. Most

importantly, virtually every enzyme and retinoid binding protein in retinoic acid

metabolism are potentially regulated by RARs in that they contain RAREs and/or

RXREs upstream in their genes (Mangelsdorf, et al, 1994). Thus, activation of CRBP-I

by atRA (Smith, et al, 1991) could potentiate synthesis of atRA or 9cRA. Conversely,

induction of the CRABP-II gene by RARs a, P, or y could lead to an increase in atRA

breakdown, thereby reducing the signal for atRA responses in the cell (Durand, et al,

1992), and acting as a negative regulatory mechanism in retinoid signalling.

1.5 RETINOIDS, RARs AND DISEASE

Since retinoids exert their effects in cells through processes involved in

differentiation, it is not surprising that many studies have been undertaken to ascertain

their roles in disease states, such as carcinogenesis. In tissue culture studies, retinoids


have been shown to promote differentiation of a wide variety of cells, including

keratinocytes, chondrocytes, preadipocytes, haemopoietic cells and tumour cells such as

F9 teratocarcinoma cells (reviewed in Amos and Lotan, 1990). Studies in animal

models have shown that retinoids are useful chemopreventitive agents against cancers in

different tissues such as mammary gland, skin, lung, urinary bladder, pancreas,

digestive tract, liver and prostate (reviewed in Moon, et al, 1994). Furthermore,

epidemiological studies have examined dietary retinoid uptake and its involvement in

various human cancers derived from lung, oesophagus, head/neck, stomach, pancreas,

colon/rectum, bladder, urinary tract, prostate, cervix and breast (reviewed in Hong and

Itri, 1994). In general, there appears to be an inverse relationship between intake of

retinoids and development of cancer.

The resulting in vitro studies, experiments in animal models, as well as the

epidemiological data have suggested retinoids to be a prime candidate in anti-cancer

treatment. Indeed, clinical trials have been undertaken which show that atRA induced

complete remission in patients with acute promyelocytic leukaemia, as well as showing

high response rates for 13-ci.y-retinoic acid in the treatment of squamous cell carcinomas

of the skin and repression of premalignant oral leukoplakia (Hong, et al, 1986; Huang,

et al, 1988; Lippman and Meyskens, 1989; Castaigne, et al, 1990). Furthermore,

combinations of atRA or 13-cw-retinoic acid with interferon-a-2a produced striking

response rates in patients with squamous cell carcinoma of the skin and cervix

(Lippman, et al, 1992a; Lippman, et al, 1992b). Recently, clinical trials have also

been conducted against Kaposi's sarcoma, as these cells are sensitive to atRA in vitro

(Miller, 1998).

21


Although retinoids demonstrate anti-carcinogenic effects, exceeding

physiological concentrations of atRA can lead to teratogenic results in the developing

embryo. In the mouse model, exposure to atRA during days 8-10 (post-implantation

period) can interfere with development of the brain and other craniofacial organs

(Webster, et al, 1986). Moreover, treatment with atRA later in gestation at days 11-13

led to limb, palate and genitourinary defects (Kochhar, 1973). Conversely, deficiency

(or excess) of atRA in the 1 day old chick embryo resulted in prevention of proliferation

and differentiation in mesenchymal cells, thereby prohibiting the formation of the early

vascular system (Thompson, et al, 1969).

In relation to their growth-inhibitory effects, retinoids have been shown to

induce apoptosis in various cell lines, including those derived from neuroblastoma,

breast, ovarian, lung, head and neck, cervical and malignant haemopoietic cancers

(Delia, et al, 1993; Mariotti, et al, 199A; Kalernkerian, et al, 1995; Oridate, et al,

1995, 1996; Sheikh, et al, 1995; Supino, et al, 1996). The mechanisms by which

retinoids induce apoptosis are not well understood. However, in contrast to the classical

pathway of nuclear receptor gene activation, the apoptotic effects of retinoids have been

reported to involve both RAR-dependent and RAR-independent mechanisms (Sun, et

al, 1999 and references therein). Such a receptor-independent pathway involves

retinoid activation of reactive oxygen species that has been shown to mediate apoptosis.

Caspase-3, part of the family of cysteinyl aspartate-specific proteinases, is involved in

the early stages of apoptosis and has been shown to be activated by retinoids through

induction of RARP expression, in synergism with a cyclic AMP analogue (Srivastava,

et al, 1999). Thus, further understanding of retinoids and their receptors in their

involvement in apoptosis offers clinical potential in treatment of human malignancies.

22


As R A R s are the ultimate mediators of the anti-carcinogenic effects of retinoic

acid, it is feasible to postulate that altered expression patterns of these receptors may be

involved in promotion and/or development of neoplasia. Indeed, the finding that the

integration site of the hepatitis B virus in human hepatocellular carcinoma was within

the RARp gene provided the first link between alterations in RARs and disease (Dejean,

et al, 1986; Brand, et al, 1988). Studies of RAR expression patterns in various tissues

have been performed and these patterns appear to be tissue-type specific.

In HT29 human colon carcinoma cells, growth inhibition can occur on treatment

with atRA, although this can be blocked by the RARa-selective receptor antagonist, Ro

40-6055. Growth inhibitory effects were not observed with RARp or RARy-selective

antagonists, suggesting that growth inhibition by retinoids in colon carcinoma is most

likely mediated by RARa (Nicke, et al, 1999). The growth inhibitory effect of atRA

was also found in gastric cancer cell lines, which was due to the up-regulation of RARa

mRNA, but sufficient levels of RARa were required for the observed anti-proliferative

effects (Liu, et al, 2001). Furthermore, in clinical samples, premalignant and malignant

gastric tissues had significantly lower levels of RARa expression as detected by in situ

hybridisation, compared to normal gastric mucosa. This suggests a causal role for

lowered RARa expression in gastric carcinoma (Jiang, et al, 1999). Probably the best

evidence that disruption of RAR signalling can lead to neoplasia involves acute

promyelocytic leukaemia (APL). This disease, characterised by an early block in

myeloid maturation, is associated with a specific t(15;17) chromosomal translocation

with breakpoints within the RARa gene and the PML (promyelocytic leukaemia) gene,

a putative transcription factor (Larson, et al, 1984; de The, et al, 1990a; Kakizuka, et

al, 1991). Recently, several rarer variant chromosomal translocations have been

23


identified in APLs that also involve R A R a . These include a t(ll;17) translocation

involving a fusion of RARa and a Kruppel-like zinc finger protein (PLZF -

promyelocytic leukaemia zinc finger), a t(5;17) translocation involving an RNA-binding

nucleolar phosphoprotein (NPM - nucleophosmin) and another t(ll;17) translocation

involving the nuclear mitotic apparatus protein (NuMA) (Chen, et al, 1993; Redner, et

al, 1996; Wells, et al, 1997). Molecular analysis revealed that the breakpoint in the

RARa gene resulting in the PML/RARa chimeric product occurs within the second

intron (Diverio, et al, 1992). Thus, the PML/RARa fusion protein still retains its

DNA-binding (domain C) and ligand binding (domain E) activities, which may explain

why APL patients are highly sensitive to atRA treatment. Indeed, it has been reported

that the unliganded PML/RARa fusion protein is directly involved in either inhibiting

differentiation or promoting survival of myeloid precursor cells in vitro, giving rise to

the APL phenotype (Grignani, et al, 1993). Although the t(15;17) translocation is

specific for APL and is not found in other leukaemias (Morosetti, et al, 1996), patients

who have complete remission of APL following atRA treatment can relapse even

though they are continuously treated with atRA (Warrell, 1993). This acquired atRA

resistance could be due to a progressive reduction of the retinoid concentration in the

plasma, or further mutations in the E (ligand-binding) domain of the RARa component

of the PML/RARa chimeric gene (Muindi, et al, 1992; Imaizumi M, et al, 1998).

Aberrations in the patterns of expression of RARp have been observed in

various human cancers. In histological specimens containing dysplastic and malignant

head and neck squamous cell carcinomas, there is a significant loss of expression of

RARP compared to adjacent normal and hyperplastic lesions (Xu, et al, 1994a, 1994b).

Furthermore, in oesophageal tissue, there is a gradual loss of RARp expression from

24


well-differentiated oesophageal carcinoma to moderately-differentiated and poorly-

differentiated specimens, all of which had significantly lower detection levels of RARp

than normal oesophageal tissues (Qiu, et al, 1999). This suggests that loss of RARp

expression is an early event in squamous cell carcinoma of head and neck tissues and

may be involved in its progression through mechanisms that are currently unknown.

Interestingly, studies of head and neck squamous cell carcinoma cell lines have shown

that while reduced or loss of RARP mRNA is observed, this does not correlate with

atRA-induced growth inhibition (Klaassen, et al, 2001; Zou, et al, 2001).

Furthermore, it has been shown that atRA treatment of oral squamous cell carcinoma

cell lines induced a switch from being angiogenic to anti-angiogenic, thus blocking

tumour cell migration. This did not occur by decreased levels of inducers of

angiogenesis such as interleukin-8, but rather by causing tumour cells to secrete

inhibitor(s) of angiogenesis which are presently unknown (Lingen, et al, 1996). Other

studies have shown selective suppression of RARp in human pancreatic carcinoma

tissues and cells, as well as in human lung cancer cell lines (Rosewicz, et al, 1995; Xu,

et al, 1996; Zhang, et al, 1996). This suppression may be a result of hypermethlyation

in the promoter of the RARp gene, as has been shown in gastric carcinoma tissues and

cells (Hayashi, et al, 2001), thereby rendering this receptor in an inactive state.

Some in vitro evidence has linked RARy with control of cancer. It has been

shown that BE(2)-C neuroblastoma cells overexpressing RARy had a reduced growth

rate compared to control cells (Marshall, et al, 1995). The same study also showed that

RARy mRNA levels were significantly higher in early stage neuroblastoma tissue

samples compared to advanced or disseminated tumours. Similarly, overexpression of

RARy in NT2/D1 embryonal carcinoma cells resulted in terminal differentiation that


was not observed in cells overexpressing R A R a or R A R p (Moasser, et al, 1995). A

strong correlation between RARy mRNA expression and atRA-induced growth

inhibition also has been reported in head and neck squamous cell carcinoma cell lines

(Klaassen, et al, 2001).

Thus it appears that aberrations in RARs, whether it is a simple loss of

expression or by mutation, are a frequent event in many cancers. However, the

abnormal patterns are not uniform and it seems that each RAR may be involved in

specific diseases.

1.6 RETINOIDS, RARs AND MAMMARY CARCINOMA

Retinoids have been extensively studied in terms of their possible therapeutic

benefits in human disease, and much research has been done in relation to breast cancer.

Based on the assumption that cancer is a multi-step process, beginning with an initiation

step (i.e., mutation of a target cell), followed by promotion (i.e., transformation and

immortalisation) and then progression (i.e., metastasis), administration of exogenous

materials (e.g., from diet) may be able to modify the cancer phenotype. This concept

would therefore be independent of the causation of the initiating event. Indeed,

chemoprevention studies showed a reduction of 52% in mammary carcinoma in rats

treated with the carcinogen 7, 12-dimethylbenz (a) anthracene (DMBA) and retinyl

acetate per day versus rats treated with DMBA and a placebo diet (Moon, et al, 1976).

Studies of dietary intake of preformed vitamin A compounds in 89, 494 women

showed a significantly inverse relationship with the risk of breast cancer (/?=0.03), but

not with carotenoid intake (Hunter, et al, 1993). However, in another study, it was

found that increased P-carotene intake was associated with increasingly well-

26


differentiated tumours in patients with mammary carcinoma (Ingram, et al, 1992). A

follow-up study revealed that those patients with a high intake level of P-carotene in

their diet had significantly reduced breast cancer mortality (/?=0.0093) (Ingram, 1994).

These diet studies, together with the aforementioned animal studies, have prompted

various clinical trials for the use of retinoids in treatment of human mammary

carcinoma, although not all retinoids have resulted in significant responses (Cassidy, et

al, 1982). The synthetic retinoid iV-(4-hydroxyphenyl)retinamide (fenretinide) has been

used in clinical trials because of its low toxicity compared to atRA and its preferential

accumulation in the breast rather than liver (Sporn and Newton, 1979). A large,

randomised study of 2969 women with stage I breast cancer in Milan, Italy, was

initiated in 1987 to evaluate the efficacy of fenretinide in the prevention of second

primaries in breast cancer patients (Costa, et al, 1994). This consisted of 1474 controls

(no retinoid treatment) and 1494 patients who received 200 mg of fenretinide per day

for 5 years. The analysis at a median of 97 months showed no significant differences in

the incidence of a second primary between the two groups, although a beneficial trend

was observed in premenopausal women (Veronesi, et al, 1999). Within this group of

women who were in phase III of the trial, those who received fenretinide after 180 days

had 1.73 times higher natural killer (NK) cell activity than those given a placebo (Villa,

et al, 1993). This indicates that retinoids can modulate cellular immunity and

cytotoxicity by NK cells and offers a mechanism of lowering cancer risk in patients

through natural immunoenhancement. A phase I/II trial in women with metastatic

breast cancer using the anti-oestrogen, tamoxifen, plus fenretinide showed no significant

adverse effects and improvement or stabilisation of disease in 80% of cases (Cobleigh,

et al, 1993). Recently, a phase I/II trial was reported in the use of atRA in combination

27


with tamoxifen in patients with advanced breast cancer (Budd, et al, 1998). However,

improvement or disease stability for more than 6 months was only found in 9 of 25

patients.

In addition to the additive anti-proliferative effects of atRA and anti-oestrogens

as shown in breast cancer cell lines (Wetherall and Taylor, 1986), the combination of

retinoids and interferons (IFNs) a or y produces synergistic anti-proliferative effects on

cultured breast cancer cells (Marth, et al, 1993). Furthermore, both types of IFNs were

shown to increase the expression of RARy mRNA in breast cancer cells, but only in

combination with a retinoid (Widschwendter, et al, 1995, 1996; Fanjul, et al, 1996).

The growth inhibition observed by synergism of retinoids and IFNs may be due to an

anti-angiogenic effect, as shown in T47D breast cancer cells (Majewski, et al, 1995).

It has long been known that oestrogens are important for the growth and

differentiation of normal mammary tissue, and carefully regulates these events (Topper

and Freedman, 1980). Conversely, in breast cancers which are oestrogen-dependent

(ER+), the fine-tuning of oestrogen-stimulated growth and differentiation is lost,

resulting in uncontrolled proliferation. It has been shown that atRA and other retinoids

can inhibit the growth of ER+ breast cancer cell lines, such as MCF-7, T47D and ZR-

75-1, but affects only some ER- cell lines, such as BT-20, SKBR-3 and Hs578T (van

der Burg, et al, 1993; Rubin, et al, 1994; Rishi, et al, 1996; Kazmi, et al, 1996). This

suggests that there is a correlation between retinoid growth inhibition of mammary

carcinoma and oestrogen receptor status. Indeed, the growth inhibition in breast cancer

cells by retinoids has been attributed to an anti-oestrogenic effect, whereby retinoids can

directly or indirectly inhibit binding of ER to its response element (Demirpence, et al,

1992, 1994). Furthermore, transforming growth factor a (TGFa), a potent inducer of

28


proliferation, is up-regulated by oestradiol but its synthesis is inhibited by atRA in

MCF-7 cells (Fontana, etal, 1992).

Analysis of RARs in mammary carcinoma cell lines has been extensive. Trends

in RAR expression in these cell cultures have been found in addition to the effects

observed between retinoid growth inhibition and oestrogen receptor positivity. The

levels of RARa mRNA are significantly higher in ER+ breast cancer cell lines than in

ER- cells (Roman, et al, 1992; van der Burg, et al, 1993). This was also observed in

ER+ breast cells grown in steroid-depleted media (Roman, et al, 1993; Sheikh, et al,

1993). The levels of RARP expression appears to be low in tumour cell lines compared

to normal breast cell cultures, and are predominantly found in ER- cells (Roman, et al,

1992; van der Burg, et al, 1993; Swisshelm, et al, 1994; Zhang, et al, 1996).

Interestingly, induction of RARp expression by atRA in ER+ cells correlated with

growth inhibition as well as induction of apoptosis (Liu, et al, 1996). Furthermore,

exposure of the DNA methylation inhibitor 5-aza-2'-deoxycytidine resulted in

significant anti-neoplastic activity in the ER- cell line, MDA-MB-231, and increased

expression of RARp and ERa (Bovenzi and Momparler, 2001). Expression levels of

RARy are variable in normal and breast tumour cells and are independent of ER status

(Roman, etal, 1992; Swisshelm, etal, 1994).

Oestrogen was found to up-regulate mRNA levels of RARa in a time- and

concentration-dependent manner and analysis of the RARa promoter revealed an

imperfect, half-palindromic oestrogen response element (Rishi, et al, 1995; van der

Leede, et al, 1995). However, oestradiol could not up-regulate RARp or y gene

expression (Roman, et al, 1993). Furthermore, when the ER- cell line MDA-MB-231,

a retinoid unresponsive cell, was transfected with ER cDNA, it expressed higher levels


of R A R a m R N A and its growth was inhibited by atRA (Sheikh, et al, 1993). These

authors later found that transfection of RARa cDNA into two retinoid-resistant, ER-

breast cancer cell lines (MDA-MB-231 and MDA-MB-468) led to growth inhibition by

retinoids (Sheikh, et al, 199A). Thus, the association of RARa with oestrogen action

has led authors to implicate this receptor to be predominantly involved in retinoid

inhibition of mammary carcinogenic growth.

Much of the in vitro data described above has provided many insights as to how

RARs are involved in retinoid growth inhibition of breast cancer cells. However, it is

obvious that the effects observed in this system do not necessarily reflect the in vivo

events. Therefore, studies in mammary carcinoma biopsies are required to gain a better

understanding of the involvement of RARs in human breast cancer. Unfortunately,

there is little information published in this area. Furthermore, the studies in some cases

are relatively small, which has resulted in several conflicting conclusions. For example,

in 9 ER+ and 9 ER- breast carcinoma biopsy specimens, Northern blotting revealed that

RARa mRNA was significantly higher in ER+ samples that in ER- samples (Shao, et

al, 199A). This finding was supported by another study which analysed RARa mRNA

levels in a modest sample size consisting of 94 ER+ and 22 ER- primary mammary

carcinoma biopsies, also by Northern blotting (Roman, et al, 1993). A third group,

however, analysed RARa mRNA levels, using in situ hybridisation, in 43 ER+

mammary carcinomas and 19 ER- mammary carcinomas and found no correlation

between RARa and ER status (Xu, et al, 1997). Interestingly, two independent

laboratories analysed the protein levels of RARa in breast carcinoma specimens and

arrived at different results. One group analysed 10 ER- and 9 ER+ breast cancer

biopsies using immunohistochemistry and found approximately 2-fold higher levels of


R A R a protein in E R + breast cancers than in ER- specimens (Han, et al, 1997). This

correlated with previous findings that RARa mRNA levels are higher in ER+ than ER-

mammary carcinomas. However, in another study consisting of 33 breast carcinoma

specimens, RARa protein levels were higher in tumours with greater proliferative

activity, but were independent of oestrogen status, as determined by

immunohistochemistry (van der Leede, et al, 1996). This implies that the RARa

mRNA levels may not be reflected in protein levels in breast cancer, although the data

are conflicting.

The expression pattern of RARp in breast cancer specimens indicates a loss of

its mRNA and supports the previous in vitro work. By Northern analysis in one study,

only 1 of 18 breast carcinoma samples expressed RARp mRNA (Shao, et al, 199A),

while another group found RARP mRNA in 1 of 14 mammary carcinomas using in situ

hybridisation (Widschwendter, et al, 1997). Others using a semi-quantitative RT-PCR

method (Pasquali, et al, 1997) have detected RARp mRNA at higher levels in breast

cancer samples with low ER and progesterone receptor levels. They also showed that

RARa expression was not related to ER status. However, their method involved the use

of a housekeeping gene as a reference and such genes may not always be constitutively

expressed in malignant tissue (Ostrowski, et al, 1989). This effect may be further

exacerbated when semi-quantitation is performed on RT-PCR samples. Furthermore,

only 18 breast cancer specimens were analysed. Although administration of the DNA

methylation inhibitor 5-aza-2'-deoxycytidine increased RARp expression in vitro

(Bovenzi and Momparler, 2001), hypermethlyation does not account for frequent loss of

RARp in breast tumour tissues (Yang, et al, 2001). However, Yang et al. (2001)

examined the promoter region of the RARp2 gene for the presence of methylation,


while others have detected methylation in the first exon of RAJRJ32 (Widschwendter, et

al, 2000). This suggests that RARp gene silencing by methylation may occur at

multiple sites.

The expression level of RARy mRNA appears to be high in breast biopsy

specimens from patients with mammary carcinoma and is independent of ER status

(Shao, et al, 199A; Xu, et al, 1997). A slight decrease of RARy expression was

observed in mammary carcinomas compared to normal breast tissues, but was

insignificant, probably due to the low sample size used in the study (Widschwendter, et

al, 1997). Surprisingly, only one study has looked at whether mutations in RARs exist

in human malignant breast cancer specimens (Widschwendter, et al, 1997). However,

the study analysed only the RARp promoter region and found no mutations in this area

of the gene in 14 breast cancer biopsies. Thus, clinical information of RAR expression

patterns in breast cancer is scant and most of the available data may be unreliable given

the low sample sizes used in such studies. Clearly, more work is needed in this area of

research.

1.7 RESEARCH PROPOSAL, AIMS AND HYPOTHESIS OF THIS STUDY

There is extensive information in the literature gathered from in vitro

experiments on retinoic acid receptors and their involvement in breast cancer.

However, there is little information of in vivo perturbations of RARs in human breast

cancer, and most of what is available is statistically unreliable due to the low number of

samples studied. To date, only two papers have been published that investigated the

status of RARs in normal breast tissue (Widschwendter, et al, 1997; Xu, et al, 1997)

using the semi-quantitative technique of in situ hybridisation. There has been very little


work done in analysing R A R protein levels in normal and malignant breast tissue (van

der Leede, et al, 1996; Han, et al, 1997), and the results obtained are contradictory.

Moreover, there is no information on RARp and y protein levels in human breast

tissues. Analysis of RAR mutations in mammary carcinoma is virtually non-existent,

yet they are present in other neoplastic diseases, such as acute promyelocytic leukaemia

(Larson, et al, 1984).

The hypothesis of this thesis is that decreased expression of RARs and the

presence of polymorphisms in their mRNA are associated with the development

and/or progression of mammary carcinoma.

Therefore the aims of this study are:

1. To determine the expression patterns and levels of mRNA for RARa, p and y in

a large number of malignant breast tissue samples obtained from patients with

mammary carcinoma, and to compare these findings in normal breast specimens.

2. To analyse protein levels of RARa, P and y in a group of patient-matched

normal breast and mammary carcinoma specimens.

3. To screen for any polymorphisms in the coding sequences of RARa, P and y.

To gain a better understanding of the role of RARs in human breast cancer, breast tissue

specimens were used rather than focusing on in vitro studies of established mammary

carcinoma cell lines. Studies on mRNA expression patterns were performed using

reverse transcription-polymerase chain reaction (RT-PCR), and mRNA expression

33


levels quantitated by competitive RT-PCR (Becker-Andre and Hahlbrock, 1989). In

this technique, a synthetic internal RNA standard was used as a reference, in that its

sequence is recognised by the same PCR primers used to amplify the wild-type mRNA,

thereby "competing" for all reagents used in the RT-PCR process (Figure 1.7.1). This

in effect controls for variations in the reaction efficiencies in both the reverse

transcription and PCR amplification processes that occur across reaction samples, thus

providing a highly sensitive and accurate method of RNA quantitation. Data obtained

from these experiments were correlated with classical phenotype markers of mammary

carcinoma, including oestrogen receptor (ER) status and tumour grade, to determine if

any associations exist.

Analysis of the RAR protein levels was performed by Western immunoblotting

and results were compared with the mRNA levels. As proteins are the ultimate

mediators of RAR signalling, it is important to determine if RAR protein levels

correlate with trends observed in RAR mRNA levels between normal and mammary

carcinoma tissues. Screening for polymorphisms in RARs in mammary carcinoma was

achieved using the single-strand conformation polymorphism technique followed by

sequencing of polymorphic bands. Polymorphisms were also assessed in a normal

population and their frequencies compared with those found in mammary carcinoma.

The data presented in this thesis provides further understanding of RARs and

their involvement in mammary carcinoma by assessing changes in their expression

level, correlations with other disease indicators, and associations with polymorphisms in

retinoic acid receptors. This information may be a useful springboard for further studies

on the aetiology and treatment of breast cancer.

34


wild-type c D N A 5 '- •3'

5'-wild-type R N A (target)

•3'

deletion

5- • 3' competitor D N A

m v.7ro transcription

! •

5- • 3' competitor R N A

reverse transcription

3'-target c D N A

•5'

3-competitor c D N A

•5'

I P C R amplification 5'H

primer 1

3'

1-3'

primer 2 5'

primer 1

primer 2

•• agarose gel electrophoresis

target

competitor

CHAPTER TWO

MATERIALS

AND

METHODS

Chapter 2 Materials And Methods

2.1 SPECIMENS

2.1.1 Specimens used for PCR Analysis

Fresh breast tissue specimens were obtained from PathCentre, Sir Charles

Gairdner Hospital and from the Mount Hospital, West Perth, Western Australia, which

included eight uninvolved breast tissues taken from mastectomies that were

macroscopically and microscopically regarded as normal. A total of 140 primary

mammary carcinomas were examined, with oestrogen receptor status determined by

routine screening in the Department of Clinical Biochemistry, PathCentre. Screening

for oestrogen receptors was performed by the ABBOTT ER-EIA monoclonal antibody

kit, according to the maufacturer's instructions (ABBOTT Laboratories, IL, USA). All

tissues were frozen as soon as possible following surgical excision and stored at -70°C.

For competitive PCR analysis, 59 primary mammary carcinomas were obtained

from the Mount Hospital, West Perth, Western Australia. These samples also had

oestrogen receptor levels assayed by routine screening in the Department of Clinical

Biochemistry, PathCentre. Twelve normal breast samples obtained from PathCentre

were also used in this study.

