AN ESTROGENICALLY REGULATED POTENTIAL TUMOR SUPPRESSOR

GENE, TYROSINE γ (PTPγ), IN HUMAN BREAST

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Suling Liu, B.S.

* * * * *

The Ohio State University

2003

Dissertation Committee: Approved by Dr. Young C. Lin, Adviser Dr. Robert W. Brueggemeier ______Dr. Yasuko Rikihisa Adviser Dr. Pui-kai Li Veterinary Biosciences Graduate Program

ABSTRACT

Except for skin cancer, breast cancer is the most common cancer among women

and second only to lung cancer as the primary cause of cancer deaths in women. Among

the endocrine factors associated with breast cancer, are considered to play a

central role in human breast carcinogenesis. In 1988, Henderson et al. showed that breast

cancer risks are increased by long-term exposure to estrogens, such as -17β (E2).

Zeranol (Z) (Ralgro®) is a agent with estrogenic activities and used as a growth promoter in the U.S. beef, veal and lamb industries. We showed E2 and Z induced

human breast epithelial cell neoplastic transformation with the similar potency in the

long-term exposure through the redox-pathway, in which metabolites undergo

redox-cycling and produce free radicals which directly induce DNA damage leading to

tumor initiation, and/or ERβ-mediated pathway and they have the similar potency in

stimulating and inhibiting some target gene expressions in human breast cancer cells.

Protein tyrosine phosphatase γ (PTPγ) is a member of the receptor-like family of

tyrosine-specific and has been implicated as a tumor suppressor gene in

kidney and lung cancers. Little is known about PTPγ expression in human breast. In our

study, we found that PTPγ was mainly immunolocalized to the epithelium and PTPγ

mRNA expression was lower in cancerous than in normal breast tissues, both E2 and Z suppressed PTPγ levels to a much greater degree in cultured normal human breast tissues

ii or epithelial cells co-cultured with stromal cells in comparison with the cultured epithelial

cells alone. The results indicated that both E2 and Z downregulate PTPγ expression in

human breast and that epithelial-stromal cell interaction is important in the regulation of

PTPγ expression by estrogenically active agents. Furthermore, we showed that lower

PTPγ was associated with higher ERα in cancerous human breast tissues, and the

estrogenic downregulation of PTPγ expression by E2 and Z in human breast is associated

with α (ERα). Based on these findings, we hypothesize that PTPγ is a potential estrogen-regulated tumor suppressor gene in human breast cancer that may be involved in neoplastic processes of human breast epithelium, which lead us to examine the action of PTPγ on in vitro growth of MCF-7 human breast cancer cells and compare the estrogenic responses of human breast cancer cells with different expression levels of

PTPγ. Our results showed that PTPγ overexpression decreased the doubling time of

human breast cancer cell line MCF-7 and led to a decrease in the anchorage-independent

growth of MCF-7 cells in soft agar. Furthermore, we described that PTPγ overexpression

could reduce the estrogenic responses of MCF-7 cell proliferation to E2 and Z. Our data

suggested that PTPγ is able to inhibit proliferation and anchorage-independent growth of

human breast cancer cells in vitro and has anti-estrogenic activities in human breast

cancer cells. The PTPγ may function as a potential tumor suppressor gene in regulating

the process of tumorigenesis in human breast.

Collectively, these studies provide new insight into the role of an endocrine

disruptor, , in the neoplastic transformation of human breast and the importance

of PTPγ in the suppression of human breast carcinomas, and shed a light on the tight

relationship among estrogen, PTPγ and breast cancer. The investigations not only further

iii our understanding of human breast cancer biology but also provide rationales for the development of therapeutic approaches targeting on PTPγ as a breast cancer suppressor.

iv

Dedicated to my parents

v ACKNOWLEDGMENTS

I must first thank my advisor Dr. Young C. Lin, whose scientific guidance, support and mentorship has been invaluable. Dr. Lin, over the past four and a half years you have been a constant source of encouragement, advice and financial support for my study and research. I am especially grateful to my PhD committee members Drs. Robert W. Brueggemeier, Yasuko Rikihisa, and Pui-kai Li who provide me advice and direction throughout my graduate studies. I also show my appreciation to Drs. Yasuro Sugimoto and Samuel K. Kulp for their technical support and suggestions. Thanks also go to all members in Dr. Lin’s laboratory for their sincere friendship. To my parents, brother and sister, thank you for your continued support. Finally, to Xi, thank you for your encouragement, support, and love.

vi VITA

October 17, 1974...... Born – Liuyang, Hunan, China

1997...... B.S., Biology University of Science and Technology of China Hefei, Anhui, China

1997 – 1999 ...... Research Associate University of Science and Technology of China Hefei, Anhui, China

1999-present...... Graduate Research Associate, Department of Veterinary Biosciences The Ohio State University Columbus, Ohio

PUBLICATIONS

Research Publications

S. Liu, Y. Sugimoto, S.K. Kulp, J. Jiang, H.L. Chang, K.Y. Park, Y. Kashida, and Y.C. Lin. Estrogenic down-regulation of Protein Tyrosine Phosphatase γ (PTPγ) in human breast is associated with Estrogen Receptor α. Anticancer Research 22: 3917-3924, 2002.

S. Liu, S.K. Kulp, Y. Sugimoto, J. Jiang, H.L. Chang, and Y.C. Lin. Involvement of breast epithelial stromal interactions in the regulation of Protein Tyrosine Phosphatase γ (PTPγ) mRNA expression by estrogenically active agents. Breast Cancer Research and Treatment 71: 21-35, 2002.

S. Liu, S.K. Kulp, Y. Sugimoto, J. Jiang, H.L. Chang, and Y.C. Lin. The (-)-enantiomer of gossypol possesses higher anticancer potency than racemic gossypol in human breast cancer. Anticancer Research 22: 33-38, 2002.

vii S. Liu, Y. Sugimoto, C. Sorio, P. Melotti, and Y.C. Lin. Function analysis of estrogenically regulated Protein Tyrosine Phosphatase γ (PTPγ) in human breast cancer cell line MCF-7. Oncogene (submitted), 2003.

S. Liu, and Y.C. Lin. Transformation of human breast epithelial cells MCF-10A by environmental disruptor, zeranol (Z), and estradiol-17β (E2). The Breast Journal (accepted, in press), 2003.

S. Liu, Y. Sugimoto, H.L. Chang, J. Jiang, K.Y. Park, C. Sorio, P. Melotti, K. Huebner, and Y. C. Lin. Function analysis of estrogenically regulated Protein Tyrosine Phosphatase γ (PTPγ) in vitro in human breast cancer. American Association for Cancer Research 94th Annual Meeting: A4962, 2003.

S. Liu, Y. Sugimoto, S.K. Kulp, J. Jiang, H.L. Chang, K.Y. Park, and Y.C. Lin. Estrogenically active agents downregulate Protein Tyrosine Phosphatase γ (PTPγ) expression in human breast through estrogen receptor α not estrogen receptor β. American Association for Cancer Research 93rd Annual Meeting: A5237, 2002.

S. Liu, S.K. Kulp, J. Jiang, Y. Sugimoto, and Y.C. Lin. Importance of breast epithelial stromal interactions in the regulation of Protein Tyrosine Phosphatase γ (PTPγ) mRNA expression by estrogenically active agents. American Association for Cancer Research 92nd Annual Meeting Proceedings: A4737, 2001.

Y.C. Lin, S. Liu, S.K. Kulp, J. Jiang, Y. Sugimoto, and M.K. Dowd, P. Wan. The (-) – enantiomer of gossypol (GP) is a potent inhibitor of normal and cancerous breast cell growth. American Association for Cancer Research 92nd Annual Meeting Proceedings 42: A387, 2001.

S.K. Kulp, S. Liu, Y. Sugimoto, R.W. Brueggemeier, Y.C. Lin. Localization of Protein Tyrosine Phosphatase γ (PTPγ) to the mammary epithelium of ACI rats” Biol Reprod, 62(suppl 1): A191, 2000.

H.L. Chang, Y. Sugimoto, S. Liu, J. Jiang, S.K. Kulp, K.Y. Park, Y. Kashida, and Y.C. Lin. Regulation of Protein Tyrosine Phosphatase γ (PTPγ) mRNA expression by estrogenically active agents in canine prostate. Biol Reprod, 64 (suppl 1), 2002.

J. Jiang, S.K. Kulp, S. Liu, H.L. Chang, Y. Sugimoto, and Y.C. Lin. Estrogenic action elevates cyclin D1 mRNA expression in canine prostate. Biol Reprod, 63 (suppl 1), 2001.

H.L. Chang, Y. Sugimoto, S. Liu, J. Jiang, S.K. Kulp, R.W. Brueggemeier, and Y.C. Lin. Development of an in vitro model for the screening of biologically active keratinocyte growth factor (KGF/FGF-7) receptor antagonists. Biol Reprod, 63 (suppl 1), 2001.

M.P. Wick, Y. Sugimoto, S. Liu, and Y.C. Lin. Cellular and molecular functions of cultured myogenic cells derived from a day old calf. In vitro Cellular and Developmental viii Biology-Animal (Submitted), 2003.

Y. Kashida, Y. Sugimoto, S. Liu, and Y.C. Lin. Protein Tyrosine Phosphatase γ (PTPγ) mRNA expression in the non-tumorous or tumorous uterus tissues. Biol Reprod, 68(suppl 1): A167, 2003.

K.Y. Park, Y. Sugimoto, S. Liu, H.L. Chang, W. Ye, L.S. Wang, and Y.C. Lin. Involvement of adipocytes on local estrogen level. Biol Reprod, 68(suppl 1): A444, 2003. H.L. Chang, Y. Sugimoto, K.Y. Park, S. Liu, W. Ye, L.S. Wang, Y.W. Huang, and Y.C. Lin. Regulation of estrogen receptor-α mRNA expression by keratinocyte growth factor (KGF) in MCF-7 cells. Biol Reprod, 68(suppl 1): A659, 2003.

J. Jiang, S.K. Kulp, Y. Sugimoto, S. Liu, H.L. Chang, and Y.C. Lin. Effects of estrogens and age on the growth of canine prostate cells. Biol Reprod, 68(suppl 1): A176, 2003.

H.L Chang, Y. Sugimoto, K.Y. Park, S. Liu, W. Ye, L.S. Wang, Y.W. Huang, and Y.C. Lin. Regulation of estrogen receptor α mRNA by Keratinocyte growth factor (KGF) in MCF-7 cells. Biol Reprod, 68(suppl 1): A659, 2003.

J. Jiang, Y. Sugimoto, S.K. Kulp, S. Liu, H.L. Chang, K.Y. Park, and Y.C. Lin. Inhibitory effects of gossypol on human prostate cancer cells-PC3 are associated with transforming growth factor β signal transduction pathway. Anticancer research. 2003 (submitted).

W. Ye, A. Murakami, J. Jiang, S. Liu, H.L. Chang, K.Y. Park, Y. Sugimoto, and Y.C. Lin. Anti-proliferative effects of 1’-acetoxychavicol acetate and auraptene in human breast cancer cells. American Association for Cancer Research 94th Annual Meeting: A2704, 2003.

J. Jiang, S.K. Kulp, Y. Sugimoto, S. Liu, and Y.C. Lin. Cell-specific and age-dependent changes in canine prostatic cell function and growth. The FASEB Journal 15 (5): 22A, 2001.

J. Jiang, S.K. Kulp, Y. Sugimoto, S. Liu, Y.C. Lin. The effects of Gossypol on the invasiveness of MAT-Lylu Cells and MAT-Lylu Cells from the metastasized longs of MAT-Lylu-bearing copenhagen rats. Anticancer research 20: 4591-4598, 2000.

J. Jiang, S. Kulp, Y. Sugimoto, S. Liu, and Y.C. Lin. Antimetastatic effects of gossypol in prostate cancer. 91st AOCS Annual Meeting & Expo. April 25-28. San Diego, California, 2000.

D. Ruan, J. Chen, W. Zhao, S. Liu, W. Gu. Effects of Lead on L-LTP and PPF in Hippocampal Dentate Gyrus. Guangdong Weiliang Yuansu Kexue 5: 25-29, 1998.

J. Jiang, P.K. Gosh, S.K. Kulp, Y. Sugimoto, S. Liu, J. Czekajewski, H.L. Chang, and

ix Y.C. Lin. Effects of gossypol an O2 consumption and CO2 production in human prostate cancer cells. Anticancer research 22: 1491-1496, 2002.

FIELDS OF STUDY

Major Field: Veterinary Biosciences

Studies in Reproductive and Molecular Endocrinology.

x TABLE OF CONTENTS

Page Abstract...... ii

Dedication...... v

Acknowledgments...... vi

Vita...... vii

List of Tables ...... xiii

List of Figures...... xiv

Abbreviations...... xvii

CHAPTERS

1. Introduction

Breast cancer...... 1 Estrogen and breast cancer...... 2 Environmental estrogens and breast cancer...... 3 Zeranol ...... 4 Estrogen receptors (ERs) ...... 6 Protein tyrosine phosphatase γ (PTPγ) ...... 7 Overview of chapters 2 through 5...... 9

2. Transformation of a normal human breast epithelial cell line MCF-10A by an environmental disruptor, Zeranol (Z), and estradiol-17β (E2)

Abstract...... 16 Introduction...... 17 Materials and Methods...... 20 Results...... 25 Discussions ...... 28

xi 3. Regulation of Protein Tyrosine Phosphatase γ (PTPγ) by estrogenically active agents, estradiol-17β (E2) and Zeranol (Z), in human breast

Abstract...... 38 Introduction...... 39 Materials and Methods...... 43 Results ...... 49 Discussions ...... 53

4. Estrogenic downregulation of Protein Tyrosine Phosphatase γ (PTPγ) in human breast is associated with estrogen receptor α (ERα)

Abstract...... 76 Introduction...... 77 Materials and Methods...... 79 Results ...... 85 Discussions ...... 88

5. Functional analysis of estrogenically regulated Protein Tyrosine Phosphatase γ (PTPγ) in human breast cancer cell line MCF-7

Abstract...... 105 Introduction...... 107 Materials and Methods...... 109 Results ...... 115 Discussions ...... 119

6. Conclusions

Perspectives...... 141 Future works and clinical implications...... 145 Concluding remarks...... 147

Literature cited...... 148

xii LIST OF TABLES

Table Page

2.1 Comparison of MCF-10A cell proliferation by doubling time assay ...... 32

xiii LIST OF FIGURES

Figure Page

1.1 Summary of factors influencing breast carcinogenesis ...... 10

1.2 Zeranol (α-zearalanol) and related compounds ...... 11

1.3 The diagram of structure between ERα and ERβ ...... 12

1.4 The diagram for E2 action in human breast cells through ER-mediated pathway...... 13

1.5 The family of receptor-like protein tyrosine phosphatases...... 14

1.6 The diagram for PTPγ structure...... 15

2.1 Effects of E2 or Z on anchorage-independent growth of MCF-10A cells in soft agar...... 33

2.2 Effects of E2 and Z on ERβ expression in transformed MCF-10A cells ...... 35

2.3 Potency comparison of E2 and Z on pS2 mRNA expression in human breast cancerous cell line MCF-7 ...... 36

2.4 Potency comparison of E2 and Z on PTPγ mRNA expression in human breast cancerous cell line MCF-7 ...... 37

3.1 Isolation of epithelial cells and stromal cells from normal human breast tissues and phase contrast photomicrograph of cultured human breast epithelial cells and stromal cells...... 60

3.2 Modified in vitro co-culture assay system...... 61

3.3 Comparison of PTPγ mRNA expression by RT-PCR in normal human breast tissues and cancerous human breast tissues ...... 62

xiv 3.4 Comparison of PTPγ mRNA expression by RT-PCR in normal human breast tissues, breast epithelial cells and stromal cells ...... 64

3.5 Immunocytochemical staining of primary cultured normal human breast epithelial cells and stromal cells ...... 66

3.6 Regulation of PTPγ mRNA expression in normal human breast tissues by E2 and Z...... 68

3.7 Regulation of PTPγ mRNA expression in normal human breast epithelial cells by E2 and Z ...... 70

3.8 Immunohistochemical localization and reactivity of PTPγ in normal and cancerous human breast tissues, and the regulation of PTPγ immunohistochemical reactivity by E2 and Z in normal human breast tissue ...... 72

3.9 Regulation of PTPγ mRNA expression E2 and Z in normal human breast epithelial cells and stromal cells in co-culture assay system ...... 74

4.1 Comparison of estrogen receptors (α and β) mRNA expression in MCF-7 cells and MDA-MB-231 cells as determined by RT-PCR...... 92

4.2 Comparison of estrogen receptors (α and β) protein expression in MDA- MB-231 cells, MDA-MB-231-ERα-1000 cells and MCF-7 cells as determined by western blotting assay...... 94

4.3 Effects of E2, Z and ICI 182,780 on PTPγ mRNA expression in MCF-7 cells as determined by RT-PCR...... 95

4.4 Effects of E2, Z and ICI 182,780 on PTPγ mRNA expression in MDA- MB-231 cells as determined by RT-PCR ...... 97

4.5 Effects of E2, Z and ICI 182,780 on PTPγ mRNA expression in MDA- MB-231-ERα-1000 cells as determined by RT-PCR...... 99

4.6 Comparison of ERα mRNA expression in human breast as determined by RT-PCR...... 101

4.7 Comparison of ERβ mRNA expression in human breast as determined by RT-PCR...... 102

4.8 Comparison of immunoreactivities of ERα, ERβ and PTPγ in normal and cancerous human breast tissues as determined by immunohistochemical staining...... 103 xv 5.1 The procedures for the establishment of stably transfected cell lines ...... 122

5.2 The sequences for the inserted PTPγ cDNAs ...... 124

5.3 Identification and confirmation of successfully transfected cell lines by RT-PCR...... 125

5.4 Identification and confirmation of successfully transfected cell lines by RT-PCR...... 127

5.5 The expression of PTPγ mRNA in transfected MCF-7 cells...... 129

5.6 Immunohistochemical staining for PTPγ in transfected MCF-7 cells ...... 131

5.7 Effects of PTPγ on the proliferation of MCF-7 cells...... 133

5.8 Optimization of cell numbers for anchorage-independent growth of MCF- 7 cells ...... 135

5.9 Effects of PTPγ on anchorage-independent growth of MCF-7 cells ...... 137

5.10 Effects of PTPγ on estrogenic activities of E2 and Z on MCF-7 cell proliferation...... 139

xvi ABBREVIATIONS

CA: -like domain

CD: Cell doubling

CoR: Co-repressor

CYP:

DMEM/F12: Dulbecco’s Modified Eagle Medium/Ham’s F12

DBD: DNA binding domain

DCC: Dextran-coated charcoal

DES:

DMSO: Dimethylsulphoxide

DNA: Deoxyribonucleic acid

E2: 17β-Estradiol

EC: Epithelial Compartment

EDTA: Ethylenediaminetetraacetic acid

ER: Estrogen receptor

ERE: Estrogen-responsive element

FBS: Fetal Bovine Serum

Fn-III: Fibronectin type III-like repeat

HBEC: Human breast epithelial cell

xvii HS: Horse serum

HSP: Heat-shock protein

LBD: Ligand binding domain

LOH: Loss of heterozigosity

NLS: Nucleus localization sequence

PBS: Phosphate buffered saline

PTPases: Protein tyrosine phosphatases

PTPγ: Protein Tyrosine Phosphatase γ

RPTPases: Receptor-like protein tyrosine phosphatases

RT-PCR: Reverse Transcriptase- Chain Reaction

SC: Stromal Compartment

SD: Standard deviation

Z: Zeranol

ZL:

xviii CHAPTER 1

INTRODUCTION

Breast cancer

Excluding cancers of the skin, breast cancer is the most common cancer among women, accounting for nearly one of every three cancers diagnosed in American women.

[American Cancer Society, 2001]. Currently, one out of nine American women will develop breast cancer in her lifetime. The rate of breast cancer incidence has increased 2-

3% per year over the past decade in both premenopausal and postmenopausal women.

Every year approximately 200,000 American women are diagnosed with breast cancer, of which 40,000 will die from the disease. The mortality rate from breast cancer has not risen as dramatically due to early detection and treatment of breast cancer patients.

Breast cancer results from a combination of many factors including inherited mutations or polymorphism of cancer susceptibility genes, environmental agents that influence the acquisition of somatic genetic changes and several other systemic and local factors (Figure 1.1). Like most cancers, breast cancer arises from alterations in expression

1 and/or regulation of genes that control normal cellular proliferation and differentiation.

Two major classes of genes that are altered in cancers are proto-oncogenes and tumor suppressor genes. Proto-oncogenes control normal growth functions including cell signaling and cell cycle progression. Upon mutation, proto-oncogenes become oncogenes that may act in a dominant-negative manner to permit uncontrolled cellular proliferation.

To maintain controlled cellular proliferation, the cell contains innate growth suppressors encoded by tumor suppressor genes. Like proto-oncogenes, tumor suppressors also maintain cellular homeostasis at many different levels. Alteration of tumor suppressor expression and/or regulation results in uncontrolled proliferation, which can lead to tumor formation. A major obstacle in the cure for cancer is elucidation of the mechanisms by which proto-oncogenes and tumor suppressors are functionally regulated and how they control cellular proliferation.

Estrogen and breast cancer

Among the endocrine factors associated with breast cancer, estrogens are considered to play a central role in human breast carcinogenesis. Support for the critical role of ovarian estrogens in breast tumorigenesis include the associations between increased risk of breast cancer and the early onset of menarche, late onset of menopause, and higher age at first birth [Bernstein and Rose, 1993; Colditz et al., 1996]; as well as the significantly lower frequency at which breast cancer occurs in men and in women without functional ovaries during her lifetime than in women with intact ovaries

[Henderson et al., 1988; Lippman and Dickson, 1989]. Approximately 70% of all breast cancer patients have hormone-dependent breast cancer, which contains estrogen receptors

2 and requires estrogen for tumor growth. This importance of estrogens in breast cancer is a

basis for concern about exposure to environmental estrogens.

Environmental estrogens and breast cancer

Endogenous estrogens are not the only sources of estrogen suspected of

promoting breast cancer. Many compounds found in the environment, such as ,

herbicides, and industrial byproducts, are able to bind to the receptor and have estrogenic

activities in vitro. Exposure to environmental estrogens has been attributed as one

possible cause for the observed increase in breast cancer incidence since 1940.

Phytoestrogens are plant-derived compounds (, , chalcones, and

) that also activate the estrogen receptor. These compounds display a range of

binding affinities and agonistic activities with the estrogen receptor [Collins et al., 1997;

Miksicek et al., 1995; Kuiper et al., 1998; Zava et al., 1997]. Compounds with the highest

affinities for ERα, , zearalenone, , and have relative binding

affinities of 20, 7, 4, and 0.1%, respectively, as compared to estradiol at 100% [Kuiper et

al., 1998]. In addition, genistein and coumestrol were shown to bind to ERβ with a 7-fold

and 2-fold higher affinity than ERα, respectively [Kuiper et al., 1997]. act as estrogen agonists in most tissues. Soy products and coumestrol alone have been shown to lower serum cholesterol and prevent bone loss [Dodge et al., 1996; Carroll et al.,

1991]. Recent research suggests that the consumption of soy products, which contain large amount of phytoestrogens, may be linked to the lower rates of breast cancer incidence seen in women living in China and Japan. Women in these countries have a four- to six-fold lower risk of breast cancer than women in western countries [Ziegler et

3 al., 1993; Lee et al., 1991]. The antitumor and chemopreventive activity of these compounds may be derived from non-estrogen receptor activity of these compounds. In vitro, genistein has been shown to inhibit tyrosine activity, DNA II activity, and angiogenesis [Fotsis et al., 1993; Akiyama et al., 1987; Peterson et al.,

1995]. In addition, some compounds inhibit the metabolizing , and 17β-hydroxysteroid , which regulate local estrogen concentrations [Makela et al., 1995; Kellis et al., 1984].

Zeranol (Z)

Zeranol (Z) (Ralgro) (Figure 1.2), which is marketed as Ralgro (Schering-

Plough Corp, Kenilworth, NJ 07033), is synthesized from the mycotoxin Zearalenone

(ZL, a nonsteroidal, resorcyclic acid lactone compound produced by fungi of the genus

Fusarium) and is a nonsteroidal agent with estrogenic activities that is used as a growth promoter in the U.S. beef, veal and lamb industries to accelerate weight gain, improve feed conversion efficiency and increase the lean meat-to-fat ratio. Thus, Z is not an environmental contaminant per se. Rather, people are exposed to Z as a result of the introduction of the compound into food animals by veterinary professionals on behalf of beef industry farmers. Given the documented estrogenicity of ZL and its potential hazard to human health, it is interesting that a commercially produced derivative of ZL is currently utilized to great advantage as a growth promotant in beef cattle, veal calves and sheep in the U.S.. Like its precursor ZL, Z has been shown to have estrogenic activities.