2.1.2 Specimens used for Protein Analysis

Frozen breast tissue specimens were obtained from the tumour bank provided by

the University of Western Australia Department of Pathology, Sir Charles Gairdner

Hospital. Eight patients provided matched histologically normal and mammary

carcinoma tissue. Oestrogen receptor levels were also obtained from tumour tissue in

these specimens, performed by routine screening in the Department of Clinical

Biochemistry, PathCentre as noted in 2.1.1.

38


2.1.3 Specimens used for SSCP Analysis

Fresh breast tissue specimens were obtained from PathCentre, Sir Charles Gairdner

Hospital, including eleven from mastectomies that were macroscopicaliy and

microscopically regarded as normal. Seventy-two primary mammary carcinomas were

also examined. All tissues were frozen immediately following surgical excision and

stored at -70°C prior to RNA processing. Whole blood was taken from 21 human

females who had no clinical breast cancer at the time of collection and were regarded as

normal. A further 40 samples of serum taken from another group of normal females,

which were stored frozen for 6 years, were used to match the number of tumour tissues

studied. To determine if polymorphisms were hereditary, a volunteer family was

selected consisting of a father and mother, three daughters (one married) and two

granddaughters and a grandson (see Figure 6.4.1a). No members of this family were

diagnosed with disease at the time of obtaining blood and were regarded as normal.

2.2 EXTRACTION OF TOTAL RNA FROM HUMAN BREAST TISSUE

2.2.1 Reagents

Solution D:

4 M guanidinium isothiocyanate

25 mM sodium citrate, pH 7

0.5% 7V-lauroylscarcosine (Sarcosyl)

0.1 M P-mercaptoethanol

2 M Sodium acetate, pH 4

DEPC-treated water-saturated phenol

49:1 chloroform:iso-amyl alcohol

Isopropanol


2.2.2 Procedure

Tissue homogenates were prepared by pulverising snap-frozen slices of specimens in a

microdismembrator. Approximately 250 mg of frozen breast tissue was cut into thin

slices with a sterile scalpel blade, with care taken to avoid fatty tissue, which was

yellow in contrast with white breast tissue. The slices were placed into a clean metal

chamber, frozen under liquid nitrogen, and then treated for 25 s in a "Mikro-

Dismembrator U" apparatus (B. Braun Biotech International). The powdered tissue was

placed into a sterile lOmL polypropylene tube and kept in ice. Total RNA was

extracted by the guanidinium isothiocyanate method as described by Chomzsynski and

Sacchi (1987). Basically, 2 mL solution D was added to the pulverised tissue, gently

vortexed, then 0.2 mL 2 M sodium acetate, pH 4, was added, followed by 2 mL DEPC-

treated water-saturated phenol and 0.4 mL 49:1 chloroform:iso-amyl alcohol. The tube

was briefly vortexed and chilled on ice for 15 min. Tubes were centrifuged at 10, 500 *

g for 30 min at 4°C. The upper aqueous phase was removed and placed into another

sterile centrifuge tube. An equal volume of isopropanol (~2 mL) was added to the tube

and RNA was precipitated at -20°C for 1 hr. Tubes were again centrifuged at 10, 500 x

g for 30 min at 4°C. The supernatant was carefully removed and the pellet was

dissolved in 0.3mL solution D, which was then transferred into an Eppendorf tube. An

equal volume of isopropanol was added and RNA was re-precipitated at -20°C for 1 hr.

Tubes were centrifuged at 7000 x g for 10 min at 4°C. The supernatant was removed

and the pellet was freeze-dried for 8 min before storage at -70°C.

40


2.3 ESTIMATION AND INTEGRITY CHECK OF TOTAL RNA

2.3.1 Reagents

10x MOPS:

0.2 M 3-(AT-morpholino)propanesulfonic acid (MOPS)

0.05 M sodium acetate (anhydrous)

0.01MNa2EDTA

Solution is made up with DEPC-treated water and then autoclaved before use.

Sample buffer:

lxMOPS

20% Formaldehyde, pH >4.0

50% Deionised formamide

Loading dye:

1 mM EDTA, pH 8.0

0.25% (w/v) bromphenol blue

50% (w/v) glycerol

(this buffer is made up in DEPC-treated water)

1 mg/mL ethidium bromide, made up in DEPC-treated water

2.3.2 Procedure

Estimation of RNA was performed by dissolving the RNA pellet in sterile, DEPC-

treated water, and a 1/100 dilution used for quantitation. Using standard

spectrophotometric methods, lOOuL of diluted RNA was placed into a quartz micro-

cuvette and its absorbance read from a Carey UV-Vis spectrophotometer (Varian,

Victoria, Australia) to derive its concentration. Total RNA was quantitated assuming

41


that 1 O D equals 40 |ig/mL (Sambrook, et al, 1989). All R N A samples used in this

study had an A260/A280 ratio of at least 1.75.

The integrity of each RNA sample was checked by denaturing formaldehyde-agarose

gel electrophoresis on 1.2% agarose gels containing 1 x MOPS, 6% formaldehyde and

0.5 uL 1 mg/mL ethidium bromide. Samples were mixed in a 1:4 ratio of RNA: sample

buffer and incubated at 55°C for 15 min. Loading dye was added in a 16.6% v/v, and

samples were electrophoresed at 60V for 90 min in a 1 x MOPS running buffer.

Samples were used in the study if both the 28S and 18.9 rRNA bands were observed.

2.4 SYNTHESIS OF cDNA FROM RNA SAMPLES

2.4.1 Reagents

5* first strand synthesis buffer

100 mM dithiothreitol (DTT)

200 U/uL Moloney Murine Leukaemia Virus (M-MLV) reverse transcriptase

500 ug/mL oligo(dT)i5

40 U/uL recombinant RNasin® (Promega Corp) Ribonuclease Inhibitor

25 mM dNTPs

2.4.2 Procedure

The cDNA synthesis reaction was broadly based on the manufacturer's instructions

(GIBCO BRL, Melbourne, Australia). Each reaction contained 200 ng total RNA, 10

ng oligo(dT)i5, lx first strand buffer, 10 mM DTT, 0.5 mM dNTPs, 8 U RNasin and 40

U M-MLV reverse transcriptase. For the competitive PCR study, an additional amount

of competitor RNA was added and the total volume of sterile water was adjusted.

42


Tubes were incubated at 37°C for 1 hr, followed by an enzyme denaturation step at

90°C for 5 min, and tubes were then stored at 4°C prior to PCR analysis.

2.5 EXTRACTION OF DNA FROM WHOLE BLOOD

Fresh whole blood was centrifuged at 1200 x g for 10 min. Buffy coats were

removed for proteinase K digestion followed by phenol/chloroform/iso-amyl alcohol

treatment to extract DNA (Sambrook, et al, 1989). Following ethanol precipitation,

DNA was re-dissolved in sterile nuclease-free water and stored at 4°C.

2.6 EXTRACTION OF DNA FROM HUMAN SERUM

Extraction of DNA from serum was performed by heating five lots of 20 uL aliquots of

patient serum at 100°C in 0.5 mL microcentrifuge tubes for 10 min. Tubes were then

centrifuged at 13 500 x g for 3 min. The supernatant was removed from each tube and

pooled into a single 0.5 mL microcentrifuge tube. Ten microlitres of supernatant was

used for PCR analysis.

2.7 EXTRACTION OF PROTEIN FROM HUMAN BREAST TISSUES

2.7.1 Reagents

Lysis buffer:

50 mM Tris-HCl, pH 7.6

150 mM NaCl

2mMNa2EDTA

1% Nonidet P-40

10% bacitracin

43


10 mg/mL aprotinin

1 mg/mL leupeptin

2.7.2 Procedure

Tissue homogenates were prepared by microdismembration of frozen breast tissue as

described in Section 2.2.2. The pulverised tissues were placed in sterile 2 mL screw-

capped tubes, kept on ice and 1 mL of lysis buffer was added per tube containing 0.05%

bacitracin, 10 u.g aprotinin and 10 ug leupeptin. Tubes were left on ice for 30 min with

occasional vortexing, followed by centrifugation at 13, 500 x g for 10 min at 4°C. The

supernatant was carefully removed and placed into a sterile 1 mL Eppendorf tube and

stored at -20°C.

2.8 ESTIMATION OF TOTAL CELL PROTEIN FROM BREAST TISSUES

2.8.1 Reagents

Coomassie Brilliant Blue solution.

per litre:

100 mg Coomassie Brilliant Blue G-250

50 mL 95% ethanol

100 mL 85% phosphoric acid

-filter through Whatman No. 1 filter paper

10 mg/mL Bovine serum albumin (BSA) standard (prepared in lysis buffer, see Section

2.7.1)

2.8.2 Procedure

Estimation of protein was performed by the Bradford assay (Bradford, 1976).

The range of BSA used in quantitation of breast tissue protein was 1.25 mg, 3.75 mg,

44


5.00 mg and 10.00 mg. Each protein standard was prepared in lysis buffer as outlined

in Section 2.7.1 and 1 uL was added to 99 uL double-deionised water for the assay. For

each tissue sample protein extract, 1 uL was used and added to 99 uL double-deionised

water. For all tubes, 1 mL Coomassie stain was added and allowed to stand at room

temperature for 5 min before quantitation. Protein was quantitated at A595 by the Carey

UV-Vis spectrophotometer (Varian, Victoria, Australia) in glass cuvettes. This

spectrophotometer automatically creates a standard curve based on the absorbance of

the BSA standards and automatically calculates the protein concentration of each tissue

sample based on this standard curve.

2.9 PREPARATION OF RARa DNA COMPETITOR FOR SEMI

QUANTITATIVE COMPETITIVE POLYMERASE CHAIN REACTION

(PCR)

Complementary DNA (cDNA) was reverse-transcribed (refer to Section 2.4) from RNA

extracted from a culture of MCF-7 cells, provided by co-workers in the laboratory. The

cDNA was used as a template for RARa PCR using the primers shown in Table

2.11.2.1 to yield a 827 bp fragment, with amplification conditions described in Section

2.11.2. The PCR product was then digested with Pvu II (Promega, Sydney, Australia)

resulting in four fragments, as shown on Figure 2.9.1. Following separation of the

fragments by gel electrophoresis, the 5' and 3' fragments of the original 827 bp PCR

product were isolated from the gel. This was performed by cutting the desired bands

out of the gel, elution of DNA through sterile glass wool and centrifugation at 13 500 x

g, followed by ethanol precipitation (Sambrook, et al, 1989). These two fragments

were blunt-end ligated together with T4 DNA ligase (Promega, Sydney, Australia) to


Pvu II Pvu II Pvu II 715 836 1046

_ J I I 607 1436

Figure 2.9.1 Pvu II restriction map of the R A R a c D N A region amplified as described

in Section 2.10.2. The nucleotide positions are in reference to that given in Giguere, et

al. (1987).


give a 496 bp fragment. After P C R using the primers for R A R a as shown on Table

2.11.2.1, this competitor DNA fragment would yield a 174 bp product. The competitor

was further purified by the Wizard® Purification System (Promega, Sydney, Australia)

for future use in semi-quantitative competitive PCR reactions.

2.10 PREPARATION OF COMPETITOR RNA FOR COMPETITIVE RT-

PCR

2.10.1 Generation of Deletion RAR DNA Competitor Fragments

Before synthesising competitor RNA molecules, it was necessary to construct

deletion DNA fragments for each receptor. For RARs a and y, the DNA template was

prepared by synthesising cDNA (refer to Section 2.4) from RNA extracted from the

MCF-7 cell line, while cDNA was synthesised from RNA extracted from the Hs058T

cell line to obtain the RARP DNA template. All RNA from cell lines was acquired

from co-workers in the laboratory. The PCR conditions used to obtain the RAR DNA

templates are described in Section 2.11.2. For RARa, the deletion fragment was

prepared by digesting the PCR product with Pst I and Sma I (both from Promega,

Sydney, Australia), which gave rise to 3 fragments, as shown on Figure 2.10.1.1a. This

was followed by 5' end-filling of the Sma I sticky-end fragment using Klenow fragment

DNA polymerase (Promega, Sydney, Australia). The 5' and 3' fragments were isolated,

purified and ligated as described in Section 2.9. For RARs p and y, the deletion

47


a) RARa

Pst I Sma I 1156 1309 I I

607 1436

b) RARp

Pvu II Pvu II 844 912 I I

800 1571

c) RARy

Pvu II Pvu II 1032 1077

930 1517

Figure 2.10.1.1 Restriction maps of the RAR cDNA regions amplified as described in

Section 2.11.2. a) Cut sites used in generation of RARa RNA competitor, b) Pvu II

sites used in generation of RARP RNA competitor, c) Pvu II sites used in generation of

RARy RNA competitor. The nucleotide positions are in reference to that given in

Giguere, et al. (1987), de The, et al. (1987) and Krust, et al. (1989) for RARs a, p and

y, respectively.


fragment was prepared by digesting the P C R product with Pvu II (Promega, Sydney,

Australia), each of which gave rise to 3 fragments as shown on Figures 2.10.1.1b and c.

Similar to RARa, the 5' and 3' fragments were isolated, purified and ligated as

described in Section 2.8. Using the PCR conditions described in Section 2.11.4, the

deletion DNA fragments yielded the following sizes: RARa: 674 bp; RARp: 255 bp

and; RAR: 482 bp. The competitors were further purified by the Wizard® Purification

System (Promega, Sydney, Australia) prior to cloning.

2.10.2 Cloning of Deletion RAR DNA Competitor Fragments

Each purified RAR competitor DNA fragment was inserted into the nMOSBlue

T-vector (Amersham Pharmacia Biotech, Sydney, Australia) according to the

manufacturer's instructions. These vectors were transformed into JM101 E. coli

competent cells and successful transformants were selected by blue/white colony

screening (Sambrook, et al, 1989). One colony for each receptor was picked and

directly amplified under PCR conditions described in Section 2.10.4 to confirm the

presence of the deletion competitor fragment. These colonies were then cultured

overnight at 37°C in LB broth containing ampicillin and tetracycline. Plasmid DNA

was extracted from culture by the Wizard® Plasmid Purification System (Promega,

Sydney, Australia) and then linearized with^va I restriction enzyme (Promega, Sydney,

Australia). The linearized plasmid was then purified by phenol/chloroform extraction

(Sambrook, etal, 1989).

2.10.3 Construction of RAR RNA Competitors

The pM0S5/we T-vector contains a T7 promoter site 5' from the multiple

cloning site where the RAR competitor DNA fragments were inserted. Using the TNT®

Coupled Reticulocyte Lysate System (Promega, Sydney, Australia), RNA was

49


transcribed from the linearized plasmid according to the manufacturer's instructions.

Following treatment with DNase I, RNA was extracted as described in Section 2.2 and

quantitated as described in Section 2.3. Specificity of the RNA was confirmed by

reverse transcribing the RNA under PCR conditions described in Section 2.11.4 before

it was used for competitive PCR.

2.11 PCR DETECTION OF RARs IN HUMAN BREAST TISSUE cDNA

2.11.1 Reagents and Equipment

10 x reaction buffer (supplied with Tth Plus DNA Polymerase):


166 mM (NH4)S04

4.5% Triton X-100

2mg/mL gelatin

25 mM magnesium chloride (supplied with Tth Plus DNA Polymerase)

5.5 U/uL Tth Plus DNA polymerase (Fischer-Biotech, Perth, Australia)

10 x reaction buffer (supplied with AmpliTaq Gold DNA Polymerase):


500 mM KC1

AmpliTaq Gold™ DNA polymerase 5 U/uL (Perkin-Elmer, Melbourne, Australia)

1 U/uL Perfect Match® PCR Enhancer (Stratagene, California, USA)

5 mM dNTPs (Promega, Sydney, Australia)

10 pmol/uL PCR primers - refer to each section

MJ-Research PTC-60 Thermal Cycler with hot bonnet (Geneworks, Sydney, Australia)

Sterile wax beads:

50


prepared personally using autoclaved paraffin wax and an electronic automatic step-

pipettor to pipette 20 uL volume beads.

Nuclease-free sterile water

2.11.2 PCR Detection of RARs in Human Breast Tissues

Specific oligonucleotide regions used to detect RARs a, P and y are shown on

Table 2.11.2.1. These primers spanned large introns and therefore would not amplify

genomic DNA. A reaction mix of 30 uL was used to amplify cDNA, which consisted

of: 10 uL cDNA, 0.2 mM dNTP, 1.25 mM MgCl2, 1 x reaction buffer, 1.65 U Tth plus

DNA polymerase (Fischer-Biotech, Perth, Australia) and 5 pmol of primers. The PCR

cycles began with 4 min at 94°C, 1 min at 70°C (63°C for RARs p and y) and 2 min at

72°C. This was followed with 35 cycles of 1 min at 94°C, 1 min at 70°C (63°C for

RARs p and y) and 2 min at 72°C, and a final extension step at 72°C for 5 min. As a

positive control, cDNA synthesised from total RNA extract of the MCF-7 cell line was

used for RARs a and y, whereas cDNA synthesised from total RNA extract of the

Hs578T cell line was used for RARp. The RNA from these mammary carcinoma cells

lines were provided by co-workers in the laboratory. A tube containing the PCR

cocktail plus nuclease-free water and no sample was used in each run as a blank control.

2.11.3 Semi-quantitative Competitive PCR Detection of RARa in Human Breast

Cancer Tissue

Estimation of RARa expression levels was done using a semi-quantitative

competitive PCR technique. In this approach, the PCR cocktail was spiked with 7uL of

1:1000 RARa competitor DNA as described in Section 2.9. The PCR mix was identical

to that as described in Section 2.11.2 for RARa, except that 1 mM MgCl2 and the

51


Table 2.11.2.1. Sequences of Oligonucleotide Primers

c D N A Transcript* Sequence (5' -» 3') Genomic Location 609-626 1436-1419

800-817 1571-1552

930-947 1517-1500

Fragment Size (bp)

827

771

587

RARa sense GCC CAA GCC CGA GTG CTC anti-sense CTA CAG CTG CCT GGC GGG RARP sense AGG AGA CTT CGA AGC AAG anti-sense GTC AAG GGT TCA TGT CCT TC RARy sense GGA AGA AGG GTC ACC TGA anti-sense CGG CGC CGG GCG TAC AGC

* R A R a sequences were derived from Giguere, et al. (1987) R A R p sequences were derived from de The, et al. (1987) RARy sequences were derived from Krust, et al. (1989)


following primers were used per reaction; sense: 5'- G G A A G A A G G G T C A C C T G A

-3' and anti-sense: 5'- CTC CGC AGA TGA GGC AGA TGG -3'. The amplification

cycle was as described in Section 2.11.2. A tube containing the PCR cocktail plus

nuclease-free water and no sample or competitor was used in each run as a blank

control. For quantitation, lOuL of PCR sample was loaded on a 12% polyacrylamide

gel and stained with ethidium bromide following electrophoresis at 100V for 2 hr. The

polyacrylamide gel was then used for image analysis, with the sample RARa fragment

migrating 505 bp and the competitor DNA fragment at 174 bp. The amount of RARa

in each sample was expressed as a ratio and calculated by measuring the intensity of the

sample fragment relative to the intensity of the competitor DNA PCR product (refer to

Section 2.18).

2.11.4 Competitive PCR Detection of RARs in Human Breast Tissue

Using this quantitative PCR method for detection of RAR levels in breast tissue,

known amounts of competitor RNA are added into tubes containing a constant amount

of sample RNA for cDNA synthesis. During the cDNA synthesis step, 100 ng sample

RNA was placed into 4 separate cDNA reaction tubes. For RARa and p quantitation,

600 fg, 800 fg, 5 pg or 10 pg competitor RNA was added into those tubes, whereas for

RARy, 10 pg, 30 pg, 50 pg, or 70 pg was used. Each PCR reaction mix consisted of the

following components: 0.2 mM dNTP, 2.0, 2.7 or 3.2 mM MgCl2 for RARa, p or y,

respectively, 1 x reaction buffer and 10 pmol of primers. The primers used for

competitive PCR are shown on Table 2.11.4.1. For RARa and p, 1.65 U Tth plus DNA

polymerase (Fischer-Biotech, Perth, Australia) was used, whereas for RARy, 1 U

AmpliTaq Gold DNA polymerase was used with the appropriate reaction buffer and

underwent a 10 min hot-start as instructed by the company. Also, 0.3 U Perfect Match®


Table 2.11.4.1. Sequences of Oligonucleotide Primers for Competitive RT-PCR

cDNA Transcript*

Sequence (5' —> 3') Genomic Fragment Competitor Location Size (bp) Size (bp)

RARa sense GCC CAA GCC CGA GTG CTC anti-sense CTA CAG CTG CCT GGC GGG RARp sense AGG AGA CTT CGA AGC AAG anti-sense TGC AAA TTC TAA GAA TCA GGA TG RARy sense GGA AGA AGG GTC ACC TGA anti-sense CGG CGC CGG GCG TAC AGC

609-626 1436-1419

800-817 1123-1101

930-947 1517-1500

827

323

587

674

255

482

* RARa sequences were derived from Giguere, et al. (1987)

RARp sequences were derived from de The, et al. (1987)

RARy sequences were derived from Krust, et al. (1989)


was used for R A R a to increase specificity and incorporated the use of wax beads for the

hot-start procedure. As a result of this, all reagents and sample minus the polymerase

and sterile water were added in the RARa PCR mix before placing a wax bead into

each tube. Tubes were then heated at 94°C for 1 min in the thermal cycler followed by

a ramp-down to 18°C for 2 min to re-solidify the paraffin wax. The remaining volume

of sterile water plus thermostable DNA polymerase was then added on top of the wax

layer. The PCR reaction consisted of the following cycles: 4 min at 94°C, 1 min at 66°C

for RARa (63°C for RARs p and y) and 2 min at 72°C. This was followed by 35 cycles

(34 for RARp) of 1 min at 94°C, 1 min at 66°C for RARa (63°C for RARs p and y) and

2 min at 72°C, and a final extension step at 72°C for 5 min. In each run, a blank control

was used containing nuclease-free water and PCR reaction mix.

2.11.5 PCR Detection of RAR Polymorphisms in Human Breast Tissue

Specific oligonucleotide primers were designed to detect the entire coding

regions of RARs a, P and y, which are shown on Tables 2.11.5.1a, 2.11.5.2a and

2.11.5.3a, respectively. A reaction mix of 25uL was used for PCR amplification, which

consisted of lOuL cDNA, 1 x reaction buffer, 0.2 mM dNTPs, 1.1 U Tth plus DNA

polymerase (Fischer-Biotech, Perth, Australia) and 5 pmol of primers. The

concentration of MgCl2 varied depending on the primer pair and is detailed in Tables

2.11.5.lb, 2.11.5.2b and 2.11.5.3b. The PCR reaction began with an initial hot start at

94°C for 5 min prior to adding Tth plus DNA polymerase (except for RARy). This was

followed by 35 cycles at 94°C for 1 min, with variable annealing temperature (see

Tables 2.11.5.1b, 2.11.5.2b and 2.11.5.3b) for 1 min and 2 min at 72°C. As a positive

control, cDNA synthesised from total RNA extract of the MCF-7 cell line was used for

RARs a and y, whereas cDNA synthesised from total RNA extract of the Hs578T cell


Table 2.11.5.1a. Sequences of Oligonucleotide Primers for R A R a Primer Sequence (5' -»3') Coding

position* Fragment size (bp)

AlphalF AlphalR Alpha2F Alpha2R Alpha3F Alpha3R Alpha4F Alpha4R Alpha5F Alpha5R

5'-ACT GTG GCC GCT TGG CAT GGC C-3' 5'-CTT GTA GAT GCG GGG TAG AGG GG-3' 5'-GCC TTG CTT TGT CTG TCA GGA CAA G-3' 5'-GCG TGT AGC TCT CAG AGC ACT CGG-3' 5'-GGA GCT CAT TGA GAA GGT GCG-3' 5'-GCG TGT ACC GCG TGC AGA TCC-3' 5 '-CCG AGC AGG ACA CCA TGA CCT T-3' 5'-CCG CAC GTA GAC CTT TAG CGC C-3' 5'AAG CGG AGG CCC AGC CGC CCC CAC A-3' 5'-CGG TCA CGG GGA GTG GGT GGC CGG GCT-3'

87->108 338->360 360->384 617->640 657->677 917->937 938->959 1173->1194 1195->1219 1468-^1494

274

281

281

257

300

Coding position corresponds with the D N A position number published in Giguere, et al. (1987).

Table 2.11.5.1b. P C R Conditions for Amplification of R A R a c D N A Primer — — LMgCl2J used

in PCR (mM) Annealing temperature

Alphal Alpha2 Alpha3 Alpha4 Alpha5

0.5 0.8 1.5 0.8 0.5

68°C 68°C 66°C 68°C 68°C


Table 2.11.5.2a. Sequences of Oligonucleotide Primers for R A R P Primer Sequence (5' ->3') Coding

position* Fragment Size (bp)

BetalF BetalR Beta2F Beta2R Beta3F Beta3R Beta4F Beta4R Beta5F Beta5R

5'-GAT C A T GTT T G A CTG TAT G G A TGT T-3' 5'- TGT CCT GGC AGA CGA AGC AGG GT-3' 5 '-ATC ATC AGG GTA CCA CTA TGG-3' 5 '-CAA GTC AGC TGT CAT TTC ATA-3' 5'-GAC GAT CTC ACA GAG AAG AT-3' 5 '-TGC AAA TTC TAA GAA TCA GGA TG-3' 5 '-CCA GGT ATA CCC CAG AAC AAC A-3' 5 '-GTG AGG CTT GCT GGG TCG TCT T-3' 5'-ATG TTT CCA AAG ATC TTA ATG AA-3' 5'-ATG TCT TAT TGC ACG AGT GGT GA-3'

318->342 558->581 582->602 832-»852 853->872 1101̂ .1123 1124-̂ 1145 1395->1416 1417̂ -1439 1651->1673

264

271

271

293

257

* Coding position corresponds with the D N A position number published in de The, et al. (1987).

Table 2.11.5.2b. Primer

Betal Beta2 Beta3 Beta4 Beta5

PCR Conditions for Amplification [MgCl2] used inPCR(mM)

1.4 2.0 2.7 2.0 1.4

of RARP cDNA Annealing temperature

64°C 64°C 64°C 64°C 56°C


Table 2.11.5.3a. Sequences of Oligonucleotide Primers for RARy Primer Sequence (5'-> 3') Coding

position* Fragment Size (bp)

Gamma IF GammalR Gamma2F Gamma2R Gamma3F Gamma3R Gamma4F Gamma4R Gamma5F Gamma5R

5'-TGC CAT G G C CAC CAA TAA G G A GCG-3' 5'-GGC TTG TAG A C C CGA G G A G G C GGA-3' 5'-ATG CTT CGT GTG CAA TGA CAA GTC C-3' 5'-CTG TCA G G T G A C CCT TCT TCC TTC A-3' 5'-CTA TGA GCT G A G CCC TCA GTT A G A A-3' 5 '-TCT AGG CAG GCA GCT TTG AGC AGA GT-3' 5'-TAT CCT GAT GCT GCG TAT CTG CAC A-3' 5 -TCC AGC AGT GGC TCC TGC AGC TT-3' 5'-AGC CCT GAG GCT GTA CGC CCG-3' 5'-CTG GTC AGG CTG GGG ACT TCA-3'

411-̂ 434 657->680 681-V705 926->950 951->975 1195-̂ 1220 1221->1245 1468-̂ 1490 1491-»1511 1763̂ -1783

270

270

270

270

293

Coding position corresponds with the D N A position number published in Krust, et al. (1989).

Table 2.11.5.3b. P C R Conditions for Amplification of RARy c D N A Primer [MgCl2] used Annealing

in P C R (mM) temperature Gamma 1 Gamma2 Gamma3 Gamma4 Gamma5

"OX 0.8 0.8 0.8 0.8

"63T" 65°C 65°C 65°C 65°C


line was used for RARp. A tube containing the PCR cocktail plus nuclease-free water

and no sample was used in each run as a blank control.

2.11.6 PCR Detection of RARy Polymorphism in Serum and Buffy Coats

Ten microlitres of serum supernatant or luL buffy coat DNA was used for PCR

analysis. The primers used to amplify RARy were as follows; sense: 5'-AGA TGG

AGA TTC CAG GCC CGA TGC-3'; antisense: 5'-CTG GTC AGG CTG GGG ACT

TCA-3'. Each 25uL PCR mix consisted of 1 x reaction buffer, 5 pmol primers and 0.2

mM dNTPs. Samples underwent a 5 min hot start at 94°C prior to adding 1.1 U Tth

plus DNA polymerase, followed by a 60 cycle amplification run comprising of 94°C for

1 min, 67°C for 1 min and 72°C for 2 min. A blank containing PCR mix without

sample DNA (substituted with nuclease-free water) was used in each run.

2.12 AGAROSE GEL ELECTROPHORESIS OF PCR PRODUCTS

2.12.1 Reagents

5 mg/mL Ethidium bromide

Bromphenol Blue Loading Dye:

0.25% Bromphenol blue

25% Ficoll Type 400 in H20

10 x Tris/Borate/EDTA Buffer (TBE)

0.89 M Tris-base

0.89 M Boric acid

0.2 M EDTA-Na2, pH 8.0

123 bp DNA Standard ladder (GIBCO-BRL, Melbourne, Australia)

59


2.12.2 Procedure

Electrophorese 5 uL of PCR product plus 2 uL of bromphenol blue loading dye in a

1.2% agarose gel containing 0.15ug ethidium bromide, in 1 x TBE buffer using a

BioRad Wide Mini-Sub Cell GT (BioRad, NSW, Australia) tank. Electrophorese at

100V for 1 hr. For competitive PCR products, 10 uL of sample was loaded plus 2 uL

bromphenol blue loading dye in a 2% agarose gel containing 0.15 ug ethidium bromide,

in the same type of gel tank. Gels were electrophoresed at 100V for 50 min for RARa

and P, and for 1 hr 20 min for RARy. All gels contained lug of 123 bp DNA ladder.

The gel was then visualised on an UV transilluminator in the dark room. The image

was captured by a Kodak DC 120 digital camera (cl- Amersham Pharmacia Biotech,

NSW, Australia), saved on the computer and printed out on an Epson Stylus 800

(Epson, NSW, Australia) inkjet printer using photo-quality paper.

2.13 PREPARATION OF POLYACRYLAMIDE GELS FOR SSCP ANALYSIS

2.13.1 Reagents

35:1 acrylamide:Af, AT-fo's-acrylamide solution

Glycerol

10 x Tris/Borate/EDTA Buffer (TBE) -refer to section 2.12.1

10% ammonium persulphate

N, N, N', AT-tetramethylethylenediamine (TEMED)

2.13.2 Procedure

A 13% polyacrylamide gel with 5% glycerol was prepared as described by

Ausubel, et al. (1994). Gel solutions were poured onto a Mini-Protean II gel system

(BioRad, NSW, Australia) and allowed to set for at least one hour.

60


2.14 SSCP SCREENING AND SEQUENCE ANALYSIS OF POLYMORPHIC

RARS

Denaturation of the RAR PCR products to produce single stranded DNA for SSCP

analysis was performed by diluting 2uL PCR product with 8uL formamide loading

buffer (95% formamide, 10 mM sodium hydroxide, 0.05% bromphenol blue) and

heating at 95°C for 10 min. Denatured samples were then loaded on a 13% non-

denaturing polyacrylamide solution (35:1 acrylamide: N, Ar-to-acrylamide) containing

5% glycerol. Electrophoresis was carried out in cold-room temperature at 4°C under

constant current of 60mA for 3 hr, with a change of cold 1 x TBE running buffer at the

half-way point of the run. Detection of DNA bands was achieved by silver staining

based on a method described by Soong and Iacopetta (1997). This involved soaking

each gel in 10% ethanol for 3 min, 1% nitric acid for 3 min, followed by 3 x 10 sec

washes with Milli-Q water and then soaking in staining solution (0.05 g silver nitrate

and 75 uL formaldehyde in 50 mL Milli-Q water) for 10 min. Following further 3x10

sec washes with Milli-Q water, bands were observed by soaking in developing solution

(1.5 g sodium carbonate, 75 uL formaldehyde and 50 uL sodium thiosulphate (2

mg/mL) in 50 mL Milli-Q water) for 2-3 min and fixed in 10% acetic acid for 5 min.

Gels were finally soaked for 15 min in a 30% methanol/3% glycerol solution before

dried overnight using the GelAir drying system (BioRad, NSW, Australia).

Polymorphic bands seen on the gels were repeated for confirmation by re-synthesising

cDNA from the relevant patient's RNA, followed by PCR amplification using the

appropriate primers. Confirmed polymorphic bands were then excised with a clean

scalpel blade. Gel slices were destained by washing five times in a 0.5mL solution of

2.5% sodium thiosulphate and 1% potassium hexacyanoferrate for 5 min each.

61


Following a final rinse with distilled water, 30uL distilled water was added and D N A

was eluted from the gel by heating at 80°C for 15 min (Hara, et al, 1994). A luL

aliquot was then re-amplified in a PCR reaction mix using primers specific for that RAR

region of interest to enrich the polymorphic sequence. After checking for amplification

by agarose gel electrophoresis, the remaining PCR sample was purified using the

Wizard® PCR Preps DNA purification system (Promega, Sydney, Australia), and then

subjected to direct sequencing in an ABI Prism automated DNA sequencing system

(Perkin-Elmer, Melbourne, Australia).

2.15 PROTEIN SEPARATION ON SDS-POLYACRYLAMIDE GELS (SDS-

PAGE)

2.15.1 Reagents

4 x Tris-HCl/SDS. pH 6.8:

5 mM Tris-base, pH 6.8

0.4 % sodium dodecyl sulphate (SDS) (w/v)

4 x Tris-HCl/SDS. pH 8.8:

0.375 M Tris-base, pH 8.8

0.4% SDS (w/v)

37.5:1 (40%) acrylamide N, AT-Zn's-acrylamide solution (BioRad, NSW, Australia)

10% ammonium persulphate

N, N, N', TV'-tetramethylethylenediamine (TEMED)

5 x SDS electrophoresis buffer:

0.125 M Tris-base

0.96 M glycine


0.5% SDS

2 x SDS sample buffer

lx Tris-HCl/SDS, pH 6.8

20% glycerol

4% SDS

0.001% bromphenol blue

P-mercaptoethanol

Rainbow™ coloured protein molecular weight markers (Amersham Pharmacia Biotech,

Sydney, Australia)

2.15.2 Procedure

Preparation of the SDS/polyacrylamide gel was adopted from the methods

described by Ausubel, et al (1994). An 8% stacking/resolving gel was used over a

3.9% stacking gel. For protein separation, 50 ug of sample was mixed with 2 x SDS

sample buffer and P-mercaptoethanol was added freshly to a final concentration of 5%

v/v. For each gel, 7 uL marker was used and treated with sample buffer and P-

mercaptoethanol as stated previously. In order to compare protein levels across gels on

a semi-quantitative basis, 50 |ig of one protein sample was loaded as a standardising

control onto every gel. This sample was previously checked by Western blotting to

ensure it had expressed all RAR proteins, and aliquots were made to prevent protein

degradation over time. Prior to electrophoresis, the protein mixes (including standard)

were denatured by heating to 100°C for 10 min. After loading all samples onto the gel,

empty lanes were loaded with blank sample buffer to ensure even running. Gels were

electrophoresed at 100V for 2 hr 45 min in cold-room temperature of 4°C, using 1 x

SDS electrophoresis buffer.

63


2.16 ELECTROBLOT TRANSFER OF PROTEINS ONTO MEMBRANES

2.16.1 Reagents

1 x transfer buffer:

25 mM Tris-base

192 mM glycine

20% methanol

0.5% Ponceau S stain (in 1% acetic acid)

2.16.2 Procedure

Preparation of the electroblotting tank using the BioRad (NSW, Australia) mini

blotter was carried out as described by Ausubel, et al. (1994). Protein on the gel was

electroblotted onto Hybond-P (PVDF) membranes (Amersham Pharmacia Biotech,

Sydney, Australia) at either 380 mA for 45 min, or 30 V overnight at 4°C, using cold 1 x

transfer buffer. Control of loading was checked by soaking membranes in 0.5%

Ponceau S stain for 5 min, followed by several washes with Milli-Q water and then

visualised for uniformity of load. The membrane was destained further in Milli-Q water

for 5 min prior to immunoblotting.

2.17 IMMUNOBLOTTING (WESTERN BLOT) OF BREAST TISSUE

PROTEIN FOR RARs

2.17.1 Reagents

10 x TBST:

0.1 M Tris-base

1.5 M NaCl

0.2%) Tween 20 (polyoxyethylene sorbitol monolaureate) " " 64


Blocking solution:

5 % non-fat milk powder in 1 x TBST

ECL detection system (Amersham, Sydney, Australia)

2.17.2 Procedure

The PVDF membranes containing separated breast tissue protein were blocked

for 1 hr with blocking solution before incubation with antibody. The RAR polyclonal

antibodies, obtained from Santa Cruz Biotechnology, Inc., Ca, USA, were all derived

from rabbit and the following dilutions were used: RARa: 1:2000 (C-20, sc-551);

RARP: 1:500 (C-19, sc-552); and RARy: 1:500 (C-19, sc-550). Primary antibodies

were incubated with the membranes in blocking solution for 1 hr, followed by 3 x io

min washes with blocking solution. A 1:5000 goat anti-rabbit IgG conjugated with

horseradish peroxidase secondary antibody (Sigma, NSW, Australia) was incubated

with the membrane in blocking solution for 1 hr, followed by 3 x 10 min washes in 1 x

TBST buffer.

Specific binding of primary antibodies was visualised by the enhanced

chemiluminescence (ECL) detection system (Amersham, Sydney, Australia) after

exposure to Hyperfilm (Amersham, Sydney, Australia) for 5 min.

2.18 QUANTITATION OF RAR COMPETITIVE PCR FRAGMENTS AND

PROTEIN BANDS

All competitive PCR gels and Western blot film/membranes were quantitated on

an IBM PC computer using the freeware program Scion Image, version Beta 4.0.1, with

the 'Gel plot 2' macro setting in accordance to the software developer's instruction

manual. For quantitation of competitive PCR, the intensity of the two fragments was

65


measured and expressed as a ratio of sample R A R (target) to competitor R A R intensity.

After plotting the ratio of target: competitor versus amount of competitor RNA, the

amount of RAR in the sample RNA was determined where the ratio equalled one.

Samples that were not in the range of this ratio were appropriately diluted (or, the

amount of competitor used was lowered) and underwent another competitive RT-PCR

run as described in section 2.11.4. The final amount of receptor was adjusted by a

correction factor to account for the difference in size of the competitor/target. The

correction factors are as follows: RARa: 0.815; RARp: 0.789; RARy: 0.821. All results

was statistically analysed by InStat software, v3.01 (GraphPad Software, Inc.).

66

CHAPTER THREE

PCR DETECTION

OF RARs IN

BREAST TISSUE

Chapter 3 P C R Detection of R A R s in Breast Tissue

3.0 Introduction

Considerable in vitro research into the associations between RARs and breast

cancer has been published, but surprisingly, there is little information on breast cancer

biopsies. As a result of this and the use of small sample groups, current clinical

research is conflicting, with some data confirming established in vitro results, while

other have been contradictory. This chapter provides more data on RARs in clinical

breast cancer biopsy samples, by analysing using RT-PCR a large sample of ER+ and

ER- mammary carcinoma biopsies and comparing the RAR status in these tissues with

normal breast tissues. Comparative analyses using normal mammary tissue are also

lacking in the literature. The hypothesis that is being proposed is that RARs are present

in mammary carcinoma, but their mRNA expression levels are lower in neoplastic than

normal tissues, thereby resulting in abrogated retinoid signalling. Furthermore, the

RAR status in mammary carcinoma biopsy samples has been compared with ER status

and tumour grade to determine any associations, as in vitro studies have shown that ER

and RARs, particularly RARa, are correlated (Roman, et al, 1992; Sheikh, et al, 1993;

van der Burg, et al, 1993). The development of a semi- and a fully quantitative RT-

PCR technique are also described and documented thus providing an extremely

sensitive and specific method for determining RAR mRNA expression levels in breast

tissue samples.

3.1 PCR Detection of RARs in Breast Tissue

3.1.1 Detection of RARs in Normal and Malignant Mammary Tissue

The presence of RARs in breast tissues has been detected by RT-PCR (see

Section 2.11.2), producing fragment sizes of 827 bp, 771 bp and 587 bp, for RARs a, p

68


and y, respectively. One hundred and forty primary mammary carcinomas, of which 97

were found to be ER+ and 43 ER-, and 18 normal breast tissue samples were analysed

in this study. The expression pattern of RARs in normal tissues is shown in Table

3.1.1.1, while the patterns observed in primary mammary carcinomas is shown in Table

3.1.1.2. In all the tissue categories, RARa was present in at least 70% of samples, with

15/18 in normal breast, 69/97 in ER+ mammary carcinomas and 37/43 in ER-

mammary carcinomas. There was no significant difference in the presence of RARa in

normal and neoplastic tissue, irrespective of ER status (p=0.3687, Fisher's exact test).

RARP mRNA was detected in 13/18 normal breast tissue samples, 45/97 ER+

mammary carcinomas and 23/43 ER- mammary carcinomas, with no difference

between tumour and normal tissue (p=0.0193, Fisher's exact test). Similarly, there was

no significant difference between the presence of RARa or p and oestrogen receptor

status (p=0.0861 and 0.4680 for RARa and p, respectively, by Fisher's exact test) in

mammary carcinomas. This suggests that the expression of oestrogen receptor is

independent of RARa or p expression. However, there was a significant absence of

RARy mRNA in mammary carcinomas compared to normal tissues (p=0.0018, Fisher's

exact test). Moreover, the prevalence of RARy was significantly less in ER+ tumours

than in ER- tumours (p=0.0016, Fisher's exact test), which indicates an inverse

relationship between oestrogen receptor status and the presence of RARy.

69

Chapter 3 P C R Detection of RARs in Breast Tissue

Table 3.1.1.1. Detection of RARs in Normal Human Breast Tissues

Receptor

RARa RARP RARy

RAR Positive

15/18 13/18 18/18

%

89 72 100

70

Chapter 3 PCR Detection of RARs in Breast Tissue

Table 3.1.1.2. Detection of RARs in Primary Mammary Carcinomas

Receptor

R A R a RARP RARya

R A R Positive (ER+) %

69/97 71 45/97 46 57/97 59

R A R Positive (ER-) %

37/43 86 23/43 53 37/43 86

a Fisher's exact test,/?=0.0016 between ER+ and ER- mammary carcinomas


3.1.2 RAR Status and Tumour Grading in Mammry Carcinoma

Of the 140 primary mammary carcinoma biopsies studied, histopathology

records were available for 63 samples. From these, 6 were classified as Grade I, 23

were Grade II and 34 were Grade III infiltrating ductal carcinomas by the Nottingham

grading system (Elston and Ellis, 1991). The distribution of RARs among these

tumours is shown on Table 3.1.2.1. RARa was detected in 3/6 Grade I, 18/23 Grade II

and 33/34 Grade III mammary carcinomas. RARp mRNA was found in 1/6 Grade I,

12/23 Grade II and 15/34 Grade III mammary carcinomas. There were no significant

relationships between the presence of RARa or RARP and tumour grade. However,

there was a significant difference in RARy expression in Grade II tumours (6/23)

compared to Grade III tumours (24/34) (p=0.0013, Fisher's Exact test). When Grade I

and Grade II tumour categories were combined (i.e., 8/29 Grades I and II compared

with 24/34 Grade III), this was also significant (p=0.001, Fisher's Exact test),

suggesting that there is a low prevalence of RARy in low-grade tumours compared to

high-grade tumours.

72


Table 3.1.2.1. R A R Status and Tumour Grading of Mammary Carcinoma Biopsies

RARa

RARp

RARy0

GRADE I

3/6

1/6

2/6

%

50

17

33

G R A D E II

18/23

12/23

6/23

%

78

52

26

G R A D E III

33/34

15/34

24/34

%

97

44

71

a Fisher's Exact Test,p=0.0013 between Grade II and III tumours


3.2 Quantitation of R A R s Using R T - P C R

3.2.1 Evaluation of the Competitive RT-PCR Method

3.2.1.1 Exponential Amplification of RARa, p and y mRNA

In order to quantitate mRNA levels by competitive PCR, the exponential range

of the PCR reaction before the plateau phase must be determined for each receptor, so

that there is a linear relationship between the amount of competitor amplified compared

to the amount of target amplified. For each receptor, 100 ng of total RNA from a

patient known to express each RAR was co-reverse-transcribed with 40 pg, 2 pg or 40

pg of RARa, P or y competitor RNA, respectively. For RARa and y, the cDNA

samples were amplified for 31 to 40 cycles, whereas for RARp, the sample was

amplified for 29 to 40 cycles, with PCR conditions described in Section 2.11.4.

Following separation by agarose gel electrophoresis and measuring the intensity of each

band, these values were plotted against the number of cycles amplified. The graphs

obtained from these values for RARa, P and y, respectively, are shown in Figures

3.2.1.1.1, 3.2.1.1.2 and 3.2.1.1.3. The exponential range of RARa amplification was

found between 32 and 39 cycles and the ratio of target/competitor remained stable

within that range, as shown on Figure 3.2.1.1.4. For RARP, the exponential range fell

between 31 and 36 cycles and the target/competitor ratio remained stable between 31

and 35 cycles, as shown on Figure 3.2.1.1.5. The exponential range for RARy was

determined to be between 31 and 37 cycles. However, the ratio of target/competitor

was only stable between 31 and 36 cycles (Figure 3.2.1.1.6) before the amplification

rate of the competitor decreased relative to the rate of target amplification. From these

results, the number of cycles used for competitive PCR of RARs in breast tissues was

determined to be 35 cycles for RARa and y, and 34 cycles for RARp.

74


31 32 33 34 35 36 37 38 39 40

w 1100

(0

<

200

100

0

» W MTypeRARa

I CompetiDrRAR a

4

31 32 33 34 35 36 37 38 39 40

Number of Cycles

Figure 3.2.1.1.1. Exponential amplification range for RARa using 100 ng total sample

RNA and 40 pg competitor RNA. Both RNA species were co-reverse-transcribed and

then amplified for various numbers of cycles. Following agarose gel electrophoresis,

the bands were scanned and their intensities measured. These values were plotted on

the y-axis and expressed as arbitrary units.

75


29 30 31 32 33 34 35 36 37 38 39 40

Figure 3.2.1.1.2. Exponential amplification range for R A R P using 100 ng total sample





76


31 32 33 34 35 36 37 38 39 40

31 32 33 34 35 36 37 38 39 40

Number of Cycles

Figure 3.2.1.1.3. Exponential amplification range for RARy using 100 ng total sample





tn

co

n I. <


1.4 1

1.2

° 1 ;M *3 &0.8 E o O 0.6 Si CU S> 0.4 (0 1-

0.2

o|

^& ' — # — •

J 1 1 1 1 1 1

A •

1

A •

- — K —

4 •

— i

31 32 33 34 35 36 37 N u m b e r of Cycles

38 39 40

Figure 3.2.1.1.4. Stability of P C R amplification over a number of cycles for R A R a

using 100 ng total sample RNA and 40 pg competitor RNA. Both RNA species were

co-reverse-transcribed and then amplified for various numbers of cycles. Following

agarose gel electrophoresis, the bands were scanned and their intensities measured. The

ratio of target/competitor was plotted on the y-axis.


29 30 31 32 33 34 35 36 37 38 39 40

Number of Cycles

Figure 3.2.1.1.5. Stability of PCR amplification over a number of cycles for RARp

using 100 ng total sample R N A and 2 pg competitor RNA. Both R N A species were co-

reverse-transcribed and then amplified for various numbers of cycles. Following



79


fc.

o CJ) Q.

E o

o 1-•5

1

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

31 32 33 34 35 36 37 Number of Cycles

Figure 3.2.1.1.6. Stability of P C R amplification over a number of cycles for R A R y

using 100 ng total sample RNA and 40 pg competitor RNA. Both RNA species were

co-reverse-transcribed and then amplified for various numbers of cycles. Following




3.2.1.2 Linearity of Competitive RT-PCR

As a result of the size difference between the target and competitor template, the

amplification kinetics of DNA synthesis may be slightly different between these two

cDNA species, especially when the amount of each starting material is different. It was

therefore necessary to assess a range of concentrations of competitor against a constant

amount of target RNA for which there was a linear relationship. For each receptor, 100

ng was co-reverse transcribed with decreasing amounts of competitor RNA. Figure

3.2.1.2.1 shows a linear response for RARa RT-PCR with constant target RNA and 0.5,

1, 3, 5, 7 and 10 pg competitor RNA (^=0.9371). In Figure 3.2.1.2.2, a linear

relationship was observed for RARP amplification over the range of 0.2, 1, 3, 5, 7 and

10 pg competitor RNA (^=0.9491). For RARy, a linear range was observed with 5, 10,

30, 50, 70 and 100 pg competitor RNA (^=0.9461), as shown in Figure 3.2.1.2.3.

These results demonstrate that the competitive PCR reaction for each receptor can

reliably detect the amount of RAR over a wide range of concentrations.

3.2.2 Comparison between Semi-Quantitative and Competitive RT-PCR

A semi-quantitative competitive PCR method was developed for RARa as described in

Section 2.11.3. This method allowed quantitation of RNA relative to a DNA competitor

of known concentration, which was constant for every sample assayed. Analysis was

performed on 59 breast tissue samples and Figure 3.2.2.1 illustrates representative

samples using semi-quantitative competitive RT-PCR, separated out on a 12% non-

denaturing polyacrylamide gel. The amount of RARa determined by this method was

expressed as an arbitrary unit, calculated as a ratio of the intensity of sample RARa

against the intensity of the competitor fragment. To validate this method, the 59 breast

81


0 2 4 6 8 10

Amount of Competitor (pg)

Figure 3.2.1.2.1. Linearity assay for R A R a competitive RT-PCR. A constant 100 ng

total R N A was co-reverse-transcribed with 0.5, 1, 3, 5, 7 and 10 pg competitor RNA.

Following amplification and gel electrophoresis, the bands were quantitated and

recorded and expressed as a ratio of competitor/target. A linear relationship was

observed over this range of competitor concentration (^=0.9371).


1.6 J

1.4

1.2

petitor/Target

p

bo

-»•

E 0.6 o

o 0.4

0.2

0 I •/

•

•

• /

i

/ •

i i i

0 2 4 6 8 10


12

Figure 3.2.1.2.2. Linearity assay for R A R p competitive RT-PCR. A constant 100 ng

total R N A was co-reverse-transcribed with 0.2, 1, 3, 5, 7 and 10 pg competitor R N A .

Following amplification and gel electrophoresis, the bands were quantitated and


observed over this range of competitor concentration (^=0.9491).


0 20 40 60 80 100


120

Figure 3.2.1.2.3. Linearity assay for RARy competitive RT-PCR. A constant 100 ng

total R N A was co-reverse-transcribed with 5, 10, 30, 50, 70 and 100 pg competitor

RNA. Following amplification and gel electrophoresis, the bands were quantitated and


observed over this range of competitor concentration (r2=0.9461).


Figure 3.2.2.1. Semi-quantitative competitive RT-PCR of R A R a using a D N A

competitor. Samples were taken from human mammary carcinoma biopsies and

underwent PCR using a constant amount of DNA competitor as described in Section

2.11.3. Samples were run on a non-denaturing 12% polyacrylamide gel. Top band

represents the target (sample) fragment and the lower band the competitor fragment.

85


tissue samples also underwent competitive RT-PCR for RARa by the methods

described in Section 2.11.4. Figure 3.2.2.2 shows a plot of the values determined by

both competitive RT-PCR methods. There was no correlation between the two methods

(^=0.0051, p=0.592A, Pearson's correlation), which showed that this semi-quantitative

method, although requiring fewer reaction steps, gave highly variable results compared

to the more stringent competitive RT-PCR method.