This evidence includes: (1) Z-induced increases in uterine weight and synthesis of uterine induced protein [Katzenellenbogen et al., 1979]; (2) stimulation of beef heifer [Moran et

4 al., 1991] and mouse mammary gland growth by Z with, in the latter case, a potency

similar to that of estradiol [Sheffield and Welsch, 1985]; (3) specific binding of Z to

uterine cytosolic and nuclear estrogen receptors [Katzenellenbogen et al., 1979]; (4) Z-

stimulated proliferation of cultured normal human breast cells [Lin et al., 1996] and the

estrogen-sensitive human breast cancer cell line, MCF-7 [Welshons et al, 1990]; (5) Z-

induced stimulation of estrogen-inducible gene expression (KGF, ) in

cultured normal human breast cells and human breast cancer cell line, MCF-7 [Zhang et

al., 1999; Zheng et al., 2000] and primary cultured human breast cancer cells [Lin et al.,

1996]; (6) Z-induced reduction of estrogen-suppressible gene expression (protein tyrosine

phosphatase-γ) in cultured normal and cancerous human breast cells and human breast

cancer cell lines [Zheng et al., 2000] and (7) induction by Z of hepatotoxicity and

subsequent hepatic neoplasia in the Armenian hamster, an animal that is especially

susceptible to exogenous estrogen-induced liver damage, and prevention of this

carcinogenic process by the , [Coe et al., 1992]. Furthermore, in

studies that examined the estrogenic activities of both ZL and Z, the synthetic product, Z,

was found to have the higher affinity for the estrogen receptor and, not coincidentally, to

be the more active compound [Katzenellenbogen et al., 1979; Welshons et al, 1990].

Furthermore, we have shown that meat and serum from Z-implanted cattle possess heat-

stable mitogenicity for cultured breast cells, and that both normal and cancerous human breast cells exhibit estrogenic responses to Z [Lin et al., 1996; Lin et al., 2000; Irshaid et al., 1999].

5 Estrogen receptors (ERs)

About 10 years after the cloning of the estrogen receptor (ER), a novel member of

ERs, termed ERβ, has been identified in cDNA libraries from rat prostate. The rat ERβ is a protein of 485 amino acids residues with a calculated molecular weight of 54.2 kDa (rat

ERα: 595 a.a., 66 kDa). ERβ is highly homologous to ERα, particularly in the DNA- binding domain and in the C-terminal ligand-binding domain. ERα and ERβ are functionally homologous and both bind to estrogen with high affinity. Some differences exist between ERα and ERβ such as aspects of regulation and they may response to some antagonists in different ways due to the lack of homology in the amino-terminal domains of these where the activation function-1 (AF-1) resides (Figure 1.3).

As we have known, the mechanisms responsible for estrogen-stimulated carcinogenesis are still unclear. One possible mechanism is that E2 acts through an ER-

mediated pathway (Figure 1.4). E2 enters the ER-positive cells just by simple diffusion

and binds to the ERs in the cells. Binding of E2 to ERs results in the conformational

changes in structure that convert the receptor from an inactive to an active conformation

and the subsequent release of the associated heat-shock proteins. The changes result in

the formation of “activated” or “transformed” ER-estradiol complex, dimerization and

phosphorylation of the complex that has a high affinity for the specific estrogen response

elements (EREs) in the estrogen targeted genes in nucleus. The binding of E2-ER complex to the regulatory elements usually involves a variety of other proteins such as other DNA binding proteins (coactivators) or transcription factors and results in gene activation, i.e., transcription of the gene by RNA polymerase to produce mRNA. The transcribed mRNA in nucleus is translocated to the cytoplasm and translated on

6 cytoplasmic ribosomes to produce the appropriate protein, which alters cell function, growth, or differentiation and result in estrogenic effects in breast cells. Except for ER- mediated pathway, evidence indicates potential tumorigenic mechanisms of estrogen, such as direct genotoxic effects of estrogen metabolites and estrogen-induced expression of genes encoding growth and transcription factors [Li and Li, 1987; Lippman and

Dickson, 1989].

Protein tyrosine phosphatase γ (PTPγ)

Previous reports showed that the of PTPases comprises some

50 members which display a bewildering variety of amino acid motifs fused to the catalytic PTPase domains [Barford et al., 1995; Brady-Kalnay et al., 1995]. The PTPase family can be divided in two main classes based on their subcellular localization, namely

(i) intracellular, cytosolic or nuclear PTPases that contain only one PTPase domain, and

(ii) receptor-like, transmembrane PTPases that have one or two tandemly repeated catalytic domains. The receptor-like PTPases (RPTPases) can be classified in seven subtypes based on their extacellular motifs (Figure 1.5). Two other protein families are as well capable of dephosphorylating tyrosine-phosphorylated proteins: the structurally distinct low molecular-weight PTPases and the structurally related dual specificity phosphatases which catalyze the dephosphorylation of phospho-tyrosine, threonine and serine residues [Barford et al., 1995;].

PTPγ is a member of the receptor-like family of tyrosine-specific phosphatases originally cloned from human brain stem and placental cDNA libraries using probes derived from the intracellular domain of CD45 [Kaplan et al., 1990] or Drosophila

7 PTPase cDNA clone, DPTP12 [Kaplan et al., 1990], respectively. The structure of the receptor-like PTPs (RPTPs) includes an extracellular, a transmembrane, and one or two tandemly repeated catalytic domains (Figure 1.6). This structure implies ligand-binding capability which may modulate enzymatic activity. The putative ligands for most of the

PTPs with receptor-like structures are yet to be identified. One isoform of leucocyte common antigen (LCA) [CD45RO], a type I RPTP, has been shown to functionally interact with CD22β [Stamenkovic et al., 1991]. Two type II RPTPs, RPTPκ and RPTPµ, participate in homophilic binding, although this self-binding has not been shown to affect the PTP activity of either [Zondag et al., 1995]. Two type V PTPs have been reported to date: PTPγ and PTPζ [Barnea et al., 1993; Levy et al., 1993; Krueger et al.,

1992]. These receptor-like PTPs each have a carbonic anhydrase-homologous amino terminus followed by a fibronectin type three domain, a variable length cysteine-free domain, a transmembrane domain and tandem intracellular PTPase catalytic domains

[Barnea et al., 1993; Levy et al., 1993; Krueger et al., 1992]. According to mRNA analysis [LaForgia et al., 1991; Barnea et al., 1993], PTPγ is a broadly expressed enzyme that exists in many tissues, including human lung, stomach, esophagus, colon, liver, spleen, and kidney [Tsukamoto et al., 1992]. In contrast, the closely related PTPζ/RPTPβ is expressed solely in specific regions of the central nervous system and can bind to the extracellular matrix protein tenascin [Levy et al., 1993; Krueger et al., 1992]. Based on the chromosomal location of the PTPγ gene (3p14.2) [LaForgia et al., 1991; LaForgia et al., 1993] and studies showing loss of heterozygosity of the gene in kidney tumors

[Lubinski et al., 1994], PTPγ has been implicated as a candidate tumor suppressor gene.

8 Overview of Chapters 2 through 5

The next four chapters will investigate the estrogenic regulation and the function

of PTPγ in human breast. Chapter 2 looks at the estrogenic potency of Z in comparison

with E2 in human breast and provides evidences that Z has the ability to induce human

breast carcinogenesis, which is just as potent as E2. Chapters 3 and 4 focus primarily on the estrogenic regulation of PTPγ expression in human breast. Chapter 3 elucidates that breast carcinogenesis can reduce PTPγ expression and both E2 and Z can downregulate

PTPγ expression in human breast. Then, chapter 4 defines the mechanism of the

downregulation of PTPγ expression by E2 and Z, which is associated with ERα. Until

now, we have known that there are tight relationships between estrogen and breast

cancer, estrogen and PTPγ, breast cancer and PTPγ, which indicate that PTPγ might play

an important role in human breast carcinogenesis. Finally, chapter 5 addresses the critical

issue of PTPγ’s anti-tumorigenesis in human breast, which is examined using an in vitro

system. Together, these studies provide a better understanding of the estrogenic

regulation and function of PTPγ in human breast and shed a light on the tight relationship

among estrogen, PTPγ and breast cancer, which suggest that PTPγ is a potential tumor

suppressor gene in human breast.

9

…… Chemicals, radiation, … Genetic alterations

Family history Risk Factors Diets

Systemic factors Endocrine disruptors (hormones, GFs,… ) (Zeranol,… )

Figure 1.1. Summary of factors influencing breast Carcinogenesis.

10

Figure 1.2. Zeranol (α-zearalanol) and related compounds.

All are beta resorcyclic acid lactones and each can be metabolized/converted into all the other compounds, albeit with different efficiencies (Leffers et al., 2001).

11

Homology rERß 16.5% 95.5% 28.9% 59.7% 16.7%

rERα 100% 100% 100% 100% 100%

A/B C D E F Modulator DBD Hinge LBD

NH2 - - COOH

AF-1 Zn++ CTE CoR-box AF-2 NLS

Transactivation DNA binding Ligand binding Dimerization Dimerization , transactivation Hsp binding, NLS Hsp binding, NLS Coactivator and Corepressor binding

Figure 1.3. The diagram of structure homology between ERα and ERβ.

The figure shows the comparison of rat (r) ERα and ERβ proteins and percent amino acid homology in the functional regions. The ERs consist of six functional domains (A-F).

DBD: DNA binding domain; LBD: ligand binding domain; AF: activation function; HSP: heat-shock protein; NLS: nucleus localization sequence; CoR: co-repressor. Adapted with modification from reference [Macgregor et al., 1998].

12

Cell membrane Nuclear membrane

E2 Diffusion Diffusion Diffusion

E2 E2

ER E2-ER dimerization

ERE DNA

Processing

AUG UAA 5’ 3’ mRNA

Altered cell function

Export New protein Translation

Figure 1.4. The diagram for E2 action in human breast cells through ER-mediated pathway.

ERE: estrogen response element

13

Figure 1.5. The family of receptor-like protein tyrosine phosphatases.

The figure shows the distinct subtypes of RPTPases based on their extracellular motifs.

The classification is according to Brady-Kalnay and Tonks [1995], except for ChPTPλ, which can be regarded as a CD45-type RPTPase. Only the largest isoform of each member is shown. The identified structural features are indicated in the open box. Fn-III: fibronectin type III-like repeat; Ig: immunoglobulin-like domain; MAM: domain with homology to meprin, the A5 glycoprotein, and RPTPµ; CA: carbonic anhydrase-like domain; PTP: protein tyrosine phosphatase domain. Adapted with modification from reference [Schaapveld et al, 1997].

14

CA-like CA: Carbonic anhydrase Extracellular Fn III: Fibronectin type III Fn III-like

Membrane Transmembrane

PTP1 PTP: protein tyrosine Intracellular phosphatase PTP2

Figure 1.6. The diagram for PTPγ structure.

The figure shows that PTPγ is a member of RPTPases and it has an extracellular domain, a transmembrane domain and an intracellular domain. In the intracellular domain, PTP1 is active, whereas PTP2 is inactive due to the replacement of cysteine by aspartic acid.

Adapted with modification from reference [Schaapveld et al, 1997].

15 CHAPTER 2

TRANSFORMATION OF A NORMAL HUMAN BREAST EPITHELIAL CELL LINE

MCF-10A BY AN ENVIRONMENTAL DISRUPTOR, ZERANOL (Z), AND

ESTRADIOL-17β (E2)

ABSTRACT

Among the endocrine factors associated with breast cancer, estrogens are

considered to play a central role in human breast carcinogenesis. Breast cancer risks are

increased by long-term exposure to estrogens. Zeranol (Z) (Ralgro) is a nonsteroidal agent with estrogenic activities that is used as a growth promoter in the U.S. beef, veal and lamb industries. To determine whether Z and estradiol-17β (E2) play a role in the

neoplastic transformation of human breast and to compare the estrogenic potency of Z to

that of E2 in human breast, we treated the immortalized human breast epithelial cell MCF-

10A with different doses of Z or E2 for 10 repeated treatment cycles (three days per

cycle). By utilizing doubling time assay, soft agar assay and Reverse Transcriptase-

Polymerase Chain Reaction (RT-PCR) assay, we showed that ten repeated E2- or Z-

16 treatment cycles to MCF-10A cells decrease the doubling time of the cells by 30% ~

40%, and stimulate colony formation in soft agar and induce estrogen receptor β (ERβ)

mRNA expression, all of which are not dose-related in our tested dose range.

Furthermore, we described that Z and E2 have the similar potency in the stimulation and

inhibition of gene expressions in human breast cancer cell line MCF-7 by RT-PCR.

These results indicate that both Z and E2 can induce human breast epithelial cell neoplastic transformation with the similar potency in the long-term exposure through the redox-pathway, in which estrogen metabolites undergo redox-cycling and produce free radicals which directly induce DNA damage leading to tumor initiation, and/or ERβ- mediated pathway.

INTRODUCTION

Except for skin cancer, breast cancer is the most common cancer among women and is second only to lung cancer as the primary cause of cancer deaths in women

[American Cancer Society, 2001]. Among the endocrine factors associated with breast cancer, estrogens are considered to play a central role in human breast carcinogenesis.

This importance of estrogens in breast cancer is a basis for concern about exposure to environmental estrogens.

Endogenous estrogens are not the only sources of estrogen suspected of promoting breast cancer. Zeranol (Z) (Ralgro), which is marketed as Ralgro (Schering-

17 Plough Corp, Kenilworth, NJ 07033), is synthesized from the mycotoxin Zearalenone

(ZL, a nonsteroidal, resorcyclic acid lactone compound produced by fungi of the genus

Fusarium) and is a nonsteroidal agent with estrogenic activities that is used as a growth

promoter in the U.S. beef, veal and lamb industries as described in Chapter 1. Thus, Z is

not an environmental contaminant per se. Rather, people are exposed to Z as a result of

the introduction of the compound into food-producing animals by veterinary

professionals on behalf of beef industry farmers.

Evidence indicates potential tumorigenic mechanisms of estrogen, such as direct

genotoxic effects of estrogen metabolites and estrogen-induced expression of genes

encoding growth and transcription factors [Li and Li, 1987; Lippman and Dickson,

1989]. However, despite the clear importance of estrogens in the etiology of breast

cancer, the mechanisms responsible for estrogen-stimulated carcinogenesis remain undefined. In the receptor-mediated pathway, estradiol and its metabolites bind to the estrogen receptor and activate gene expression to promote cell proliferation, which makes estrogens excellent tumor promoters after the initial cellular damage is induced. In the redox-mediated pathway, estrogen metabolites undergo redox-cycling and produce free radicals which directly induce DNA damage leading to tumor initiation. E2, under the

effect of 17β- is continuously interconverted to (E1), and both are

hydroxylated at C-2, C-4, or C-16 positions by cytochrome P450 isoenzymes, i.e.

CYP1A1, CYP1A2, or CYP1B1, to form catechol estrogens [Liehr et al., 1986; Roy and

Liehr, 1988; Yan and Roy, 1997; Ball and Knuppen, 1980; Zhu et al., 1994; Ashburn et

al., 1993]. The metabolic activation of estrogens can be mediated by various cytochrome

P450 (CYP) complexes, generating through this pathway reactive intermediates that elicit

18 direct genotoxic effects by increasing mutation rates. Estrogen and estrogen metabolites exert direct genotoxic effects that might increase mutation rates, or compromise the DNA repair system, leading to the accumulation of genomic alterations essential to tumorigenesis [Liehr et al., 1986; Roy and Liehr, 1988; Yan and Roy, 1997; Ball and

Knuppen, 1980; Zhu et al., 1994; Ashburn et al., 1993]. Previous work showed that short-term treatment of MCF-10F cells with physiological doses of E2 induces anchorage

independent growth and colony formation in agar methocel, which indicate neoplastic

transformation of human breast [Russo et al., 2002]. The fact that the MCF-10F cells are

both ERα- and ERβ-negative, are in favor of a metabolic activation of estrogens

mediated by various cytochrome P450 (CYP) complexes, generating through this

pathway reactive intermediates that elicit direct genotoxic effects by increasing mutation

rates [Russo et al., 2002].

Two estrogen receptor types, named ERα and ERβ, have been found to be the

major mediators of a variety of biological functions of estrogens [Warner et al., 1999;

Gupta et al., 2001; Ogawa et al., 2000; Hilakivi, 2000]. ERβ is similar to ERα with

approximately 96% and 60% homology in the DNA-binding domains and ligand-binding

domains, respectively. However, their exact roles are still poorly elucidated, especially in

the case of the recently discovered ERβ [Warner et al., 1999]. On the other hand, it is

becoming increasingly clear that both receptor types are responsible for different

biological functions, as indicated by their specific expression patterns and various

consequences in gene knockouts [Lubahn et al., 1993; Krege et al, 1998; Couse et al.,

1999; Dupont et al., 2000]. It is thought that ERβ may also have distinct functions in the

biology of breast cancer. Moreover, both ERs work as either homo- or heterodimers, 19 when inducing transcription from gene promoters equipped with estrogen response

elements (ERE) [Kumar and Chambon, 1988; Cowley et al., 1997; Pace et al., 1997;

Ogawa et al., 1998]. In addition, estrogens and anti-estrogens can induce differential

activation of ERα and ERβ to control transcription of genes that are under the control of Activating Protein 1 (AP1) sites [Paech et al., 1997].

In the present study, we have shown that MCF-10A cells treated with either E2 or

Z have acquired tumor cell properties including anchorage-independent cell growth and increased cell proliferation rate, and ERβ mRNA expression is increased in the transformed MCF-10A cells treated with E2 or Z. In addition, E2 and Z have the same

potency in the neoplastic transformation of MCF-10A cells.

MATERIALS AND METHODS

Cell culture. Both MCF-10A and MCF-7 cell lines were purchased from

American Type Culture Collection (ATCC, Manassas, VA). MCF-10A cells were

cultured in phenol red-free low calcium DMEM/F12 (0.04 mM CaCl2) supplemented

with Chelex-100 (Bio-Rad Laboratories, Richmond, CA) – treated Fetal Bovine Serum

(FBS) (10%, Atlanta Biologicals, Norcross, GA). MCF-7 cells were cultured in phenol

red-free high-calcium DMEM/F12 (1.05 mM CaCl2) supplemented with 5% FBS. Both

cell lines were plated separately in 75-cm2 culture flasks in a humidified incubator (5%

CO2: 95% air, 37°C). The media of both human breast cell lines were changed every two

20 days. When the cells grew to 85-90% confluence, cells were washed twice with calcium-

and magnesium-free Phosphate Buffered Saline (PBS, pH7.3), and then trypsinized with

0.5% - 5.3 mM EDTA (GibcoBRL) in PBS for 10 minutes at 37°C. The

trypsinization was stopped by addition of culture medium with 5% or 10% serum. After

centrifugation, The dissociated cell viability (about 99%) were determined by using a

hemacytometer and the cells were resuspended in the same medium and subcultured into

75-cm2 culture flasks at a ratio of 1 flask to 5 flasks.

Cell treatment and RNA extraction. Human breast epithelial cell line MCF-10A

was plated in 75-cm2 culture flasks (1 × 105 viable cells/flask) and treated with 0.1, 1, 10,

100 nM of E2, Z, or vehicle as a control in phenol red-free low-calcium DMEM/F12

supplemented with Dextran-Coated Charcoal (DCC) (Dextran T-70; Pharmacia; activated

charcoal; Sigma)-stripped Chelex-100-treated HS (10%) for 3 days. Treatments were

repeated for 10 times. The cells were passaged at the end of every two treatment cycles.

At the end of both sixth treatment cycle and tenth treatment cycle, partial of the cells

were collected for reverse transcription-polymerase chain reaction assay and cell

proliferation assay. In addition, human breast cancer cell line MCF-7 was plated in 25-

cm2 culture flasks (2 × 105 viable cells/flask) and cultured overnight. The media was

changed to phenol red-free high-calcium DMEM/F12 supplemented with Dextran-Coated

Charcoal (DCC) (Dextran T-70; Pharmacia; activated charcoal; Sigma)-stripped FBS

(5%). After 24 hours, cells were treated with 7.5, 15, 30, 60, 120 nM of E2 or Z or vehicle

as controls in phenol-red-free high-calcium DMEM/F12 supplemented with 5% DCC-

treated FBS for 24 hours. Total RNA was isolated in 2.5 ml TRIZOL Reagent

(GibcoBRL) according to manufacturer’s instructions.

21 Reverse transcription-polymerase chain reaction (RT-PCR). RT-PCR was

performed in a gradient mastercycler (Eppendorf ®). PCR conditions were optimized for

MgCl2 concentration, annealing temperature and cycle number for the amplification of

each PCR product [ERα, ERβ, pS2, protein tyrosine phosphatase γ (PTPγ) and 36B4].

Under optimal conditions, the amounts of PCR products generated fell within the linear

portion of the PCR amplification curve between twenty-six and thirty-nine amplification

cycles. Briefly, 1 µg of total RNA from cultured cells or tissues was reverse transcribed

with 200 U M-MLV Reverse Transcriptase (GibcoBRL) at 42°C for 1 hour in the

presence of 5 mM each of dATP, dCTP, dGTP and dTTP, 4 µl 5X 1st strand buffer

(GibcoBRL), 0.01M DDT, 1 U RNA Guard RNase inhibitor (Pharmacia Biotech,

Uppsala, Sweden), and 2.5 mM random hexamers in a total volume of 20 µl. The reaction

was terminated by heating to 95°C for 3 minutes. The newly synthesized cDNAs were

used as templates for PCR after adjusting reagent concentrations to 1.5 mM (pS2), 1.0

mM (ERα), 2.5 mM (ERβ) or 3.5 mM (PTPγ and 36B4) MgCl2, 2.5 µl 10X PCR Buffer

(GibcoBRL), 1 U Platinum® Taq DNA polymerase (GibcoBRL), and 0.24 µM primers.

The reactant was incubated at 95 °C for 5 minutes. Then, thirty-five cycles (ERα and

ERβ) or thirty cycles (pS2, PTPγ and 36B4) of amplification were performed with each

cycle consisting of denaturation at 95°C for 1 minute, annealing at 63°C for 1 minute,

and extension at 72°C for 1 minute. For pS2, the primer sequences were 5′_TTT GGA

GCA GAG AGG AGG CAA TGG_3′ (sense) and 5′_TGG TAT TAG GAT AGA AGC

ACC AGG G_3′ (antisense); for ERα, they were 5′_TAC TGC ATC AGA TCC AAG

GG_3′ (sense) and 5′_ATC AAT GGT GCA CTG GTT GG_3′ (antisense); for ERß,

22 they were 5′_TGA AAA GGA AGG TTA GTG GGA ACC_3′ (sense) and 5′-TGG TCA

GGG ACA TCA TCA TGG_3′ (antisense); for PTPγ, they were 5′_GCG CAG CGA

CTT TAG CCA GAC GA _3′ (sense) and 5′ _GCT CCC GCT CCC CAT CCT CAC TC

_3′ (antisense); for 36B4, they were 5′ _AAA CTG CTG CCT CAT ATC CG _3′ (sense)

and 5′ _TTG ATG ATA GAA TGG GGT ACT GAT G_3′ (antisense). The final PCR

products (10 µl) mixed with 1 µl of 10 × loading buffer were separated on a 1.0 ~ 1.5%

agarose gel containing ethidium bromide. The specific bands were quantified by

ImageQuaNT software (Molecular Dynamics, Sunnyvale, CA). The results are presented

as the ratio of PTPγ to 36B4. 36B4 is a cDNA clone for human acidic ribosomal

phosphoprotein PO [Masiakowski et al, 1982], for which mRNA levels have been shown

to be unmodified by estradiol treatment [Laborda, 1991].

Doubling time Assay. MCF-10A cells collected at the end of both sixth treatment

and tenth treatment with 0.1, 1, 10, 100 nM of E2, Z or vehicle and wild type MCF-10A cells were plated separately at a density of 5000 in 24-well plates in a volume of 1 ml/well. After cells are attached to the wells, the medium was replaced with 2 ml of fresh low-calcium DMEM/F12 with 10% HS. At the same time (time 0 hour), a group of cells were counted. Cells were grown for 2 days and counted every 4 hours. Adherent cells were detached by rapid trypsinization (about 5 minutes incubation in 1 ml of 0.5% trypsin - 5.3 mM EDTA). An adequate volume of medium containing 50% trypan blue was added. Then cells were counted by use of a hemacytometer. Experiments were performed in 4 replicate culture wells for each group, and each experiment was repeated twice. Based on the counted cell numbers at different time points, a cell proliferation curve was generated. Cell doubling (CD) was calculated using the formula ln (Nj-Ni)/ln 2 23 where Nj or Ni are the cell numbers at different time points Tj or Ti (Tj > Ti) in the growth log phase of the cells. Doubling time (DT) was consequently obtained by dividing the time interval (Tj > Ti) by CD [Poliseno et al., 2002].