3.2.3 Quantitation of RARs in Normal and Malignant Breast Tissues using

Competitive RT-PCR

The expression of RAR mRNA was examined in 59 mammary carcinoma

biopsy samples by competitive RT-PCR as described in Section 2.11.4. In addition, 12

normal breast tissues were analysed in this study and all tissues demonstrated

expression of all three RARs as determined in Section 3.1.1. A representative

competitive RT-PCR can be seen on the gel photographs shown in Figures 3.2.3.1,

3.2.3.2 and 3.2.3.3 for RARa, P and y, respectively, using constant amount of target

RNA co-reverse-transcribed with increasing amount of competitor RNA. Analysis of

the amount of RARs found in normal breast tissue, compared to primary mammary

carcinomas, is summarised as a semi-logarithmic plot in Figure 3.2.3.4. For RARa, the

mean value (± S.E.M.) found in normal tissues was 2.27 ± 0.55 pg, while in tumour

tissues, the mean value was 3.66 ± 0.50 pg. No significant difference was found for

RARa expression in normal and tumour breast tissues using the 2-tailed Mann-Whitney

test (p=0.AA16). The average level of RARp mRNA in normal tissues was 4.72 ± 1.05

pg and was significantly higher compared to the levels in tumour tissues, which was

0.72 ± 0.08 pg (pO.0001, Mann-Whitney test). For RARy, the mean level of

86


20000

18000

E

3 6000

0.00 1.00 2.00 3.00 4.00 5.00 6.00

Seml-Quantltatlve Values (Arbitrary Unit)

7.00 8.00 aoo

Figure 3.2.2.2. Comparison between semi-quantitative and competitive RT-PCR.

Human mammary carcinoma biopsies taken from 59 patients underwent both PCR

methods as described in Sections 2.11.3 and 2.11.4 for RARa. Following quantitation,

values were plotted on a scattergram to assess the correlation between the two methods.

No such correlation was found, as ^=0.0051 and there was no significance by Pearson's

test (p=0.592A).


Figure 3.2.3.1. Detection of R A R a m R N A levels by competitive RT-PCR.

Representative sample separated on a 2% agarose gel containing ethidium bromide.

Top band in each lane represents the sample (target) fragment, whereas the lower band

represents the competitor fragment with increasing amount from left to right of the gel,

as described in Section 2.11.4.

88


Figure 3.2.3.2. Detection of R A R p m R N A levels by competitive RT-PCR.





89


Figure 3.2.3.3. Detection of R A R y m R N A levels by competitive RT-PCR.






R A R a R A R p RARy

Figure 3.2.3.4. Quantitation of RAR mRNA levels by competitive RT-PCR in normal

breast and primary mammary carcinoma tissue. Bar graphs correspond to mean mRNA

levels and error bars representing standard error of the mean. No significant change

was observed for RARa (p=0.AA16, Mann-Whitney test) between normal and tumour

tissue. There were significant decreases in expression of RARp and y (pO.0001 and

/?=0.001 for RARp and y, respectively, Mann-Whitney test) in tumour tissue compared

to normal breast tissue.


expression in normal breast tissue was 110.93 ±31.27 pg, whereas in tumour tissue, the

mean level was 30.02 ± 2.05 pg. There was also a significant difference between the

levels of RARy expression in normal and tumour breast tissues (p=0.00l, Mann-

Whitney test). Thus, it appears that there is lowered expression of RARp and y in

mammary carcinoma.

3.2.4 Comparison between RAR mRNA Levels and ER Status in Mammary

Carcinoma

Oestrogen receptor status had been determined on the 59 mammary carcinoma

specimens used for competitive RT-PCR analysis of RARs and comparisons were made

between these two receptors. Scatter diagrams comparing ER status and RARa, p and y

mRNA levels are shown in Figures 3.2.4.1, 3.2.4.2 and 3.2.4.3, respectively.

Statistically significant positive correlations were observed with ER status and RARa

(r2=0.2647, /?<0.0001) and y (r2=0.1325, p=0.005A) mRNA levels, by Pearson's

correlation. There was no relationship between RARp mRNA levels and ER status by

Pearson's correlation (^=0.0127,^=0.4042).

92


0 50 100 150 200 250 300 350 400 450

Amount of ER (fmol/mg protein)

Figure 3.2.4.1. Relationship between E R status and R A R a m R N A levels in mammary

carcinoma tissues. The amount of RARa in each tissue was determined using

competitive RT-PCR and the ER levels assayed routinely as described in Section 2.1.1.

A significant correlation was found between ER status and RARa mRNA levels using

Pearson's correlation test (r2=0.2647,/?<0.0001).


3000

2500

< z DC

•a 2000

I e o gl 1500

5 1000 o E <

500

50 100 150 200 250 300


350 400 450

Figure 3.2.4.2. Relationship between E R status and R A R p m R N A levels in mammary

carcinoma tissues. The amount of RARp in each tissue was determined using


No significant correlation was observed between ER status and RARp mRNA levels

using Pearson's correlation test (r^O.0127,^0.4042).


0 50 100 150 200 250 300 350 400 450


Figure 3.2.4.3. Relationship between E R status and RARy m R N A levels in mammary

carcinoma tissues. The amount of RARy in each tissue was determined using


A significant correlation was observed between ER status and RARy mRNA levels

using Pearson's correlation test (r =0.1325,/?=0.0054).


3.3 Discussion

Although there has been considerable research on the expression of RARs in

human breast tissue, progress in this area has been limited, primarily because tumour

biopsies are small in size and the extraction of RNA may sometimes be difficult.

Furthermore, assays such as Northern blotting require a substantial amount of RNA and

may not be as sensitive as RT-PCR, which was the method of choice in this study. One

can rule out the possibility that the PCR results presented here were obtained from

contaminating genomic DNA, as all PCR primers were designed to span introns. These

primers were also tested for amplification of human genomic DNA, which either

produced a faint fragment of larger size, or failed to amplify (data not shown). Normal

breast tissues excised distant from the site of the tumour were also used in this study

instead of "normal" tissue located at the immediate vicinity of the neoplasm. This was

done to avoid the possibility of any neoplastic cells contaminating the samples used for

the RNA extraction process, which may have influenced results of the PCR analyses.

The use of different molecular biology techniques in the literature may have contributed

to the conflicting data over the status of RARs in mammary carcinoma, although this

may also be due in part to low sample size used in some studies.

The results presented here show no significant change in RARa expression in

mammary carcinoma compared to normal breast tissue. There also appears to be no

significant relationship between ER status and the occurrence of RARa expression,

which is in agreement with a previous study using in situ hybridisation (Xu, et al,

1997). Using Northern analysis, it was shown by others that there was a significant

relationship between ER and RARa co-expression in human breast cancers, by %2

contingency analysis (/?=0.0052) (Roman, et al, 1993). This discrepancy may be due to

96


the use of RT-PCR, which is more sensitive than Northern hybridisation. The sample

size used in this study comprised a large cohort of ER+ and ER- human mammary

carcinoma biopsies (97 and 43, respectively), and was comparable to the sample size

used by Roman et al. (1993). However, approximately twice as many ER- mammary

carcinoma tissues were used in this study, which provides more statistically reliable

results. Previous studies in vitro have shown that atRA inhibits the growth of ER+

mammary carcinoma cells, whereas ER- cells are refractory to this effect (van der Burg

et al, 1993). The present data, however, suggest mechanisms not involving RARa may

contribute to atRA sensitivity in ER+ mammary carcinomas.

The frequency of RARp in mammary carcinoma appeared to be lower when

compared to normal breast tissue. This is, however, not significant, although others

have found a significant loss of RARp expression in cultured mammary carcinoma cells

(Swisshelm, et al, 1994). It has been demonstrated by in situ hybridisation that RARp

expression was absent in 13 of 14 breast tumour tissues, but not in adjacent normal

breast tissue, in formalin-fixed, paraffin-embedded specimens (Widschwendter, et al,

1997). This information, together with the data shown in this study, suggests a possible

role for RARp in mammary carcinogenesis via its absence of expression.

A significant loss of RARy expression was observed in mammary carcinoma

compared to normal tissue, using PCR detection. Furthermore, the frequency of RARy

expression in these neoplasms was significantly less in ER+ tumours than in ER-

tumours. Other studies have found no correlation between ER status and RARy

expression in breast cancer cells (Roman, et al, 1992, 1993; Shao, et al, 199A; Xu et

al, 1997) and in another study, there was a slight loss of RARy expression in breast

tumour tissue compared to normal breast tissue (Widschwendter, et al, 1997).


However, in the latter study, the researchers only analysed 14 formalin-fixed, paraffin-

embedded tissue samples. The results presented here show for the first time a

significant loss of RARy expression in mammary carcinoma and suggest an

involvement in the initiation or progression of the malignancy. Indeed, in BE(2)-C

neuroblastoma cells, overexpression of RARy resulted in a reduced growth rate

(Marshall, et al, 1995). Moreover, overexpression of RARy in NT2/D1 embryonal

carcinoma cells led to terminal differentiation (Moasser, et al, 1995).

It has been reported that the expression of RARp was progressively lost in

human mammary carcinoma as the neoplasm became increasingly aggressive (Xu et

al, 1997). Their data found a significant drop in RARp expression from normal breast,

to ductal carcinoma in situ (DCIS) and finally to low- and high-grade invasive tumours.

The results obtained here show no apparent relationship between RARa and p

expression and tumour grade. However, a significant decrease in the prevalence of

RARy expression was observed in grade I and II tumours compared to normal breast

tissue. This was followed by a significant increase in presence of RARy expression in

poorly differentiated grade III neoplasms, compared to the low-grade mammary

carcinomas. This also relates to the findings in this chapter that RARy expression was

present more significantly in ER- tumours, of which are usually classified with high

grade neoplasms. Thus, the data suggest that loss of RARy expression may be involved

in the early stages of mammary carcinogenesis and once the neoplasm has become

highly aggressive, this loss of expression may not be of importance for any further

tumour progression. A study of RARy expression in pre-malignant DCIS tissues and its

comparison with normal mammary tissue and invasive breast may substantiate this

theory.


The results described in Section 3.1 reveal relevant trends in the expression

patterns of RARs in mammary carcinoma and normal breast tissue simply through their

presence or absence. However, more meaningful results could be obtained from a

quantitative viewpoint rather than qualitative assessment. Thus, methods were designed

to estimate mRNA levels of RARs using a PCR approach. This has obvious advantages

over Northern and other blot analyses, such as higher sensitivity and avoiding the use of

radioactive material. Furthermore, many breast tumours collected in this study were

small in size, therefore using Northern blotting on these samples was not achievable, as

these samples contained very low quantities of total RNA. The PCR method was

further warranted in this instance, as it does not require RNA in microgram quantities

for observable amplification. There have been numerous reports on methods of

quantitating RNA using RT-PCR by incorporating an internal standard cDNA of known

amount in the PCR reaction (Zhao, et al, 1995; Corry, et al, 1996; Schieldrop, et al,

1996; Vidal, et al, 1996). Some studies even use housekeeping genes, such as

glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or p-actin, as internal standards

for quantitation (Zhao, et al, 1995; Pasquali, et al, 1997). However, in tumour cells,

constant expression of such genes may not necessarily be the case (Ostrowski, et al,

1989). Competitive PCR (Becker-Andre and Hahlbrock, 1989) is a more reliable

approach, where sample DNA (target) competes with a synthetic standard (competitor),

of known amount, for components in the same PCR reaction mix, including the same

primer pair. This method, however, has the disadvantage in that quantitation of each

sample requires titration of decreasing amounts of competitor that is co-amplified with a

constant amount of target, thus making the procedure laborious. Therefore, a semi

quantitative competitive PCR protocol was devised where each sample is quantitated in


a single tube containing a known amount of sample cDNA and a known amount of

competitor DNA, both of which are constant for every sample. A relative amount of

expression is obtained by analysing the ratio of target intensity against the intensity of

the competitor fragment. This procedure has been published by others (Jiang et al,

1996) and has the advantage over conventional competitive RT-PCR in that it requires

fewer tubes and therefore less sample RNA and laboratory preparation. This method

was tested among 59 mammary carcinoma samples for RARa and validation attempted

by conventional competitive RT-PCR on the same samples. The result of each method

provided disparate values when compared to one another, for two possible reasons.

Firstly, the semi-quantitative approach does not control for the efficiency of the cDNA

synthesis reaction, which is likely to be variable for different samples. Secondly, the

large difference in size between target and competitor in the semi-quantitative method

may affect the efficiency of PCR amplification (McCulloch, et al, 1995). However,

others dispute this latter point, claiming that amplification efficiency primarily depends

on the primer sequences (which are identical for target and competitor in this study) and

not template size (Siebert and Larrick, 1992). Since the semi-quantitative approach

could not be validated, the conventional but laborious competitive RT-PCR method was

chosen for quantitation of RAR mRNA.

To effectively calculate levels of RAR mRNA using competitive RT-PCR, the

amplification conditions must be optimised. Quantitation was performed during the

exponential phase of the amplification procedure. Some authors have reported that this

is unnecessary and that it is possible to quantitate during the plateau phase (Siebert and

Larrick, 1992). However, as there may be some variation in amplification efficiency

between different sized templates (i.e., target and competitor), one template may plateau


before the other, which may result in a false indication of the amount of target. The

linearity assays showed that competitive RT-PCR could detect very low quantities of

RNA, which can be extrapolated further by linear regression. Thus, although

competitive RT-PCR provides a powerful tool in detecting and quantitating very small

amounts of specific mRNA, other disadvantages were found in this technique besides its

laborious nature. As described in Section 2.11.4, competitive RT-PCR protocols varied

for each type of RAR to be analysed. This was done to ensure that the two PCR

fragments amplified for each sample (i.e., target and competitor), as observed following

agarose gel electrophoresis, were absolutely clean and did not contain feint smears or

non-specific fragments, which would affect the quantitation procedure. Hence,

competitive RT-PCR can be a tedious procedure, especially during the optimisation

stage where clean PCR fragments are a necessity, and this can be seen in the

competitive PCR protocol for RARa, which required extra components to optimise the

procedure.

The results of competitive RT-PCR of RARs in breast tissues indeed provide

further information than the qualitative data discussed earlier. Variations in levels of

expression of the RARs, as represented by the standard error of the mean, were

expected and this may be due to differences found between females with respect to age,

time of menstruation (or lack of) and lifestyles, which may affect hormonal status.

There was no significant difference in the mean expression levels of RARa between

normal breast and mammary carcinoma tissues, which implies that this receptor alone

may not have a strong involvement in the development or progression of breast cancer.

Interestingly, although a trend in loss of RARp expression was observed in mammary

carcinoma compared to normal breast tissue, competitive RT-PCR analysis of RARp

101


expression levels reveal that it is 6.5 times lower in breast tumours compared to normal

tissue. This highly significant decrease in R A R p expression may suggest a causal role

in mammary carcinogenesis. Indeed, it has been shown in vitro that R A R p induction by

atRA correlates with growth inhibition in breast cancer cells, and this anti-proliferative

effect was attributed to promotion of apoptosis by R A R p (Liu, et al, 1996). A low

concentration of R A R p m R N A in breast tumour tissue may prevent such growth

inhibitory and apoptotic events to occur using atRA treatment.

Quantitation of R A R y in breast tissue showed that its expression levels were

much higher compared to R A R a and R A R p . This was also observed by another group

studying 18 breast carcinoma biopsy specimens using Northern analysis (Shao, et al,

1994). In addition, the mean level of expression in normal tissues was significantly

higher compared to mammary carcinoma tissues, which correlates well with the

qualitative analysis discussed previously, as the prevalence of R A R y expression is also

significantly lower in mammary carcinoma compared to normal breast tissue. It has

been shown that the combination of retinoids and IFNa or y produces a synergistic anti

proliferative effect in breast cancer cells (Marth, et al, 1993). This combination

treatment also increased the expression of R A R y m R N A in breast cancer cells

(Widschwendter, et al, 1995, 1996; Fanjul, et al, 1996) and more recently in mouse

renal carcinoma cells (Hara, et al, 2001). As the data here show decreased expression

levels of R A R y m R N A in breast tumour tissue, combination therapy using retinoids and

IFNs could possibly bring R A R y m R N A back to levels usually found in normal breast

tissue. As a result, this may allow responsiveness of mammary carcinoma tissue to the

growth-inhibitory effects of atRA. Thus, a combination treatment of retinoids and IFNs

could potentially offer a new approach to treatment of breast cancer.

~ 102


The use of the highly sensitive competitive RT-PCR technique has also

confirmed an existing trend as well as a new insight in the relationship between ER

status and RAR mRNA levels. As discussed earlier, no relationship was observed

between ER status and the occurrence of RARa expression. However, a highly

significant positive correlation was found between ER status and RARa mRNA levels.

This verifies previous studies using Northern blotting in the detection of RARs in

malignant breast tumours, which also found a significant relationship between ER status

and RARa mRNA levels (Roman, et al, 1993; Shao, et al, 1994). It also has been

shown that oestrogen up-regulates RARa expression in a time- and concentration-

dependent manner (Roman, et al, 1993; van der Leede, et al, 1995). However, it was

also observed that the growth-inhibitory effects of atRA in cultured mammary

carcinoma cell lines correlated with RARa status, irrespective of the ER status

(Fitzgerald, et al, 1997). Thus, although RARa is present in ER- and ER+ mammary

carcinomas, adequate levels of its expression are probably required to acquire the

growth-inhibitory effects of atRA, as shown in ER+ mammary carcinoma cell culture

studies (van der Burg, et al, 1993). However, studies at the protein level need to be

performed to validate these results, as they are the ultimate mediators of RAR

signalling.

Determining the type of RAR that exerts the effects of atRA in mammary

carcinoma is further complicated by findings in this study that RARy mRNA expression

significantly correlates with ER status. This significance has not been previously

reported. Studies in vitro have shown that RARy mRNA expression levels were similar

in both ER+ and ER- mammary carcinoma cell lines, and a previous study examining

18 mammary carcinoma biopsy specimens found no significant relationship (Roman, et


al, 1992; Shao, et al, 1994). In normal mouse cervical and vaginal tissues, oestrogen

was found to induce RARy expression and was suggested to be involved in the growth

and development of these tissues (Celli, et al, 1996). It was shown in this study that

RARy expression occurs significantly less frequently in ER+ mammary carcinomas than

in their ER- counterparts. Thus, analysis of the RARy status and expression levels in

mammary carcinomas may provide an indication of the effectiveness of treatment using

retinoid therapy in these patients.

Unfortunately, data on tumour grade was not available for the 59 mammary

carcinoma specimens used for competitive RT-PCR, thereby preventing any

conclusions to be drawn about trends in RAR expression levels and progression of

mammary carcinoma. In summary, the data presented here may clarify the conflict of

existing information found in the literature. The data have been determined on a large

sample size, through the use of a highly sensitive and specific technique. It is more than

likely that combinatory functional losses of at least RARp and y are involved in

mammary carcinogenesis instead of a singular type of receptor.

3.4 Conclusions

From the studies using PCR techniques described in this chapter, the following

conclusions can be made with respect to RARs and mammary carcinoma:

• RARa, p and y mRNA are found in normal breast and mammary carcinoma tissues,

and although there was a decrease in trend in RARp expression in mammary

carcinoma, this was not significant. Furthermore, there was a significant loss of

RARy expression in mammary carcinomas.

104


• The prevalence of RARa and P expression are not associated with ER status,

whereas RARy is present more significantly in ER- mammary carcinomas than in

ER+ mammary carcinomas. Furthermore, although there were no associations

between RARa and P expression and tumour grade, RARy expression was

significantly lower in low-grade mammary carcinomas compared with high-grade

mammary carcinomas. As ER- status correlates with high-grade tumours, the data

suggests that loss of RARy expression may be involved in the early stages of

mammary carcinogenesis.

• By competitive RT-PCR analysis, mRNA levels of RARp and y, but not a, are

significantly lower in mammary carcinomas than in normal breast tissues. This

suggests that low concentrations of RARs may be involved in mammary

carcinogenesis as a result of insufficient signalling of the normal retinoid response.

• There is a positive correlation between RARa mRNA and ER levels in mammary

carcinoma, which has been reported previously in the literature. Surprisingly, a

positive correlation was also observed with RARy mRNA and ER levels in

mammary carcinoma, suggesting that oestrogen may up-regulate levels of RARa

and y expression.

• It is more than likely that a combined loss of expression of at least RARp and y,

rather than only one type, is involved in mammary carcinogenesis.

105

CHAPTER FOUR

ANALYSIS OF RETINOIC ACID

RECEPTOR PROTEIN LEVELS IN

BREAST TISSUES

"106

Chapter 4 Analysis of Retinoic Acid Receptor Protein Levels in Breast Tissues

4.0 Introduction

In Chapter 3, RT-PCR methods were described which analysed the mRNA

expression patterns of RARs in breast tissue specimens. Receptor proteins are the

eventual molecules involved in RAR signalling and since these levels do not necessarily

reflect the mRNA levels, it is additionally important to analyse RAR protein patterns.

At the beginning of this study, human RAR antibodies were not commercially available

and thus the RT-PCR approach was chosen initially to analyse RAR expression patterns

in breast tissue. Only RARa protein levels in mammary carcinoma specimens have

been analysed in the literature, however no comparisons with normal breast tissue were

reported (van der Leede, et al, 1996; Han, et al, 1997). The hypothesis in this chapter

is that the RAR protein levels in breast tissue correlates with the mRNA expression

levels. This chapter investigates the protein expression patterns of RARa, p and y in

mammary carcinoma biopsy specimens, as well as in normal breast tissue from the same

patient, and will be compared against the RAR mRNA expression patterns as described

in Chapter 3. Due to refinements in the early diagnosis of breast cancer, it has been

difficult to obtain sufficient biopsy material of normal and tumour tissue to undertake

this work, because often the lesions that are being sampled are only a few mm in

diameter.

4.1 Distribution of RAR Protein Levels in Breast Tissues

Eight females were examined for RAR protein levels, using both mammary

carcinoma biopsies as well as normal breast tissue from the same patient, which was

obtained distant from the site of the tumour. Using Western immunoblotting, all tissue

extracts analysed contained RARa, P and y protein and each receptor detected was «51


tn* Nl Tl

97.4-1

66-1

46-

30-

N 2 T2 N3 T3 N 4 T4

kDa N 5 T5 N 6 T6 N 7 T7 N 8 T8 N 4

Figure 4.1.1. Detection of R A R a protein in normal breast and mammary carcinoma

tissue. The RARa protein was detected as a «51 kDa fragment following Western

immunoblotting and chemiluminescent detection (see arrow). The numbers in each lane

represent matched normal breast (N) and mammary carcinoma (T) tissue taken from the

same patient. Sample N4 was loaded onto every gel as a standardising control for semi

quantitative analysis.

108


Nl Tl N 2 T2 N3 T3 N 4 T4

kDa N 5 T5 N 6 T6 N 7 T7 N 8 T8 N 4

Figure 4.1.2. Detection of R A R P protein in normal breast and mammary carcinoma

tissue. The RARp protein was detected as a «51 kDa fragment following Western

irnmunoblotting and chemiluminescent detection (see arrow). The numbers in each lane





Nl Tl N 2 T2 N3 T3 N 4 T4

N 5 T5 N 6 T6 N 7 T7 N 8 T8 N 4

Figure 4.1.3. Detection of RARy protein in normal breast and mammary carcinoma

tissue. The RARy protein was detected as a «51 kDa fragment following Western

immunoblotting and chemiluminescent detection (see arrow). The numbers in each lane




110


kDa in size (Gaub, et al, 1989; Rochette-Egly, et al, 1991). Representative blots of

RARa, P and y probing of normal and corresponding tumour breast tissue are shown in

Figures 4.1.1, 4.1.2 and 4.1.3, respectively, as detected by chemiluminescence. Due to

the polyclonal nature of each RAR antibody, some samples exhibited several non

specific protein bands. The expected 51 kDa fragment of the RAR protein was detected

in all samples and its size was determined relative to pre-stained protein markers. RAR

protein levels in normal and neoplastic breast tissue are summarised graphically in

Figure 4.1.4. Protein values were determined relative to a standardising control sample

that was loaded onto each gel (refer to Section 2.15.2). The mean level ± S.E.M. (in

arbitrary units) of RARa protein in normal tissue was 10.46 ± 0.77, while in tumour

tissue the mean level was 11.69 ± 1.26. The mean RARp protein levels in normal tissue

were 10.37 ± 1.34, and 6.14 ± 1.06 in neoplastic tissue. The mean levels of RARy

protein were 9.19 ± 1.09 in normal tissue and 6.10 ± 0.64 in mammary carcinoma.

Using the Wilcoxon matched-pairs, signed ranks test, no significant difference was

observed for RARa protein levels between normal and matched breast tumour tissues

(p=0.23). However, RARp and RARy protein levels in normal breast tissue were

significantly greater than in breast tumour tissue (p=0.0A, p=0.02 for RARp and RARy,

respectively).

4.2 Comparison of RAR and Oestrogen Receptor Protein Levels

All breast samples had oestrogen receptor levels routinely determined in their

tumour tissue and these results were compared with their corresponding RAR levels in

tumour tissue. Figures 4.2.1, 4.2.2 and 4.2.3 show scatter diagrams of ER levels and

111


R A R a R A R P RARy

Figure 4.1.4. Comparison of RAR protein levels in normal breast and mammary

carcinoma tissues matched for the same patient. Bar graphs correspond to mean protein

levels and error bars representing standard error of the mean. Significantly higher levels

of RARp and y protein were detected in normal tissues compared to mammary

carcinoma tissues (p=0.0A, p=0.02 for RARp and y, respectively, as shown by an

asterisk), but not for RARa (p=0.23), using Wilcoxon matched-pairs, signed ranks

testing.


a> 3 (0 to

o> E o E >*-

a> > 4) _l C CD *H

2 Q. DC UJ

1000

900

800

7on

«()()

500

400

300

200

100

0

Figure 4.2.1. Relationship between R A R a protein levels and E R protein levels in

mammary carcinoma tissues. N o significant relationship was found between these two

receptors (r2=0.11 l,p=0.A2, Pearson's correlation).