Soft agar assay for colony formation. MCF-10A cells collected at the end of both sixth treatment and tenth treatment with 0.1, 1, 10, 100 nM of E2, Z or vehicle and wild type MCF-10A cells were cultured in 6-well plates first covered with an agar layer (1.0 ml of phenol red-free low-calcium DMEM/F12 with 0.5% agar and 10% HS). The middle layer contained 8000 cells in 1.0 ml of phenol red-free low-calcium DMEM/F12 with 0.35% agar and 10% HS. The top layer, consisting of 1 ml of medium, was added to prevent drying of the agar in the plates. The plates were incubated for 30 days and another 1 ml of medium will be added to the top layer at day 15. After 30 days’ incubation, the plates were stained in 0.5 ml of 0.005% crystal violet for >1 hour and the cultures were inspected and photographed.

Statistical analysis. The results for doubling time assay were presented as the mean ± standard deviation (SD) for 4 replicate culture wells as one cell group. Analysis was performed using Minitab statistical software for Windows (Minitab Inc., State

College, PA, USA). Statistical differences were determined by using one-way ANOVA for independent groups. P-values of less than 0.05 were considered statistically significant.

24

RESULTS

Effects of long-term exposure to E2 or Z on the proliferation of MCF-10A cells.

Exposure to estrogen is known to be a contributing factor in the development of breast

and endometrial cancers [Henderson et al., 1988]. Pre-natal exposure to estrogens has

been linked to the development of vaginal, ovarian, testicular, and prostate cancers.

Several risk factors associated with an increased risk of breast cancer reflect an increased

exposure to estrogens. For example, both early age at menarche and old age at

menopause increase a woman’s risk of developing breast cancer [Kelsey et al., 1993]. We

have shown that Z-treatment of cultured human breast cancer cells and human breast

cancer cell lines results in the elevation of mRNA levels of estrogen-regulated genes

involved in the control of normal and cancerous human breast cell proliferation such as

keratinocyte growth factor (KGF/ FGF –7), cyclin D1 and cathepsin [Lin et al., 2000]. To

examine whether long-term exposure to E2 or Z has a mitogenic effect on MCF-10A

cells, a proliferation study was performed by using doubling time assay. Our results

showed that six repeated E2- or Z-treatment cycles have no effects on the doubling time of MCF-10A cells (data not shown), but ten repeated E2- or Z-treatment cycles decrease

the doubling time of MCF-10A cells by 30% ~ 40%, which is not dose-related in our

tested dose range (Table 2.1). These results suggested that long-term exposure to E2 or Z

can stimulate MCF-10A breast epithelial cell growth, and both E2 and Z play important

roles in neoplastic transformation of breast cancer.

25

Effects of E2 or Z on anchorage-independent growth of MCF-10A cells. Short- term treatment of MCF-10F cells with physiological doses of E2 induces anchorage

independent growth and colony formation in agar methocel, which is indicative of

neoplastic transformation [Russo et al., 2002]. In the present work, evaluation of colony

formation at the end of the tenth treatment cycle of E2 or Z reveals that MCF-10A cells

formed colonies in soft agar, and there is no significant difference in the size and colony

efficiency in different doses of E2 or Z-treatment groups and the potency of E2 and Z in

the stimulation of colony formation in MCF-10A cells are similar (Figure 2.1). Whereas,

there are no formed colonies found in the end of sixth treatment of any doses of E2 or Z

(data not shown).

Effects of E2 and Z on ERβ expression in MCF-10A cells. Previous work showed

that ERβ expression can be induced in chemical carcinogen-transformed human breast epithelial cell MCF-10F, and the more transformed cells showed higher levels of ERβ expression, regardless of which chemical carcinogens were initially used for cell transformation [Hu et al., 1998]. As shown in Figure 2.2, both wild type MCF-10A and the transformed MCF-10A do not express ERα and pS2. ERβ is expressed at a very low

level in both wild type MCF-10A and controls, and an elevated level of ERβ expression

is observed in the transformed MCF-10A cells at the end of the tenth different doses of

E2- or Z-treatment and the degree of ERβ expression induction is similar in treatments

with different doses of E2 and Z in our tested range (Figure 2.2). Whereas, ERβ

expression is not changed in the MCF-10A cells in the end of the sixth different doses of

E2- or Z-treatment (data not shown). Furthermore, the expression of protein tyrosine 26 phosphatase γ (PTPγ), an estrogenically regulated tumor suppressor gene, is not changed

by the E2 or Z transformation of MCF-10A cells (Figure 2.2).

Potency comparison of E2 and Z in human breast cell lines. The data presented above showed that E2 and Z had the similar potency in the stimulation of MCF-10A cell

proliferation, anchorage-independent growth of MCF-10A cells and the induction of ERβ

expression in MCF-10A cells. In order to determine whether Z has the similar potency to

E2 in human breast cancer cell lines, we treat estrogen receptor (ER)-positive human

breast cancer cell line MCF-7 with different doses of E2 and Z to determine the dose-

response curve of some gene expression to E2 and Z in MCF-7 cells. As we know, the

pS2 promoter region contains an estrogen response element and pS2 is an estrogen-

regulated gene through estrogen receptor in human breast. Previously, we suggested that

both E2 and Z regulate PTPγ expression in human breast [Liu et al., 2002a], which is

associated with ERα [Liu et al., 2002b].

Our current results show that E2 and Z can upregulate pS2 mRNA expression and

downregulate PTPγ mRNA expression in a dose-response manner, as low as 15 nM of E2 and Z can upregulate pS2 mRNA expression by ~ 30% and downregulate PTPγ mRNA expression by ~ 26% (Figure 2.3, Figure 2.4). These results indicate that E2 and Z have

the similar potency in the regulation of some target gene expressions in ER-positive

MCF-7 cells.

27

DISCUSSIONS

Epidemiological evidence of human breast cancer indicates that white American women have a 5-fold greater risk for breast cancer than Asian women in China and

Japan. Furthermore, the risk of acquiring breast cancer among Asian women immigrants in the U.S. approaches to that of American women after 1 to 2 generations [Kelsey and

Berkowitz, 1988]. These observations indicate a strong environmental, rather than genetic, component in the etiology of this disease [Buell, 1973]. It has been speculated that dietary factors may contribute to this ethnic difference in human breast cancer incidence [Committee on Diet, 1982]. The growth promoter, Z, is a federally approved agent used primarily in the beef, veal and lamb industries in the U.S. Based on our published data generated from the use of cultured normal and cancerous human breast cells and human breast cancer cell lines, the estrogenicity of Z in terms of the induction of estrogen-regulated genes is comparable to the natural estrogen, estradiol-17β (E2), and the synthetic estrogen, diethylstilbestrol (DES).

In our present work, we have demonstrated the transformation of human breast epithelial cell (HBEC) line MCF-10A by an environmental disruptor, Z, for the comparison with some characteristics induced by the natural E2. Long-term treatment of these cells with Z or E2 induces post-confluent foci formation (data not shown), anchorage-independent growth and colony formation in soft agar. The immortalized

HBEC MCF-10A has been claimed as an ER-negative cell line. However, we found that

28 there is low level of ERβ mRNA expression but no ERα mRNA expression in these

cells. In our previous work, we described that both Z and E2 downregulate PTPγ

expression is associated with ERα [Liu et al., 2002b]. All of these findings suggested

that both Z and E2 might transform MCF-10A cells through the redox-mediated pathway

not estrogen receptor-mediated pathway. Some reactive intermediates are generated

through this pathway, which elicit direct genotoxic effects by increasing mutation rates.

Genomic analysis revealed that short-term E2-treated cells exhibited loss of

heterozigosity (LOH) in chromosome 11 [Russo et al, 2002]. In the future, genomic

analysis to the cells treated with Z or E2 is necessary to investigate whether long term and

low doses of Z or E2 can damage the genomic DNA of the treated cells and cause HBEC transformation through the redox-mediated pathway.

Furthermore, our results indicated that expression of ERβ could be induced in both estrogen- and environmental disruptor-transformed MCF-10A cells, which suggested that expression of ERβ might contribute to the initiation and progression of Z- or E2-induced neoplastic transformation. The regulatory mechanisms of inducing ERβ

expression need to be investigated in the future. ERβ can mediate estrogen-induced

biological responses by forming heterodimers with ERα as well as homodimers in a

manner similar to ERα [Cowley et al., 1997]. Since we could not find any ERα

expression in MCF-10A cells, we could speculate that the transformation of MCF-10A

cells by Z and E2 might also be mediated through ERβ homodimer signaling pathway.

However, the role of ERβ-mediated estrogen signaling pathways in the pathogenesis of

malignant diseases is still not quite clear. To provide a quantitative and more functionally

29 relevant evaluation of changes in ERβ expression levels, we have purchased several

commercial ERβ antibodies and conducted both immunohistochemical staining assay and

western blotting assay for ERβ protein. Unfortunately, none of these antibodies have

been found suitable for our experiments.

In our study, we have noticed that there is no dose effect on those tested

biological markers, and we think that this can be due to the following two possibilities.

The dose range we used in our experiment is from 0.1 nM to 100 nM. There is no dose

effect of these biological markers in our tested dose range, but we cannot exclude the

possibility that there will be dose effect if we lower the treatment dose in our experiment.

For example, we could choose 10-4 nM to 0.1 nM as our test dose range in a future

experiment. The other possibility is that ten treatment cycles might have been too long,

so that the effect for each dose has reached the maximum and thus, we did not observe a

dose effect. If this is the case, we might be able to see a dose effect if we terminated the

treatment at the seventh, eighth or ninth treatment cycle.

As we have known, pS2 was cloned from a poly (A) RNA library from the

hormone-dependent, breast cancer cell line, MCF-7 [Masiakowski et al., 1982] and the

pS2 promoter region contains an estrogen response element that varies from the

consensus sequence by one base pair [Berry et al., 1989]. pS2 gene expression is

regulated by the estrogen in breast tissues. Here, we compared the estrogenic potency of

E2 and Z on the regulation of pS2 or PTPγ expression. Surprisingly, we found that the

potency of Z on these targeted gene expression is almost the same as that of E2. One possible explanation is the potential potency of Z is much larger than its actual concentrations suggest (up to 50 times) since most exogenous hormone-like chemicals,

30 including Z and the other synthetic growth promoting hormones, exhibit limited or no

binding to carrier proteins, such as binding globulin (SHBG) [Mastri et al.,

1985; Shrimanker et al., 1985; Nagel et al., 1998]. In our study, we showed that MCF-

10A doesn’t express pS2 and we couldn’t examine the estrogenic effects of E2 and Z on

this estrogenic targeted gene expression. In the future, we can use another estrogen

targeted gene PR as the biomarker in our experiments.

In conclusion, we have demonstrated that both Z and E2 can induce human breast epithelial cell line transformation and can induce ERβ expression in human breast epithelial cells by long term and low dose exposure, which might be mediated through the redox-pathway and/or ERβ-mediated pathway needed to be confirmed in the future

experiments. Most interestingly, Z and E2 showed the similar potency in these experiments.

31

Treatments DT (hours) Wild type MCF-10A 17.8 ± 1.2 Control (0.1% DMSO) 18.0 ± 1.0

0.1 nM E2 12.1 ± 0.8

1 nM E2 11.4 ± 0.9

10 nM E2 11.8 ± 0.6

100 nM E2 12.0 ± 0.7 0.1 nM Z 11.6 ± 0.8

1 nM Z 12.0 ± 0.5

10 nM Z 12.0 ± 0.6

100 nM Z 11.4 ± 0.4

Table 2.1. Comparison of MCF-10A cell proliferation by doubling time assay.

MCF-10A cells were treated with different doses of E2 or Z (0.1 nM, 1 nM, 10 nM or 100 nM) or vehicle (0.1% DMSO) as controls for 10 repeated treatment cycles and 0.5 × 104

cells were plated into 24-well plates in low-calcium DMEM/F12 with 10% HS. Cells

were grown for 2 days and counted every 4 hours. The results represented the mean ± SD

of 4 replicate wells. Each experiment was repeated twice. DT: Doubling Time. Statistical

differences were determined by using one-way ANOVA for independent groups. P-

values of less than 0.05 were considered statistically significant.

32

Figure 2.1. Effects of E2 or Z on anchorage-independent growth of MCF-10A cells in

soft agar.

Cells were cultured in 6-well plates first covered with an agar layer [phenol red-free low-

calcium DMEM/F12 with 0.5% agar and 10% Horse Serum (HS)]; the middle layer

contained 8000 cells (wild type MCF-10A or MCF-10A treated with DMSO, E2 or Z for ten repeated treatment cycles) in phenol red-free low-calcium DMEM/F12 with 0.35% agar and 10% HS; the top layer, consisting of cell culture medium, was added to prevent drying of the agarose gels. The plates were incubated for 30 days and another 1 ml of medium was added to the top layer at day 15. After 30 days’ incubation, the plates were stained in 0.5 ml of 0.005% crystal violet for >1 hour and the cultures were inspected and photographed. Bar: 500µm.

A: Wild type MCF-10A; B: 0.1% DMSO-treated MCF-10A; C: 0.1 nM E2-treated MCF-

10A; D: 1 nM E2-treated MCF-10A; E: 10 nM E2-treated MCF-10A; F: 100 nM E2-

treated MCF-10A; G: 0.1 nM Z-treated MCF-10A; H: 1 nM Z-treated MCF-10A; I: 10

nM Z-treated MCF-10A; J: 100 nM Z-treated MCF-10A.

Results showed that E2 or Z-treated MCF-10A cells formed colonies in soft agar, and

there is no significant difference in the size and colony efficiency in our tested different

doses of E2 or Z-treatment groups and the potency of E2 and Z in the stimulation of colony formation in MCF-10A cells are similar.

33

Figure 2.1

34 pS2 (220 bp) ERα (650 bp) ERβ (528 bp) PTPγ (492 bp) 36B4 (563 bp) 2 2 2 2 l Z Z Z Z o E E E E pe y MCF-7 1 nM Marker Contr 1 nM 10 nM 0.1 nM 10 nM 100 nM Wild t 0.1 nM 100 nM

MCF-10A

Figure 2.2. Effects of E2 and Z on ERβ expression in transformed MCF-10A cells.

MCF-10A cells were treated with 0.1, 1, 10, 100 nM of E2, Z or vehicle as controls for

ten repeated treatment cycles. At the end of the tenth treatment cycle, partial of the cells

were collected for RT-PCR assay. Ethidium bromide-stained PCR products were

separated in a 1.5% agarose gel.

Our results showed that both wild type MCF-10A and the transformed MCF-10A does

not express ERα and pS2. ERβ is expressed at a very low level in both wild type MCF-

10A and controls, and an elevated level of ERβ expression is observed in the transformed

MCF-10A cells at the end of the tenth different doses of E2- or Z-treatment and the degree of ERβ expression induction is similar in treatments with different doses of E2 and

Z in our tested range. Furthermore, the expression of protein tyrosine phosphatase γ

(PTPγ), an estrogenically regulated tumor suppressor gene, is not changed by the E2 or Z

transformation of MCF-10A cells. 35 A. M

pS2 (220 bp)

36B4 (563 bp)

B. 2.50

estradiol-17β 2.25 Zeranol

2.00 pression 4)

1.75 /36B 2 (pS mRNA ex

e 1.50 tiv a 1.25 Rel

1.00 0.0 7.5 15 30 60 120 Concentrations (nM)

Figure 2.3. Potency comparison of E2 and Z on pS2 mRNA expression in human

breast cancerous cell line MCF-7.

MCF-7 cells were treated with 7.5, 15, 30, 60, 120 nM of E2 or Z or vehicle as controls

for 24 hours. Then, cells were collected for RT-PCR assay. A. Ethidium bromide-stained

PCR products were separated in a 1.5% agarose gel. B. The mRNA ratios of pS2 to 36B4

as measured by densitometry.

Our results showed that E2 and Z can upregulate pS2 mRNA expression in a dose-

response manner, as low as 15 nM of E2 and Z can upregulate pS2 mRNA expression by

~ 30%.

36 A. M PTPγ (492 bp) 36B4 (563 bp) B. 0.4

estradiol-17β Zeranol

0.3 ssion e r p 4) x

/36B 0.2 γ NA e P R T m (P ve i 0.1 lat Re

0.0 0.0 7.5 15 30 60 120 Concentrations (nM)

Figure 2.4. Potency comparison of E2 and Z on PTPγ mRNA expression in human breast cancerous cell line MCF-7.

MCF-7 cells were treated with 7.5, 15, 30, 60, 120 nM of E2 or Z or vehicle as controls for 24 hours. Then, cells were collected for RT-PCR assay. A. Ethidium bromide-stained

PCR products were separated in a 1.5% agarose gel. B. The mRNA ratios of PTPγ to

36B4 as measured by densitometry.

Our results showed that E2 and Z can downregulate PTPγ mRNA expression in a dose- response manner, as low as 15 nM of E2 and Z can downregulate PTPγ mRNA expression by ~26%.

37 CHAPTER 3

REGULATION OF PROTEIN TYROSINE PHOSPHATASE γ (PTPγ) BY

ESTROGENICALLY ACTIVE AGENTS, ESTRADIOL-17β (E2) AND ZERANOL (Z),

IN HUMAN BREAST

ABSTRACT

Protein tyrosine phosphatase γ (PTPγ) has been implicated as a tumor suppressor gene in kidney and lung cancers. Our previous results indicate that estradiol-17β (E2)-

induced suppression of PTPγ may play a role in mammary tumorigenesis. Zeranol (Z), a

nonsteroidal growth promoter with estrogenic activities that is used by the U.S. meat

industries, induces estrogenic responses in primary cultured breast cells and breast cancer

cell lines. PTPγ mRNA expression in human breast tissues and cells isolated from

surgical specimens of mammoplasty and breast cancer patients were detected and

quantified by RT-PCR. Immunohistochemical staining was used to localize PTPγ in

human breast tissues. Breast epithelial and stromal cells were isolated and co-cultured to

determine the involvement of cell-cell interactions in the regulation of PTPγ mRNA

38 expression by E2 and Z. PTPγ mRNA expression was lower in cancerous than in normal

breast tissues. Both E2 and Z suppressed PTPγ mRNA levels in cultured normal breast tissues by ~80%, but had a lesser effect in the cultured epithelial cells isolated from normal breast tissues. In the co-culture system, both E2 and Z suppressed PTPγ mRNA to

a greater degree in epithelial cells than in stromal cells. In whole breast tissues, PTPγ was

immunolocalized to the epithelium. Treatment with E2 or Z diminished PTPγ staining

indicating reductions in PTPγ at the protein level. The results indicate that both E2 and Z

downregulate PTPγ expression in human breast and that epithelial-stromal cell

interactions are important in the regulation of PTPγ expression by estrogenically active

agents.

INTRODUCTION

Protein tyrosine phosphatases (PTPases) are a family of proteins of which the first

one, PTP1B, was discovered in 1988 by Fischer and co-workers [Tonks et al., 1988].

PTPases get their names from the enzymatic roles they perform inside cells, namely the

removal of phosphate groups from phosphotyrosine residues of specific target proteins.

The phosphorylation of these tyrosine residues is catalyzed by the protein tyrosine

(PTKs) and regulates important cellular processes like metabolism, gene

expression, cell division and differentiation, development, transport, and locomotion.

Since PTPases act on the same kind of biochemical switches as the PTKs, they are 39 thought to play an equally important biological role [Barford et al., 1995; Sun et al.,

1994]. In support with this contention, Klarlund, already in 1985, showed that addition of the PTPase-inhibitor vanadate to cells in culture leads to increased amounts of phosphotyrosine-containing proteins and cellular transformation [Klarlund et al., 1985].

Thus, a delicate balance between PTK and PTPase action is essential for normal functioning of cells. The V-Src, a transforming principle of the chicken Rous sarcoma virus, was determined to have activity as it phosphorylates both itself and other proteins on tyrosine residues [Hunter et al., 1980]. Several other viral oncogenes have been shown to be tyrosine kinases [Bishop, 1985]. In addition, the epidermal growth factor receptor was also shown to have tyrosine kinase activity [Ushiro et al., 1980].

Protein tyrosine phosphatases (PTPs) play an essential role in the regulation of cell activation, proliferation and differentiation, since they counterbalance the growth- promoting effects of protein tyrosine kinases (PTKs) [Shock et al., 1995]. Therefore, alterations in PTPs activity might affect cell growth, neoplastic processes and transformation [Gaits et al., 1995].

PTPγ is a member of the receptor-like tyrosine-specific phosphatase family originally cloned from human brain stem and placental cDNA libraries using probes derived from the intracellular domain of CD45 [Kaplan et al., 1990] or Drosophila

PTPase cDNA clone, DPTP12 [Kaplan et al., 1990], respectively. According to mRNA analysis [LaForgia et al., 1991; Barnea et al., 1993], PTPγ is a broadly expressed enzyme that exists in many tissues, including human lung, stomach, esophagus, colon, liver, spleen, and kidney [Tsukamoto et al., 1992]. Based on the chromosomal location of the

PTPγ gene (3p14.2) [LaForgia et al., 1991; LaForgia et al., 1993] and studies showing

40 loss of heterozygosity of the gene in kidney tumors [Lubinski et al., 1994], PTPγ has

been implicated as a candidate tumor suppressor gene.

Human breast tumors exhibit enhanced tyrosine kinase activity relative to benign

breast disease or normal breast tissues [Hennipman et al., 1989]. In breast cancer,

HER2/neu overexpression is an important prognostic indicator, and constitutes a

therapeutic target [Ross et al., 1999], but other receptor tyrosine kinases are also

overexpressed [Ghoussoub et al., 1998]. The growth rate of a large proportion of breast

cancers is influenced by sex steroid hormones, and both steroid hormones and protein tyrosine phosphorylation are demonstrated to play important roles in cell proliferation

[van Biesen et al., 1995].

Zeranol (Z) (described in Chapter 1) is a nonsteroidal agent with estrogenic activities and used as a growth promoter in the U.S. beef, veal and lamb industries.

The estrogen receptor (ER)-positive MCF-7 human breast cancer cell line shows estrogen-dependent growth in vitro, as well as estrogen-dependent tumorigenicity in vivo

[Katzenellenbogen et al., 1987]. Previous work suggests that PTPs may also be important in the growth of breast cancer and may be affected by estrogenic agonists and antagonists. PTP activity can be increased with O-phospho-L-tyrosine with a consequent

S-phase block of breast cancer cell line growth and an enhancement of the effects of chemotherapy on cell killing [Mishra et al., 1993]. Anti-estrogens such as 4-hydroxy- tamoxifen increase the activity of membrane associated PTPs in ER positive breast cancer cell lines, but not in those that are ER negative. This increased PTP activity correlates with decreased growth rates, and the tamoxifen effect on growth rates can be blocked with the PTP antagonist sodium vanadate [Freiss et al., 1994].

41 Consistent with these findings are our previous results that show lower PTPγ mRNA expression levels in diethylstilbestrol (DES)-induced kidney tumors in hamsters than in normal hamster kidney [Lin et al., 1994]. Also, we have shown that, in ACI rats,

PTPγ localizes to the mammary epithelium and Z can suppress PTPγ mRNA levels in mammary glands [Kulp et al., 2000]. Furthermore, we have reported that PTPγ is expressed in normal and malignant human breast epithelium, and that PTPγ mRNA levels can be suppressed by estrogens through an estrogen receptor-mediated mechanism

[Zheng et al., 2000]. These findings suggest that PTPγ is a potential estrogen-regulated tumor suppressor gene in human breast cancer which may play an important role in neoplastic processes of human breast epithelium.

Complex interactions between epithelium and mesenchyme play an essential role in epithelial cell proliferation and differentiation during normal breast development [Van

Roozendaal et al., 1996]. Stromal influences upon epithelia are part of a continuum of cellular interactions that begins at fertilization and extends into adulthood [Donjacour et al., 1991]. A considerable amount of evidence supports the role of stromal cells and their factors in the development and growth of breast cancer. Conditioned media of stromal cells derived from breast tumors has been shown to stimulate the proliferation of several human breast cell lines and primary cultured human breast cancer epithelial cells via the secretion of not-yet-defined factors [Van Roozendaal et al., 1992; Hofland et al., 1995].

In the current study, we demonstrate that the nonsteroidal agent, Z, induces estrogenic effects in human breast tissues, that both E2 and Z regulate PTPγ expression in human breast and that epithelial-stromal cell interactions are important in the regulation of PTPγ expression by estrogenically active agents. 42

MATERIALS AND METHODS

Human breast tissues. Normal human breast tissues and breast cancer tissues were obtained through the Tissue Procurement Program of The Ohio State University Hospital and Riverside Methodist Hospital in Columbus, Ohio. At the time of procurement, the tissue samples were placed in a mixture of Dulbecco's Modified Eagle's Medium and

Ham's F12 Medium (1:1) (DMEM/F12) without phenol red (Sigma Chemical Co., St.

Louis, MO) and stored at 4°C before transfer to the laboratory.

Tissue dissociation. Tissues were sterilized in 70% ethanol for 30 seconds, and then washed three times with fresh DMEM/F12. Tissue samples were minced, and then dissociated overnight at 37°C with 0.1% (GibcoBRL, Bethesda, MD) in phenol red-free DMEM/F12 medium (1 gram tissue/ml) supplemented with 5% fetal bovine serum (FBS) (Atlanta Biologicals, Norcross, GA) and antibiotic-antimycotic (100 unit/ml penicillin G sodium, 100 mg/ml streptomycin sulfate and 0.25 mg/ml amphotericin B) (GibcoBRL, Bethesda, MD).

Cell culture. The digested mixture was centrifuged at 200 × g (~ 3200 rpm) for 5 min at 25°C. The cell pellet was re-suspended and allowed to settle by gravity for about

10 minutes. The supernatant (containing mostly stromal cells) was then centrifuged at

200 × g for 5 min at 25°C and the pelleted stromal cells were re-suspended in phenol red- free high-calcium DMEM/F12 (1.05 mM CaCl2) supplemented with 5% FBS. The initial

43 sedimented cells (containing mostly epithelial cells) were washed three times with 20 ml

of phenol red-free DMEM/F12 medium and allowed again to settle by gravity for about

10 minutes. Then, the sedimented epithelial cells were re-suspended in phenol red-free

low calcium DMEM/F12 (0.04 mM CaCl2) supplemented with Chelex-100 (Bio-Rad

Laboratories, Richmond, CA) – treated FBS (10%) (Figure 3.1). Sometimes, adipose

stromal cells were also isolated from normal human breast tissues. After initial

centrifugation of the digested tissue, yellow adipose fraction was collected and suspended

in phenol red-free high-calcium DMEM/F12 (1.05 mM CaCl2) supplemented with 5%

FBS. All three isolated specific cell types were plated separately in 75-cm2 culture flasks in a humidified incubator (5% CO2: 95% air, 37°C). The media of all primary cultured

human breast cells and human breast cancer cells were changed every two days. When

the cells grew to ~ 85-90% confluence, cells were washed twice with 10 ml of calcium-

and magnesium-free Phosphate Buffered Saline (PBS, pH7.3), and then trypsinized with

1 ml of 0.5% trypsin - 5.3 mM EDTA (GibcoBRL) in PBS for 10 minutes at 37°C. The

trypsinization was stopped by addition of 10 ml of culture medium with 5% FBS. After

centrifugation, the dissociated cells were resuspended in the culture medium with 5%

FBS and subcultured into 75 cm2 culture flasks at a ratio of 1 flask to 5 flasks.

Immunocytochemistry was used to determine the purity of the isolated primary cultured human breast epithelial cells and stromal cells. Stromal cells were distinguished from epithelial cells based on not only their characteristic morphology but also their specific gene expression, which was confirmed by immunocytochemical staining for multi- cytokeratin and vimentin. Epithelial cells should be positive for the presence of cytokeratin and stromal cells should be positive for the presence of vimentin. Human

44 breast epithelial cells and stromal cells were cultured in multichamber slides (Nunc Inc.,

Naperville, IL). After -10°C methanol fixation for 5 minutes, cells were stained for multi-

cytokeratin (4/5/6/8/10/13/18) (Cat. No. NCL-C11, Vector Laboratories, Inc.,

Burlingame, CA) or vimentin (Cat. No. NCL-VIM, Vector Laboratories, Inc.,

Burlingame, CA) following the manufacturer’s instructions (VECTASTAIN Universal

Quick Kit, Cat. No. PK-8800, and DAB Kit, Cat. No. SK-4100, Vector

Laboratories, Inc., Burlingame, CA). The optimal primary antibody dilution was 1:10 and

1:100 for multi-cytokeratin and vimentin, respectively. Omission of primary antibody

served as a negative control.

Cell treatment and total RNA extraction. Treatments and total RNA extractions

were performed on cells not propagated beyond the third passage, and the viabilities of

each cell type were greater than 95% as determined by the trypan blue dye exclusion

method [Tennant et al., 1964]. Epithelial cells were plated in 25-cm2 culture flasks (2 ×

105 viable cells/flask) and cultured overnight. The media was changed to phenol red-free

low-calcium DMEM/F12 supplemented with Dextran-Coated Charcoal (DCC) (Dextran

T-70; Pharmacia; activated charcoal; Sigma) -stripped Chelex-100-treated FBS (5%).

After 24 hours, cells were treated with 30 nM E2, 30 nM Z or vehicle as controls in

phenol-red-free low-calcium DMEM/F12 supplemented with 5% DCC-treated FBS for

24 hours. The treatment dose was chosen based on the dose-response curve and our

previous experiments. Total RNA was isolated in 2.5 ml TRIZOL Reagent (GibcoBRL) according to manufacturer’s instructions.

Tissue culture. Normal human breast tissues were sterilized in 70% ethanol for 30 seconds, and then were washed three times with DMEM/F12. Tissue samples were cut

45 into ~ 1mm3/piece pieces which were cultured (5-6 pieces per group, about 60 mg total)

overnight on collagen hemostatic sponges (Integra LifeSciences, Plainsboro, NJ) in 6-

well plates (Corning Costar, Corning, NY). Then, the media was changed to phenol red-

free high-calcium DMEM/F12 supplemented with 5% DCC-treated FBS. After 24 hours,

the tissues were treated with 30 nM E2, 30 nM Z or vehicle in phenol red-free high-

calcium DMEM/F12 supplemented with 5% DCC-treated FBS for 24 hours. All tissues

within a treatment group were pooled and total RNA was then isolated. Tissues were

rapidly frozen in liquid nitrogen and then homogenized with a mortar and pestle in the

presence of TRIZOL Reagent (1 ml/group). Subsequent steps for RNA isolation

followed the manufacturer’s instructions. In our condition, we got about 5-8 µg RNA per

group.

Co-culture system. Co-cultures of epithelial and stromal cells were performed by

using flat-bottomed cell culture plates (Figure 3.2). Stromal cells (0.5 × 106 cells/well) were plated on the nucleopore polycarbonate membrane (0.4 mm pore size) of the cell culture inserts (upper chamber). Epithelial cells (1.0 × 106 cells/well) were seeded on the

bottom plates (lower chamber). The cells were cultured in their specific culture media

overnight, which were replaced with phenol red-free high-calcium DMEM/F12

supplemented with 5% DCC-treated FBS. After 24 hours, the cells were treated with 30

nM E2, 30 nM Z or vehicle in the same medium for 24 hours. Total RNA was then

isolated using 5 ml TRIZOL Reagent following manufacturer’s instructions.

Reverse transcription-polymerase chain reaction (RT-PCR). RT-PCR was

performed in a gradient mastercycler (Eppendorf ®). PCR conditions were optimized for

MgCl2 concentration, annealing temperature and cycle number for the amplification of 46 each of the PCR products (PTPγ and 36B4). Under optimal conditions, the amounts of

PCR products generated fell within the linear portion of the PCR amplification curve

between twenty-six and thirty-nine amplification cycles. Briefly, 1 µg of total RNA from

cultured cells or tissues was reverse transcribed with 200 U M-MLV Reverse

Transcriptase (GibcoBRL) at 42°C for 1 hour in the presence of 5 mM each of dATP,

dCTP, dGTP and dTTP, 4 µl 5× 1st strand buffer (GibcoBRL), 0.01M DDT, 1 U RNA

Guard RNase inhibitor (Pharmacia Biotech, Uppsala, Sweden), and 2.5 mM random

hexamers in a total volume of 20 µl. The reaction was terminated by heating to 95°C for

3 minutes. The newly synthesized cDNAs were used as templates for PCR after adjusting

reagent concentrations to 3.5 mM MgCl2, 2.5 µl 10X PCR Buffer (GibcoBRL), 1 U

Platinum® Taq DNA polymerase (GibcoBRL), and 0.24 µM primers. The reactant was incubated at 95 °C for 5 minutes. Then, thirty cycles of amplification were performed with each cycle consisting of denaturation at 95°C for 1 minute, annealing at 63°C for 1 minute, and extension at 72°C for 1 minute. The primer sequences for PTPγ were

5´_GCG CAG CGA CTT TAG CCA GAC GA _3´ (sense @ 51 to 73) and 5´_GCT

CCC GCT CCC CAT CCT CAC TC _3´ (antisense @ 542 to 519). The primer sequences for 36B4 were 5´_AAA CTG CTG CCT CAT ATC CG _3´ (sense @ 306 to

325) and 5´_TTG ATG ATA GAA TGG GGT ACT GAT G_3´ (antisense @ 868 to

848). The final PCR products (10 µl) mixed with 1 µl 10 × loading buffer were separated on a 1.5% agarose gel containing ethidium bromide. The lengths of the PCR products were 492 bp for PTPγ and 563 bp for 36B4. The specific bands were quantified by

ImageQuaNT software (Molecular Dynamics, Sunnyvale, CA). The results are presented as the ratio of PTP γ to 36B4. 47 Immunohistochemical staining. Normal human breast tissues were cultured and treated as described above. All tissues within a treatment group were fixed, dehydrated, embedded in paraffin, and sectioned at 5-micron thickness for immunohistochemical staining of PTP γ. As a reference antibody, we used an affinity-purified goat polyclonal antibody raised against a peptide corresponding to amino acids 1421-1438 mapping at the carboxy terminus of the PTPγ precursor of human origin (C-18; Cat. No. sc-1111, Santa

Cruz Biotechnology, Santa Cruz, CA). According to the manufacturer’s instructions, the

C-18 anti-PTPγ serum reacts with PTPγ of mouse, rat and human origin by Western blotting and immunohistochemistry. The goat ABC staining system (Cat. No. sc-2023,

Santa Cruz Biotechnology, Santa Cruz, CA) was used to stain PTPγ in paraffin- embedded tissue sections following the manufacturer’s instructions. An optimal primary antibody concentration of 1.60 µg/ml was determined by titration in this system, and omission of the primary antibody served as a negative control. Furthermore, the specificity of the PTPγ antibody was confirmed by elimination of specific binding after preincubation of the antibody with the PTPγ blocking peptide. Briefly, following the manufacture’s instructions, the PTPγ primary antibody was combined with five-fold (by weight) excess of PTPγ blocking peptide (Cat. No. sc-1111p, Santa Cruz Biotechnology,

Santa Cruz, CA) in a small volume of PBS and incubated for 2 hours at room temperature. The antibody/peptide mixture was then diluted to the predetermined optimal primary antibody concentration (1.60 µg/ml) and the immunohisto-chemistry procedure described above was performed.

48 RESULTS

Comparison of PTPγ mRNA expression levels in normal and cancerous human breast tissues. PTPγ expression has been previously documented in human lung, stomach, esophagus, colon, liver, spleen, and kidney tissues [Tsukamoto et al., 1992].

Also, previous work in our laboratory showed that PTPγ mRNA expression was lower in mixed cell populations from cancerous human breast tissues than in mixed cell populations from normal human breast tissues [Zheng et al., 2000]. To compare mRNA expression of PTPγ in normal and cancerous human breast tissues, PTPγ mRNA levels were determined by RT-PCR in breast tissues from 3 reduction mammoplasty patients

(20-30 years of age) and 3 breast cancer patients (50-70 years of age). Normal breast tissues had 50 – 60% higher levels of PTPγ mRNA than the breast cancer tissues (Figure

3.3). Also, PTPγ mRNA expression was greater in isolated epithelial cells than in stromal cells from the same patient (Figure 3.4) suggesting a cell-specific expression pattern.

The purity of the human breast stromal and epithelial cell preparations were indicated by morphology and confirmed by immunohistochemistry. Stromal cells were distinguished from epithelial cells based on not only their characteristic morphology but also their specific gene expression, which was confirmed by immunocytochemical staining (Figure 3.5). Epithelial cells tended to grow in characteristic rounded shapes, while stromal cells exhibited typical spindle-shaped morphology. Immunocytochemical staining revealed that more than 95% of the cultured epithelial cells were positive for the presence of cytokeratin (Figure 3.5B), while almost no expression of vimentin was 49 detected (Figure 3.5C). Similarly, the majority of stromal cells (>95%) were

immunopositive for the presence of vimentin (Figure 3.5F) and no expression of

cytokeratin was detected (Figure 3.5E), which confirmed the fibroblastic nature of the

stromal cells.

Regulation of PTPγ mRNA expression in normal human breast tissues and

epithelial cells by E2 and Z. We have shown that PTPγ mRNA expression is inhibited by

E2 in a dose-dependent manner in primary cultured human breast cells [Zheng et al.,

2000], and both normal and cancerous human breast cell exhibit estrogenic responses to

Z [Irshaid et al., 1999]. To investigate the effects of E2 and Z on PTPγ mRNA expression

in normal human breast tissues, RT-PCR was used to determine PTPγ mRNA levels in

cultured breast tissue after treatment with E2 or Z. Levels of PTPγ mRNA in the E2- and

Z-treated tissues were significantly suppressed in comparison to levels in the control

tissues (approximately an 80% reduction) (Figure 3.6). Since epithelial cells were shown

to express higher levels of PTPγ than stromal cells (Figure 3.4 and 3.8), the sensitivity of

PTPγ mRNA expression level to estrogenic action in epithelial cells was examined and

compared to that in cultured breast tissues. After treatment of normal breast epithelial

cells with 30 nM E2 or Z, PTPγ mRNA levels were reduced by 30% (Figure 3.7), as

opposed to the 80% reduction observed in E2 or Z-treated cultured breast tissues (Figure

3.6). Thus, E2 and Z suppressed PTPγ mRNA levels to a greater degree in cultured breast

tissues than in epithelial cells. Also, under the in vitro conditions described, Z was

capable of suppressing PTPγ mRNA levels to a degree identical to that of E2.

Immunohistochemical localization and reactivity of PTPγ in normal and

50 cancerous human breast tissues, and the regulation of PTPγ immunohistochemical reactivity by E2 and Z in normal human breast tissue. The expression of PTPγ is severely reduced (>50%) in lung tumors and ovarian tumors in comparison to normal tissues [van

Niekerk et al., 1999]. We have shown PTPγ is localized to the mammary epithelium of

ACI rats [Kulp et al., 2000]. The present results show that PTPγ was primarily localized to the glandular epithelium while staining was almost absent from the stromal compartment in both normal and cancerous breast tissues; however, the degree of staining was observably diminished in cancerous human breast tissue (Figure 3.8E) when compared to normal human breast tissue (Figure 3.8D). In Z-treated cultured human breast tissue (Figure 3.8H), PTPγ was also immunolocalized to the epithelium; however, the degree of staining was observably diminished in comparison to the control tissue

(Figure 3.8F). Comparable results were obtained for the E2-treated breast tissues (Figure

3.8G) in which PTPγ staining was diminished in most of the epithelium observed.

Densely immunopositive glandular components were occasionally observed, but these were sparsely distributed throughout the normal tissues. Thus, these results indicate that in human breast tissues, immunoreactive PTPγ is localized to the glandular epithelium and is not detected in the stromal compartment. The expression of PTPγ is reduced in cancerous human breast tissue compared to normal tissue. Furthermore, treatment with

E2 or Z apparently does not alter the epithelial localization of PTPγ, but the degree of immunoreactivity is diminished by these treatments. These results are consistent with those presented in Figure 3.6 above in which treatment of cultured normal human breast tissues with 30 nM of E2 or Z resulted in the suppression of PTPγ mRNA expression as

51 determined by RT-PCR.

Effects of E2 and Z on PTPγ mRNA expression on normal human breast epithelial

cells and stromal cells in co-culture systems. It is known that stromal cells and their

factors play a very important role in the development and growth of breast cancer [Van

Roozendaal et al., 1996; Donjacour et al., 1991; Van Roozendaal et al., 1992; Hofland et

al., 1995]. Based on the results above showing that E2 and Z suppressed PTPγ mRNA expression to a greater degree in breast tissues than in breast epithelial cells, we hypothesized that stromal cells were important for the greater suppression of PTPγ mRNA expression by E2 and Z in normal human breast tissues. To test this hypothesis,

co-culture systems containing epithelial and stromal cells were used to determine effects

of the presence of stromal cells on E2- or Z-induced suppression of PTPγ mRNA in

epithelial cells. The results showed that both E2 and Z (30 nM; 24 h) suppressed PTPγ

mRNA expression to a greater degree in epithelial cells (80-90% reduction) than in

stromal cells (20-30% reduction) (Figure 3.9). Also, the degree of suppression observed

in the co-cultured epithelial cells was greater than that seen in epithelial cells cultured

alone (80-90% vs 30% reductions) (Figures 3.9 and 3.7), and comparable to that observed

in cultured breast tissues (Figure 3.6). Furthermore, we found that the conditioned

medium from the co-culture model also increased the suppression of PTPγ expression by

E2 and Z in the epithelial cells cultured alone, and the degree of suppression is similar to that in the co-culture model. Combined together, these results suggested that stromal cells might be able to produce some factors to enhance the estrogenic effects of E2 and Z on

PTPγ expression in human breast.

52

DISCUSSIONS

Previous work has indicated an intriguing link between PTPγ and cancer. In renal and lung cancers, PTPγ has been implicated as a tumor suppressor gene [LaForgia et al.,

1991] and its expression is reduced in lung and ovarian tumors [van Niekerk et al., 1999].

Our own findings have shown that PTPγ mRNA levels are lower in breast tissues from breast cancer patients than from normal patients (Figure 3.3; [Zheng et al., 2000]). We also showed that E2 suppresses PTPγ mRNA expression via an ER-mediated mechanism

[Zheng et al., 2000]. The findings reported here extend our previous work by showing that the epithelial compartment is the primary site of PTPγ expression in human breast, and demonstrating the importance of stromal cells in the suppression of epithelial PTPγ expression by estrogenic agents. We also demonstrate that a nonsteroidal agent with estrogenic activities, Z, suppresses PTPγ mRNA expression to a degree comparable to that of E2.

Using cultured epithelial and stromal cells isolated from normal human breast tissues acquired from 3 different reduction mastectomy patients, we showed that

PTPγ mRNA levels are approximately 2-fold greater in epithelial cells than in stromal cells as determined by RT-PCR (Figure 3.4). This finding is supported by the results of immunohistochemistry that reveal PTPγ immunoreactivity to be localized exclusively to the epithelium in cultured normal human breast tissues (Figure 3.8). Similarly, our

53 previous results immunolocalized PTPγ to the glandular epithelium in mammary glands

of ACI rats, an animal model used for the study of estrogen-induced mammary

tumorigenesis [Kulp et al., 2000]. The absence of observable PTPγ immunoreactivity in

the stroma of cultured breast tissues (Figure 3.8), despite detection of PTPγ mRNA in

isolated stromal cells by RT-PCR (Figure 3.4), may indicate the presence of epithelial

contamination in the stromal cell isolated. This possibility seems unlikely since

immunohistochemistry failed to reveal the presence of cytokeratin-positive cells in the

isolated stromal cell population (Figure 3.5). Indeed, the isolated epithelial and stromal

cell populations were shown to be greater than 95% pure based on cytokeratin and

vimentin immunoreactivity (Figure 3.5). Alternatively, the results may suggest that

immunohistochemistry was not sufficiently sensitive to detect low expression levels of

PTPγ protein in the cultured breast tissues, or indicate the presence of a suppressive

epithelial influence in stromal PTPγ expression that is active in cultured tissues, but

absent in the culture of isolated stromal cells. These possibilities are speculative and

require further work to clarify.

Previous work from our laboratory demonstrated substantial suppression of PTPγ

mRNA expression by E2 treatment in mixed breast cell populations that contained both

epithelial and stromal components [Zheng et al., 2000]. Utilizing cultured normal human

breast tissues, the present study produced similar results. Treatment with 30 nM E2 resulted in an approximately 80% reduction in PTPγ mRNA expression (Figure 3.6).

Immunohistochemical staining of these cultured breast tissues supported this finding by revealing a noticeable reduction in the intensity of PTPγ immunoreactivity in the epithelial compartment (Figure 3.8G). However, identical treatment of cultured epithelial 54 cells isolated from normal human breast tissues suppressed PTPγ mRNA levels to a much

smaller degree (only a 30% reduction; Figure 3.7). These findings led to the hypothesis

that stromal cells are important for full suppression of PTPγ mRNA expression by E2.

The results of subsequent experiments using co-culture systems of isolated breast epithelial and stromal cells supported this hypothesis. Treatment of epithelial cells co- cultured with stromal cells from normal human breast tissues with 30 nM E2 resulted in a

level of suppression of PTPγ mRNA expression (80 – 90% reduction) that was nearly

identical to that observed in cultured breast tissues (Figures 3.6 and 3.9). Because the co-

culture system utilized for this study eliminated the influence of cell-to-cell contact

(Figure 3.2), the findings suggest the presence of a soluble factor(s) of stromal cell origin

that affects the response of epithelial cells to E2-mediated suppression of PTPγ mRNA

expression. Paracrine interactions between stromal and epithelial cells are known to play

an essential role in epithelial and stromal cell proliferation and differentiation during

normal breast development [Clark et al., 1992]. A disruption or perturbation of stromal-

epithelial interaction of normal cells might lead to abnormal cell growth and tumor

development [Donjacour et al., 1991]. Therefore, understanding the regulation and

mechanisms of stromal-epithelial interactions with respect to proliferation of normal

breast cells is critically important. It has been reported that normal breast epithelial cells

significantly inhibited, while normal breast stromal cells significantly stimulated, breast

cancer cell growth [Dong-LeBourhis et al., 1997; Hu et al., 1995]. Furthermore, it was

shown that primary cultured normal breast epithelial cells significantly inhibited the

proliferation of normal breast stromal cells and adipose stromal cells in co-culture, and

that TGF-β1 may play an important role in mediating normal human breast stromal- 55 epithelial interactions [Zhang et al., 1999]. Growth factors have been implicated as autocrine/paracrine mediators of epithelial-stromal interactions [Aaronson et al., 1991].

Human breast stromal cells secret peptide growth factors including insulin-like growth factor-I and -II [Cullen et al., 1991] and transforming growth factor-α [Cunha et al.,

1994] to regulate breast epithelial cell functions. Perhaps some of these known growth factors and/or other not-yet-defined growth factors from stromal cells play an important role in the regulation of PTPγ expression by estrogenically active agents in human breast.

Our previous work demonstrating estrogen-induced suppression of PTPγ in ER- positive breast cancer cell lines and primary cultured human breast cells was one of only two reports describing the pharmacological modulation of PTPγ expression [Zheng et al.,

2000; Schumann et al., 1998]. The demonstrated estrogenic regulation of PTPγ expression in human breast tissues is consistent with our previous work using the diethylstilbestrol (DES)-induced hamster kidney tumor model of estrogen-dependent tumorigenesis [Lin et al., 1994]. Kidney tumors in DES-treated hamsters contained lower levels of PTPγ mRNA than kidneys from control hamsters. These findings are intriguing in light of previous studies that showed changes in PTPs and PTP activity in human breast cancer. Hydroxy-tamoxifen increases the activity of a membrane-associated

PTP(s) in ER-positive breast cancer cell lines, but not in those that are ER-negative

[Freiss et al., 1994], a pattern that is complementary to that which we find for PTPγ, a transmembrane PTP. Nonetheless, a definitive role for PTPγ in breast cancer has not yet been established. Based on our results, however, it appears that estrogen might promote breast cell growth at least in part by manipulating the expression of PTPγ. This notion is supported by findings of our further work using MCF-7 cells transfected with an 56 expression vector containing a full-length PTPγ cDNA insert. As determined by RT-PCR,

the PTPγ-transfectants express PTPγ mRNA levels that are approximately 150% higher

than those in the mock-transfected and wild-type MCF-7 cells. Also, the 3H-thymidine

incorporation assay indicated that the proliferation rate of the PTPγ-transfectants is

approximately 70% lower than the mock-transfected and wild-type MCF-7 cells (data

shown in Chapter 5).