2 4 6 8 10 12 14 16 18 20

RARa Protein Level (Arbitrary Unit)


IV 3 (0 (A "P O) E o

1 a> > OJ _i c o HH»

E 0. 0. UJ

1000

900

800

/OO

600

500

400

300

200

100

0 4 6 8 10

RARp Protein Level (Arbitrary Unit) 12

Figure 4.2.2. Relationship between R A R p protein levels and E R protein levels in

mammary carcinoma tissues. N o significant relationship was found between these two

receptors (r2=0.006,/>=0.86, Pearson's correlation).


1000 i

900

--. 800 -Cv 3 (0 .2 700 -

o> E 1 is 600-o 1 E >*-- 500l » > a> -i 400 -c '3 «w g 300-Q.

oc LU 200

100

0

4+/

, , •/•» •

•

2 4 6 8

RARy Protein Level (Arbitrary Unit)

10

Figure 4.2.3. Relationship between R A R y protein levels and E R protein levels in

m a m m a r y carcinoma tissues. A significant relationship was observed between these

two receptors (r2=0.540,/?=0.04, Pearson's correlation).


levels of RARa, p and y, respectively. There was no significant relationship between

ER status and RARa (r2=0.111,/?=0.42) or RARp (r2=0.006, p=0.86) protein levels by

Pearson's correlation test. However, a significant correlation was observed between ER

status and RARy (r2=0.540, p=Q.0A) protein levels. As shown on the scatter diagrams,

one patient had a relatively high concentration of ER protein (900 fmol/mg tissue)

compared with the other patients. Although this may be considered as an outlier,

removal of this data value in the statistical analysis did not alter the relationship

observed between ER and RAR protein levels.

4.3 Comparison of RAR Protein Levels and Grading of Mammary Carcinoma

Tissue

Histopathology records were available for all samples and were graded

according to the criteria of Elston and Ellis (1991). Malignant tumours included 7

infiltrating ductal carcinomas and one infiltrating tubulo-lobular carcinoma, of which 2

were classified as grade I, 3 were grade II and 3 were grade III neoplasms. The average

level of RAR protein ± standard error of the mean detected among the 3 tumour grades

is summarised in Table 4.3.1. Using one-way analysis of variance (ANOVA), no

significant trends were observed between each RAR isotype protein levels and tumour

grading (p=0.01, p=0.53 and;?=0.91 for RARa, P and y, respectively). However, there

appears to be an increasing trend in RAR protein levels (for all 3 types) as the tumour

progresses from grade I to III, although this may reflect in the increasing tumour cell

load.

116


Table 4.3.1. Analysis of R A R Protein Levels and Tumour Grading of Mammary

Carcinoma Biopsies0

RARa6

RARpc

RAR/

GRADE I

9.41 ± 2.07

4.00 ± 0.22

5.53 ± 0.29

GRADE II

9.73 ±1.36

6.29 ±1.32

6.26 ±1.43

GRADE III

15.18± 1.34

7.44± 2.53

6.32 ±1.26

a Protein levels are expressed as average arbitrary units ± standard error of the mean.

b One-way ANOVA, p=0.01, across all three tumour grades

c One-way ANOVA, p=0.53, across all three tumour grades

d One-way ANOVA, ^=0.91, across all three tumour grades


4.4 Discussion

Using Western immunoblotting, RARp and y protein levels, but not RARa,

were significantly lower in mammary carcinoma than normal breast tissue. These

results confirm the competitive RT-PCR studies detailed in Chapter 3, where it was

demonstrated that RARp and y mRNA levels were significantly lower in mammary

carcinoma compared with normal breast tissue. However, there were no significant

differences in RARa mRNA expression levels between normal breast and mammary

carcinoma tissue. This finding is also reflected in the protein levels reported in this

chapter. It is important to note that different patient samples were used in the protein

analysis to those used in Chapter 3. This was necessary because of the small amounts

of biopsy samples available for analysis. It was demonstrated in Chapter 3 that RARy

mRNA levels were much higher than RARa and p, as has been shown by others (Shao,

et al., 1994). Due to the use of different antibodies in the Western immunoblots, which

may have variable affinity for tissue protein, comparisons between protein levels of

each type of RAR cannot be made. However, as the protein levels bear a similar trend

to the mRNA data presented in Chapter 3, it appears that decreased expression at the

mRNA level for RARp and y, but not RARa, are associated with mammary carcinoma.

This connection may arise from impairment of normal retinoid signalling through

insufficient levels of these receptors, which would have adverse effects on

differentiation and proliferation of normal breast cells.

Analysis of a relationship between ER status and RAR mRNA expression in

Chapter 3 showed a significantly positive correlation between ER levels and RARa and

y mRNA levels, but not with RARp. However, when comparing RAR protein levels, a

significant correlation was observed only with ER status and RARy. Although the


RARa mRNA levels do not correlate with protein levels, others have shown by

immunohistochemistry that there was no relationship between ER status and RARa

protein levels (van der Leede, et al., 1996). It has been postulated that discrepant results

observed between ER status and RARa mRNA and protein levels may arise because of

the use of whole breast tumour tissues, which include connective tissue adjacent to

tumour cells (van der Leede, et al, 1996). This could give rise to false positive results

for RARa status in mammary carcinoma cells. However, in this study, normal breast

tissues were analysed and compared with mammary carcinoma tissues. Furthermore, all

breast tissues used either in the mRNA or protein analysis of RARs underwent identical

procedures in sample pulverisation for extraction of either total RNA or protein. The

data in Chapter 3 suggest that oestrogen may regulate RARa and y expression at the

mRNA level, but at the protein level, RARa may be degraded at an enhanced rate,

thereby nullifying the up-regulatory effects of oestrogen. Recently, it was shown that

addition of atRA in the ER+ MCF-7 cell line resulted in a rapid breakdown of RARa

and y protein levels through a process that involved ubiquitination. Conversely, there

was an accumulation of RARa and y mRNA levels (Tanaka, et al, 2001). Similar to

the correlation found between ER status and RARy mRNA levels in Chapter 3, there is

no indication in the literature of a correlation between ER and RARy protein levels.

The discovery of this relationship reported in this chapter may have important

ramifications on the use of RARy-selective retinoids in the treatment of breast cancer.

Although the design and use of such selective retinoids may prove beneficial in

minimising side effects often observed with non-selective retinoids, the measurement of

the level of RARy expression in each patient would be required prior to treatment,

particularly in ER- breast tumours where RARy expression may be low. However,

" " 119


more research is needed to determine if a single RAR, or a combination of RARs, is

required for a response to retinoid treatment, and whether a threshold in RAR protein

levels is necessary to obtain the anti-cancer effects of retinoids.

The data reported here demonstrated that there was no statistically significant

relationship between RAR protein levels and tumour grading. Identical observations

were found in Chapter 3 for RARa and p mRNA expression, although there was a

significant increase in the prevalence of RARy mRNA from low- to high-grade

mammary carcinomas. Unfortunately, no information on tumour grade could be

obtained in mammary carcinoma specimens used in the competitive RT-PCR study.

Therefore, no comparisons can be made concerning trends in RAR mRNA, protein

levels and tumour grading in mammary carcinoma. However, the qualitative PCR

results, together with observations of significantly lower RARP and y protein levels in

mammary carcinomas compared to normal breast tissue, indicate that a loss or reduced

expression of RARs may be an early event in mammary carcinogenesis. It is possible

that investigations on RAR protein levels in pre-malignant breast lesions may provide

some clarification.

In summary, the work presented here provides evidence of reduced

concentrations of RAR protein levels in mammary carcinoma. The findings agree with

most of the mRNA studies discussed in Chapter 3. However, there is not a strong

association between ER status and RARa as previously reported (Roman, et al, 1992,

1993; Sheikh, et al, 1993; van der Burg et al, 1993). The studies detailed by these

authors analysed mRNA levels and not protein levels as has been described in this

chapter. The discrepancy in these results indicate that it is important to determine the

protein levels and not simply the mRNA levels, because mRNA processing can vary


and may not accurately reflect the protein concentrations in mammary carcinoma.

Since RARs are critical in the control of cellular growth, determining their protein

levels would be more advantageous than merely analysis of their mRNA levels. As

there may be clinical ramifications for treatment of breast cancer patients using retinoid

therapy, it is important the results described in this chapter are replicated in other

laboratories.

4.5 Conclusions

The following conclusions can be made in relation to the protein levels of RARs in

breast tissues of normal and neoplastic origin:

• RARp and y protein levels, but not RARa, are significantly lower in mammary

carcinoma than normal breast tissue. This is in agreement with the RAR mRNA

levels described in Chapter 3, and suggests that lowered expression of RARs may be

associated with the development of mammary carcinoma.

• There is no relationship between RARa or p protein levels and ER status in

mammary carcinoma. However, a significant correlation exists with ER and RARy

protein levels, which was also observed at the mRNA level described in Chapter 3.

This finding has not been previously reported.

• There is no relationship between RAR protein levels and grading of mammary

carcinoma, suggesting that decreased RAR expression may be an early event in

mammary carcinogenesis.

• Routine assessment of RAR protein levels, for example by immunohistopathology,

should be considered in patients with mammary carcinoma to determine if therapy

with retinoids is warranted. The use of RARy-selective retinoids may prove useful

in treating ER+ breast cancer patients.

' 121

CHAPTER FIVE

IDENTIFICATION OF

RETINOIC ACID RECEPTOR

POLYMORPHISMS

122

Chapter 5 Identification of Retinoic Acid Receptor Polymorphisms

5.0 Introduction

Retinoids are involved in many signalling processes in the body and anomalies in these

pathways may lead to the development and progression of neoplasia. Indeed, it is well

known that a specific translocation of the RARa gene (t(15;17)) is a characteristic of

acute promyelocytic leukemia (Larson, et al, 1984; Kastner, et al, 1992). Moreover,

down-regulation of RARp has been observed in several types of human malignancy of

epithelial and neuroectodermal origin (Torma, et al, 1993; Ferrari, et al, 1994; Lotan,

et al, 1995; Xu, et al, 1996), including several breast cancer cell lines (Swisshelm, et

al., 1994). Although it has been shown that retinoids inhibit the growth of human breast

cancer cells and their receptors detected in human breast cancer tissue (Roman, et al,

1993; van der Leede, et al, 1996), whether these receptors are present as fully

functional proteins in malignant cells is yet unknown.

To investigate the role that RARs may play, the presence of any polymorphism or

mutation in RARs a, P and y was determined in human breast tissue, both of normal and

neoplastic origin, and in serum and buffy coats from normal females. The hypothesis to

be tested in this chapter is that polymorphisms in the coding sequence of RARs are

associated with mammary carcinoma. Through the use of the polymerase chain reaction

(PCR), reverse transcription PCR and the single strand conformation polymorphism

(SSCP) technique, the occurrence of an RARy polymorphism was demonstrated in

serum and buffy coat DNA from normal females. Furthermore, this polymorphism as

well as another type was identified in mRNA from primary mammary carcinoma as

well as in DNA from serum of the same patients.

123


5.1 Absence of Polymorphisms in RARa and p mRNA in Normal and

Malignant Breast Tissue

Analysis was performed on 27 specimens, which comprised of samples from six

normal breast and 21 primary breast carcinomas and evidence for polymorphisms in the

entire coding regions of R A R s a and P assessed. Figures 5.1.1, 5.1.2, 5.1.3, 5.1.4 and

5.1.5 demonstrate typical band patterns seen for each of the primer sets used for R A R a

(refer to Table 2.11.5.1 a). For R A R p , representative gels for each of the P C R primer

pairs (refer to Table 2.11.5.2a) are in Figures 5.1.6, 5.1.7, 5.1.8, 5.1.9 and 5.1.10,

following SSCP and silver staining. Band patterns seen for each region of the R A R a

receptor amplicons appeared identical when both tissue types were compared with one

another. Likewise, identical band patterns were seen in the R A R p receptor when

normal samples were compared against primary tumour samples. The absence of

polymorphisms in the R A R a or p receptor suggests that mutations in these receptors are

unlikely to be involved in the development and progression of mammary carcinoma.

5.2 Detection of Polymorphisms in RARy mRNA in Normal and Malignant

Breast Tissue

The SSCP band patterns for the R A R y receptor derived from each of the primer

sets listed in Table 2.11.5.3a are represented in Figures 5.2.1, 5.2.2, 5.2.3, 5.2.4 and

5.2.5. Screening of eleven normal breast tissues failed to identify any band shifts.

However, in samples of primary mammary carcinoma, two types of polymorphisms

were found in R A R y m R N A . Representative bands of each polymorphism are shown in

Figure 5.2.6a. These polymorphic bands were detected with the Gamma5 primer set

(Table 2.11.5.3a). Confirmation of these polymorphisms by repeat c D N A synthesis and

~~ 124


Figure 5.1.1. Representative SSCP band patterns observed for R A R a breast tumour

cDNA samples using Alphal primer set (refer to Section 2.11.5). Band patterns for this

region of the receptor appeared identical in all tissues studied.



cDNA samples using Alpha2 primer set (refer to Section 2.11.5). Band patterns for this






127

Chapter 5 Identification of Retinoic Acid Receptor Polymorphi




129


Figure 5.1.6. Representative SSCP band patterns observed for R A R p breast tumour

cDNA samples using Betal primer set (refer to Section 2.11.5). Band patterns for this


130



cDNA samples using Beta2 primer set (refer to Section 2.11.5). Band patterns for this


131





132

Chapter 5 Identification of Retinoic Acid Receptor Polymorphi




133


Figure 5.2.1. Representative SSCP band patterns observed for R A R y breast tumour

cDNA samples using Gammal primer set (refer to Section 2.11.5). Band patterns for

this region of the receptor appeared identical in all tissues studied.

135



cDNA samples using Gamma2 primer set (refer to Section 2.11.5). Band patterns for






137





138



cDNA samples using Gamma5 primer set (refer to Section 2.11.5). These band patterns

represent the 'normal' profile and do not include polymorphic band shifts.


Figure 5.2.6. Band shift patterns seen for RARy in mammary tumour tissues. A: Lane 1

represents the normal phenotype. Lane 2 shows the S427L polymorphism band profile

(shift marked by arrow). Lane 3 shows the G443G polymorphism band profile (marked

by arrow). B: Sequencing data showing point mutations for each polymorphism. 1:

TCG->TTG base change at amino acid 427 (triangle); 2: GGG->GGA base change at

amino acid 443 (triangle).

140


PCR of the relevant sample showed the identical band shift as seen in the previous

sample run. Upon sequencing the PCR samples in which a polymorphism was detected

(Figure 5.2.6b), one type was found to be a silent mutation at codon 443, involving a

GGG-»GGA base change (glycine-»glycine). The other polymorphism was shown to

be a serine—^leucine amino acid change at codon 427, which involved a TCG-»TTG

point mutation. A total of 72 primary mammary carcinoma cases were screened for

polymorphisms within this region of RARy, of which 17/72 (23.6%) cases expressed the

G443G polymorphism, while 7/72 (9.7%) were found to express the S427L

polymorphism. These polymorphisms were mutually exclusive; none of the cases

studied expressed both.

5.3 Detection of RARy Polymorphism in Serum from Females with Mammary

Carcinoma

Upon finding the S427L polymorphism in seven mammary carcinoma biopsies,

DNA was extracted from these patients to determine whether the polymorphism was

genetic. Serum was the only available source of DNA from these patients and RARy

was successfully amplified by PCR. Analysis by SSCP of these serum samples also

detected the S427L polymorphism in all seven patients, thus confirming that the

polymorphism is genetic and not acquired through oncogenesis.

5.4 Detection of RARy Polymorphism in Serum and Whole Blood from Normal

Females

Forty serum samples from normal females were analysed for RARy

polymorphisms after it was revealed that the S427L polymorphism could be detected in


DNA in the serum from the seven patients with mammary carcinoma. It was found that

1/40 of these normal serum samples showed the S427L polymorphism. An additional

21 samples of whole blood taken from normal females were used to match the number

of tumour cases studied. It was found that 2/21 of these DNA samples showed the

S427L polymorphism. Upon pooling these results with the 11 normal breast tissue

RNA studied above, 3/72 (4.2%) of normal females, compared with 9.7% in females

with mammary carcinoma, had the S427L polymorphism. However, Fischer's exact

test showed no significant differences in these two groups (p=0.3260).

5.5 Heritability of RARy S427L Polymorphism

Analysis of DNA taken from whole blood was done on a three-generation family

(Figure 5.5.1a), using PCR/SSCP analysis (Figure 5.5.1b). The grandson's DNA

(sample 8, Figure 5.5.1a) failed to amplify, and so that result was excluded from this

study. It was found that the mother and two of her daughters exhibited the S427L

genotype, following sequencing of re-confirmed samples that were affected (samples 2,

4 and 5). Since the married daughter (sample 6) did not have this genotype (hence her

offspring did not have the S427L polymorphism) it was not possible to follow it through

three generations. However, the data suggest that this polymorphism is heritable from

parent to offspring.

142


B

Figure 5.5.1. Detection of R A R y S427L polymorphism in a three-generation family. A:

family tree of persons analysed in this study, all of whom were regarded as normal at

the time of bleeding. B: SSCP gel showing band patterns for family members.

Numbers in A correspond to lane numbers seen in B. Lane 8 could not be shown as the

sample failed to amplify. Following confirmation of band shifts and sequencing, lanes

2, 4 and 5 showed the S427L polymorphism, suggesting that it is inherited (see arrow in

B).

143


5.6 Discussion

To our knowledge, the results reported here indicate for the first time that RARy

polymorphisms exist in human tissues. We have identified two polymorphisms, a silent

base change and an amino acid change at codon 427. Using mammary carcinoma as a

disease model, we were able to show the S427L polymorphism was present in 9.7% of

mammary tumours studied and less than half of that incidence in normal samples

(4.2%). The S427L polymorphism was identified at the genomic level and was not

acquired spontaneously through oncogenesis. It was found that the prevalence of the

S427L polymorphism in normal compared to tumour samples were not statistically

significant. However, those normal females with this polymorphism could not be

followed up to determine if they had developed breast cancer. Furthermore, analysis of

a family possessing the S427L polymorphism showed that it could be inherited.

Although polymorphisms were not identified in the coding sequences of RARa and p,

the possibility that such polymorphisms occur cannot be ruled out, particularly since a

low sample size of 27 specimens was used in this study.

We were able to detect human DNA in serum samples using PCR. Since these

samples were 6 years old, it is interesting that all 47 samples studied had detectable

human DNA. This offers an alternative source of DNA for retrospective studies. The

number of amplification cycles was important for detection of the genomic RARy PCR

product. When 37 cycles were used, only some samples amplified, while at 49 cycles, a

majority but not all samples amplified. This optimisation procedure involved the use of

the seven samples that were identified with the S427L RARy polymorphism through

RT-PCR of their respective RNA extracts (data not shown). Using 60 cycles, all

samples amplified. In all PCR runs performed using this method, only twice did the

144


blank appear positive for RARy. However, when new reagents were used, the blank did

not contain any PCR products. Amplification of DNA has been achieved for serum

samples with between 35 and 45 cycles (Chen, et al, 1996; Nawroz, et al, 1996; De

Kok, et al, 1997; Hibi, et al, 1998; Esteller, et al, 1999). The number of cycles we

used may reflect on the age of the serum specimens. To test this, DNA was extracted

from 12 freshly collected serum samples and amplified for RARy for 40 cycles without

an initial hot start (Figure 5.6.1a). All samples amplified, which indicated that

degradation of DNA in the older frozen serum samples could be the reason why more

PCR cycles were required to amplify the RARy product. Interestingly, none of the 12

matching plasma samples that were amplified for 40-cycles for the RARy PCR reaction

contained any detectable DNA (Figure 5.6.1b). This indicated that in serum samples

DNA is released from cells during the clotting process. Plasma samples were collected

in heparinised Vacutainer tubes and the effect of heparin has been previously reported

to inhibit the PCR process (Beutler, et al, 1990). However, DNA has been successfully

analysed from whole blood collected in heparinsed tubes in the detection of markers for

Huntington's disease (Beilby, et al, 199A). The PCR method reported here used a very

small amount of serum (2-3 uL) as a source of DNA. It is important to note that the

supernatant must be removed from the serum precipitate, as this inhibited the PCR

reaction when it was amplified directly (Figure 5.6.2). An alternative method of

extracting DNA from serum has been proposed using a microwave oven (Sandford and

Pare, 1997). Based on the optimal times relative to the power of the microwave oven

(in our case, 700W for 3 min), we were unable to extract DNA from serum with this

method. As shown in Figure 5.6.2, DNA could only be detected by PCR in serum that

145


369bp -246bp -

123bp -

369bp 246bp

123bp

Figure 5.6.1. Agarose gel showing P C R products of RARy. (A) Amplification in serum

(lanes 1-12) and (B) matching plasma (lanes 13-24) samples. Lane 25 is the blank

control.


Figure 5.6.2. Agarose gel of R A R y P C R amplification comparing extraction methods.

Each run was performed in duplicate, using the same serum sample for all runs. Lanes

1 and 2: micro waved serum with supernatant removed. Lanes 3 and 4: micro waved

serum without removal of supernatant. Lanes 5 and 6: boiled serum with supernatant

removed. Lanes 7 and 8: boiled serum without removal of supernatant. Lane 9: blank

control.


was boiled and the supernatant removed. The differences in age and quality of the

serum used may contribute to the discrepancy seen between these two methods.

Thus, we have shown that the human RARy gene can be successfully amplified from

serum stored frozen at -70°C for 6 years, and that RARy can be amplified from serum

but not plasma. This method has the advantages that it requires very low sample

volumes, is simple, cheap and efficient. As a result of this, we propose that this method

could be used as an initial approach to extracting human DNA from serum. It will be of

great value in the field of genotyping markers for human disease and where cells have

not been collected and only serum is available.

Alterations in the RARy gene may affect the normal cell cycle. Indeed, in F9

embryonal carcinoma cells, a study has shown that disruptions in the RARa and y genes

altered regulation of the expression of differentiation-specific genes (Boylan, et al,

1995). Point mutations in the E domain of the RARy receptor have resulted in impaired

ligand binding and transactivation (Renaud, et al, 1995). The polymorphisms

identified here are located in the F domain, a region of the receptor that is poorly

understood. Deletion studies have shown that in RARa, the F domain was not required

for high-affinity ligand binding (Lefebvre, et al, 1995). However, in the oestrogen

receptor, the F domain was found to modulate gene transcription in its liganded form, as

well as in ascertaining the efficacy of antiestrogens (Montano, et al, 1995). These data

led to the conclusion that the conformation of the ligand-receptor complex differs

between the wild type form and a truncated receptor lacking the F domain. This could

subsequently affect the ability of the oestrogen receptor to interact with co-repressors or

co-activators. It is possible that the S427L amino acid change in RARy we have

detected could have a similar effect in altering the ligand-receptor complex, or perhaps


other functional regions of the receptor, such as the AF-2 region or the D N A binding

domain. This could possibly arise as a result of impaired binding of coregulator

molecules. W e were unable to perform a ligand-binding assay with the S427L protein

in tissue samples. This was primarily due to the amount of tissue being a limiting factor

and each tissue sample also contained normal R A R y m R N A (Fig. 5.6.2a, c.f. lane 1 and

lane 2), which would interfere with the binding assay.

Polymorphisms have been identified in the R A R a receptor (Larson, et al,

1984). Such mutations were found in patients with acute promyelocytic leukemia

(APL), where a translocation occurs in the R A R a gene (t(15;17)) that results in a

P M L / R A R a fusion protein. T w o missense mutations were found in the R A R a IE

domain of the chimeric gene in A P L patients showing clinical resistance to atRA

therapy. This suggested that mutations in the P M L / R A R a chimeric gene may give rise

to atRA resistance (Imaizumi M, et al, 1998). However, the authors reported that they

only detected 2/23 patients with mutated P M L / R A R a m R N A . Other investigators did

not find any mutations in the P M L / R A R a gene using SSCP (Morosetti, et al, 1996),

which implies that there is a low prevalence of alterations in this chimeric gene. Our

results showed an absence of mutations in the R A R a and p transcripts, suggesting that

other mechanisms must be involved with the role of R A R s in the development and

progression of mammary carcinoma.

Although mutations in the receptor can affect the retinoid signaling process, the

mere absence of these R A R s in the cell may also give rise to faulty cellular function. It

has been shown that reduced expression of endogenous R A R y in neuroblastoma cells

can lead to the malignant phenotype (Marshall, et al, 1995). Conversely, over-

expression of RARy, and not R A R a or R A R P , induces terminal differentiation in the

" " 149


DI human embryonal carcinoma cell line (Moasser, et al, 1995). We conclude that in

mammary tumours, aberrant expression of RARy is present and may be involved in its

carcinogenesis. As we could not follow up the health status of those normal females

with the S427L polymorphism, further work needs to be done to determine its

usefulness as a prognostic marker for hereditary breast cancer. Whether these mutations

do affect the functionality of RARy also remains to be answered.

150


5.6 Conclusions

The following conclusions can be made about RAR polymorphisms in human breast

tissue and blood:

• Polymorphisms could not be found in RARa or p mRNA in breast tissues by SSCP

analysis. However, two polymorphisms were identified in RARy mRNA: G443G

and S427L. Both of these polymorphisms are located in the F domain of the

receptor.

• The S427L polymorphism occurred at the genomic level and was found to be

genetic and not acquired through oncogenesis. Furthermore, this polymorphism was

found to be heritable in a family study.

• The frequency of the S427L polymorphism in normal breast tissue and blood was

not statistically different from that observed in mammary carcinoma.

• Follow-up studies could not be done on those normal specimens that contained the

S427L polymorphism. Thus, it is not clear as to whether this polymorphism can

predispose to disease.

• The functional status of the S427L RARy protein is unclear.

151

CHAPTER SIX

GENERAL

DISCUSSION

152

Chapter 6 General Discussion

6.0 General Discussion

In this thesis it was hypothesised that expression of retinoic acid receptors may

be decreased in mammary carcinoma, and that polymorphisms may exist in their coding

sequences, and together these perturbations contribute to the development and/or

progression of breast cancer. As there has been considerable in vitro work performed

on RAR abnormalities, human breast tissue biopsies were used in experiments to test

this hypothesis because few such reports are available in the scientific literature despite

the obvious relevance. Thus, the aims focused on the determination of the mRNA and

protein expression patterns of RARa, p and y in a large cohort of mammary carcinoma

biopsies and compare with similar estimations in normal breast tissue samples.

Oestrogen status and tumour grading, which are commonly assessed in biopsies of

mammary carcinoma, were also analysed in conjunction with RAR expression to

determine if correlations exist. Finally, breast tissue and leucocyte DNA were screened

for polymorphisms in the coding sequences of RARa, P and y to elucidate if an

association existed with a predisposition to mammary neoplasia.

Breast cancer is a leading cause of morbidity and mortality in women. In

Australia, approximately 1 in 11 women are diagnosed with the disease, irrespective of

age1". Studies investigating the causes of mammary carcinoma have been extensive and

advancements have been made in the earlier diagnosis of the disease. However, despite

this progress, the mortality rate of patients with breast cancer remains high, with

approximately 2,600 Australian women dying from breast cancer every yearf. It is clear

that new approaches are required that will assist with prevention, early diagnosis,

prognosis and treatment of this disease. In addition, markers that may

f Data obtained from Cancer in Australia, 1991-1994 with projections to 1999 (1998) published

by the Australian Institute of Health and Welfare


indicate those w o m e n w h o are at greater risk of developing breast cancer would be of

great significance.

Natural or synthetic retinoids are an appropriate candidate for chemoprevention

of malignancy as they possess anti-proliferative, differentiative and apoptotic

characteristics. Clinical trials using retinoids in treatment of breast cancer are on-going.

Promising in vitro results have been shown for the synergistic effects of retinoid

treatment in combination with other agents, such as the anti-oestrogen tamoxifen, or the

interferons (Wetherall and Taylor, 1986; Marth, et al, 1993). Therefore it would be

beneficial for women if current clinical trials prove effective using a combined

treatment modality (Budd, et al, 1998; Decensi and Costa, 2000). Indeed, retinoid

treatment has been effectively used in skin cancers and in other malignancies, such as

acute promyelocytic leukaemia (Huang et al, 1988; Lippman and Meyskens, 1989;

Castaigne, et al, 1990). However, the information present in the literature on clinical

trials using retinoids in breast cancer treatment show that they are not always effective

and may only be beneficial to certain subgroups, such as premenopausal women

(Cassidy, et al, 1982; Veronesi, et al, 1999). Thus, further understanding of the nature

of the retinoic acid receptors, which are the mediators of retinoid action, is required in

both malignant and normal breast tissues. It may be, for example, that RARs are

present at levels too low to mediate the retinoid-triggered responses in mammary

carcinoma.

The RARs are considered to be regulators of the homeostatic state of cells in

view of their profound effects in promoting cellular differentiation and inhibition of cell

growth, as well as their involvement in regulating transcription of a myriad of genes as

discussed in Section 1.3.3. The work presented in this thesis demonstrates that there is

abnormal expression of the RARs in mammary carcinoma. Alteration of retinoid


signalling pathways, which are required for normal cellular growth and differentiation,

may arise as a result of abnormal expression and thus may be implicated in the

causation of mammary neoplasia. Very few studies have analysed expression patterns

of all 3 RAR types in tissues derived from normal breast and mammary carcinoma

(Shao, etal, 1994; Pasquali, etal, 1997; Widschwendter, etal, 1997; Xu, etal, 1997).

It was shown in Chapter 3 that there was an absence of RARy expression in mammary

carcinoma compared to normal breast. Moreover, analysis of the mRNA levels of

RARs in breast tissue using competitive RT-PCR demonstrated there was significantly

reduced expression of RARp and y in mammary carcinoma tissue compared to normal

breast tissue. This was further supported by the data in Chapter 4 by the findings that

protein levels of RARp and y were significantly decreased in mammary carcinoma

compared with patient-matched normal breast tissue. Thus, the decreased expression of

RARs, particularly p and y, in mammary carcinoma provide support for the hypothesis

described in this thesis. As RARs are able to autoregulate their expression due to

inherent RAREs in the promoter sequences (de The, et al, 1990b; Leroy, et al, 1991b;

Lehmann, et al, 1992), it is likely that functional losses of all three RARs may be

involved in mammary carcinogenesis instead of a single type of RAR. Consequently,

these neoplastic cells may not have adequate levels of RARs for normal retinoid

signalling, thus favouring the evolution and progression of mammary carcinoma. It was

also shown in Chapter 4 that there were no significant differences between tumour

grading and RAR protein levels. This suggests that decreased RAR expression may be

an early event in mammary carcinogenesis and could possibly be involved in its

initiation; further work on this aspect, however, is required.

155


M a m m a r y carcinogenesis is associated with decreased R A R expression and this

may be an important early event in the development of breast cancer as these receptors

modulate the expression of many other genes. Studies in vitro have shown that the

addition of atRA in MCF-7 cells resulted in greater than 70% inhibition of oestrogen-

stimulated synthesis and secretion of transforming growth factor a (TGFa), which is

known to be a potent inducer of proliferation (Fontana, et al, 1992). Thus, decreased

expression of RARs may not be sufficient to antagonize oestrogen stimulation of growth

in mammary cells. Furthermore, an imbalance of the cross-coupling between RARs and

AP-1 can lead to tumourigenesis. Human skin studies in vivo have shown that UV

irradiation activates AP-1 and concomitantly results in reduced expression of RARy

mRNA and protein, which gave rise to photo-aging and development of skin cancer

(Fisher, et al, 1996; Wang et al, 1999). However, pre-treatment of skin with atRA

followed by UV exposure resulted in amelioration of loss of RARy compared with non-

treated, UV-exposed skin (Wang et al, 1999). Thus, the protective effect of atRA pre-

treatment in skin may prevent an imbalance of the cross-talk between AP-1 and RARs,

which would therefore diminish the involvement of AP-1 in the development of

neoplasia in skin.

Another inducer of proliferation, TGFpi, is known to contain three AP-1

binding sites in two of the promoter regions of this gene. Repression of this gene has

been demonstrated in HepG2 hepatocellular carcinoma cells upon treatment with atRA,

which was hormone dependent and a function of RAR concentration (Salbert, et al,

1993). Carcinogenesis can also arise through prevention of apoptosis and it has been

shown that caspase-3 can be activated by retinoids through induction of RARp

(Srivastava, et al, 1999). Thus, deficiencies in RAR expression can result in atRA


unresponsiveness in cells and may promote molecular events that lead to initiation

and/or progression of neoplasia.

The data presented in this thesis demonstrates that a reduction in RAR

expression occurs in mammary carcinoma tissue compared to normal breast tissue,

although it is unknown what causes this lowered expression. An explanation of this

could be related to dietary uptake of retinoids, as it has been demonstrated in our

laboratory that P-carotene intake is significantly less in patients with breast cancer in

contrast with findings obtained from age-matched control patients (Ching, et al, 2002).

This observation, together with the RAR data in this thesis, show that consumption of

foods rich in retinoids may be protective against the formation of mammary carcinoma.

Another possible involvement of RARs in mammary carcinogenesis could be

because of the expression of defective receptors. In Chapter 6, a polymorphism that

altered an amino acid type was identified in the F domain of the human RARy gene.

This S427L polymorphism was observed in both normal females and in mammary

carcinoma tissue. Although there was more than twice the prevalence of the

polymorphism in subjects with breast cancer, the difference was not significant.

However, follow-up studies could not be done on the normal females with the S427L

polymorphism to determine if they had developed mammary carcinoma. Given that

breast cancer has no age onset, it would be highly improbable to perform such a follow-

up task. As this polymorphism is heritable, it may play a role in the development of

familial breast cancer. Unfortunately it was not possible to develop this part of the

project further. Nevertheless, there is some support for the second part of the

hypothesis and it is clear that more research is required to determine if this

polymorphism can serve as a biomarker for disease. A model describing how


disruptions in R A R processing may be involved in the initiation and/or progression of

mammary carcinoma is shown in Figure 6.0.1 and could provide the basis for future

investigations.

In conclusion, it can be postulated that effective retinoid signalling can be

achieved when levels of RARs are adequate. This theory has been suggested in the

past, but was based on in vitro studies (van der Burg et al, 1993). Therefore, to

provide a suitable approach for chemoprevention of mammary carcinoma by retinoid

treatment, detection of sufficient levels of RARs at the site of the tumour should be

performed, and this would especially be useful if selective retinoids are to be used in

chemoprevention. In essence, this approach would be similar to the routine detection of

oestrogen receptor levels in mammary carcinomas to determine whether hormone

treatment is warranted. The detection of RARs can be achieved at the rnRNA level

using competitive RT-PCR as described in this thesis. However, the threshold level of

each type of RAR, or whether a combination of all three receptors is required for

effective mediation by retinoids, remains to be determined. Whether the S427L

polymorphism of the RARy gene could serve as a marker for hereditary breast cancer

also remains to be elucidated.

158


6.1 Future W o r k

The research achieved in this thesis opens new avenues of further investigations

into the role of RARs in breast cancer. Some likely possibilities are enumerated below:

1) Development of a protocol to determine cut-off concentrations of RARs required

for the effects of retinoids. Such an approach can be performed at the mRNA level

using competitive RT-PCR, or at the protein level by antibody detection. The latter

approach could possibly be adopted from current methods in determining ER status in

breast tumours using commercially available antibody kits, or by

immunohistopathology.

2) Screening a larger population of normal females and those with mammary

carcinoma for the S427L polymorphism to determine if it is significantly associated

with breast cancer development.

3) Determine if the S427L RARy protein is functionally active.

4) Examination of the promoter region of RARs for polymorphisms that may be

associated with its de - or down-regulation.

159


?Methylation O"Normal"

Absence of RARs in cells ^RAR mRNA/protein R A R polymorphism

?DCIS

Mammary carcinoma

TAP-I tTGFa, p 4<Apoptosis

?DCIS Mammary carcinoma

IRAR mRNA/protein

Mammary carcinoma

Figure 6.0.1. Model of mammary carcinogenesis due to disruptions in RAR signalling.

This could occur from their absence, inadequate levels, or non-functional receptors due

to polymorphisms.

REFERENCES

161

References

Alvarez, R., de Andres, J., Yubero, P., Vinas, O., Mampel, T., Iglesias, R., Giralt, M.,

and Villarroya, F. (1995) A novel regulatory pathway of brown fat

thermogenesis. Retinoic acid is a transcriptional activator of the mitochondrial

uncoupling protein gene. J. Biol Chem. 270: 5666-73.

Amos, B. and Lotan, R. (1990) Retinoid-sensitive cells and cell lines. Methods

Enzymol 190: 217-25.

Angel, P. and Karin, M. (1991) The role of Jun, Fos and the AP-1 complex in cell-

proliferation and transformation. Biochim. Biophys. Acta. 1072: 129-57.

Arens, J. F. (1946) Synthesis of some compounds possessing vitamin A activity.

Nature. 157: 190-1.

Astrom, A., Tavakkol, A., Pettersson, U., Cromie, M., Elder, J. T., and Voorhees, J. J.

(1991) Molecular cloning of two human cellular retinoic acid-binding proteins

(CRABP). J. Biol Chem. 266: 17662-6.

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A.,

and Struhl, K. Current protocols in molecular biology: John Wiley & Sons, Inc.,

1994.

Baniahmad, A., Kohne, A. C, and Renkawitz, R. (1992) A transferable silencing

domain is present in the thyroid hormone receptor, in the v-erbA oncogene

product and in the retinoic acid receptor. EMBO J. 11: 1015-23.

Baniahmad, A., Thormeyer, D., and Renkawitz, R. (1997) xAh c/Af-2 of the thyroid

hormone receptor relieves silencing of the retinoic acid receptor silencer core

independent of both T4 activation function and full dissociation of corepressors.

Mol Cell Biol. 17: 4259-71.

162

References

Bashor, M. M., Toft, D. O., and Chytil, F. (1973) In vitro binding of retinol to rat tissue

components. Proc. Natl. Acad. Sci. USA. 70: 3483-87.

Bavik, C.-O., Ericksson, U., Allen, R. A., and Peterson, P. A. (1991) Identification and

partial characterization of a retinal pigment epithelial membrane receptor for

plasma retinol-binding protein. J. Biol Chem. 266: 14978-85.

Becker-Andre, M. and Hahlbrock, K. (1989) Absolute mRNA quantification using the

polymerase chain reaction (PCR). A novel approach by a PCR aided transcript

titration assay (PATTY). Nucleic Acids Res. 17: 9437-46.

Becker-Andre, M., Andre, E., and DeLamarter, J. F. (1993) Identification of nuclear

receptor mRNAs by RT-PCR amplification of conserved zinc-finger motif

sequences. Biochem. Biophys. Res. Commun. 194: 1371-9.

Beilby, J. P., Chin, C, Porter, I., Walpole, I., and Goldblatt, J. (1994) Diagnosis of

Huntington's disease in a Western Australian population by analysis of an

intragenic trinucleotide repeat expansion. Aust. Med. J.. 161: 356-60.

Benbrook, D., Lernhardt, E., and Pfahl, M. (1988) A new retinoic acid receptor

identified from a hepatocellular carcinoma. Nature. 333: 669-72.

Beutler, E., Gelbart, T., and Kuhl, W. (1990) Interference of heparin with the

polymerase chain reaction. BioTechniques. 9: 166.

Blanco, J. C, Minucci, S., Lu, J., Yang, X. J., Walker, K. K, Chen, H., Evans, R. M.,

Nakatani, Y., and Ozato, K. (1998) The histone acetylase PCAF is a nuclear

receptor coactivator. Genes Dev. 12: 1638-51.

Blaner, W. S., Dixon, J. L., Moriwaki, H., Martino, R. A., Stein, O., Stein, Y., and

Goodman, D. S. (1987) Studies on the in vivo transfer of retinoid from

parenchymal to stellate cells in the rat liver. Eur. J. Biochem. 164: 301-7.

References

Blaner, W. S. (1989) Retinol-binding protein: the serum transporter protein for vitamin

A. Endocr. Rev. 10: 308-16.

Blaner, W. S. and Olson, J. A. Retinol and retinoic acid metabolism. In: M. B. Sporn,

A. B. Roberts, and D. S. Goodman (eds.), The Retinoids: Biology, Chemistry

and Medicine, 2nd edition. New York: Raven Press, 1994.

Bouterfa, H. L., Piedrafita, F. J., Doenecke, D., and Pfahl, M. (1995) Regulation of HI0

gene expression by nuclear receptors through an unusual response element:

implications for regulation of cell proliferation. DNA Cell Biol. 14: 909-19.

Bovenzi, V., and Momparler, R. L. (2001) Antineoplastic action of 5-aza-2 '-

deoxycytidine and histone deacetylase inhibitor and their effect on the

expression of retinoic acid receptor p and estrogen receptor a genes in breast

carcinoma cells. Cancer Chemo. PharmacolAH: 71-6.

Boylan, J. F. and Gudas, L. J. (1991) Overexpression of the cellular retinoic acid

binding protein-I (CRABP-I) results in a reduction in differentiation-specific

gene expression in F9 teratocarcinoma cells. J. Cell Biol 112: 965-79.

Boylan, J. F., Lufkin, T., Achkar, C. C, Taneja, R., Chambon, P., and Gudas, L. J.

(1995) Targeted disruption of retinoic acid receptor alpha (RARa) and RARy

results in receptor-specific alterations in retinoic acid-mediated differentiation

and retinoic acid metabolism. Mol Cell. Biol. 15: 843-51.

Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram

quantities of protein utilizing the principle of protein dye binding. Anal.

Biochem. 72: 248-54.

Brand, N., Petkovich, M., Krust, A., and Chambon, P. (1988) Identification of a second

human retinoic acid receptor. Nature. 332: 850-53.

References

Breen, J. J., Matsuura, T., Ross, A. C , and Gurr, J. A. (1995) Regulation of thyroid-

stimulating hormone beta-subunit and growth hormone messenger ribonucleic

acid levels in the rat: effect of vitamin A status. Endocrinol 136: 543-9.

Breitman, T. R., Collins, S. J., and Keene, B. R. (1982) Terminal differentiation of

human promyelocytic leukemic cells in primary culture in response to retinoic

acid. Blood. 57: 1000-4.

Brockes, J. P. (1989) Retinoids, homeobox genes, and limb morphogenesis. Neuron. 2:

1285-94.

Budd, G. T., Adamson, P. C, Gupta, M., Homayoun, P., Sandstrom. S. K., Murphy, R.

F., McLain, D., Tuason, L., Peereboom, D., Bukowski, R. M., and Ganapathi, R.

(1998) Phase I/II trial of all-trans retinoic acid and tamoxifen in patients with

advanced breast cancer. Clin. Cancer Res. 4: 635-42.

Bulens, F., Ibanez-Tallon, I., Van Acker, P., De Vriese, A., Nelles, L., Belayew, A., and

Collen, D. (1995) Retinoic acid induction of human tissue-type plasminogen

activator gene expression via a direct repeat element (DR5) located at -7

kilobases. J. Biol. Chem. 270: 7167-75.

Carlberg, C, Hooft van Huijsduijnen, R., Staple, J. K., DeLamarter, J. F., and Becker-

Andre, M. (1994) RZRs, a new family of retinoid-related orphan receptors that

function as both monomers and homodimers. Mol Endocrinol. 8: 757-70.

Cassidy, J., Lippman, M., Lacroix, A., and Peck, G. (1982) Phase II trial of 13-cis-

retinoic acid in metastatic breast cancer. Eur. J. Cancer Clin. Oncol. 18: 925-8.

Castaigne, S., Chomienne, C, Daniel, M. T., Ballerini, P., Berger, R., Fenaur, P., and

Degos, L. (1990) All-trans retinoic acid as a differentiation therapy for acute

promyelocytic leukemia. I. Clinical results. Blood. 76: 1704-9.

165

References

Celli, G., Darwiche, N., and De Luca, L. M. (1996) Estrogen induces retinoid receptor

expression in mouse cervical epithelia. Exp. Cell Res. 226: 273-82.

Chambon, P. (1994) The retinoid signaling pathway: molecular and genetic analyses.

Semin. Cell Biol. 5: 115-25.

Chen, J. D. and Evans, R. M. (1995) A transcriptional co-repressor that interacts with

nuclear hormone receptors. Nature. "¥11: 454-7.

Chen, X.-Q., Stroun, M., Magnenat, J. L., Nicod, L. P., Kurt, A.-M., Lyautey, J.,

Lederrey, C, and Anker, P. (1996) Microsatellite alterations in plasma DNA of

small cell lung cancer patients. Nat. Med. 2: 1033-4.

Chen, Z., Brand, N. J., Chen, A., Chen, S. J., Tong, J. H., Wang, Z. Y., Waxman, S.,

and Zelent, A. (1993) Fusion between a novel Kriippel-like zinc finger gene and

the retinoic acid receptor-alpha locus due to a variant t(l 1 ;17) translocation

associated with acute promyelocytic leukaemia. EMBO J. 12: 1161-7.

Ching, S., Ingram, D., Hahnel, R., Beilby, J., Rossi, E. (2002) Serum levels of

micronutrients, antioxidants and total antioxidant status predict risk of breast

cancer in a case control study. J. Nutr. 132: 303-6.

Chomszynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid

guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 162:

156-9.

Chytil, F. and Ong, D. E. (1987) Intracellular vitamin A-binding proteins. Annu. Rev.

Nutr. 7: 321-35.

Clark, A. R., Wilson, M. E., London, N. J., James, R. F., and Docherty, K. (1995)

Identification and characterization of a functional retinoic acid/thyroid hormone-

response element upstream of the human insulin gene enhancer. Biochem. J.

309: 863-70.

References

Cobleigh, M. A., Dowlatshahi, K., Deutsch, T. A., Mehta, R. G., Moon, R. C , Minn, F.,

Benson III, A. B., Rademaker, A. W., Ashenhurst, J. B., Wade III, J. L., and

Wolter, J. (1993) Phase I/II trial of tamoxifen with or without fenretinide, an

analog of vitamin A, in women with metastatic breast cancer. J. Clin. Oncol. 11:

474-7.

Collingwood, T. N., Urnov, F. D., and Wolffe, A. P. (1999) Nuclear receptors:

coactivators, corepressors and chromatin remodelling in the control of

transcription. J. Mol. Endocrinol. 23: 255-75.

Corry, D. B., Ampel, N. M., Christian, L., Locksley, R. M., and Galgiani, J. N. (1996)

Cytokine production by peripheral blood mononuclear cells in human

coccidioidomycosis. J. Infect. Dis. 11 A: 440-3.

Costa, A., Formelli, F., Chiesa, F., Decensi, A., De Palo, G., and Veronesi, U. (1994)

Prospects of chemoprevention of human cancers with the synthetic retinoid

fenretinide. Cancer Res. 54: 2032s-7s.

Crestani, M., Sadeghpour, A., Stroup, D., Galli, G., and Chiang, J. Y. L. (1998)

Transcriptional activation of the cholesterol 7-alpha-hydroxylase gene

(Cyp7a) by nuclear hormone receptors. J. Lipid Res. 39: 2192-200.

De Kok, J. B., Van Solinge, W. W., Ruers, T. J. M., Roelofs, R. W. H. M., Van Muijen,

G. N. P., Willems, J. L., and Swinkels, D. W. (1997) Detection of tumour DNA

in serum of colorectal cancer patients. Scand. J. Lab. Invest. 57: 601-4.

De Luca, L. M. (1991) Retinoids and their receptors in differentiation, embryogenesis,

and neoplasia. FASEB J. 5: 2924-33.

de The, H., Marchio, A., Tiollas, P., and Dejean, A. (1987) A novel steroid thyroid

hormone receptor-related gene inappropriately expressed in human

hepatocellular carcinoma. Nature. 330: 667-70.

References

de The, H., Chomienne, C, Lanotte, M., Degos, L., and Dejean, A. (1990a) The

t(15;17) translocation of acute promyelocytic leukemia fuses the retinoic acid

receptor a gene to a novel transcribed locus. Nature. 347: 558-61.

de The, H., del Mar, V.-R. M., Tiollas, P., Stunneberg, H., and Dejean, A. (1990b)

Identification of a retinoic acid responsive element in the retinoic acid receptor P

gene. Nature. 343: 177-80.

Decensi, A. and Costa, A. (2000) Recent advances in cancer chemoprevention, with

emphasis on breast and colorectal cancer. Eur. J. Cancer. 36: 694-709.

Dejean, A., Bougueleret, L., Grzeschik, K. H., and Tiollas, P. (1986) Hepatits B virus

DNA integration in a sequence homologous to v-erb-A and steroid receptor

genes in a hepatocellular carcinoma. Nature. 322: 70-2.

Delia, D., Aiello, A., Lombardi, L., Pelicci, P. G., Grignani, F., Formelli, F., Menard,

S., Costa, A., Veronesi, U., and Pierotti, M. A. (1993) N-(A-

hydroxyphenyl)retinamide induces apoptosis of malignant hematopoietic cell

lines including those unresponsive to retinoic acid. Cancer Res. 53: 6036-41.

Demirpence, E., Pons, M., Balaguer, P., and Gagne, D. (1992) Study of an

antiestrogenic effect of retinoic acid in MCF-7 cells. Biochem. Biophys. Res.

Commun. 183: 100-6.

Demirpence, E., Balaguer, P., Trousse, F., Nicolas, J. C, Pons, M., and Gagne, D.

(1994) Antiestrogenic effects of all-trans-retinoic acid and 1,25-

dihydroxyvitamin D3 in breast cancer cells occur at the estrogen response

element level but through different molecular mechanisms. Cancer Res. 54:

1458-64.

168

References

Diverio, D., Lo Coco, F., D'Adamo, F., Biondi, A., Fagioli, M., Grignani, F., Rambaldi,

A., Rossi, V., Awisati, G., Petti, M. C, Testi, A., Liso, V., Specchia, G.,

Fioritoni, G., Recchia, A., Frassoni, F., Ciolli, S., and Pelicci, P. (1992)

Identification of DNA rearrangements at the retinoic acid receptor-alpha (RAR-

a) locus in all patients with acute promyelocytic leukemia (APL) and mapping

of APL breakpoints within the RAR-alpha second intron. Italian Cooperative

Study Group "GIMEMA". Blood. 79: 3331-6.

Doerksen, L. F., Bhattacharya, A., Kannan, P., Pratt, D., and Tainsky, M. A. (1996)

Functional interaction between a RARE and an AP-2 binding site in the

regulation of the human HOX A4 gene promoter. Nucleic Acids Res. 24: 2849-

56.

Duester, G., Shean, M. L., McBride, M. S., and Stewart, M. J. (1991) Retinoic acid

response element in the human alcohol dehydrogenase gene ADH3: implications

for regulation of retinoic acid synthesis. Mol. Cell. Biol. 11: 1638-46.

Durand, B., Saunders, M., Leroy, P., Leid, M., and Chambon, P. (1992) All-trans and 9-

cis retinoic acid induction of CRABPII transcription is mediated by RAR-RXR

heterodimers bound to DR1 and DR2 repeat motifs. Cell. 71: 73-85.

Eckhoff, C, Collins, M. D., and Nau, H. (1991) Human plasma all-trans-, 13-cis- and

13-cw-4-oxoretinoic acid profiles during subchronic vitamin A supplementation:

comparison to retinol and retinyl ester plasma levels. J. Nutr. 121: 1016-25.

Eichele, G. (1989) Retinoids and vertebrate limb pattern formation. Trends Genet. 5:

246-51.

169

References

Elston, C. W . and Ellis, I. O. (1991) Pathological prognostic factors in breast cancer. I.

The value of histological grade in breast cancer: experience from a large study

with long-term follow-up. Histopathol. 19: 403-10.

Emerick, R. J., Zile, M., and H.F., D. (1967) Formation of retinoic acid from retinol in

the rat. Biochem. J. 102: 606-11.

Esteller, M., Sanchez-Cespedes, M., Rosell, R., Sidransky, D., Baylin, S. B., and

Herman, J. G. (1999) Detection of aberrant promoter hypermethylation of tumor

suppressor genes in serum DNA from non-small cell lung cancer patients.

Cancer Res. 69: 67-70.

Evans, R. (1988) The steroid and thyroid hormone receptor superfamily. Science. 240:

889-95.