If the suppression of PTPγ expression plays a role in estrogen-induced breast cell

growth and/or tumorigenesis, the question arises as to whether other estrogenic agents,

such as environmental estrogens/, induce similar changes in PTPγ. One

such compound is Z, a nonsteroidal anabolic growth promoter with estrogenic activities.

Human exposure to Z occurs via the consumption of food products, particularly beef,

veal and lambs, derived from food animals treated with Z. Our results show that, in

cultured normal human breast tissues, isolated human breast epithelial cells and breast

epithelial cells in co-culture with stromal cells, Z at 30 nM significantly reduced PTPγ

mRNA levels (Figures 3.6, 3.7 and 3.9). This reduction is reflected in the

immunohistochemical results, which show markedly reduced PTPγ immunoreactivity in

the epithelium of Z-treated breast tissues (Figure 3.8H). The ability of Z treatment to

stimulate pS2 mRNA expression levels (data not shown) confirms its estrogenic activities

and indicates the absence of a generalized toxic effect of Z on the cultured breast tissues.

Of particular significance is the finding that the degree of Z-induced reduction of PTPγ

mRNA levels in these in vitro systems was equivalent to the reductions induced by E2.

Thus, exposure to equimolar amounts of E2 or Z results in identical effects in estrogen-

sensitive tissues, at least with respect to PTPγ expression in human breast under the 57 conditions described. These findings may have implications for human health,

particularly in regards to breast cell growth and perhaps breast cancer, by suggesting the

presence of a health risk from the consumption of beef or other food products derived

from Z-treated animals. The relevance of this potential relationship to human health is

strengthened by our preliminary findings, which revealed that Z levels, as measured by

High Performance Liquid Chromatography (HPLC), in edible tissues from Z-implanted

beef cattle were much lower than the permissible limits of free Z established by the FDA

[Code of Federal Regulations, 1991]. In particular, muscle tissue (meat) was shown to

contain an HPLC-detectable Z level that was approximately 29-times lower than the FDA

established limit (5.16 ± 0.46 ng/gm vs 150 ppb). Despite this low level of HPLC-

detectable Z, extracts of this meat containing 0.34, 1.70 and 8.50 ng Z/ml, stimulated 3H- thymidine incorporation by cultured human breast cells [Lin et al., 2000]. Effects of the

long-term consumption of low levels of Z in foods derived from Z-implanted animals are

unknown. Exploration of the potential risk to human health posed by this mode of

exposure is an active area of our research.

The experiments described here yield new and potentially important information

regarding the regulation of PTPγ, a potential tumor suppressor, in human breast by

estrogenic agents. The results show that the breast epithelium is the primary site of PTPγ

expression and that full estrogenic suppression of PTPγ mRNA expression involves

epithelial-stromal interactions. Furthermore, Z, a nonsteroidal agent with estrogenic

activities and a potential source of environmental estrogen exposure, reduces PTPγ

expression to a degree identical to that induced by the natural estrogen, E2. The

identification of PTPγ’s ligands, intracellular substrates and function, which are yet 58 unknown, and the complete elucidation of the mechanisms by which its expression and activity are regulated are essential for establishing the importance of PTPγ in the processes of estrogen-induced breast cell growth and/or breast tumorigenesis.

59

Tissue

Mince tissue Digest tissue in collagenase

Cultured in high Ca2+ Cultured (1.05 mM) DMEM/F12 human breast stromal cells Centrifuge

Cultured Cultured in low Ca2+ Wash (5 times, human breast (0.04 mM) DMEM/F12 epithelial cells allow settle by gravity)

Figure 3.1. Isolation of epithelial cells and stromal cells from normal human breast tissues and phase contrast photomicrograph of cultured human breast epithelial cells and stromal cells.

Epithelial cells and stromal cells from normal human breast specimens were isolated as described in Materials and Methods section. Normal human breast epithelial cells and stromal cells were cultured for 5 days. Epithelial cells grow in sheets and have round prominent nuclei; stromal cells have a typical spindle shape.

60

Medium level

Upper Chamber (0.5x106) Stromal cells

Polycarbonate membrane Epithelial cells 6 (pore size: 0.4µm) (1.0x10 )

Lower Chamber

Figure 3.2. Modified in vitro co-culture assay system.

The 70 mm TranswellTM system was used for co-culture assay. The diameters of the upper chamber and lower chambers were 7.5 cm and 8.5 cm, respectively. The bottom of upper chamber is a polycarbonate membrane with a 0.4 µm pore size. Stromal (0.5×106) cells were seeded in the upper chamber and epithelial cells (1.0×106) were seeded in the lower chamber. Since the size of human breast stromal cell is larger than that of human breast epithelial cell, we seeded less number of stromal cells in the upper chamber.

61

Figure 3.3. Comparison of PTPγ mRNA expression by RT-PCR in normal human breast tissues and cancerous human breast tissues.

Normal human breast tissues and breast cancer tissues were obtained through the Tissue

Procurement Program of The Ohio State University Hospital and Riverside Methodist

Hospital in Columbus, Ohio. A. Ethidium bromide-stained PCR products separated in a

1.5% agarose gel. Tissues were not treated and total RNA was isolated from each tissue sample separately. Lane 1-3: normal human breast tissues from 3 different patients. Lane

4-6: cancerous human breast tissues from 3 different patients. 36B4 was used as internal standard. B. The mRNA ratios of PTPγ to 36B4 as measured by densitometry.

The result showed that normal breast tissues had 50 – 60% higher levels of PTPγ mRNA

than the breast cancer tissues.

62

A 123 456 36B4 (563 bp)

PTPγ (492 bp)

B

0.8

0.6 Expression /36B4) γ 0.4 mRNA (PTP e

0.2 Relativ

0.0 123 4 56 Normal Tissue Cancerous Tissue

Figure 3.3

63 Figure 3.4. Comparison of PTPγ mRNA expression by RT-PCR in normal human breast tissues, epithelial cells and stromal cells.

As described in Materials and Methods, normal human breast epithelial cells and stromal cells were isolated from the normal human breast tissue, which were obtained through the

Tissue Procurement Program of The Ohio State University Hospital and Riverside

Methodist Hospital in Columbus, Ohio. Epithelial cells were cultured in phenol red-free low calcium DMEM/F12 (0.04 mM CaCl2) supplemented with Chelex-100 (Bio-Rad

Laboratories, Richmond, CA) – treated FBS (10%). Stromal cells were cultured in phenol red-free high-calcium DMEM/F12 (1.05 mM CaCl2) supplemented with 5% FBS. A.

Ethidium bromide-stained PCR products separated in a 1.5% agarose gel. 1, 2 and 3 represent patient 1, patient 2 and patient 3, respectively. Tissues and specific cell types were not treated and total RNA was isolated from each group of tissues or cells separately. 36B4 was used as internal standard. B. The mRNA ratios of PTPγ to 36B4 as measured by densitometry.

The result showed that PTPγ mRNA expression was greater in isolated epithelial cells than in stromal cells from the same patient.

64

A B Relative mRNA Expression (PTPγ/36B4) 0. 0. 0. 0. 0. 0. 0. 0. 0. 0 1 2 3 4 5 6 7 8 1

2 Tissues

Tis 31 s ue

Epi 23 Figure 3.4 Epit t h h

65 e lial C el ia l C e lls e

* lls 1 2 S S tr tr o o ma ma 3 l l C C el ell ls s 36B4 (563bp) PT Patient 3 Patient 2 Patient 1 P γ (492 bp)

Figure 3.5. Immunocytochemical staining of primary cultured human breast epithelial cells and stromal cells.

Human breast epithelial cells and stromal cells were cultured in multichamber slides

(Nunc Inc., Naperville, IL). After -10°C methanol fixation for 5 minutes, cells were stained for multi-cytokeratin (4/5/6/8/10/13/18) (Cat. No. NCL-C11, Vector Laboratories,

Inc., Burlingame, CA) or vimentin (Cat. No. NCL-VIM, Vector Laboratories, Inc.,

Burlingame, CA) following the manufacturer’s instructions (VECTASTAIN Universal

Quick Kit, Cat. No. PK-8800, and DAB Substrate Kit, Cat. No. SK-4100, Vector

Laboratories, Inc., Burlingame, CA). The optimal primary antibody dilution was 1:10 and

1:100 for multi cytokeratin and vimentin, respectively. Omission of primary antibody served as a negative control.

Brown staining represents cytokeratin or vimentin immunoreactivity; A. Negative control in epithelial cells; B. Cytokeratin staining in epithelial cells; C. Vimentin staining in epithelial cells; D. Negative control in stromal cells; E. Cytokeratin staining in stromal cells; F. Vimentin staining in stromal cells.

These staining results showed that more than 95% of the cultured epithelial cells were positive for the presence of cytokeratin, and the majority of stromal cells (>95%) were immunopositive for the presence of vimentin, which indicated the high purity in the isolated specific cells.

66

A D

5µm 5µm

B E

5µm 5µm

C F

5µm 5µm

Figure 3.5

67

Figure 3.6. Regulation of PTPγ mRNA expression in normal human breast tissues by E2 and Z.

Normal human breast tissues were sterilized in 70% ethanol for 30 seconds, and then were washed three times with DMEM/F12. Tissue samples were cut into ~ 1 mm3 /piece pieces which were cultured (5-6 pieces per group) overnight on collagen hemostatic sponges (Integra LifeSciences, Plainsboro, NJ) in 6-well plates (Corning Costar, Corning,

NY). Then, the media was changed to phenol red-free high-calcium DMEM/F12 supplemented with 5% DCC-treated FBS. After 24 hours, the tissues were treated with 30 nM E2, 30 nM Z or vehicle in phenol red-free high-calcium DMEM/F12 supplemented

with 5% DCC-treated FBS for 24 hours. All tissues within a treatment group were

pooled, and total RNA was then isolated and RT-PCR was performed.

A. Ethidium bromide-stained PCR products separated in a 1.5% agarose gel. 1 and 2

represent patient 1 and patient 2, respectively. Normal human breast tissues were treated

with 30 nM E2, 30 nM Z or vehicle in phenol red-free high-calcium DMEM/F12

supplemented with 5% DCC-treated FBS for 24 hours. Total RNA was isolated from

each treatment group separately. 36B4 was used as internal standard. B. The mRNA

ratios of PTPγ to 36B4 as measured by densitometry.

The result showed that 30 nM of E2 or Z can downregulate PTPγ mRNA expression by

~80% in normal human breast tissues.

68

A ß 17 l- ol SO radio an M st er D E Z

0.03% 30 nM 30 nM 112212 36B4 (563 bp)

PTPγ (492 bp) B

0.8 Patient 1

on Patient 2 essi 0.6 4) B Expr /36 γ 0.4 mRNA (PTP e

tiv 0.2 a Rel

0.0

ol SO 17ß M l- an D radio 0.03% st 30 nM Zer M E

30 n

Figure 3.6

69

Figure 3.7. Regulation of PTPγ mRNA expression in normal human breast epithelial

cells by E2 and Z.

Epithelial cells were plated in 25-cm2 culture flasks (2 × 105 viable cells/flask) and cultured overnight. The media was changed to phenol red-free low-calcium DMEM/F12 supplemented with Dextran-Coated Charcoal (DCC) (Dextran T-70; Pharmacia; activated charcoal; Sigma) -stripped Chelex-100-treated FBS (5%). After 24 hours, cells were treated with 30 nM E2, 30 nM Z or vehicle as controls in phenol-red-free low-calcium

DMEM/F12 supplemented with 5% DCC-treated FBS for 24 hours. Total RNA was isolated and RT-PCR was performed.

A. Ethidium bromide-stained PCR products separated in a 1.5% agarose gel. 1 and 2 represent patient 1 and patient 2, respectively. Normal human breast epithelial cells were treated with 30 nM E2, 30 nM Z or vehicle in phenol red-free low-calcium DMEM/F12

supplemented with 5% DCC-treated FBS for 24 hours. Total RNA was isolated from

each treatment group separately. 36B4 was used as internal standard. B. The mRNA

ratios of PTPγ to 36B4 as measured by densitometry.

The result showed that 30 nM of E2 or Z can downregulate PTPγ mRNA expression by

~30% in normal human breast epithelial cells.

70

B Relative mRNA Expression A (PTPγ/36B4) 0. 0. 0. 0. 0 2 4 6 11

0.03% 0.03% 22 DM D SO M SO

30 n

M Figure 3.7 Est 30 nM

71 r adio E st l- rad 17ß io l 1 -17ß

30 nM 30 n

M 2 Z Zer er ano a nol l PT 36 P P B4 (563 P a a γ ti ti (492 bp) e e n n t 2 t 1 bp) Figure 3.8. Immunohistochemical localization and reactivity of PTPγ in normal and cancerous human breast tissues, and the regulation of PTPγ immunohistochemical reactivity by E2 and Z in normal human breast tissue.

Normal human breast tissues were cultured and treated as described in figure 3.6. All

tissues within a treatment group were fixed, dehydrated, embedded in paraffin, and

sectioned for immunohistochemical staining of PTPγ. As a reference antibody, we used

an affinity-purified goat polyclonal antibody raised against a peptide corresponding to

amino acids 1421-1438 mapping at the carboxy terminus of the PTPγ precursor of human

origin (C-18; Cat. No. sc-1111, Santa Cruz Biotechnology, Santa Cruz, CA). The goat

ABC staining system (Cat. No. sc-2023, Santa Cruz Biotechnology, Santa Cruz, CA) was

used to stain PTPγ in paraffin-embedded tissue sections.

Brown staining represents PTPγ immunoreactivity; Blue staining represents nuclei. EC:

Epithelial compartment; SC: Stromal compartment. A. Negative control; B. Peptide

neutralization in normal human breast tissue without treatment; C. Peptide neutralization

in cancerous human breast tissue without treatment; D. Normal human breast tissue

without treatment; E. Cancerous human breast tissue without treatment; F. Normal

human breast tissue treated with 0.03% DMSO for 24 hours; G. Normal human breast

tissue treated with 30 nM E2 for 24 hours; H. Normal human breast tissue treated with Z

for 24 hours.

These results indicated that in human breast tissues, immunoreactive PTPγ is localized to

the glandular epithelium. The PTPγ level is reduced in cancerous human breast tissue

compared to normal tissue. Furthermore, the degree of immunoreactivity is diminished

by the treatment with E2 or Z. 72 A E

SC EC

EC

10µm 10µm

B F

SC EC EC SC 10µm 10µm

C G

EC

EC SC

10µm 10µm

D H

SC

EC EC SC

10µm 10µm

Figure 3.8

73

Figure 3.9. Regulation of PTPγ mRNA expression E2 and Z in normal human breast

epithelial cells and stromal cells in co-culture assay system.

Stromal (0.5×106) cells were seeded in the upper chamber and epithelial cells (1.0×106)

were seeded in the lower chamber in the co-culture system. The cells were cultured in

their culture media specific for epithelial cells or stromal cells overnight, which were

replaced with phenol red-free DMEM/F12 supplemented with 5% DCC-treated serum.

After 24 hours, the cells were treated with 30 nM E2, 30 nM Z or vehicle in the same medium for 24 hours. Total RNA was then isolated and RT-PCR was performed.

A. Ethidium bromide-stained PCR products separated in a 1.5% agarose gel. 1, 2 and 3 represent patient 1, patient 2 and patient 3, respectively. Cells were treated with 30 nM

E2, 30 nM Z or vehicle in phenol red-free low-calcium DMEM/F12 supplemented with

5% DCC-treated FBS for 24 hours. Total RNA was isolated from epithelial cells and

stromal cells in each treatment group separately. 36B4 was used as internal standard. B.

The mRNA ratios of PTPγ to 36B4 as measured by densitometry.

The results showed that the degree of suppression by both E2 and Z observed in the co-

cultured epithelial cells was greater than that seen in epithelial cells cultured alone (80-

90% vs 30% reductions) .

74

A lls s e ll e l C C lia l e th ma i ro Ep St 112233 36B4 (563 bp)

PTPγ (492 bp) B 0.6

0.3% DMSO n 30 nM Estradiol-17ß

io 0.5 30 nM Zeranol * = P<0.05 ess r 0.4 Exp

/36B4) 0.3 γ mRNA

(PTP 0.2 e * * v i t

la *

e 0.1 * R

0.0 s ll lls e e C l C al a li m the o Str Epi

Figure 3.9

75 CHAPTER 4

ESTROGENIC DOWNREGULATION OF PROTEIN TYROSINE PHOSPHATASE γ

(PTPγ) IN HUMAN BREAST IS ASSOCIATED WITH ESTROGEN RECEPTOR α

ABSTRACT

We have reported PTPγ expression was downregulated by 17β-estradiol (E2) and

Zeranol (Z), and PTPγ may function as an estrogen-regulated cancer suppressor in human breast. We utilized RT-PCR to examine expression of estrogen receptor α (ERα) and

β (ERβ) mRNA in MCF-7 and MDA-MB-231 cells and to investigate the regulation of

PTPγ expression by E2 and Z in the absence or presence of ICI 182,780 (ICI) in both cells, and immunohistochemistry to examine ERα and ERβ protein in normal and cancerous human breast. Results show that MCF-7 express both ERα and ERβ, and

MDA-MB-231 express only ERβ. Both E2 and Z (30nM; 24 h) suppressed PTPγ by

~56% in MCF-7 cells, and these effects were completely blocked by 1 µM of ICI. In contrast, E2, Z and ICI had no effects on PTPγ expression in MDA-MB-231 cells.

Interestingly, both E2 and Z suppressed PTPγ by ~45% in MDA-MB-231 cells 76 transfected with ERα, and these effects were completely blocked by 100 nM of ICI. Both

RT-PCR and immunohistochemical staining showed that ERα expression was significantly higher in cancerous human breast than in normal breast, while ERβ was higher in normal human breast than in cancerous breast. In combination with our previous findings of greater PTPγ expression levels in normal human breast than cancerous breast, current results show that there are lower ERα expression levels in normal human breast than cancerous breast, which indicate that lower PTPγ was associated with higher ERα in cancerous human breast tissues. In conclusion, our results indicate that Z induces estrogenic effects in human breast relative to PTPγ expression and the estrogenic downregulation of PTPγ expression in human breast is associated with

ERα.

INTRODUCTION

Protein tyrosine phosphatases (PTPs) play an essential role in the regulation of cell activation, proliferation and differentiation, since they counterbalance the growth- promoting effects of protein tyrosine kinases (PTKs) [Shock et al., 1995]. Therefore, alterations in PTPs activity might affect cell growth, neoplastic processes and transformation [Gaits et al., 1995]. Based on the chromosomal location of the PTPγ gene

(3p14.2) [LaForgia et al., 1991; LaForgia et al., 1993] and studies showing loss of

77 heterozygosity of the gene in kidney tumors [Lubinski et al., 1994], PTPγ has been

implicated as a candidate tumor suppressor gene.

The growth rate of a large proportion of breast cancer cells is influenced by sex

steroid hormones, and both steroid hormones and protein tyrosine phosphorylation are

demonstrated to play important roles in cell proliferation [van Biesen et al., 1995].

Zeranol (Z) (Ralgro) (described in Chapter 1) is a nonsteroidal agent with estrogenic activities that is used as a growth promoter in the U.S. beef, veal and lamb industries. Our previous results demonstrated lower PTPγ mRNA expression levels in the diethylstilbestrol (DES)-induced kidney tumors in hamsters than in normal hamster kidneys [Lin et al., 1994]. Also, we have shown that, in ACI rats, PTPγ localizes to the

mammary epithelium and Z can suppress PTPγ mRNA levels in mammary glands [Kulp

et al., 2002]. Recently, we have shown that both E2 and Z can downregulate PTPγ

expression in both human breast tissues and human breast cells [Liu et al., 2002a]. These

findings implicate that PTPγ may be considered as a potential estrogen-regulated tumor

suppressor gene in human breast cancer which may play an important role in neoplastic

processes of human breast epithelium.

The estrogen receptors (ERs) are member of a superfamily of nuclear receptors

which are ligand-regulated and sequence-specific transcription factors. Until now, two

ER subtypes have been identified: ERα and ERβ. Both of ERα and ERβ have five major

well-conserved structural domains (A/B, C, D, E and F). The C-domain (DNA binding

domain, DBD) is the most conserved region between ERα and ERβ, which suggests that

they can both bind to estrogen response elements (EREs); and conservation in regions

within the DBD required for dimmer formation suggests that the two receptors may 78 heterodimerize. Therefore, ERα and ERβ may influence or affect each other’s transcriptional activity if expressed in the same cells. The E-domain (Ligand binding domain, LBD) is also highly conserved between ERα and ERβ, indicating that they might bind similar ligands. In contrast, the A/B- and D-domains are not as conserved. It has been reported that both ERα and ERβ have high affinity to E2, and Z can interact with both subtypes with similar affinities [Kuiper et al., 1997].

In the present study, we investigated the distributions of ERα, ERβ and PTPγ expression in human breast; and we examined whether estrogenically active agent, such as Z, downregulates PTPγ expression via ERα or ERβ.

MATERIALS AND METHODS

Human breast tissues. Normal human breast tissues and cancerous human breast tissues were obtained through the Tissue Procurement Program of The Ohio State

University Hospital and Riverside Methodist Hospital in Columbus, Ohio. At the time of procurement, the tissue samples were placed in a mixture of Dulbecco's Modified Eagle's

Medium and Ham's F12 Medium (1:1) (DMEM/F12) without phenol red (Sigma

Chemical Co., St. Louis, MO) and stored at 4°C before transfer to the laboratory for process.

Tissue dissociation. Tissues were sterilized in 70% ethanol for 30 seconds, and 79 then washed three times with fresh DMEM/F12. Then, tissues were dissociated as

described previously in Chapter 3.

Cell culture. Specific cell types (epithelial cells and stromal cells) were separated

from the digested mixture and cultured as described previously in Chapter 3. To

determine the purity of the isolated human breast epithelial cells and stromal cells,

immunocytochemical staining was used for multi-cytokeratin (4/5/6/8/10/13/18) or

vimentin [Liu et al., 2002a]. Both MCF-7 cells and MDA-MB-231 cells were purchased

from American Type Culture Collection (ATCC, Manassas, VA). MDA-MB-231 cells

stably transfected with ERα (MDA-MB-231-ERα-1000) was a gift from Dr. Robert

Brueggemeier at College of Pharmacy, The Ohio State University. All of those cell lines

were cultured in phenol red-free high-calcium DMEM/F12 (1.05 mM CaCl2)

supplemented with 5% FBS and were plated separately in 75-cm2 culture flasks in a

humidified incubator (5% CO2: 95% air, 37°C). The media of all human breast cell lines and primary cultured human breast cells were changed every two days. When the cells grew to 85-90% confluence, cells were washed twice with calcium- and magnesium-free

Phosphate Buffered Saline (PBS, pH7.3), and then trypsinized with 0.5% trypsin - 5.3 mM EDTA (GibcoBRL) in PBS for 10 minutes at 37°C. The trypsinization was stopped by addition of culture medium with 5% FBS. After centrifugation, the dissociated cells were resuspended in the same medium and subcultured into 75-cm2 culture flasks at a

ratio of 1 flask to 5 flasks. The concentration of geneticin used for maintenance of MDA-

MB-231-ERα-1000 cells was only one half (500 µg/ml) of the concentration (1000

µg/ml) which was used for the selection every 2-3 passages.

Cell treatment and total RNA extraction. Each human breast cancer cell line was

80 plated in 25-cm2 culture flasks (2 × 105 viable cells/flask) and cultured overnight. The

media was changed to phenol red-free high-calcium DMEM/F12 supplemented with

Dextran-Coated Charcoal (DCC) (Dextran T-70; Pharmacia; activated charcoal; Sigma) -

stripped FBS (5%). After 24 hours, cells were treated with 30 nM of E2 or Z, different

doses of ICI 182,780 (10 nM, 100 nM, or 1 µM) in the presence or absence of 30 nM of

E2, Z or vehicle as controls in phenol-red-free high-calcium DMEM/F12 supplemented with 5% DCC-treated FBS for 24 hours. Total RNA was isolated in 2.5 ml TRIZOL

Reagent (GibcoBRL) according to manufacturer’s instructions.

Reverse transcription-polymerase chain reaction (RT-PCR). RT-PCR was performed in a gradient mastercycler (Eppendorf®, Eppendorf Scientific, Inc., Westbury,

NY). PCR conditions were optimized for MgCl2 concentration, annealing temperature

and cycle number for the amplification of each of the PCR products (ERα, ERβ, PTPγ

and 36B4). Under optimal conditions, the amounts of PCR products generated from the

experimental samples were fallen within the linear portion of the PCR amplification

curve between twenty and thirty-nine amplification cycles. Briefly, 1 µg of total RNA

from cultured normal and cancerous human breast cells or tissues was reverse transcribed

with 200 U M-MLV Reverse Transcriptase (GibcoBRL) at 42°C for 1 hour in the

presence of 5 mM each of dATP, dCTP, dGTP and dTTP, 4µl 5× 1st strand buffer

(GibcoBRL), 0.01M DDT, 1 U RNA Guard RNase inhibitor (Pharmacia Biotech,

Uppsala, Sweden), and 2.5 µM random hexamers in a total volume of 20 µl. The reaction

was terminated by heating to 95°C for 3 minutes. For ERα, the newly synthesized

cDNAs were used as templates for PCR after adjusting reagent concentrations to 1.0 mM

® MgCl2, 2.5µl 10X PCR Buffer (GibcoBRL), 1 U Platinum Taq DNA polymerase 81 (GibcoBRL), and 0.24 µM primers. The reactant was incubated at 95 °C for 5 minutes.

Then, thirty-five cycles of amplification were performed with each cycle consisting of denaturation at 95°C for 1 minute, annealing at 60°C for 45 seconds, and extension at

72°C for 1 minute. For ERβ, the newly synthesized cDNAs were used as templates for

PCR after adjusting reagent concentrations to 2.5 mM MgCl2, 2.5µl 10X PCR Buffer

(GibcoBRL), 1 U Platinum® Taq DNA polymerase (GibcoBRL), and 0.24 µM primers.

The reactant was incubated at 95 °C for 5 minutes. Then, thirty-five cycles were performed with each cycle consisted of denaturation at 95°C for 1 min, annealing at 56°C for 45 seconds, and extension at 72°C for 1 minute. For PTPγ and the housekeeping gene,

36B4, the conditions were described before [Liu et al., 2002a]. The primer sequences for

ERα were 5′-TAC TGC ATC AGA TCC AAG GG-3′ (sense) and 5′-ATC AAT GGT

GCA CTG GTT GG-3′ (antisense), for ERβ, they were 5′-TGA AAA GGA AGG TTA

GTG GGA ACC-3′ (sense) and 5′-TGG TCA GGG ACA TCA TCA TGG-3′ (antisense), and the primer sequences for PTPγ and 36B4 were listed previously [Liu et al., 2002a].

The final RT-PCR products (10 µl) were run on a 1.5 % agarose gel containing ethidium bromide. The specific bands were quantified by ImageQuant software (Molecular

Dynamics, Sunnyvale, CA). The results are presented as the ratio of targeted genes

(ERα, ERβ, or PTPγ) to 36B4 or cyclin D1 to 36B4.

Western blotting assay. MDA-MB-231 cells, MDA-MB-231-ERα-1000 cells and

MCF-7 cells were plated in separated 75-cm2 flasks (1.5×106 cells/flask) and cultured overnight before cell lysis. Cells were rinsed with cold 1× PBS and 0.5 ml of M-PERTM

Mammalian Protein Extraction Reagent (Cat. No: 78501, PIERCE, Rockford, IL) was added to each flask and flasks were kept gently shaking at 4 °C for 5 minutes. Then, the 82 cell lysate was collected into a microcentrifuge tube and samples were centrifuged at

27,000 x g, 4 °C for 10 minutes to pellet the cell debris. Finally, the supernatant was transfer to a clean tube for further analysis. The concentration of total protein was determined by using Micro BCATM Protein Assay (Cat. No: 23232, PIERCE, Rockford,

IL) according to manufacturer’s instructions. 50 µg of protein from each sample was denatured in Laemmli sample buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 25% glycerol,

0.01% Bromophenol Blue) (Cat. No: 161-0737, Bio-Rad Laboratories, Hercules, CA) at

99 °C for 5 minutes and loaded onto ready gel 4-15% Tris-HCl Gel for SDS-PAGE (Cat.

No: 161-1158, Bio-Rad Laboratories, Hercules, CA) and run at 100 volts for 1.5 hours in

1× Tris/Glycine/SDS buffer (Cat. No: 161-0732, Bio-Rad Laboratories, Hercules, CA).

The proteins were then transferred to a PVDF membrane in Semi-Dry Blotting Unit (Cat.

No: FB-SDB-2020, Fisher Scientific, Pittsburgh, PA) for 2.5 hours according to the manufacturer’s instructions. The membrane was detected by using ECL+Plus Western blotting detection reagents (Cat. No: RPN 2132, Amersham pharmacia biotech,

Piscataway, NJ) according to the manufacturer’s instructions. Briefly, the membrane was rinsed with PBS-T (PBS + 0.1% Tween-20) and blocked in a PBS-T solution containing

10% dried milk. After 1 hour, the membrane were rinsed with PBS-T and probed with an affinity-purified rabbit polyclonal antibody raised against a peptide mapping at the carboxy terminus of the ERα of human origin (HC-20; Cat. No. sc-543, Santa Cruz

Biotechnology, Santa Cruz, CA) at 1:500 dilution for 1 hour. Then, the membrane was rinsed with PBS-T and probed with a secondary anti-rabbit antibody linked to horseradish (NA 934, Amersham pharmacia biotech, Piscataway, NJ) at a 1:5000 dilution for 1 hour. The membrane was rinsed with PBS-T, detected with ECL+Plus Western

83 blotting detection reagents and exposed to Hyperfilm™ MP (Amersham pharmacia biotech, Piscataway, NJ). Finally, the film was developed and scanned. To detect β-actin protein expression, the membrane was incubated in stripping buffer (100 mM 2- mercaptoethanol, 2% SDS, 62.5 mM Tris-HCl, pH 6.7) at 50 °C for 30 minutes with occasional agitation and rinsed in PBS-T twice for 10 minutes each. Then, the membrane was detected for β-actin protein expression by using ECL+Plus Western blotting detection reagents as described above. Unfortunately, since there is no commercial ERβ antibody available for western blotting, we couldn’t conduct the western blotting experiment to compare ERβ expression in these cell lines.

Immunohistochemical staining. Both normal human breast tissues and cancerous human breast tissues were fixed, dehydrated, embedded in paraffin according to protocols used for routine histopathological sample preparations and sectioned for immunohistochemical staining of ERα, ERβ, and PTPγ. As reference antibodies, we used an affinity-purified goat polyclonal antibody raised against a peptide corresponding to amino acids 1421-1438 mapping at the carboxy terminus of the PTPγ precursor of human origin (C-18; Cat. No. sc-1111, Santa Cruz Biotechnology, Santa Cruz, CA), an affinity- purified rabbit polyclonal antibody raised against a peptide mapping at the carboxy terminus of the ERα of human origin (HC-20; Cat. No. sc-543, Santa Cruz

Biotechnology, Santa Cruz, CA), and an affinity-purified rabbit polyclonal antibody raised against a peptide mapping at the amino acid terminus of the ERβ of human origin

(H-150; Cat. No. sc-8974, Santa Cruz Biotechnology, Santa Cruz, CA). The goat or rabbit ABC staining system (Cat. No. sc-2023 or sc-2018, Santa Cruz Biotechnology,

Santa Cruz, CA) was used to stain PTPγ, ERα, or ERβ in paraffin-embedded tissue 84 sections following the manufacturer’s instructions. An optimal primary antibody

concentration of 1.60 µg/ml was determined by titration in this system, and omission of primary antibody served as a negative control. Furthermore, the specificities of antibodies were confirmed by elimination of specific binding after preincubation of the antibodies with blocking peptides (sc-1111 P; sc-543 P; Santa Cruz Biotechnology, Santa Cruz,

CA).

RESULTS

Comparison of ERα and ERβ expression in MCF-7 cells and MDA-MB-231 cells.

As we know, MCF-7 cells have been claimed as an estrogen receptor (ER)-positive

human breast cancer cell line and MDA-MB-231 cells have been claimed as an ER-

negative human breast cancer cell line. About 10 years after the cloning of the ER (now it

is referred to as ERα), a novel member of ERs, termed ERβ, has been identified in cDNA

libraries from rat prostate [Kuiper et al., 1997; Katzenellenbogen et al., 1997]. To

compare mRNA expressions of ERα and ERβ in MCF-7 cells and MDA-MB-231 cells,

mRNA levels were determined by RT-PCR. The results showed that MCF-7 cells express

both ERα and ERβ; and MDA-MB-231 cells express only ERβ (Figure 4.1). Also, we

have determined ERα expression at protein level in MDA-MB-231, MDA-MB-231-

ERα-1000 and MCF-7 cells by western blotting assay, the results further support that

ERα protein is expressed in MDA-MB-231-ERα-1000 and MCF-7 cells not in MDA- 85 MB-231 cells (Figure 4.2). Unfortunately, there is no commercial ERβ antibody available for western blotting assay.

Regulation of PTPγ expression by E2, Z and ICI 182,780 (ICI) in MCF-7 cells,

MDA-MB-231 cells and MDA-MB-231-ERα-1000 cells. Estrogen diffuses through the plasma and nuclear membranes of the cells and binds to the ER in the nucleus. Once estrogen binds to the ER, heat shock proteins dissociate and a change in conformation and dimerization occurs [Macgregor et al., 1998]. ICI 182,780 is a pure antiestrogen without intrinsic estrogenic activities such as tamoxifen. Numerous reports demonstrate that pure antiestrogen-ER complexes can bind to estrogen response elements (EREs) but the transcriptional unit is inactive [Pink et al., 1996; Pham et al., 1991]. To investigate the effects of E2 and Z on PTPγ mRNA expression in these human breast cancer cell lines,

RT-PCR was used to determine PTPγ mRNA levels in cultured human breast cancer cell lines after the treatments with E2, or Z in the absence or presence of ICI. The expression levels of PTPγ mRNA in the E2- and Z-treated (30 nM, 24 h) MCF-7 cells were significantly suppressed compared to the PTPγ mRNA levels in the controls

(approximately a 60% reduction) and these effects were completely blocked by 1 µM of

ICI, however ICI alone had no effects on PTPγ mRNA expression (Figure 4.3). In contrast, E2, Z and ICI had no effects on PTPγ expression in MDA-MB-231 cells (Figure

4.4). Furthermore, PTPγ mRNA expression in the E2- and Z-treated (30 nM, 24 h) ERα- transfected MDA-MB-231 cells (MDA-MB-231-ERα-1000) were significantly suppressed in comparison to the PTPγ mRNA levels observed in the controls

(approximately an 50% reduction) and these effects were completely blocked by 100 nM

86 of ICI, however ICI alone had no blocking effects on PTPγ mRNA expression (Figure

4.5). These results suggested that downregulation of PTPγ expression by E2 and Z was intimately related with ERα status in human breast.

Comparison of ERα, ERβ and PTPγ expression levels in normal and cancerous human breast. Previous works showed that both ERα and ERβ expressed in many estrogen-targeted tissues [Kuiper et al., 1997]. To determine the relative distribution of

ERα and ERβ mRNA in normal and cancerous human breast tissues and cells, total RNA was isolated from both normal and cancerous human breast tissues and cells and RT-PCR was performed with specific primers for each ER subtype. Our results showed that

ERα mRNA expression was significantly higher in cancerous human breast tissues than in normal human breast tissues (Figure 4.6), while ERβ and PTPγ [Liu et al., 2002a] mRNA expression was higher in normal human breast tissues than in cancerous human breast tissues (Figure 4.7). In immunohistochemical staining study, the results showed that ERα protein level significantly higher in cancerous human breast tissues than in normal human breast tissues, while ERβ and PTPγ protein levels were higher in normal human breast tissues than in cancerous human breast tissues (Figure 4.8). These results indicated that lower PTPγ expression was associated with higher ERα expression in human breast cancerous tissues.

87 DISCUSSIONS

Previous reports indicated that ERα and ERβ share about 95% homology in the

DNA binding domain and 55% homology in the ligand binding domain, both bind to a consensus ERE [Tremblay et al., 1997] and exhibit similar ligand binding properties

[Kuiper et al., 1997]. Although the relative expression of ERα and ERβ varies in different cell types, their ligand binding, DNA binding, and transactivation properties are rather similar to one another. As we know, MDA-MB-231 cells have been traditionally classified as an ER-negative breast cancer cell lines. But, in our study, we found that

MCF-7 cells expressed both ERα and ERβ, and MDA-MB-231 cells expressed only

ERβ. Based on our results, the MDA-MB-231 cells we employed in the experiment should be recognized as an ERβ-positive breast cancer cell line.

In our present study, we investigated the effects of E2 and Z on PTPγ expression in both ERα (+) human breast cell lines and ERα (-) human breast cell lines, the results showed that both E2 and Z could downregulate PTPγ expression in only ERα (+) cell lines but not in ERα (-) cell lines and the estrogenic effects were blocked by ICI 182,780, which indicated that ERα played an important role in the regulation of PTPγ expression by estrogenically active agents in human breast and Z downregulated PTPγ expression in human breast through the similar pathway as E2 (ER pathway). Estrogens induce gene transcription both by binding to the classic estrogen response element(s) and by signaling through an AP-1 enhancer element requiring the products of c-fos and c-jun [Gaub et al.,

88 1990]. ERα and ERβ may have similar effects on gene transcription mediated via the

EREs but opposite effects on promoters containing AP-1 [Paech et al., 1997]. Estradiol

activates AP-1-mediated gene transcription when bound to ERα but inhibits promoter

activity when bound to ERβ [Paech et al., 1997]. We have briefly analyzed the promoter

region of human PTPγ gene, but we couldn’t find any consensus ERE in that region.

Therefore, we hypothesize that the regulation of PTPγ expression by E2 and Z might not

be through the EREs, which need to be determined by further experiments. Report

showed that ERα and ERβ expressed both in vitro and in vivo, formed heterodimers,

which bind to DNA with an affinity similar to that ERα and greater than that of ERβ

homodimers [Cowley et al., 1997]. Furthermore, the heterodimers, like the homodimers,

are capable of binding the steroid receptor co-activator-1 when bound to DNA and

stimulating transcription of a reporter gene in transfected cells [Cowley et al., 1997].

Based on our investigations, we are still not sure whether ERα alone or a combination of

ERα and ERβ are essential in the downregulation of PTPγ by estrogenically active agents

in human breast. For the future studies, we need to construct ERα- or ERβ- or both

transfected human breast cancer cell lines without any endogenous ER and use these

transfected human breast cancer cell lines to determine whether E2 and Z downregulate

PTPγ expression through ERα homodimers or ERα-ERβ heterodimers.

We have demonstrated that ERα expression level was significantly higher in

cancerous human breast than in normal human breast, while ERβ and PTPγ mRNA

expression was significantly higher in normal human breast than in cancerous human

breast. Our previous report indicated that PTPγ is a potential estrogen-regulated tumor

89 suppressor gene in human breast cancer that may play an important role in neoplastic

processes of human breast epithelium. Altogether, our results shed a light on the tight

relationship among estrogen, human breast cancer and PTPγ. But, the left question is

whether PTPγ can really inhibit human breast tumor formation. Some experiments about

the functional analysis of PTPγ in human breast need to be done in the future.

It is worth noting that the non-steroidal agent with estrogenic activities, Z, has

been employed in the current study. Z has been used as a growth promoter in the U.S.

beef industry to improve feed efficiency, weight gains and carcass quality. The U.S. Food

and Drug Administration (FDA) has a restricted legal limit at 150 parts per billion (ppb)

for the Z residues contained in edible beef for human consumption. At the present time,

there is no proved scientific evidence to suggest that the consumption of beef from the Z-

implanted cattle cause any risk of breast cancer. Our in vitro experimental data clearly

demonstrated that the natural estrogen (E2) and Z are capable of down regulating the

estrogen-regulated cancer suppressor gene, PTPγ in both human normal and cancerous

breast tissues as well as the cultured cells isolated from these tissues. Therefore, our

concern is whether the long-term exposure of extremely low levels of Z residues from the

consumption of beef from Z-implanted cattle has any health risk to the consumers. The

extrapolation of our in vitro findings to the linkage of potential risk in in vivo is not

appropriate. More studies in this important topic are warranted. In combination with the

results from other laboratory which showed that ERß could be a potent proliferation gatekeeper as well as an inhibitor of cell motility and invasion and the decreased

expression of ERß observed between normal and cancerous breast could be one of the

events leading to an uncontrolled proliferation of the cells, our results suggested that ERβ

90 might possess tumor suppressor activity in human breast, which might serve as a new concept in breast cancer research.

91

Figure 4.1. Comparison of estrogen receptors (α and β) mRNA expression in MCF-7 cells and MDA-MB-231 cells as determined by RT-PCR.

MCF-7 cells and MDA-MB-231 cells were cultured in 25-cm2 flasks (3×105 cells/flask)

without treatment. Total RNA was isolated from each cell line and RT-PCR was

performed.

A. Ethidium bromide-stained PCR products separated in a 1.5% agarose gel.

(M: Marker)

B. The mRNA ratios of ERα or ERβ to 36B4 as measured by densitometry.

The results showed that MCF-7 cells express both ERα and ERβ mRNA; and MDA-MB-

231 cells express only ERβ mRNA.

92

A. M

ERα (650 bp)

ERβ (528 bp)

36B4 (563 bp)

B. 0.6

0.5 ER α ER β on 0.4 6B4) pressi

x /3 β NA e ER 0.3

R or α ve m ti 0.2 (ER a Rel 0.1

0.0 7 F- -231 MC

MDA-MB

Figure 4.1

93

ERα (66 KDa)

β-actin (43 KDa)

1 2 3

Figure 4.2. Comparison of estrogen receptor α protein expression in MDA-MB-231 cells, MDA-MB-231-ERα-1000 cells and MCF-7 cells as determined by western blotting assay.

MDA-MB-231 cells, MDA-MB-231-ERα-1000 cells and MCF-7 cells were cultured in

75-cm2 flasks (1.5×106 cells/flask) without treatment. Total protein was isolated from

each cell line and western blotting assay was performed.

1: MDA-MB-231 cells; 2: MDA-MB-231-ERα-1000 cells; 3: MCF-7 cells

The result showed that ERα protein is expressed in MDA-MB-231-ERα-1000 and MCF-

7 cells not in MDA-MB-231 cells.

94

Figure 4.3. Effects of E2, Z and ICI 182,780 on PTPγ mRNA expression in MCF-7 cells as determined by RT-PCR.

MCF-7 cells were cultured in 25-cm2 flasks (3×105 cells/flask) and treated with 30 nM of

E2 or Z in the absence or presence of 10 nM, 100 nM or 1 µM of ICI 182,780, or ICI

182,780 alone for 24 hours. Total RNA was isolated from each treatment group and RT-

PCR was performed.

A. Ethidium bromide-stained PCR products separated in a 1.5% agarose gel.

(M: Marker)

B. The mRNA ratios of PTPγ to 36B4 as measured by densitometry.

These results showed that the expression levels of PTPγ mRNA in the E2- and Z-treated

MCF-7 cells were significantly suppressed compared to the PTPγ mRNA levels in the

controls (approximately a 60% reduction) and these effects were completely blocked by 1

µM of ICI, however ICI alone had no effects on PTPγ mRNA expression.

95

A. M PTPγ (492 bp)

36B4 (563 bp)

B.

0.40

0.35

0.30 on i s es r 0.25 4) exp 36B / NA

γ 0.20 P T

(P 0.15 ve mR i t a

Rel 0.10

0.05

0.00 l I I I I I Z tro CI ICI IC n M IC IC IC IC ICI n M I M M ICI Co nM E230 nM nM µ nM nM µΜ 30 0 0 1 0 µΜ 0 0 nM 1 1 10 10 n 0 + + 1 Z + 2 + 10 + 1 + 1 2 Z 2 E Z M M E M E nM 0 n n 0 nM 30 n 3 0 nM 3 0 30 3 3

Figure 4.3

96

Figure 4.4. Effects of E2, Z and ICI 182,780 on PTPγ mRNA expression in MDA-

MB-231 cells as determined by RT-PCR.

MDA-MB-231 cells were cultured in 25-cm2 flasks (3×105 cells/flask) and treated with

30 nM of E2 or Z in the absence or presence of 10 nM, 100 nM or 1 µM of ICI 182,780,

or ICI 182,780 alone for 24 hours. Total RNA was isolated from each treatment group

and RT-PCR was performed.

A. Ethidium bromide-stained PCR products separated in a 1.5% agarose gel.

(M: Marker)

B. The mRNA ratios of PTPγ to 36B4 as measured by densitometry.

The result showed that, in contrast to figure 4.3, E2, Z and ICI had no effects on PTPγ

expression in MDA-MB-231 cells.

97

A. M PTPγ (492 bp)

36B4 (563 bp)

B.

1.00 on 0.75 pressi 4) x e 6B /3 NA γ P R 0.50 (PT ve m i

Relat 0.25

0.00 l I I I 2 Z I ro ICI ICI C ICI ICI nt M ICI IC IC IC M E n M M M Co 0 n 30 nM µ nM nM I nM nM µΜ 3 0 1 µΜ 0 0 10 n 10 00 1 1 10 1 1 + + + 2 + Z + 10 Z + E 2 2 Z M M E n M M E nM nM n 0 30 n 0 3 30 n 30 3 30

Figure 4.4

98

Figure 4.5. Effects of E2, Z and ICI 182,780 on PTPγ mRNA expression in MDA-

MB-231-ERα-1000 cells as determined by RT-PCR.

MDA-MB-231-ERα-1000 cells were cultured in 25-cm2 flasks (3×105 cells/flask) and

treated with 30 nM of E2 or Z in the absence or presence of 10 nM, 100 nM or 1 µM of

ICI 182,780, or ICI 182,780 alone for 24 hours. Total RNA was isolated from each

treatment group and RT-PCR was performed.

A. Ethidium bromide-stained PCR products separated in a 1.5% agarose gel.

(M: Marker)

B. The mRNA ratios of PTPγ to 36B4 as measured by densitometry.

The result showed that PTPγ mRNA expression in the E2- and Z-treated (30 nM, 24 h)

ERα-transfected MDA-MB-231 cells (MDA-MB-231-ERα-1000) were significantly

suppressed in comparison to the PTPγ mRNA levels observed in the controls

(approximately an 50% reduction) and these effects were completely blocked by 100 nM

of ICI, however ICI alone had no blocking effects on PTPγ mRNA expression

99

A.

M PTPγ(492 bp)

36B4 (563 bp)

B.

1.00 on i s 0.75 es r 4) exp 36B / NA γ P 0.50 T (P ve mR i t a

Rel 0.25

0.00 l I I I I I Z tro CI ICI IC n M IC IC IC IC ICI n M I M M ICI Co nM E230 nM nM µ nM nM µΜ 30 0 0 1 0 µΜ 0 0 nM 1 1 10 10 n 0 + + 1 Z + 2 + 10 + 1 + 1 2 Z 2 E Z M M E M E nM 0 n n 0 nM 30 n 3 0 nM 3 0 30 3 3

Figure 4.5

100

0.6

Normal breast 0.5 Cancerous breast n

io 0.4 ress ) 4

exp

6B A

/3 0.3 N α R m (ER e 0.2

Relativ 0.1

0.0 s ll sues l Ce l Cells is a a T li he rom St Epit

Figure 4.6. Comparison of ERα mRNA expression in human breast as determined by RT-PCR.

Total RNA was isolated from both normal and cancerous human breast tissues, epithelial cells and stromal cells (Normal human breast tissues and cells are from the same patients;

Cancerous human breast tissues and cells are from the same patients) and RT-PCR was performed. The mRNA ratios of ERα to 36B4 was measured by densitometry.

Our results showed that ERα mRNA expression was significantly higher in cancerous human breast tissues and cells than in normal human breast tissues and cells.

101

0.4 Normal breast Cancerous breast n o i 0.3 ss e r p 4) x

e

/36B 0.2 NA β R (E ve mR i t a l 0.1 Re

0.0 s lls lls e C l Tissue lial Ce he roma St Epit

Figure 4.7. Comparison of ERβ mRNA expression in human breast as determined by RT-PCR.

Total RNA was isolated from both normal and cancerous human breast tissues, epithelial cells and stromal cells (Normal human breast tissues and cells are from the same patients;

Cancerous human breast tissues and cells are from the same patients) and RT-PCR was performed. The mRNA ratios of ERβ to 36B4 was measured by densitometry.

Our results showed that ERβ mRNA expression was significantly lower in cancerous human breast tissues and cells than in normal human breast tissues and cells.

102

Figure 4.8. Comparison of immunoreactivities of ERα, ERβ and PTPγ in normal and cancerous human breast tissues as determined by immunohistochemical staining.

Brown staining represents ERα, ERβ, or PTPγ immunoreactivities; Blue staining

represents nuclei. A. Normal human breast tissue stained for ERα; B. Cancerous human

breast tissue stained for ERα; C. Normal human breast tissue stained for ERβ; D.

Cancerous human breast tissue stained for ERβ; E. Normal human breast tissue stained

for PTPγ; F. Cancerous human breast tissue stained for PTPγ.