Fanjul, A. N., Bouterfa, H., Dawson, M., and Pfahl, M. (1996) Potential role for retinoic

acid receptor-gamma in the inhibition of breast cancer cells by selective

retinoids and interferons. Cancer Res. 56: 1571-7.

Ferrari, N., Tonini, G. P., Briata, A., Bottini, F., and Vidali, G. (1994) Distribution of

retinoic acid receptor a, P and y mRNAs in neuroblastoma-derived cell lines and

in fresh tumours. Int. J. Oncol. 5: 1019-22.

Fisher, G. J., Datta, S. C, Talwar, H. S., Wang, Z. Q., Varani, J., Kang, S., and

Voorhees, J. J. (1996) Molecular basis of sun-induced premature skin ageing and

retinoid antagonism. Nature. 379: 335-9.

Fitzgerald, P., Teng, M., Chandraratna, R. A., Heyman, R. A., and Allegretto, E. A.

(1997) Retinoic acid receptor alpha expression correlates with retinoid-induced

growth inhibition of human breast cancer cells regardless of estrogen receptor

status. Cancer Res. 57: 2642-50.

170

References

Fleischhauer, K., Park, J. H., DiSanto, J. P., Marks, M., Ozato, K., and Yang, S. Y.

(1992) Isolation of full-length cDNA clone encoding a N-terminally variant

form of the human retinoid X receptor p. Nucleic Acids Res. 20: 1801.

Fontana, J. A., Nervi, C, Shao, Z.-M., and Jetten, A. M. (1992) Retinoid antagonism of

estrogen-responsive transforming growth factor a and pS2 gene expression in

breast carcinoma cells. Cancer Res. 52: 3938-45.

Forman, B. M., Yange, C.-R., Au, M., Casanova, J., Ghysdael, J., and Samuels, H. H.

(1989) A domain containing leucine-zipper-like motifs mediate novel in vivo

interactions between the thyroid hormone and retinoic acid rceptors. Mol.

Endocrinol 3: 1610-26.

Fujisawa, K., Umesono, K, Kikawa, Y., Shigematsu, Y., Taketo, A., Mayumi, M., and

Inuzuka, M. (2000) Identification of a response element for vitamin D3 and

retinoic acid in the promoter region of the human fructose-1,6-bisphosphatase

gene. J. Biochem. 127: 373-82.

Gaub, M. P., Lutz, Y., Ruberte, E., Petkovich, M., Brand, N., and Chambon, P. (1989)

Antibodies specific to the retinoic acid human nuclear receptors a and p. Proc.

Natl. Acad. Sci. USA. 86: 3089-93.

Giguere, V., Ong, E. S., Segui, P., and Evans, R. M. (1987) Identification of a receptor

for the morphogen retinoic acid. Nature. 330: 624-29.

Giguere, V. (1994) Retinoic acid receptors and cellular retinoid binding proteins:

complex interplay in retinoid signaling. Endocr. Rev. 15: 61-79.

Glass, C. K. (1994) Differential recognition of target genes by nuclear receptor

monomers, dimers and heterodimers. Endocr. Rev. 15: 391-407.

References

Goodman, D. S., Huang, H. S., and Shiratori, T. (1965) Tissue rat distribution and

metabolism of newly absorbed vitamin A in the rat. J. Lipid Res. 6: 390-6.

Gordon, J., Sacchettini, J. C, Ropson, I. J., Frieden, C, Li, E., Rubin, D. C,

Roth, K. A., and Cistola, D. P. (1991) Intracellular fatty acid-binding proteins

and their genes: useful models for diverse biological questions. Curr. Opin.

Lipidol 2: 125-37.

Green, S., Walter, P., Kumar, V., Krust, A., Bornert, J. M., Argos, P., and Chambon, P.

(1986) Human oestrogen receptor cDNA: sequence, expression and homology to

v-erb-A. Nature. 320: 134-9.

Green, S. and Chambon, P. (1988) Nuclear receptors enhance our understanding of

transcription regulation. Trends Genet. 4: 309-14.

Grignani, F., Ferrucci, P. F., Testa, U., Talamo, G., Fagioli, M., Alcalay, M.,

Mencarelli, A., Peschle, C, Nicoletti, I., and Pelicci, P. (1993) The acute

promyelocytic leukemia-specific PML-RAR alpha fusion protein inhibits

differentiation and promotes survival of myeloid precursor cells. Cell. 14: 423-

31.

Han, Q.-X., Allegretto, E. A., Shao, Z.-M., Kute, T. E., Ordonez, J., Aisner, S. C, Rishi,

A. K., and Fontana, J. A. (1997) Elevated expression of retinoic acid receptor-a

(RARa) in estrogen-receptor-positive breast carcinomas as detected by

immunohistochemistry. Diagn. Mol Pathol 6: 42-8.

Hanstein, B., Eckner, R., DiRenzo, J., Halachmi, S., Liu, H., Searoy, B., Kurokawa, R.,

and Brown, M. (1996) p300 is a component of an estrogen receptor coactivator

complex. Proc. Natl Acad. Sci. USA. 93: 11540-5.

172

References

Hara, F., Suzuki, T., Saito, M., Hikino, T., Fukushima, T., Oyama, T., Fukuda, T., and

Nakajima, T. (1994) Non-isotopic single strand conformation polymorphism

(PCR-SSCP) with subsequent direct sequencing after silver staining [letter].

Pathol. Int. 44: 815-6.

Hara, I., Taguchi, I., Miyake, H., Hara, S., Gotoh, A., and Kamidono, S. (2001) Effects

of retinoic acid and interferon-y on expression of the retinoic acid receptor in

mouse renal cell carcinoma. Int. J. Oncol. 19: 959-62.

Hayashi, K., Yokozaki, H., Goodison, S., Oue, N., Suzuki, T., Lotan, R., Yasui, N., and

Tahara, E. (2001) Inactivation of retinoic acid receptor beta by promoter CpG

hypermethylation in gastric cancer. Differ en. 68: 13-21.

Heery, D. M., Kalkhoven, E., Hoare, S., and Parker, M. G. (1997) A signature motif in

transcriptional co-activators mediates binding to nuclear receptors. Nature. 387:

733-6.

Heyman, R. A., Mangelsdorf, D. J., Dyck, J. A., Stein, R. B., Eichele, G., R.M., E., and

Thaller, C. (1992) 9-cz',s-retinoic acid is a high affinity ligand for the retinoid X

receptor. Cell. 68: 397-406.

Hibi, K., Robinson, C. R., Booker, S., Wu, L., Hamilton, S. R., Sidransky, D., and Jen,

J. (1998) Molecular detection of genetic alterations in the serum of colorectal

patients. Cancer Res. 58: 1405-7.

Hong, W. K., Endicott, J., Itri, L. M., Doos, W., Batsakis, J. G., Bell, R., Fofonoff, S.,

Byers, R., Atkinson, E. N., Vaughan, C, Toth, B. B., Kramer, A., Dimery, I. W.,

Skipper, P., and Strong, S. (1986) 13-cw-retinoic acid in the treatment of oral

leukoplakia. N. Engl. J. Med. 315: 1501-5.

173

References

Hong, W. K. and Itri, L. Retinoids and human cancer. In: M. B. Sporn, A. B. Roberts,

and D. S. Goodman (eds.), The Retinoids: Biology, Chemistry and Medicine.

New York: Raven Press, 1994.

Horlein, A. J., Naar, A. M., Heinzel, T., Torchia, J., Gloss, B., Kurokawa, R., Ryan, A.,

Kamei, Y., Soderstrom, M., Glass, C. K, and Rosenfeld, M. G. (1995) Ligand-

independent repression by the thyroid hormone receptor mediated by a nuclear

receptor co-repressor. Nature. 377: 397-404.

Huang, M., Ye, Y. C, Chen, S. R., Chai, J. R., Lu, J. X., Zhao, L., Gu, L. J., and Wang,

Z. Y. (1988) Use of all-trans retinoic acid in the treatment of acute

promyelocytic leukemia. Blood. 72: 567-72.

Hunter, D. J., Manson, J. E., Colditz, G. A., Stampfer, M. J., Rosner, B., Hennekens, C.

H., Speizer, F. E., and Wilier, W. C. (1993) A prospective study of the intake of

vitamins C, E, and A and the risk of breast cancer. N. Engl. J. Med. 329: 234-40.

Imaizumi M, Suzuki H, Yoshinari M, Sato A, Saito T, Sugawara A, Tsuchiya S, Hatae

Y, Fujimoto T, Kakizuka A, Konno T, and Iinuma K (1998) Mutations in the E-

Domain of RARa portion of the PML/RARa chimeric gene may confer clinical

resistance to all-trans retinoic acid in acute promyelocytic leukemia. Blood. 92:

374-82.

Ingram, D., Roberts, A., and Nottage, E. M. (1992) Host factors and breast cancer

growth characteristics. Eur. J. Cancer. 28A: 1153-61.

Ingram, D. (1994) Diet and subsequent survival in women with breast cancer. Br. J.

Cancer. 69: 592-5.

Ishikawa, T., Umesono, K., Mangelsdorf, D. J., Aburatani, H., Stanger, B. Z.,

Shibasaki, Y, Imawari, M., Evans, R. M., and Takaku, F. (1990) A functional

174

References

retinoic acid receptor encoded by the gene on human chromosome 12. Mol.


Jetten, A. M. and Jetten, M. E. R. (1979) Possible role of retinoic acid-binding protein

in retinoid stimulation of embryonal carcinoma cell differentiation. Nature. 278:

180-2.

Jiang, S.-Y., Shen S.-R., Shyu, R.-Y., Yu, J.-C, Ham, H.-J., Yeh, M.-Y., Lee M. M. S.,

and Chang, Y.-C. (1999) Expression of nuclear retinoid receptors in normal,

premalignant and malignant gastric tissues determined by in situ hybridization.

Br. J. Cancer. 80: 206-14.

Jiang, S.-Y., Davidson, L. A., Lupton, J. R., and Chapkin, R. S. (1996) Rapid

competitive PCR determination of relative gene expression in limiting tissue

samples. Clin. Chem. 42: 227-31.

Kakizuka, A., Miller, W. H. J., Umesono, K, Warrell, R. P. J., Frankel, S. R., Murty, V.

V. V. S., Dmitrovsky, E., and Evans, R. M. (1991) Chromosomal translocation

t(15;17) in human acute promyelocytic leukemia fuses RARa with a novel

putative transcription factor, PML. Cell 66: 663-74.

Kalemkerian, G. P., Slusher, R., Ramalingam, S., Gadgeel, S., and Mabry, M. (1995)

Growth inhibition and induction of apoptosis by fenretinide in small-cell lung

cancer cell lines. J. Natl Cancer Inst. 87: 1674-80.

Kamei, Y., Xu, L., Heinzel, T., Torchia, J., Kurokawa, R., Gloss, B., Lin, S.-C,

Heyman, R. A., Rose, D. W., Glass, C. K., and Rosenfeld, M. G. (1996) A CBP

integrator complex mediates transcriptional activation and AP-1 inhibition by

nuclear receptors. Cell 85: 403-14.

175

References

Kastner, P., Krust, A., Mendelsohn, C , Gamier, J.-M., Zelent, A., Leroy, P., Staub, A.,

and Chambon, P. (1990) Murine isoforms of retinoic acid receptor y with

specific patterns of expression. Proc. Natl. Acad. Sci. USA. 87: 2700-4.

Kastner, P., Mark, M., and Chambon, P. (1995) Nonsteroid nuclear receptors: what are

the genetic studies telling us about their role in real life? Cell. 83: 859-69.

Kastner, P., Perez, A., Lutz, Y., Rochette-Egly, C, Gaub, M. P., Durand, B., Lanotte,

M., Berger, R., and Chambon, P. (1992) Structure, localization and

transcriptional properties of two classes of retinoic acid receptor a fusion

proteins in acute promyelocytic leukemia (APL): structural similarities with a

new family of oncoproteins. EMBOJ. 11: 629-42.

Kato, S., Sasaki, H., Suzawa, M., Masushige, S., Tora, L., Chambon, P., and

Gronemeyer, H. (1995) Widely spaced, directly repeated PuGGTCA elements

act as promiscuous enhancers for different classes of nuclear receptors. Mol.

Cell. Biol. 15: 5858-67.

Kazmi, S. M., Plante, R. K, Visconti, V., and Lau, C. Y. (1996) Comparison of N-(4-

hydroxyphenyl)retinamide and all-trans-retinoic acid in the regulation of retinoid

receptor-mediated gene expression in human breast cancer cell lines. Cancer

Res. 56: 1056-62.

Klaassen, I., Brakenhoff, R. H., Smeets, S. J., Snow, G. B., and Braakhuis B. J. M.

(2001) Expression of retinoic acid receptor gamma correlates with retinoic acid

sensitivity and metabolism in head and neck squamous cell carcinoma cell lines.

Int. J. Cancer 92: 661-5.

Kochhar, D. M. (1973) Limb development in mouse embryos. I. Analysis of teratogenic

effects of retinoic acid. Teratol. 7: 289-98.

176

References

Krust, A., Green, S., Argos, P., Kumar, V., Walter, P., Bornert, J. M., and Chambon, P.

(1986) The chicken oestrogen receptor sequence: homology with v-erbA and the

human oestrogen and glucocorticoid receptors. EMBO J. 5: 891-7.

Krust, A., Kastner, P., Petkovich, M., Zelent, A., and Chambon, P. (1989) A third

human retinoic acid receptor, hRAR-y. Proc. Natl Acad. Sci. USA. 86: 5310-14.

Larson, R. A., Kondo, K., Vardiman, J. W., Beutler, A. E., Golomb, H. M., and Rowley,

J. D. (1984) Evidence for a 15; 17 translocation in every patient with acute

promyelocytic leukemia. Am. J. Med. 76: 827-41.

Lavrentiadou, S. N., Hadzopoulou-Cladaras, M., Kardassis, D., and Zannis, V. I. (1999)

Binding specificity and modulation of the human ApoClII promoter activity by

heterodimers of ligand-dependent nuclear receptors. Biochemistry. 38: 964-75.

Le Douarin, B., Zechel, C, Gamier, J. M., Lutz, Y., Tora, L., Pierrat, B., and Heery, D.

(1995) The N-terminal part of TIF 1, a putative mediator of the the ligand-

dependent activation function (AF-2) of nuclear receptors, is fused to B-ra/in

the oncogenic protein T18. EMBO J. 14: 2020-33.

Le Douarin, B., Nielsen, A. L., Gamier, J. M., Ichinose, H., Jeanmougin, F., Losson, R.,

and Chambon, P. (1996) A possible involvement of TIF la and TIFip in the

epigenetic control of transcription by nuclear receptors. EMBO J. 15: 6701-15.

Lee, M. O., Liu, Y., and Zhang, X. K. (1995) A retinoic acid response element that

overlaps an estrogen response element mediates multihormonal sensitivity in

transcriptional activation of the lactoferrin gene. Mol. Cell Biol 15: 4194-207.

Lefebvre, B., Rachez, C, Formstecher, P., and Lefebvre, P. (1995) Structural

determinants of the ligand-binding site of the human retinoic acid receptor a.

Biochemistry. 34: 5477-85.

References

Lehmann, J. M.s Zhang, X.-K., Graupner, G., Lee, M.-O., Hermann, T., Hoffman, B.,

and Pfahl, M. (1993) Formation of retinoid X receptor homodimers leads to

repression of T3 response: hormonal cross-talk by ligand-induced squelching.

Mol. Cell Biol. 13: 7698-707.

Lehmann, J. M., Zhang, X.-K., and Pfahl, M. (1992) RARy2 expression is regulated

through a retinoic acid response element embedded in Spl sites. Mol. Cell. Biol.

12: 2976-85.

Leid, M., Kastner, P., and Chambon, P. (1992a) Multiplicity generates diversity in the

retinoic acid signalling pathways. Trends Biochem. Sci. 17: 427-33.

Leid, M., Kastner, P., Lyons, R., Nakshari, H., Saunders, M., Zacharewski, T., Chen, J.-

Y., Staub, A., Gamier, J.-M., Mader, S., and Chambon, P. (1992b) Purification,

cloning, and RXR identity of the HeLa cell factor with which RAR or TR

heterodimerizes to bind target sequences efficiently. Cell. 68: 377-95.

Leo, M. A., Iida, S., and Lieber, C. S. (1984) Retinoic acid metabolism by a system

reconstituted with cytochrome P-450. Arch. Biochem. Biophys. 234: 305-12.

Leroy, P., Krust, A., Zelent, A., Mendelsohn, C, Harnier, J. M., Kastner, P., Dierich,

A., and Chambon, P. (1991a) Multiple isoforms of the mouse retinoic acid

receptor a gene are generated by alternative splicing and differential induction

by retinoic acid. EMBO J. 10: 59-69.

Leroy, P., Nakshatri, H., and Chambon, P. (1991b) Mouse retinoic acid receptor a2

isoform is transcribed from a promoter that contains a retinoic acid response

element. Proc. Natl. Acad. Sci. USA. 88:.

Levin, A. A., Sturzenbecker, L. J., Kazmer, S., Bosakowski, T., Huselton, C, Allenby,

G., Speck, J., Kratzeisen, C, Rosenberger, M., Lovey, A., and Grippo, J. F.

References

(1992) 9-cw-retinoic acid stereoisomer binds and activates the nuclear receptor

RXRa. Nature. 355: 359-61.

Levin, M. S., Li, E., Ong, D. E., and Gordon, J. I. (1987) Comparison of the tissue-

specific expression and developmental regulation of two closely linked rodent

genes encoding cytosolic retinol-binding proteins. J. Biol. Chem. 262: 7118-24.

Lingen, M. W., Polverini, P. J., and Bouck, N. P. (1996) Retinoic acid induces cells

cultured from oral squamous cell carcinomas to become anti-angiogenic. Am. J.

Pathol. 149: 247-58.

Lippman, S. M. and Meyskens, F. L. J. (1989) Results of the use of vitamin A and

retinoids in cutaneous malignancies. Pharmacol. Therapeut. 40: 107-22.

Lippman, S. M., Kavanagh, J. J., Paredes-Espinoza, M., Delgadillo-Madrueno, F.,

Paredes-Casillas, P., Hong, W. K, Holdener, E., and Krakoff, I. H. (1992a) 13-

cw-retinoic acid plus interferon alpha-2a: highly active systemic therapy for

squamous cell carcinoma of the cervix. J. Natl Cancer Inst. 84: 241-5.

Lippman, S. M., Parkinson, D. R., Itri, L. M., Weber, R. S., Schantz, S. P., Ota, D. M.,

Schusterman, M. A., Krakoff, I. H., Gutterman, J. U., and Hong, W. K. (1992b)

13-cw-retinoic acid and interferon alpha-2a: effective combination therapy for

advanced squamous cell carcinoma of the skin. J. Natl. Cancer Inst. 84: 235-41.

Liu, S., Wu, Q., Che, Z. M., and Su, W. J. (2001) The effect pathway of retinoic acid

through regulation of retinoic acid receptor alpha in gastric cancer cells. World

J. Gasteroenterol 7: 662-6.

Liu, Y., Chen, H., and Chiu, J. F. (1994) Identification of a retinoic acid response

element upstream of the rat alpha-fetoprotein gene. Mol. Cell Endocrinol. 103:

149-56.

References

Liu, Y., Lee, M. O., Wang, H. G., Li, Y., Hashimoto, Y., Klaus, M., Reed, J. C , and

Zhang, X. (1996) Retinoic acid receptor beta mediates the growth-inhibitory

effect of retinoic acid by promoting apoptosis in human breast cancer cells. Mol.

Cell. Biol. 16: 1138-49.

Lotan, R., Xu, X. C, Lippman, S. M., Ro, J. Y., Lee, J. S., Lee, J. J., and Hong, W. K.

(1995) Suppression of retinoic acid receptor-beta in premalignant oral lesions

and its up-regulation by isotretinoin. N. Engl. J. Med. 332: 1405-10.

Lucas, P. C, O'Brien, R. M., Mitchell, J. A., Davis, C. M., Imai, E., Forman, B. M.,

Samuels, H. H., and Granner, D. K. (1991) A retinoic acid response element is

part of a pleiotropic domain in the phosoewo/pyruvate carboxykinase gene. Proc.

Natl. Acad. Sci. USA. 88: 2184-8.

MacDonald, P. N. and Ong, D. E. (1988) Evidence for a lecithin-retinol acyltransferase

activity in the rat small intestine. J. Biol. Chem. 263: 12478-82.

MacDonald, P. N., Dowd, D. R., Nakajima, S., Galligan, M. A., Reeder, M. C,

Haussler, C. A., Ozato, K., and Haussler, M. R. (1993) Retinoid X receptors

stimulate and 9-cwretinoic acid inhibits 1,25-dihydroxyvitamin D3-activated

expression of the rat osteocalcin gene. Mol Cell Biol. 13: 5907-17.

Majewski, S., Marczak, M., Szmurlo, A., Jablonska, S., and Bollag, W. (1995)

Retinoids, interferon alpha, 1,25-dihydroxyvitamin D3 and their combination

inhibit angiogenesis induced by non-HPV-harboring tumor cell lines. RAR

alpha mediates the antiangiogenic effect of retinoids. Cancer Lett. 89: 117-24.

Mangelsdorf, D. J., Ong, E. S., Dyck, J. A., and Evans, R. M. (1990) Nuclear receptor

that identifies a novel retinoic acid response pathway. Nature. 345: 224-29.

References

Mangelsdorf, D. J., Umesono, K., and Evans, R. M. The Retinoid Receptors. In: M. B.

Sporn, A. B. Roberts, and D. S. Goodman (eds.), The Retinoids: Biology,

Chemistry and Medicine, 2nd edition, pp. 319-49. New York: Raven Press,

1994.

Mangelsdorf, D. J. and Evans, R. M. (1995) The RXR heterodimers and orphan

receptors. Cell. 83: 841-50.

Mariotti, A., Marcora, E., Bunone, G., Costa, A., Veronesi, U., Pierotti, M. A., and

Valle, G. D. (1994) 7V-(4-hydroxyphenyl)retinamide: a potent inducer of

apoptosis in human neuroblastoma cells. J. Natl. Cancer Inst. 86: 1245-7.

Marshall, G. M., Cheung, B., Stacey, K. P., Camacho, M. L., Simpson, A. M., Kwan,

E., Smith, S., Haber, M., andNorris, M. D. (1995) Increased retinoic acid

receptor gamma expression suppresses the malignant phenotype and alters the

differentiation potential of human neuroblastoma cells. Oncogene. 11: 485-91.

Marth, C, Widschwendter, M., and Daxenbichler, G. (1993) Mechanism of synergistic

action of all-trans- or 9-cis-retinoic acid and interferons in breast cancer cells. J.

Steroid Biochem. Mol. Biol. 47: 123-6.

Mattei, M.-G., de The, H., Mattei, J.-F., Marchio, A., Tiollais, P., and Dejean, A.

(1988a) Assignment of the human hap retinoic acid receptor RARp gene to the

p24 band of chromosome 3. Hum. Genet. 80: 189-90.

Mattei, M.-G., Petkovich, M., Mattei, J.-F., and Chambon, P. (1988b) Mapping of the

human retinoic acid receptor to the q21 band of chromsome 17. Hum. Genet. 80:

186-89.

181

References

McCulloch, R. K., Choong, C. S., and Hurley, D. M . (1995) A n evaluation of

competitor type and size for use in the determination of mRNA by competitive

PCR. PCR Methods Appl 4: 219-26.

McKenna, N. J., Xu, J., Nawaz, Z., Tsai, S. Y., Tsai, M. J., and O'Malley, B. W. (1999)

Nuclear receptor coactivators: multiple enzymes, multiple complexes, multiple

functions. J. Steroid Biochem. Mol. Biol. 69: 3-12.

Miller, W. H., Jr. (1998) The emerging role of retinoids and retinoic acid metabolism

blocking agents in the treatment of cancer. Cancer. 83: 1471-82.

Minucci, S. and Pelicci, P. G. (1999) Retinoid receptors in health and disease: co-

regulators and the chromatin connection. Cell Dev. Biol. 10: 215-25.

Misrahi, M., Atger, M., d'Auriol, L., Loosfelt, H., Meriel, C, Fridlansky, F., Guiochon-

Mantel, A., Galibert, F., and Milgrom, E. (1987) Complete amino acid sequence

of the human progesterone receptor deduced from cloned DNA. Biochem.

Biophys. Res. Commun. 143: 740-8.

Miyata, K. S., McCaw, S. E., Patel, H. V., Rachubinski, R. A., and Capone, J. P. (1996)

The orphan nuclear hormone receptor LXR alpha interacts with the peroxisome

proliferator-activated receptor and inhibits peroxisome proliferator signaling. J.

Biol. Chem. 271:9189-92.

Moasser, M. M., Reuter, V. E., and Dmitrovsky, E. (1995) Overexpression of the

retinoic acid receptory directly induces terminal differentiation of human

embryonal carcinoma cells. Oncogene. 10: 1537-43.

Montano, M. M., Muller, V., Trobaugh, A., and S., K. B. (1995) The carboxy-terminal

F domain of the human estrogen receptor: role in the transcriptional activity of

182

References

the receptor and effectiveness of antiestrogens as estrogen antagonists. Mol.

Endocrinol. 9: 814-25.

Moon, R. C, Grubbs, C. J., and Sporn, M. B. (1976) Inhibition of 7,12-dimethylbenz

(a) anthracene -induced mammary carcinogenesis by retinyl acetate. Cancer

Res. 36: 2626-30.

Moon, R. C, Mehta, R. G., and Rao, K. V. N. Retinoids and cancer in experimental

animals. In: M. B. Sporn, A. B. Roberts, and D. S. Goodman (eds.), The

Retinoids: Biology, Chemistry and Medicine, pp. 573-95. New York: Raven

Press, 1994.

Morosetti, R., Grignani, F., Liberatore, C, Pelicci, P. G., Schiller, G. J., Kizaki, M.,

Bartram, C. R., Miller, C. W., and Koeffler, H. P. (1996) Infrequent alterations

of the RARa gene in acute myelogenous leukemias, retinoic acid-resistant acute

promyelocytic leukemias, myelodysplastic syndromes, and cell lines. Blood. 87:

4399-403.