These results showed that ERα protein level significantly higher in cancerous human

breast tissues than in normal human breast tissues, while ERβ and PTPγ protein levels

were higher in normal human breast tissues than in cancerous human breast tissues.

103 AB

10 µm 10 µm

C D

10 µm 10 µm

EF

10 µm 10 µm

Figure 4.8

104 CHAPTER 5

FUNCTIONAL ANALYSIS OF ESTROGENICALLY REGULATED PROTEIN

TYROSINE PHOSPHATASE γ (PTPγ) IN HUMAN BREAST

CANCER CELL LINE MCF-7

ABSTRACT

Protein tyrosine phosphatase γ (PTPγ) is a member of the receptor-like family of

tyrosine-specific phosphatases and has been implicated as a tumor suppressor gene in

kidney and lung cancers. We have reported that PTPγ expression in normal human breast

was higher than that in cancerous human breast, and PTPγ expression was downregulated

through estrogen receptor α (ERα) homodimers or ERα and ERβ heterodimers by 17β-

estradiol (E2) and Zeranol (Z). Based on these findings, we hypothesize that PTPγ is a potential estrogen-regulated tumor suppressor gene in human breast cancer that may involve in neoplastic processes of human breast epithelium. The present study is to examine the action of PTPγ on in vitro growth of MCF-7 cells and compare the estrogenic responses of human breast cells with different expression levels of PTPγ. By

105 transfecting plasmid DNA with or without different cDNA inserts of PTPγ in pCR®3.1 vector into MCF-7 cells and selecting with G-418 (Geneticin), we established several stably transfected MCF-7 cell lines expressing different levels of PTPγ. By RT-PCR, we

found that full-length PTPγ-transfected MCF-7 cells [M7-PTPγ-800 (s)] had much higher

PTPγ mRNA expression than both wild type and mock-transfected MCF-7 cells [M7-

pCR®3.1-800 (s)], whereas, antisense PTPγ-transfected MCF-7 cells [M7-A24-800 (s)] had much lower PTPγ mRNA expression than wild type, M7-pCR®3.1-800 (s) and sense

PTPγ-transfected MCF-7 cells [M7-B19-800 (s)]. In doubling time assay, we demonstrated that the doubling time for M7-PTPγ-800 (s) was increased about 60%. In contrast, the doubling time for M7-A24-800 (s) was decreased about 34%. These results suggested that PTPγ is capable of inhibiting MCF-7 breast cancer cell growth and plays an important role in growth suppression of breast cancer. Our results further demonstrated that PTPγ overexpression led to a decrease in the anchorage-independent growth of MCF-7 cells in soft agar, and the reduction of PTPγ expression in MCF-7 cells resulted in an increase in the colony formation in soft agar. Furthermore, the results in the cell proliferation assay showed that PTPγ overexpression could reduce the estrogenic responses of MCF-7 cell proliferation to E2 and Z. Our data suggest that PTPγ is able to

inhibit proliferation and anchorage-independent growth of breast cancer cells in vitro and

has anti-estrogenic activities in human breast cancer cells, and PTPγ may function as an

important modulator in regulating the process of tumorigenesis in human breast. Studies

in progress are focused on examining the effect of PTPγ on the in vivo growth of human

breast cancer.

106

INTRODUCTION

Protein tyrosine phosphatases (PTPases) are a family of proteins which perform the enzymatic role to remove phosphate groups from phosphotyrosine residues of specific targets inside cells. PTPases regulate important cellular processes like gene expression, cell activation and proliferation, differentiation, development, transport, and locomotion, since they counterbalance the growth-promoting effects of protein tyrosine kinases

(PTKs), which catalyze the phosphorylation of tyrosine residues [Shock et al., 1995].

Therefore, alterations in PTPs activity might affect cell growth, neoplastic processes and transformation [Gaits et al., 1995]. Addition of the PTPase-inhibitor vanadate to cells in culture leads to increased amounts of phosphotyrosine-containing proteins and cellular transformation [Klarlund, 1985]. Therefore, a delicate balance between PTK and PTPase action is essential for normal functioning of cells.

PTPγ is a member of the receptor-like family of tyrosine-specific phosphatases as described in Chapter 3. The structure of the receptor-like PTPs (RPTPs) includes an extracellular, a transmembrane, and one or two tandemly repeated catalytic domains. This structure implies ligand-binding capability which may modulate enzymatic activity.

However, the putative ligands for most of the PTPs with receptor-like structures are not yet to be identified. Receptor-like PTPγ has a carbonic anhydrase-homologous amino terminus followed by a fibronectin type three domain, a cysteine-free domain, a 107 transmembrane domain and two intracellular PTPase catalytic domains [Barnea et al.,

1993; Levy et al., 1993; Krueger et al., 1992]. According to mRNA analysis [Barnea et al., 1993; LaForgia et al., 1991], PTPγ is a broadly expressed enzyme that exists in many tissues, including human lung, stomach, esophagus, colon, liver, spleen, and kidney

[Tsukamoto et al., 1992]. Based on the chromosomal location of the PTPγ gene (3p14.2)

[LaForgia et al., 1991; LaForgia et al., 1993] and studies showing loss of heterozygosity

(LOH) of the gene in kidney tumors [Lubinski et al., 1994], PTPγ has been implicated as a candidate tumor suppressor gene. More recently, PTPγ expression levels were shown to be reduced in ovarian and lung tumors [van Niekerk et al., 1999]. Our previous results showed lower PTPγ mRNA expression levels in diethylstilbestrol-induced kidney tumors in hamsters than in normal hamster kidney [Lin et al., 1994]. Also, we have shown that, in ACI rats, PTPγ is localized to the mammary epithelium and that zeranol (Z), a nonsteroidal agent with estrogenic activities that is used as a growth promoter in the U.S. beef, veal and lamb industries, can suppress PTPγ mRNA levels in mammary glands

[Kulp et al., 2000]. Furthermore, we have reported that PTPγ is expressed in normal and malignant human breast epithelium, and PTPγ mRNA levels can be suppressed by estrogens through an estrogen receptor-mediated mechanism [Zheng et al., 2000]. More recently, our results showed that PTPγ mRNA expression was lower in cancerous than in normal breast tissues; both 17β-estradiol (E2) and Z suppressed PTPγ mRNA levels in both cultured normal breast tissues and cultured breast cells isolated from normal breast tissues. In whole breast tissues, PTPγ was immunolocalized to the epithelium and the treatment with E2 or Z diminished PTPγ staining, which indicated reductions in PTPγ at

108 the protein level [Liu et al., 2002a]. These findings suggest that PTPγ is a potential estrogen-regulated tumor suppressor gene in human breast cancer that may play an important role in neoplastic processes of human breast epithelium, and indicate an intriguing relationship among cancer, estrogen and PTPγ.

The aim of the present study was to examine the effect of PTPγ on the growth of human breast cancer and compare the estrogenic responses of human breast cells with different expression levels of PTPγ. MCF-7 breast cancer cells were used as a model and we stably transfected a plasmid containing the whole PTPγ coding sequence or containing a small portion of PTPγ coding sequence in the antisense orientation into human MCF-7 breast carcinoma cells. PTPγ-overexpression MCF-7 cells normally show the characteristics with slow-growing and lower colony efficiency in soft agar, and show anti-estrogenic effects on human breast cell proliferation.

MATERIALS AND METHODS

Cell culture. The human breast cancer cell line MCF-7 cells were purchased from

American Type Culture Collection (ATCC, Manassas, VA). The cells were maintained in a mixture of Dulbecco's Modified Eagle's Medium and Ham's F12 Medium (1:1)

(DMEM/F12) without phenol red (Sigma Chemical Co., St. Louis, MO) supplemented with 5% Fetal Bovine Serum (FBS) (Atlanta Biologicals, Norcross, GA) and antibiotic- antimycotic (100 unit/ml penicillin G sodium, 100 mg/ml streptomycin sulfate and 0.25 109 mg/ml amphotericin B) (GibcoBRL, Bethesda, MD) and were plated separately in 75- cm2 culture flasks in a humidified incubator (5% CO2: 95% air, 37°C). The media were changed every two days. When the cells grew to about 85% confluence, cells were washed twice with calcium- and magnesium-free Phosphate Buffered Saline (PBS, pH7.3), and then trypsinized with 1 ml of 0.5% trypsin - 5.3 mM EDTA (GibcoBRL) in

PBS for 10 minutes at 37°C. The trypsinization was stopped by addition of 10 ml of culture medium with 5% FBS. After centrifugation, the viability of the dissociated cells were determined by using a hemacytometer, which showed that the viability was over

99%. The dissociated cells were resuspended in the same medium and subcultured into

75-cm2 culture flasks at a ratio of 1 flask to 5 flasks.

Transfection (Figure 5.1). The expression vector pCR®3.1 (Invitrogen, Carlsbad,

CA; Figure 5.2), containing a human full-length PTPγ cDNA (5.3 kb; Figure 5.3) in an

EcoRI site or a partial PTPγ cDNA fragment (756 bp amplified using the following primer set: 5’-CGTCACCAGTCTCCTATTGA-3’ and 5’-

GTCTGTCATGTCGTGGTTCC-3’; Figure 5.3) cloned in pCR®3.1 vector. Clones containing the construct in the sense or antisense orientation were selected and used for transfection [Sorio et al., 1995; Sorio et al., 1997]. As a control, we used an empty pCR®3.1 vector (mock). Plasmids (10 µg) were transfected into MCF-7 cells by using

Superfect (Qiagen, Valencia, CA) according to the manufacturer’s instructions. Forty- eight hours after transfection, the cells were transferred to selection medium containing different doses (0 µg/ml, 100 µg/ml, 200 µg/ml, 400 µg/ml, 600 µg/ml, 800 µg/ml, 1000

µg/ml) of G-418 (Geneticin) (Invitrogen, Carlsbad, CA) for 6 days to determine the optimal dose of G-418 for the selection. Finally, surviving single colonies from four to 110 five dishes with selection medium containing 800 µg/ml of G-418 were picked from each

stably transfected group for future culture and cells were treated with 400 µg/ml of G-418

in the culture medium every 2-3 passages to maintain the homogeneity of the cells. The

whole procedure for the establishment of cell lines is shown in Figure 5.1.

RNA isolation and Reverse transcription-polymerase chain reaction (RT-PCR).

1.5×105 viable cells from each single colony were plated in one well of 6-well plates with

5 ml of culture medium and cultured to ~ 85% confluence. Total RNA was isolated in 1 ml of TRIZOL Reagent (GibcoBRL) according to manufacturer’s instructions. RT-PCR

was performed in a gradient mastercycler (Eppendorf ®). PCR conditions were optimized

for MgCl2 concentration, annealing temperature and cycle number for the amplification

of each of the PCR products. Under optimal conditions, the amounts of PCR products

generated fell within the linear portion of the PCR amplification curve between twenty-

six and thirty-nine amplification cycles. Briefly, 1 µg of total RNA from cultured cells or

tissues was reverse transcribed with 200 U M-MLV Reverse Transcriptase (GibcoBRL)

at 42°C for 1 hour in the presence of 5 mM each of dATP, dCTP, dGTP and dTTP, 4 µl

5x 1st strand buffer (GibcoBRL), 0.01M DDT, 1 U RNA Guard RNase inhibitor

(Pharmacia Biotech, Uppsala, Sweden), and 2.5 mM random hexamers in a total volume

of 20 µl. The reaction was terminated by heating to 95°C for 3 minutes. The newly

synthesized cDNAs were used as templates for PCR after adjusting reagent

concentrations to 3.5 mM MgCl2, 2.5 µl 10× PCR Buffer (GibcoBRL), 1 U Platinum®

Taq DNA polymerase (GibcoBRL), and 0.24 µM primers. The reactant was incubated at

95 °C for 5 minutes. Then, thirty cycles of amplification were performed with each cycle

consisting of denaturation at 95°C for 1 minute, annealing at 63°C for 1 minute, and 111 extension at 72°C for 1 minute. The primer sequences generating 990 bp PCR products

for full-length PTPγ construct were 5′ _GTA TGG AGC AGT TTC AGC_ 3′ (sense @

PTPγ exon 29) and 5′ _TAG AAG GCA CAG TCG AGG_ 3′ (antisense @ pCR®3.1 reverse priming site). The primer sequences generating 583 bp PCR products for antisense PTPγ construct were 5′ _TAA TAC GAC TCA CTA TAG GG_ 3′ (sense @ T7

promoter priming site in pCR®3.1 vector) and 5′ _GAA CAC AGC ATC AAT GGC

AGG AGG_ 3′ (antisense @ PTPγ exon 4). The primer sequences generating 337 bp

PCR products for sense PTPγ construct were 5′ _ TAA TAC GAC TCA CTA TAG GG_

3′ (sense @ T7 promoter priming site in pCR®3.1 vector) and 5′ _CCT CCT GCC ATT

GAT GCT GTG TTC_ 3′ (antisense @ PTPγ exon 4). The primer sequences generating

492 bp PCR product for PTPγ were 5′ _GCG CAG CGA CTT TAG CCA GAC GA_ 3′

(sense @ PTPγ exon 8) and 5′ _GCT CCC GCT CCC CAT CCT CAC TC_ 3′ (antisense

@ PTPγ exon 11). The primer sequences generating 563 bp PCR products for 36B4 were

5′ _AAA CTG CTG CCT CAT ATC CG_ 3′ (sense @ 306 to 325) and 5′ _TTG ATG

ATA GAA TGG GGT ACT GAT G_ 3′ (antisense @ 868 to 848). The final PCR

products (10 µl) mixed with 1 µl of 10 × loading buffer were separated on a 1.5% agarose

gel containing ethidium bromide. The specific bands were quantified by ImageQuaNT

software (Molecular Dynamics, Sunnyvale, CA). The results are presented as the ratio of

each PCR product to 36B4.

Immunohistochemical staining. Wild type MCF-7 cells, mock-transfected MCF-7

cells (M7-pCR®3.1-800 (s)), PTPγ-overexpression MCF-7 cells (M7-PTPγ-800 (s)),

antisense PTPγ-transfected MCF-7 cells (M7-A24-800 (s)) and sense PTPγ-transfected

112 MCF-7 cells (M7-B19-800 (s)) were cultured on 24 × 30 mm cell culture cover slips

(Cat. No: 150067, Nalge Nunc Int., Naperville, IL) and washed in PBS for 3 times. After

–10 °C methanol fixation for 5 minutes and air-dry, cells were stained for PTPγ by using

VECTASTAIN Universal Quick Kit and DAB Substrate Kit (Cat. No: PK-8800, Cat.

No: SK-4100, Vector Laboratories, Inc., Burlingame, CA) according to the manufacturer’s instructions. The primary antibody was an affinity-purified rabbit polyclonal antibody raised against a self-designed in-house synthetic peptide (N–

SEDGEREHEEDGEKD –C) corresponding to amino acids 588-602 mapping in the extracellular domain of the human origin PTPγ and was produced by Sigma® Genosys

(The Woodlands, TX). An optimal primary antibody dilution of 1:100 was determined by titration in this system, and omission of primary antibody served as a negative control.

Doubling time assay. Wild type MCF-7 cells, M7-pCR®3.1-800 (s), M7-PTPγ-

800 (s), M7-A24-800 (s) and M7-B19-800 (s) were plated separately at a density of

0.5×104 cells / well in 24-well plates in a volume of 1 ml of fresh DMEM/F12 with 5%

FBS/well. After cells are attached to the wells, the medium was replaced with 2 ml of fresh DMEM/F12 with 5% FBS. At the same time (time 0 hour), a group of cells were counted. Cells were grown for 3 days and counted every 8 hours. Adherent cells were detached by rapid trypsinization. An adequate volume of medium containing trypan blue was added. Then cells were counted by using a hemacytometer. Experiments were performed in four replicate wells, and each experiment was repeated twice. Based on the counted cell numbers at different time points, a cell proliferation curve was generated.

Cell doubling (CD) was calculated by using the formula ln (Nj-Ni)/ln 2 where Nj or Ni is the cell numbers at different time point Tj or Ti (Tj > Ti) in the growth log phase of the 113 cells. Doubling time (DT) was consequently obtained by dividing the time interval (Tj >

Ti) by CD [Poliseno et al., 2002].

Soft agar assay for colony formation. Cells were cultured in 6-well plates first

covered with an agar layer (1 ml of phenol red-free DMEM/F12 with 0.5% agar and 10%

FBS. The middle layer contained different number of cells (2000, 4000, 8000 or 16000

cells) in 1 ml of phenol red-free DMEM/F12 with 0.35% agar and 10% FBS. The top

layer, consisting of 1 ml of culture medium, was added to prevent drying of the agar in

the plates. The plates were incubated for 21 days and another 1 ml of fresh culture

medium was added to the top layer at day 11. After 21 days’ incubation, the plates were

stained in 0.5 ml of 0.005% crystal violet for >1 hour and the cultures were inspected and

photographed. Colony efficiency (CE) was determined by a count of the number of

colonies greater than 150 µm in diameter, which was calculated as the average of

colonies counted at 40× magnification in five individual fields by using BioQuant NOVA

software.

Non-radioactive cell proliferation assay. 2×104 cells [Wild type MCF-7 cells,

M7-pCR®3.1-800 (s), M7-PTPγ-800 (s), M7-A24-800 (s) or M7-B19-800 (s)] were

seeded and cultured in each well of 24-well plates in phenol red-free high-calcium

DMEM/F12 supplemented with 5% FBS overnight. The medium was switch to phenol

red-free high-calcium DMEM/F12 supplemented with Dextran-coated charcoal (DCC)

(Dextran T-70, Pharmacia; activated charcoal, Sigma)-stripped FBS (5%). After 24

hours, cells were treated with the E2 or Z at 0, 7.5, 15, 30, 60 or 120 nM in the same fresh medium for 24 hours. Cell proliferation rate was quantified by using CellTiter 96TM

AQueous assay (Promega, Madison, WI). Briefly, at the end of treatment, the medium in

114 the wells were discarded and 250 µl of fresh medium with 50 µl of freshly combined

MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3- carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-

tetrazolium, inner salt)/PMS (phenazine methosulfate) (the ratio of MTS:PMS is 20:1))

solution was added to each well. Then, the plates were incubated for 1.5 hours and the

color density was checked every 30 minutes. Finally, 100 µl of culture medium from

each well were transferred to one well of 96-well plates and optical density was read at

490 nm (O.D.490nm) by an ELISA plate reader.

Statistical analysis. The results for doubling time assay and non-radioactive cell

proliferation assay were presented as the mean ± standard deviation (SD) for 4 replicate

culture wells as one group. The results for colony efficiency assay were presented as the

mean ± standard deviation (SD) for 5 individual fields as one well. Analysis was

performed using Minitab statistical software for Windows (Minitab Inc., State College,

PA, USA). Statistical differences were determined by using one-way ANOVA for

independent groups. P-values of less than 0.05 were considered statistically significant.

RESULTS

Comparison of PTPγ expression in stably transfected MCF-7 cells. MCF-7 cells

were transfected with pCR®3.1 vector, full-length PTPγ, PTPγ antisense construct or

PTPγ sense construct. Based on the dose-response curve of each transfected cell line to

G-418, it was determined that 800 µg/ml was the optimal dose of G-418 for the single 115 colony selection in every transfected cell line. We have selected 5 single clones from

mock-transfected MCF-7 cells, 19 single clones from PTPγ-transfected MCF-7 cells, 17

single clones from antisense PTPγ-transfected MCF-7 cells and 11 single clones from

sense PTPγ-transfected MCF-7 cells. To identify the successfully and ideally transfected

single colonies, RT-PCR with a set of specific primers was used for each group of

transfection (Figure 5.4). The results showed that 13 out of 19 clones from M7-PTPγ-800

(s) expressed exogenous PTPγ mRNA, and all of the single clones from M7-pCR®3.1-

800 (s), M7-A24-800 (s) and M7-B19-800 (s) were successful. Furthermore, PTPγ

mRNA expression was compared among those successfully transfected cell lines. Our

results showed that M7-PTPγ-800 (s) had higher PTPγ mRNA expression than any of

wild type MCF-7, M7-pCR®3.1-800 (s), M7-A24-800 (s) and M7-B19-800 (s) cell lines.

M7-A24-800 (s) had lower PTPγ mRNA expression than any of wild type MCF-7, M7-

pCR®3.1-800 (s), M7-PTPγ-800 (s) and M7-B19-800 (s) cell lines (Figure 5.5). Also, the

results from immunohistochemical staining showed that the PTPγ staining were weak and

very similar among wild type MCF-7 cell (Figure 5.6A), M7-pCR®3.1-800 (s) (Figure

5.6B) and M7-B19-800 (s) (Figure 5.6C). Furthermore, the degree of staining was

observably increased in M7-PTPγ-800 (s) (Figure 5.6D) when compared to others;

however, the degree of staining was almost diminished in M7-A24-800 (s) (Figure 5.6E)

when compared to others. Thus, these results indicate that PTPγ is overexpressed in M7-

PTPγ-800 (s) and is greatly reduced in M7-A24-800 (s) compared to other cells. These

results are consistent with those presented in Figure 5.5 above.

Effects of PTPγ on the proliferation of MCF-7 cells. To check whether PTPγ level

116 has an anti-mitogenic effect on MCF-7 cells, a proliferation study was performed by

using doubling time assay. Wild type MCF-7 cells, M7-pCR®3.1-800 (s) and M7-B19-

800 (s) had the mean cell population doubling time of 24.0 ± 2 h, 24.3 ± 3 h and 23.8 ± 1

h, respectively, but it was increased to 40.0 ± 3 h for M7-PTPγ-800 (s) (~ 60% increase)

and it was decreased to 16.4 ± 1 h for M7-A24-800 (s) (~ 34% decrease) (Figure 5.7).

These results suggested that PTPγ can inhibit MCF-7 breast cancer cell growth and plays

an important role in tumor suppression of breast cancer.

Effects of PTPγ on anchorage-independent growth of MCF-7 cells. The

acquisition of anchorage-independent growth is generally considered to be one of the in

vitro properties associated with the malignancy of cells. The transfected cells were

examined for their ability to grow in soft agar. First, we checked the optimal number of

cells for the colony formation in 6-well plates seeded with different number of wild type

MCF-7 cells, M7-pCR®3.1-800 (s) or M7-PTPγ-800 (s). The results showed that the colony number in soft agar was increased while the seeded cell number was increased for each cell type, M7-PTPγ-800 (s) formed much less and smaller colonies in soft agar than wild type MCF-7 cells and M7- pCR®3.1-800 (s) in the same seeded cell number group

(Figure 5.8). Based on these results, we determined that 8000 cells/well was the optimal

MCF-7 cell number for the future soft agar assay in our experiment. Furthermore, we

found that M7-PTPγ-800 (s) formed less and smaller colonies in soft agar than wild type

MCF-7 cells and M7- pCR®3.1-800 (s), but M7-A24-800 (s) formed more and much

larger colonies in soft agar than wild type MCF-7 cells, M7- pCR®3.1-800 (s) and M7-

B19-800 (s) (Figure 5.9). These in vitro data indicated that PTPγ is able to inhibit proliferation and anchorage-independent growth of breast cancer cells. 117 Effects of PTPγ on estrogenic activities of E2 and Z on MCF-7 cell proliferation.

As we have known, E2 is a potent mitogen for human breast cancers. Z is a non-steroidal

anabolic growth promoter with estrogenic activities. Our previous results showed that

both Z and E2 could induce normal human breast epithelial cell transformation and

stimulate human breast cell proliferation in a dose-dependent manner, which suggested

that PTPγ is a potential estrogen-regulated tumor suppressor gene in human breast cancer

that may play an important role in neoplastic processes of human breast epithelium,

which indicates an intriguing relationship among cancer, estrogen and PTPγ. The

established stably transfected cell lines with different PTPγ expression levels were used

to examine whether PTPγ has any anti-estrogenic activities on MCF-7 cell proliferation.

Our results showed that both Z and E2 could stimulate the cell proliferation of wild type

MCF-7, M7-pCR®3.1-800 (s), M7-PTPγ-800 (s), M7-B19-800 (s) and M7-A24-800 (s) in a dose-dependent manner, and the potency of Z and E2 are very similar on the stimulation

of cell proliferation (Figure 5.10). Most interestingly, the degree of Z’s or E2’s

stimulation on M7-PTPγ-800 (s) cell proliferation (Figure 5.10D) is much less than that

on the cell proliferation of wild type MCF-7, M7-pCR®3.1-800 (s) or M7-B19-800 (s) cell proliferation (Figure 5.10A, 5.10B, 5.10C); however, the degree of Z’s or E2’s

stimulation on M7-A24-800 (s) cell proliferation (Figure 5.10E) is much greater than that

on the cell proliferation of wild type MCF-7, M7-pCR®3.1-800 (s) or M7-B19-800 (s) cell proliferation (Figure 5.10A, 5.10B, 5.10C). These results indicated that PTPγ overexpression could reduce the estrogenic responses of MCF-7 cell proliferation to E2 and Z and has anti-estrogenic activities on human breast cancer cells.