Muindi, J., Frankel, S. R., Miller, W. H., Jr., Jakubowski, A., Scheinberg, D. A., Young,

C. W., Dmitrovsky, E., and Warrell, R. P., Jr. (1992) Continuous treatment with

all-frans-retinoic acid causes a progressive reduction in plasma drug

concentrations: implications for relapse and retinoid "resistance" in patients with

acute promyelocytic leukemia. Blood. 79: 299-303.

Nagy, L., Saydak, M., Shipley, N, Lu, S., Basilion, J. P., Yan, Z. H., Syka, P.,

Chandraratna, R. A., Stein, J. P., Heyman, R. A., and Davies, P. J. (1996)

Identification and characterization of a versatile retinoid response element

(retinoic acid receptor response element-retinoid X receptor response element)

References

in the mouse tissue transglutaminase gene promoter. J. Biol Chem. 271: 4355-

65.

Nawroz, H., Koch, W., Anker, P., Stroun, M., and Sidransky, D. (1996) Microsatellite

alterations in serum DNA of head and neck cancer patients. Nat. Med. 2: 1035-7.

Nicholson, R. C, Mader, S., Nagpal, S., Leid, M., Rochette-Egly, C, and

Chambon, P. (1990) Negative regulation of the rat stromelysin gene promoter by

retinoic acid is mediated by an API binding site. EMBO J. 9: 4443-54.

Nicke, B., Kaiser, A., Wiedenmann, B., Riecken, E. O., and Rosewicz, S. (1999)

Retinoic acid receptor a mediates growth inhibition by retinoids in human colon

carcinoma HT29 cells. Biochem. Biophys. Res. Commun. 261: 572-7.

Ong, D. E. (1984) A novel retinoid binding protein from rat: purification and partial

characterization. J. Biol. Chem. 259: 1476-82.

Oridate, N., Lotan, D., Mitchell, M. F., Hong, W. K, and Lotan, R. (1995) Induction of

apoptosis by retinoids in human cervical carcinoma cell lines. Int. J. Oncol. 7:

433-41.

Oridate, N., Lotan, D., Xu, X. C, Hong, W. K, and Lotan, R. (1996) Differential

induction of apoptosis by all-toms-retinoic acid and N-(A-

hydroxyphenyl)retinamide in human head and neck squamous carcinoma cell

lines. Clin. Cancer Res. 2: 855-63.

Ostrowski, L. E., Krieg, P., Finch, J., Cress, A. E., Nagle, R., and Bowden, G. T. (1989)

Expression of p-actin during progression of mouse skin tumors. Carcinogen. 10:

1439-44.

Pasquali, D., Bellastella, A., Valente, A., Botti, G., Capasso, I., del Vecchio, S.,

Salvatore, M., Colantuoni, V., and Sinisi, A. A. (1997) Retinoic acid receptors

184

References

a, P and y, and cellular retinol binding protein-I expression in breast fibrocystic

disease and cancer. Eur. J. Endocrinol. 137: 410-4.

Petkovich, M., Brand, N. J., Krust, A., and Chambon, P. (1987) A human retinoic acid

receptor which belongs to the family of nuclear receptors. Nature. 330: 444-50.

Posch, K. C, Boerman, M. H. E. M., Bums, R. D., and Napoli, J. L. (1991) Holocellular

retinol binding protein as a substrate for microsomal retinal synthesis.


Posch, K. C, Burns, R. D., and Napoli, J. L. (1992) Biosynthesis of all-trans-retinoic

acid from retinol. Recognition of retinal bound to cellular retinol binding protein

(type I) as substrate by a purified cytosolic dehydrogenase. J. Biol. Chem. 267:

19676-82.

Predki, P. F., Zamble, D., Sarkar, B., and Giguere, V. (1994) Ordered binding of

retinoic acid and retinoid-X receptors to asymmetric response elements involves

determinants adjacent to the DNA-binding domain. Mol Endocrinol. 8: 31-9.

Qiu, H. M., Zhang, W., El-Naggar, A. K, Lippman, S. M., Lin, P. Z., Lotan, R., and Xu

X. C. (1999) Loss of retinoic acid receptor-beta expression is an early event

during esophageal carcinogenesis. Am. J. Pathol. 155: 1519-23.

Ramharack, R., Wyborski, R. J., and Spahr, M. A. (1998) The apolipoprotein(a)

promoter contains a retinoid response element. Biochem. Biophys. Res.

Commun. 245: 194-7.

Redner, R. L., Rush, E. A., Faas, S., Rudert, W. A., and Corey, S. J. (1996) The t(5;17)

variant of acute promyelocytic leukemia expresses a nucleophosmin-retinoic

acid receptor fusion. Blood. 87: 882-6.

185

References

Renaud, J. P., Rochel, N., Ruff, M., Vivat, V., Chambon, P., Gronemeyer, H., and

Moras, D. (1995) Crystal structure of the RARy ligand-binding domain bound to

all-trans retinoic acid. Nature. 378: 681-9.

Richard, S. and Zingg, H. H. (1991) Identification of a reintoic acid response element in

the human oxytocin promoter. J. Biol. Chem. 266: 73-85.

Rigtrup, K. M. and Ong, D. E. (1992) A retinyl ester hydrolase activity intrinsic to the

brush border membrane of rat small intestine. Biochemistry. 31: 2920-6.

Rishi, A. K., Shao, Z. M., Baumann, R. G., Li, X. S., Sheikh, M. S., Kimura, S.,

Bashirelahi, N., and Fontana, J. A. (1995) Estradiol regulation of the human

retinoic acid receptor alpha gene in human breast carcinoma cells is mediated

via an imperfect half-palindromic estrogen response element and Spl motifs.

Cancer Res. 55: 4999-5006.

Rishi, A. K., Gerald, T. M., Shao, Z. M., Li, X. S., Baumann, R. G., Dawson, M. I., and

Fontana, J. A. (1996) Regulation of the human retinoic acid receptor alpha gene

in the estrogen receptor negative human breast carcinoma cell lines SKBR-3 and

MDA-MB-435. Cancer Res. 56: 5246-52.

Rochette-Egly, C, Lutz, Y., Saunders, M., Scheuer, I., Gaub, M.-P., and Chambon, P.

(1991) Retinoic acid receptor y: specific immunodetection and phosphorylation.

J. Cell Biol 115:535-45.

Roman, S. D., Clarke, C. L., Hall, R. E., Alexander, I. E., and Sutherland, R. L. (1992)

Expression and regulation of retinoic acid receptors in human breast cancer

cells. Cancer Res. 52: 2236-42.

186

References

Roman, S. D., Ormandy, C. J., Manning, D. L., Blarney, R. W., Nicholson, R. I.,

Sutherland, R. L., and Clarke, C. L. (1993) Estradiol induction of retinoic acid

receptors in human breast cancer cells. Cancer Res. 53: 5940-5.

Rosewicz, S., Stier, U., Brembeck, F., Kaiser, A., Papadimitriou, C. A., Berdel, W. E.,

Wiedenmann, B., and Riecken, E. O. (1995) Retinoids: effects on growth,

differentiation, and nuclear receptor expression in human pancreatic carcinoma

cell lines. Gastroenterol. 109: 1646-60.

Ross, A. C. (1982) Retinol esterification by mammary gland microsomes from the

lactating rat. J. Lipid Res. 23: 133-44.

Ross, A. C. (1993a) Cellular metabolism and activation of retinoids: roles of cellular

retinoid-binding proteins. FASEBJ. 7: 317-27.

Ross, A. C. (1993b) Overview of retinoid metabolism. J. Nutr. 123: 346-50.

Rottman, J. N., Widom, R. L., Nadal-Ginard, B., Mahdavi, V., and Karathanasis, S. K.

(1991) A retinoic acid-responsive element in the apolipoprotein AI gene

distinguishes between two different retinoic acid response pathways. Mol. Cell.

Biol. 11: 3814-20.

Rubin, M., Fenig, E., Rosenauer, A., Menendez-Botet, C, Achkar, C, Bentel, J. M.,

Yahalom, J., Mendelsohn, J., and Miller Jr, W. H. (1994) 9-cis retinoic acid

inhibits growth of breast cancer cells and down-regulates estrogen receptor RNA

and protein. Cancer Res. 54: 6549-56.

Salbert, G., Fanjul, A., Piedrafita, F. J., Lu, X. P., Kim, S. J., Tran, P., and Pfahl, M.

(1993) Retinoic acid receptors and retinoid X receptor-a down-regulate the

transforming growth factor-p 1 promoter by antagonizing AP-1 activity. Mol.


187

References

Sambrook, J., Fritsch, E. F., and Maniatis, T. Molecular Cloning, A Laboratory Manual,

2nd edition: Cold Spring Harbor Laboratory Press, 1989.

Sandford, A. J. and Pare, P. D. (1997) Direct PCR of small genomic DNA fragments

from serum. BioTechniques. 23: 890-2.

Schieldrop, P. J., Gelling, R. W., Elliot, R., Hewitt, J., Kieffer, T. J., Mcintosh, C. H. S.,

and Pederson, R. A. (1996) Isolation of a murine glucose-dependent

insulinotropic polypeptide (GIP) cDNA from a tumor cell line (STC644) and

quantification of glucose-induced increases in GIP mRNA. Biochim. Biophys.

Acta. 1308: 111-3.

Schindler, J., Matthaei, K. I., and Sherman, M. I. (1981) Isolation and characterization

of mouse mutant embryonal carcinoma cells which fail to differentiate in

response to retinoic acid. Proc. Natl. Acad. Sci. USA. 78: 1077-80.

Schrader, M., Bendik, I., Becker-Andre, M., and Carlberg, C. (1993) Interaction

between retinoic acid and vitamin D signalling pathways. J. Biol. Chem. 268:

17830-6.

Schule, R., Umesono, K., Mangelsdorf, D. J., Bolado, J., Pike, J. W., and Evans, R. M.

(1990) Jun-Fos and receptors for vitamins A and D recognize a common

response element in the human osteocalcin gene. Cell. 61: 497-504.

Schule, R., Rangarajan, P., Yang, N, Kliewer, S., Ransone, L. J., Bolado, J., Verma, I.

M., and Evans, R. M. (1991) Retinoic acid is a negative regulator of AP-1-

responsive genes. Proc. Natl. Acad. Sci. USA. 88: 6092-6.

Shao, Z.-M., Sheikh, M. S., Chen, J.-C, Kute, T., Aisner, S., Schnaper, L., and Fontana,

J. A. (1994) Expression of retinoic acid nuclear receptors (RARs) and retinoid X

References

receptor (RXR) genes in estrogen receptor positive and negative breast cancer.

Int. J. Oncol. 4: 859-63.

Sheikh, M. S., Shao, Z. M., Chen, J. C, Hussain, A., Jetten, A. M., and Fontana, J. A.

(1993) Estrogen receptor-negative breast cancer cells transfected with the

estrogen receptor exhibit increased RAR alpha gene expression and sensitivity to

growth inhibition by retinoic acid. J. Cell. Biochem. 53: 394-404.

Sheikh, M. S., Shao, Z. M., Li, X. S., Dawson, M., Jetten, A. M., Wu, S., Conley, B. A.,

Garcia, M., Rochefort, H., and Fontana, J. A. (1994) Retinoid-resistant estrogen

receptor-negative human breast carcinoma cells transfected with retinoic acid

receptor-alpha acquire sensitivity to growth inhibition by retinoids. J. Biol

Chem. 269: 21440-7.

Sheikh, M. S., Shao, Z. M., Li, X. S., Ordonez, J. V., Conley, B. A., Wu, S., Dawson,

M. I., Han, Q. X., Chao, W. R., Quick, T., and et al. (1995) N-(A-

hydroxyphenyl)retinamide (4-HPR)-mediated biological actions involve retinoid

receptor-independent pathways in human breast carcinoma. Carcinogen. 16:

2477-86.

Sheikh, M. S., Augereau, P., Chalbos, D., Garcia, M., and Rochefort, H. (1996)

Retinoid regulation of human cathepsin D gene expression. J.'Steroid Biochem.

Mol Biol. 57:283-91.

Siebert, P. D. and Larrick, J. W. (1992) Competitive PCR. Nature. 359: 557-8.

Skare, K. L., Schnoes, H. K., and DeLuca, H. F. (1982) Biliary metabolites of all-trans-

retinoic acid in the rat: isolation and identification of a novel polar metabolite.


189

References

Smith, W . C , Nakshatri, H., Leroy, P., Rees, J., and Chambon, P. (1991) A retinoic acid

response element is present in the mouse cellular retinol binding protein I

(mCRBPI) promoter. EMBO J. 10: 2223-30.

Soong, R. and Iacopetta, B. J. (1997) A rapid and nonisotopic method for screening and

sequencing of p53 gene mutations in formalin-fixed, paraffin-embedded tumors.

Mod. Pathol. 10: 252-8.

Sporn, M. B. and Newton, D. L. (1979) Chemoprevention of cancer with retinoids. Fed.

Proc. 38: 2528-34.

Srivastava, R. K., Srivastava, A. R., Cho-Chung, Y. S., and Longo, D. 1. (1999)

Synergistic effects of retinoic acid and 8-Cl-cAMP on apoptosis require caspase-

3 activation in human ovarian cancer cells. Oncogene. 18: 1755-63.

Stephanou, A. and Handwerger, S. (1995) Identification of a composite steroid hormone

response element on the human placental lactogen promoter. Mol. Cell.

Endocrinol. 112: 123-9.

Sun, S. Y., Li, W. B., Yue, P., Lippman, S. M., Hong, W. K, and Lotan, R. (1999)

Mediation of N-(4-hydoxyphenyl)retinamide-induced apoptosis in human cancer

cells by different mechanisms. Cancer Res. 59: 2493-8.

Supino, R., Crosti, M., Clerici, M., Warlters, A., Cleris, L., Zunino, F., and Formelli, F.

(1996) Induction of apoptosis by fenretinide (4HPR) in human ovarian

carcinoma cells and its association with retinoic acid receptor expression. Int. J.

Cancer. 65: 491-7.

Swisshelm, K., Ryan, K., Lee, X., Tsou, H. C, Peacocke, M., and Sager, R. (1994)

Down-regulation of retinoic acid receptor beta in mammary carcinoma cell lines

and its up-regulation in senescing normal mammary epithelial cells. Cell Growth

Different: 133-41.

References

Tanaka, T., de la Concepcion, M. L. R., and De Luca, L. M. (2001) Involvement of all-

trans-retinoic acid in the breakdown of retinoic acid receptors a and y through

proteasomes in MCF-7 human breast cancer cells. Biochem. Pharmacol. 61:

1347-55.

Thaller, C. and Eichele, G. (1990) Isolation of 3,4-didehydroretinoic acid, a novel

morphogenetic signal in the chick wing bud. Nature. 345: 815-9.

Thompson, J. N., Howell, J. M., Pitt, G. A. J., and McLauglin, C. I. (1969) The

biological activity of retinoic acid in the domestic fowl and the effects of

vitamin A deficiency on the chick embryo. Br. J. Nutr. 23: 471-90.

Tomie-Canie, M., Day, D., Samuels, H. H., Freedberg, I. M., and Blumenberg, M.

(1996) Novel regulation of keratin gene expression by thyroid hormone and

retinoid receptors. J. Biol. Chem. 271: 1416-23.

Topper, J. and Freedman, C. (1980) Multiple hormone interactions in the developmental

biology of the mammary gland. Physiol. Rev. 60: 1049-60.

Torchia, J., Rose, D. W., Inostroza, J., Kamei, Y., Westin, S., Glass, C. K, and

Rosenfeld, M. G. (1997) The transcriptional co-activator p/CIP binds CBP and

mediates nuclear-receptor function. Nature. 387: 677-84.

Torchia, J., Glass, C, and Rosenfeld, M. G. (1998) Co-activators and co-repressors in

the integration of transcriptional responses. Curr. Opin. Cell Biol. 10: 373-83.

Torma, H., Rollman, O., and Vahlquist, A. (1993) Detection of mRNA transcripts for

retinoic acid, vitamin D3, and thyroid hormone (c-erb-A) nuclear receptors in

human skin using reverse transcription and polymerase chain reaction. Acta

Dermato Venereol. 73: 102-7.

References

Urbach, J. and Rando, R. R. (1994) Isomerization of all-fraws-retinoic acid to 9-cis-

retinoic acid. Biochem. J. 299:.

van der Burg, B., van der Leede, B.-J., Kwakkenbos-Isbrucker, L., Salverda, S., de Laat,

S. W., and van der Saag, P. T. (1993) Retinoic acid resistance of estradiol-

independent breast cancer cells coincides with diminished retinoic acid receptor

function. Mol. Cell. Endocrinol. 91: 149-57.

van der Leede, B. J., Folkers, G. E., van den Brink, C. E., van der Saag, P. T., and van

der Burg, B. (1995) Retinoic acid receptor alpha 1 isoform is induced by

estradiol and confers retinoic acid sensitivity in human breast cancer cells. Mol.

Cell. Endocrinol. 109: 77-86.

van der Leede, B. M., Geertzema, J., Vroom, T. M., Decimo, D., Lutz, Y., van der Saag,

P. T., and van der Burg, B. (1996) Immunohistochemical analysis of retinoic

acid receptor-a in human breast tumors: retinoic acid receptor-alpha expression

correlates with proliferative activity. Am. J. Pathol. 148: 1905-14.

Vansant, G. and Reynolds, W. F. (1995) The consensus sequence of a major Alu

subfamily contains a functional retinoic acid response element. Proc. Natl. Acad.

Sci. USA. 92: 8229-33.

Vasios, G. W., Gold, J. D., Petkovich, M., Chambon, P., and Gudas, L. J. (1989) A

retinoic acid-responsive element is present in the 5'-flanking region of the

laminin Bl gene. Proc. Natl. Acad. Sci. USA. 86: 9099-103.

Vecchini, F., Lenoir-Viale, M. C, Cathelineau, C, Magdalou, J., Bernard, B. A., and

Shroot, B. (1994) Presence of a retinoid responsive element in the promoter

region of the human cytochrome P4501 Al gene. Biochem. Biophys. Res.

Commun. 201: 1205-12.

192

References

Veronesi, U., De Palo, G., Marubini, E., Costa, A., Formelli, F., Mariani, L., Decensi,

A., Camerini, T., Del Turco, M. R., Di Mauro, M. G., Muraca, M. G., Del

Vecchio, M., Pinto, C, D'Aiuto, G., Boni, C, Campa, T., Magni, A., Miceli, R.,

Perloff, M., Malone, W. F., and Sporn, M. B. (1999) Randomized trial of

fenretinide to prevent second breast malignancy in women with early breast

cancer. J. Natl. Cancer Inst. 91: 1847-56.

Vidal, H., Aubeouf, D., de Vos, P., Staels, B., Riou, J. P., Auwerx, J., and Laville, M.

(1996) The expression of ob gene is not acutely regulated by insulin and fasting

in human abdominal subcutaneous adipose tissue. J. Clin. Invest. 98: 251-5.

Villa, M. L., Ferrario, E., Trabattoni, D., Formelli, F., De Palo, G., Magni, A., Veronesi,

U., and Clerici, E. (1993) Retinoids, breast cancer and NK cells. Br. J. Cancer.

68: 845-50.

Wald, G. (1968) The molecular basis of visual excitation. Nature. 219: 800-7.

Wang, Z., Boudjelal, M., Kang, S., Voorhees, J. J., and Fisher, G. J. (1999) Ultraviolet

irradiation of human skin causes functional vitamin A deficiency, preventable by

all-trans retinoic acid pre-treatment. Nat. Med. 5: 418-22.

Warrell, R. P., Jr. (1993) Retinoid resistance in acute promyelocytic leukemia - New

mechanisms, strategies, and implications. Blood. 82: 1949-53.

Webster, W. S., Johnston, M. C, Lammer, E. J., and Sulik, K. K. (1986) Isotretinoin

embryopathy and the cranial neural crest: an in vivo and in vitro study. J.

Cranio. Genet. Dev. Biol. 6: 211-22.

Weinberger, C, Hollenberg, S. M., Ong, E. S., Harmon, J. M., Brower, S. T.,

Cidlowski, J., Thompson, E. B., Rosenfeld, M. G., and Evans, R. M. (1985a)

Identification of human glucocorticoid receptor complementary DNA clones by

epitope selection. Science. 228: 740-2.

References

Weinberger, C , Hollenberg, S. M., Rosenfeld, M . G., and Evans, R. M . (1985b)

Domain structure of human glucocorticoid receptor and its relationship to the v-

erb-A oncogene product. Nature. 318: 670-2.

Weinberger, C, Thompson, C. C, Ong, E. S., Lebo, R., Gruol, D. J., and Evans, R. M.

(1986) The c-erb-A gene encodes a thyroid hormone receptor. Nature. 324: 641-

6.

Wells, R. A., Catzavelos, C, and Kamel-Reid, S. (1997) Fusion of retinoic acid receptor

alpha to NuMA, the nuclear mitotic apparatus protein, by a variant translocation

in acute promyelocytic leukaemia. Nat. Genet. 17: 109-13.

Wetherall, N. T. and Taylor, C. M. (1986) The effects of retinoid treatment and

antiestrogens on the growth of T47D human breast cancer cells. Eur. J. Cancer

Clin. Oncol. 22: 53-9.

Widschwendter, M., Daxenbichler, G., Dapunt, O., and Marth, C. (1995) Effects of

retinoic acid and gamma-interferon on expression of retinoic acid receptor and

cellular retinoic acid-binding protein in breast cancer cells. Cancer Res. 55:

2135-9.

Widschwendter, M., Daxenbichler, G., Bachmair, F., Muller, E., Zeimet, A. G.,

Windbichler, G., Uhl-Steidl, M., Lang, T., and Marth, C. (1996) Interaction of

retinoic acid and interferon-alpha in breast cancer cell lines. Anticancer Res. 16:

369-74.

Widschwendter, M., Berger, J., Daxenbichler, G., Miiller-Holzner, E., Widschwendter,

A., Mayr, A., Marth, C, and Zeimet, A. G. (1997) Loss of retinoic acid receptor

P expression in breast cancer and morphologically normal adjacent tissue but no

in the normal breast tissue distant from the cancer. Cancer Res. 57: 4158-61.

References

Widschwendter, M., Berger, J., Hermann, M., Muller, H. M., Amberger, A., Zeschnigk,

M., Widschwendter, A., Abendstein, B., Zeimet, A. G., Daxenbichler, G., and

Marth, C. (2000) Methylation and silencing of the retinoic acid receptor-P2 gene

in breast cancer. J. Natl. Cancer. Inst. 92: 826-32.

Willy, P. J., Umesono, K., Ong, E. S., Evans, R. M., Heyman, R. A., and Mangelsdorf,

D. J. (1995) LXR, a nuclear receptor that defines a distinct retinoid response

pathway. Genes Dev. 9: 1033-45.

Wolbach, S. B. and Howe, P. R. (1925) Tissue changes following deprivation of fat-

soluble A vitamin. J. Exp. Med. 42: 743-78.

Wolbach, S. B. and Howe, P. R. (1933) Epithelial repair in recovery from vitamin A

deficiency. J. Exp. Med. 57: 511-26.

Xu, X. C, Clifford, J. L., Hong, W. K, and Lotan, R. (1994a) Detection of nuclear

retinoic acid receptor mRNA in histological tissue sections using nonradioactive

in situ hybridization histochemistry. Diagn. Mol Pathol. 3: 122-31.

Xu, X. C, Ro, J. Y., Lee, J. S„ Shin, D. M., Hong, W. K, and Lotan, R. (1994b)

Differential expression of nuclear retinoid receptors in normal, premalignant,

and malignant head and neck tissues. Cancer Res. 54: 3580-7.

Xu, X. C, Stier, U., Rosewicz, S., El-Naggar, A. K., and Lotan, R. (1996) Selective

suppression of nuclear retinoic acid receptor p gene expression in human

pancreatic carcinomas. Int. J. Oncol. 8: 445-51.

Xu, X. C, Sneige, N., Liu, X., Nandagiri, R., Lee, J., Lukmanji, F., Hortobagyi, G.,

Lippman, S. M., Dhingra, K., and Lotan, R. (1997) Progressive decrease in

nuclear retinoic acid receptor p messenger RNA level during breast

carcinogenesis. Cancer Res. 57: 4992-6.

References

Yang, Q. F., Sakurai, T., Yoshimura, G., Mori, I., Nakamura, M., Nakamura, Y.,

Suzuma, T., Tamaki, T., Umemura, T., and Kakudo, K. (2001)

Hypermethylation does not account for the frequent loss of the retinoic acid

receptor P2 in breast carcinoma. Anticancer Res. 21: 1829-33.

Zelent, A., Krust, A., Petkovich, M., Kastner, P., and Chambon, P. (1989) Cloning of

murine a and p retinoic acid receptors and a novel receptor y predominantly

expressed in skin. Nature. 339: 714-7.

Zelent, A., Mendelsohn, C, Kastner, P., Krust, A., Gamier, J.-M., Ruffenbach, F,

Leroy, P., and Chambon, P. (1991) Differentially expressed isoforms of the

mouse retinoic acid receptor p are generated by usage of two promoters and

alternative splicing. EMBO J. 10: 71-81.

Zhang, X. K., Liu, Y., and Lee, M. O. (1996) Retinoid receptors in human lung cancer

and breast cancer. Mutat. Res. 350: 267-77.

Zhao, J., Araki, N., and Nishimoto, S. K. (1995) Quantitation of matrix Gla protein

mRNA by competitive polymerase chain reaction using glyceraldehyde-3-

phosphate dehydrogenase as an internal control. Gene. 155: 159-65.

Zhou, X. F., Shen, X. Q.5 and Shemshedini, L. (1999) Ligand-activated retinoic acid

receptor inhibits AP-1 transactivation by disrupting c-Jun/c-Fos dimerization.

Mol. Endocrinol. 13: 276-85.

Zou, C. P., Youssef, E. M., Zou, C. C, Carey, T. E., and Lotan R. (2001) Differential

effects of chromosome 3p deletion on the expression of the putative tumor

suppressor RARP and on retinoid resistance in human squamous carcinoma

cells. Oncogene 20: 6820-7.

retinoic acid receptor expression and polymorphisms …

Documents