118 DISCUSSIONS

Transmembrane receptor tyrosine phosphatases represent a growing family of enzymes, the structural features of which suggest a role in the control of cellular phosphotyrosine balance. But the biological significance of PTPγ is only poorly understood. PTPγ has been considered as a candidate tumor suppressor gene due to the location of its gene on human chromosome 3p14.2, in a region found to be frequently deleted in certain types of renal and lung carcinomas [Barnea et al., 1993; LaForgia et al.,

1991; LaForgia et al., 1993]. Our previous results suggested that PTPγ is a potential estrogen-regulated tumor suppressor gene in human breast cancer that may play an important role in breast tumorigenesis [Lin et al., 1994; Kulp et al., 2000; Zheng et al.,

2000; Liu et al., 2002]. But, the function of PTPγ in human breast cancer is still not clear.

To examine the effect of PTPγ on the growth properties of human breast cancer cells in vitro and provide a biochemical background for understanding PTPγ function, we transfected PTPγ-expressing vector with a variant full-length PTPγ cDNA lacking the intracellular juxtamembrane exon of the reported PTPγ cDNA, which is expressed as the dominant isoform in most tissues [Sorio et al., 1995], or PTPγ antisense construct into

MCF-7 cells. It is clear from the data shown in Figure 5.7 that the relationship of link for the PTPγ and longer doubling time as well as less colony formation is intimately related.

When the breast cancer cells have higher levels of PTPγ mRNA or its protein, the proliferative ability of breast cancer cells will be greatly suppressed or blocked.

119 The results from doubling time assay and soft agar assay showed that MCF-7

cells overexpressing PTPγ has the characteristics with slow-growing and lower colony

efficiency in soft agar. Furthermore, the results from the cell proliferation assay showed

that MCF-7 cells overexpressing PTPγ can antagonize estrogenic effects of E2 and Z on

MCF-7 cell proliferation, which indicates PTPγ has anti-estrogenic effects on human breast cancer.

In combination with our previous results which showed that PTPγ mRNA

expression was significantly lower in cancerous than in normal breast tissues, and both E2 and Z suppressed PTPγ mRNA levels in cultured normal breast tissues and cultured normal breast cells isolated from the tissues, all these experimental data support our suggested concept that PTPγ is a potential estrogen-regulated breast cancer suppressor

gene. This molecular biomarker will be a powerful tool for investigating the estrogen

and/or nonsteroidal estrogenic agents in controlling the etiological process of

tumorigenesis in human breast and rodent mammary. It was claimed that the reduction of

PTPγ expression in human breast cancer is unlikely to be due to genetic events; therefore,

epigenetic mechanisms (such as methylation) might be responsible. Several genes have

been demonstrated to be hypermethylated in breast carcinomas, including HIN-1, p16, E-

cadherin, BRCA1, estrogen receptor, GSTP1 (glutathione S- P1), MDGI

(mammary-derived growth inhibitor), HoxA5 and 14-3-3s (Krop et al., 2001). To identify

the mechanisms of PTPγ reduction in human breast cancer and PTPγ downregulation by

E2 and Z, human PTPγ promoter region analysis and investigation of hypermethylation in

PTPγ need to be conducted in the future.

120 Our data are the first evidence to demonstrate that the PTPγ does exert its suppressor capability to inhibit or slow down the proliferation of MCF-7 cells and does have the anti-estrogenic activities on human breast cancers exposed to E2 or Z. Although the in vivo function of PTPγ is unknown, its reduced expression in breast carcinomas and decreased colony formation following its overexpression suggest a tumor suppressor role.

If similar results can be reproduced in athymic mouse testing model, then the presence of this potential cancer suppressor gene and its protein will significantly prolong the survivability of human breast cancer patients by inhibiting or slowing down the proliferative ability of human breast cancer cells. The signaling pathway with PTPγ involvement may provide a new target for cancer prevention and treatment. In the future,

PTPγ might be able to be applied in clinical trials for cancer prevention and cancer therapy.

121

Figure 5.1. The procedures for the establishment of stably transfected cell lines.

The details were described in Materials and Methods. M7-pCR®3.1-800-a, b, … (s) are

single colonies selected with 800 µg/ml G418 from MCF-7 cells stably transfected pCR®3.1 vectors without inserts. M7-PTPγ-800-a, b, … (s) are single colonies selected

with 800 µg/ml G418 from MCF-7 cells stably transfected with full-length PTPγ cDNA

constructs (a pCR®3.1 vector carrying full-length PTPγ cDNA in the sense direction).

M7-A24-800-a, b … (s) are single colonies selected with 800 µg/ml G418 from MCF-7 cells stably transfected with PTPγ antisense constructs (a pCR®3.1 vector carrying a 756

bp PTPγ cDNA in the antisense direction). M7-B19-800-a, b … m (s) are single colonies

selected with 800 µg/ml G-418 from MCF-7 cells stably transfected with PTPγ sense

constructs (a pCR®3.1 vector carrying a 756 bp PTPγ cDNA in the sense direction).

122

EcoR I PTPγ cDNAs M7-pCR®3.1-800-a,b, … (s) pCR®3.1 M7-PTPγ-800-a,b, … (s) M7-A24-800-a,b, … (s) M7-B19-800-a,b, … (s) Cloning

Single clone selection

Transformation 100 200 400 600 800 1000 ….…...….... …...…...…. . ….…...….... …...…...…. . ….…..….... . ….…...….... …..…...…. . Preparation of plasmid DNA Geneticin (G 418) Transfection selection (ug/ml) ….…..….... . MCF-7 cells

Figure 5.1

123

PTPγ cDNA*

Figure 5.2. Diagram of pCR®3.1 vector from invitrogen and the inserting location of

PTPγ cDNA in the vector (Adapted from Invitrogen).

124

Figure 5.3. The sequences for the inserted PTPγ cDNAs.

The whole sequence represents human protein tyrosine phosphatase γ (PTPγ) mRNA

(www.ncbi.nih.gov; Accession number: L09247). The letters in blue color represent the partial PTPγ cDNA inserted to pCR®3.1 vector in sense or antisense direction; EcoRIsite:

EcoRI site was ligated to the two ends of the inserted PTPγ cDNAs;

ATG: translation start codon; TGA: translation stop codon

125 EcoRIsite+AGGCTCGCACGGAGGCAAGAACTTATTCAACAAGTTTACCTCCCTGCTTTCCTCTTTTCGATGTGCGTTTTCG GACATGCGGAGGTTACTGGAACCGTGTTGGTGGATTTTGTTCCTGAAAATCACCAGTTCCGTGCTCCATTATGTCGTGTG CTTCCCCGCGTTGACAGAAGGCTACGTTGGGGCCCTGCACGAGAATAGACACGGCAGCGCAGTGCAGATCCGCAGGCG CAAGGCTTCAGGCGACCCGTACTGGGCCTACTCTGGTGCCTATGGTCCTGAGCACTGGGTCACGTCTAGTGTCAGCTGT GGGAGCCGTCACCAGTCTCCTATTGACATTTTAGACCAGTATGCGCGTGTTGGGGAAGAATACCAGGAACTGCAACTCG ATGGCTTCGACAATGAGTCTTCTAACAAAACCTGGATGAAAAACACAGGGAAAACAGTCGCCATCCTTCTGAAAGACG ACTATTTTGTCAGTGGAGCTGGTCTACCTGGCAGATTCAAAGCTGAGAAGGTGGAATTTCACTGGGGCCACAGCAATGG CTCAGCGGGCTCTGAACACAGCATCAATGGCAGGAGGTTTCCTGTTGAGATGCAGATTTTCTTTTACAATCCAGATGAC TTTGACAGCTTTCAAACCGCAATTTCTGAGAACAGAATAATCGGAGCCATGGCCATATTTTTTCAAGTCAGTCCGAGGG ACAATTCTGCACTGGATCCTATTATCCACGGGTTGAAGGGTGTCGTACATCATGAGAAGGAGACCTTTCTGGATCCTTT CGTCCTCCGGGACCTCCTGCCTGCATCCCTGGGCAGCTATTATCGGTACACAGGTTCCTTGACCACACCACCGTGTAGC GAAATAGTGGAGTGGATAGTCTTCCGGAGACCCGTCCCCATCTCTTACCATCAGCTTGAGGCTTTTTATTCCATCTTCAC CACGGAGCAGCAAGACCATGTCAAGTCGGTGGAGTATCTGAGAAATAACTTTCGACCACAGCAGCGTCTGCATGACAG GGTGGTGTCCAAGTCCGCCGTCCGTGACTCCTGGAACCACGACATGACAGACTTCTTAGAAAACCCACTGGGGACAGA AGCCTCTAAAGTTTGCAGCTCTCCACCCATCCACATGAAGGTGCAGCCTCTGAACCAGACGGCACTGCAGGTGTCCTGG AGCCAGCCGGAGACTATCTACCACCCACCCATCATGAACTACATGATCTCCTACAGCTGGACCAAGAATGAGGACGAG AAGGAGAAGACGTTTACAAAGGACAGCGACAAAGACTTGAAAGCCACCATTAGCCATGTCTCACCCGATAGCCTTTAC CTGTTCCGAGTCCAGGCCGTGTGTCGGAACGACATGCGCAGCGACTTTAGCCAGACGATGCTGTTTCAAGCTAATACCA CTCGAATATTCCAAGGGACCAGAATAGTGAAAACAGGAGTGCCCACAGCGTCTCCTGCCTCTTCAGCCGACATGGCCC CCATCAGCTCGGGGTCTTCTACCTGGACGTCCTCTGGCATCCCATTCTCATTTGTTTCCATGGCAACTGGGATGGGCCCC TCCTCCAGTGGCAGCCAGGCCACAGTGGCCTCGGTGGTCACCAGCACGCTGCTCGCCGGCCTGGGGTTCGGCGGTGGT GGCATCTCCTCTTTCCCCAGCACTGTGTGGCCCACGCGCCTCCCGACGGCCGCCTCAGCCAGCAAGCAGGCGGCTAGGC CAGTCCTAGCCACCACAGAGGCCTTGGCTTCTCCAGGGCCCGATGGTGATTCGTCACCAACCAAGGACGGCGAGGGCA CCGAGGAAGGAGAGAAGGATGAGAAAAGCGAGAGTGAGGATGGGGAGCGGGAGCACGAGGAGGATGGAGAGAAGG ACTCCGAAAAGAAGGAGAAGAGTGGGGTGACCCACGCTGCCGAGGAGCGGAATCAGACGGAGCCCAGCCCCACACCC TCGTCTCCTAACAGGACTGCCGAGGGAGGGCATCAGACTATACCTGGGCATGAGCAGGATCACACTGCCGTCCCCACA GACCAGACGGGCGGAAGGAGGGATGCCGGCCCAGGCCTGGACCCCGACATGGTCACCTCCACCCAAGTGCCCCCCACC GCCACAGAGGAGCAGTATGCAGGGAGTGATCCCAAGAGGCCCGAAATGCCATCTAAAAAGCCTATGTCCCGCGGGGA CCGATTTTCTGAAGACAGCAGATTTATCACTGTTAATCCAGCGGAAAAAAACACCTCTGGAATGATAAGCCGCCCTGCT CCAGGGAGGATGGAGTGGATCATCCCTCTGATTGTGGTATCAGCCTTGACCTTCGTGTGCCTCATCCTTCTCATTGCTGT GCTCGTTTACTGGAGAGGGTGTAACAAAATAAAGTCCAAGGGCTTTCCCAGACGTTTCCGTGAAGTGCCTTCTTCTGGG GAGAGAGGAGAGAAGGGGAGCAGAAAATGTTTTCAGACTGCTCATTTCTATGTGGAAGACAGCAGTTCACCTCGAGTG GTCCCTAATGAAAGTATTCCTATTATTCCTATTCCGGATGACATGGAAGCCATTCCTGTCAAACAGTTTGTCAAACACAT CGGTGAGCTCTATTCTAATAACCAGCATGGGTTCTCTGAGGATTTTGAGGAAGTCCAGCGCTGTACTGCTGATATGAAC ATCACTGCAGAGCATTCCAATCATCCAGAAAACAAGCACAAAAACAGATACATCAACATTTTAGCATATGATCACAGT AGGGTGAAGTTAAGACCTTTACCAGGAAAAGACTCTAAGCACAGCGACTACATTAATGCAAACTATGTTGATGGTTAC AACAAAGCAAAAGCCTACATTGCCACCCAAGGACCTTTGAAGTCTACATTTGAAGATTTCTGGAGGATGATTTGGGAA CAAAACACTGGAATCATTGTGATGATTACGAACCTTGTGGAAAAAGGAAGACGAAAATGTGATCAGTATTGGCCAACA GAGAACAGTGAGGAATATGGAAACATTATTGTCACGCTGAAGAGCACAAAAATACATGCCTGCTACACTGTTCGTCGT TTTTCAATCAGAAATACAAAAGTGAAAAAGGGTCAGAAGGGAAATCCCAAGGGTCGTCAGAATGAAAGGGTAGTGAT CCAGTATCACTATACACAGTGGCCTGACATGGGAGTTCCCGAGTATGCCCTTCCAGTACTGACTTTCGTGAGGAGATCC TCAGCAGCTCGGATGCCAGAAACGGGCCCTGTGTTGGTGCACTGCAGTGCTGGTGTGGGCAGAACAGGCACCTATATT GTAATAGACAGCATGCTGCAACAGATAAAAGACAAAAGCACAGTTAACGTCCTGGGATTCCTGAAGCATATCAGGACA CAGCGTAACTACCTCGTCCAGACTGAGGAGCAGTACATTTTCATCCATGATGCCTTGTTGGAAGCCATTCTTGGAAAGG AGACTGAAGTATCTTCAAATCAGCTGCACAGCTATGTTAACAGCATCCTTATACCAGGAGTAGGAGGAAAGACACGAC TGGAAAAGCAATTCAAGCTGGTCACACAGTGTAATGCAAAATATGTGGAATGTTTCAGTGCTCAGAAAGAGTGTAACA AAGAAAAGAACAGAAACTCTTCAGTTGTGCCATCTGAGCGTGCTCGAGTGGGTCTTGCACCATTGCCTGGAATGAAAG GAACAGATTACATTAATGCTTCTTATATCATGGGCTATTATAGGAGCAATGAATTTATTATAACTCAGCATCCTCTGCCA CATACTACGAAAGATTTCTGGCGAATGATTTGGGATCATAACGCACAGATCATTGTCATGCTGCCAGACAACCAGAGCT TGGCAGAAGATGAGTTTGTGTACTGGCCAAGTCGAGAAGAATCCATGAACTGTGAGGCCTTTACCGTCACCCTTATCAG CAAAGACAGACTGTGCCTCTCTAATGAAGAACAAATTATCATCCATGACTTTATCCTTGAAGCTACACAGGATGACTAT GTCTTAGAAGTTCGGCACTTTCAGTGTCCCAAATGGCCTAACCCAGATGCCCCCATAAGTAGTACCTTTGAACTTATCA ACGTCATCAAGGAAGAGGCCTTAACAAGGGATGGTCCCACCATTGTTCATGATGAGTATGGAGCAGTTTCAGCAGGAA TGTTATGTGCCCTTACCACCCTGTCCCAGCAACTGGAGAATGAAAATGCTGTGGATGTTTTCCAGGTTGCAAAAATGAT CAATCTTATGAGGCCTGGAGTATTCACAGACATTGAACAATACCAGTTCATCTATAAAGCAAGGCTTAGCTTGGTCAGC ACTAAAGAAAATGGAAATGGTCCCATGACAGTAGACAAAAATGGTGCTGTTCTTATTGCAGATGAATCAGACCCTGCT GAGAGCATGGAGTCCCTAGTGTGACTGGAATCCTGAAAGGGCACTTAATTTGTAAACTTCTGAAGACTGAGAACTTTTT TGAGGCCTTTTTTGCCAGACTCTAGGTTATACAATAACCCAGTTACTTTTTTACACTGATAAAAGTTTTGATATTTATTTT TTGCCATTTTATGTCTTAATGGTATCCTACTGAGCATTTGCACCTCTGTTCATTTCACACAGTGAAACGCAATTTTACCTA GTTTGCACTATATGATCAGTGTTACTGCCTATAATCTTATACAACAGCAAACCCTGATGTGACATTCCATGAC+EcoRISite Figure 5.3

126

Figure 5.4. Identification and confirmation of successfully transfected cell lines by

RT-PCR.

1.5×105 viable cells from wild type MCF-7 cells and each single colony were plated in one well of 6-well plates with 5 ml of culture medium separately and cultured to ~ 85% confluence, total RNA was isolated for each group, and RT-PCR was performed.

Ethidium bromide-stained PCR products separated in a 1.5% agarose gel. 36B4 was used as internal standard. Primer 305 is located in Bovine Growth Hormone (BGH) reverse priming site in pCR®3.1 vector; Primer 333 is located in the sense strand of full-length

PTPγ cDNA insert; Primer 83 is located in T7 promoter/priming site in pCR®3.1 vector;

Primer 360 is located in the antisense strand of the partial PTPγ cDNA insert; Primer 334

(complementary to primer 360) is located in the sense strand of the partial PTPγ cDNA insert. W.T.: wild type MCF-7 cells.

127

Primer pair 333 & 305 (990 bp)

36B4 (563 bp)

M W

a

.

T r abcdeafbc de ghijklmnopqrs

k

.

e r M7-pCR3.1-800 (s) M7-PTPγ-800 (s)

Primer pair: 83&360 (583 bp) Primer pair: 83&334 (337 bp)

36B4 (563 bp)

M

W

P

N

M M

P

T

e

T

7

.

a a

T P

g

-

P r r a b c d e f g h i k l m n o p q r a b c d e f g i j k m

p

a

k γ k

.

γ

t

C

e e

s

i

a

r r

v

e

R

n M7-A24-800 (s) M7-B19-800 (s)

e

n

3

t

s

i

c

.

s

e

1

o

e

-

p

n

n

8

l

t

s

0 a

r

e

o

s

0

m

l

p

(

l

s

i

a

d

)

s

m

i

d

Figure 5.4

128

Figure 5.5. The expression of PTPγ mRNA in transfected MCF-7 cells.

1.5×105 viable cells from wild type MCF-7 cells, one of single colonies from mock transfected MCF-7 cells, and three of single colonies from M7-PTPγ-800 (s), M7-B19-

800 (s) or M7-A24-800 (s) were plated in one well of 6-well plates with 5 ml of culture

medium separately and cultured to ~ 85% confluence, total RNA was isolated for each

group, and RT-PCR was performed. Top: Ethidium bromide-stained PCR products

separated in a 1.5% agarose gel. Cells were not treated and total RNA was isolated from

each cell type separately. 36B4 was used as internal standard. Bottom: The mRNA ratios

of PTPγ to 36B4 as measured by densitometry.

Our results showed that M7-PTPγ-800 (s) had higher PTPγ mRNA expression than any

of wild type MCF-7, M7-pCR®3.1-800 (s), M7-A24-800 (s) and M7-B19-800 (s) cell

lines. M7-A24-800 (s) had lower PTPγ mRNA expression than any of wild type MCF-7,

M7-pCR®3.1-800 (s), M7-PTPγ-800 (s) and M7-B19-800 (s) cell lines.

129

M 650 bp 500 bp PTPγ (492 bp) 400 bp

650 bp 36B4 (563 bp) 500 bp

0.8

on 0.6 pressi 4) x

e 6B /3 NA γ P R 0.4 (PT ve m i

Relat 0.2

0.0 s ) efkabf hkj ) ) ) CF-7 cell 00 (s 0 (s 8 3.1-800-a (s γ- R P 19-800 (s 24-80 pe M C B A y p d t - 7- il 7 M7- M W M M7-PT

Figure 5.5

130

Figure 5.6. Immunohistochemical staining for PTPγ in transfected MCF-7 cells.

Cells were cultured on 24 × 30 mm cell culture cover slips overnight and washed in PBS for 3 times. After –10 °C methanol fixation for 5 minutes and air-dry, cells were stained for PTPγ by using VECTASTAIN Universal Quick Kit and DAB Substrate Kit. Brown staining represents PTPγ immunoreactivity; Blue staining represents nuclei. A. Wild type

MCF-7 cells; B. M7-pCR®3.1-800 (s); C. M7-B19-800 (s); D. M7-PTPγ-800 (s); E. M7-

A24-800 (s); F. Negative control.

These results showed that the PTPγ staining were weak and very similar among wild type

MCF-7 cell, M7-pCR®3.1-800 (s) and M7-B19-800 (s). Furthermore, the degree of staining was observably increased in M7-PTPγ-800 (s) when compared to others; however, the degree of staining was almost diminished in M7-A24-800 (s) when compared to others.

131

A B C

D E F

Figure 5.6

132

Figure 5.7. Effects of PTPγ on the proliferation of MCF-7 cells.

5000 cells / well were seeded in 24-well plates and cells were grown for 3 days and counted every 8 hours. Experiments were performed in four replicate wells, and each experiment was repeated twice. Cell doubling (CD) was calculated by using the formula ln (Nj-Ni)/ln 2 where Nj or Ni is the cell numbers at different time point Tj or Ti (Tj > Ti) in the growth log phase of the cells. Doubling time (DT) was consequently obtained by dividing the time interval (Tj > Ti) by CD.

The results were presented as the mean ± standard deviation (SD) for 4 replicate culture wells as one group. Analysis was performed using Minitab statistical software for

Windows (Minitab Inc., State College, PA, USA). Statistical differences were determined by using one-way ANOVA for independent groups. P-values of less than 0.05 were considered statistically significant.

These results showed that M7-PTPγ-800 (s) has the longest doubling time, whereas M7-

A24-800 (s) has the shortest doubling time, which suggested that PTPγ can inhibit MCF-

7 breast cancer cell growth and plays an important role in tumor suppression of breast cancer.

133

Wild type MCF-7 cells: T1/2=24 hours

M7-pCR3.1-800 (s): T1/2=24.3 hours M7-PTP -800 (s) T =40 hours γ : 1/2 M7-B19-800 (s): T1/2=23.8 hours M7-A24-800 (s): T =16.4 hours 1/2 2.75 Wild type MCF-7 cells 2.50 M7-pCR3.1-800 (s) M7-PTPγ-800 (s) 2.25 M7-B19-800 (s) M7-A24-800 (s) )

4 2.00 0 1

x 1.75 ( r e b 1.50

1.25 ll num e C 1.00 0.75

0.50

0 8 16 24 32 40 48 56 64 72 Hours

Figure 5.7

134

Figure 5.8. Optimization of cell numbers for anchorage-independent growth of

MCF-7 cells.

A. Colony formation of different MCF-7 cell numbers in soft agar. Cells were cultured in

6-well plates first covered with an agar layer (phenol red-free high-calcium DMEM/F12

with 0.5% agar and 10% FBS). The middle layer contained different number of cells

(2000, 4000, 8000 or 16000 cells) in phenol red-free DMEM/F12 with 0.35% agar and

10% FBS. The top layer, consisting of medium, was added to prevent drying of the agar

in the plates. The plates were incubated for 21 days and another 1 ml of fresh culture

medium was added to the top layer at day 11. After 21 days’ incubation, the plates were

stained in 0.5 ml of 0.005% crystal violet for >1 hour and the cultures were inspected and

photographed. B. Comparison of colony efficiency (CE) among different MCF-7 cell

numbers. CE was determined by a count of the number of colonies greater than 15 µm in

diameter, which was calculated as the average of colonies counted at 50× magnification in five individual fields by using BioQuant NOVA software. Bar, 500 µm.

The results showed that the colony number in soft agar was increased while the seeded cell number was increased for each cell type, M7-PTPγ-800 (s) formed much less and smaller colonies in soft agar than wild type MCF-7 cells and M7- pCR®3.1-800 (s) in the

same seeded cell number group.

135

A.

Wild type MCF-7

M7-pCR®3.1-800 (s)

M7-PTPγ-800 (s)

B. 70 )

2 Wild type MCF-7 cells

m 60 M7-pCR3.1-800 (s) M7-PTPγ-800 (s) 50 * P<0.05 N=4

40 * (colonies/3.3 m 30

fficiency 20 * E * 10 * Colony 0 2000 4000 8000 16000 Cell number

Figure 5.8

136

Figure 5.9. Effects of PTPγ on anchorage-independent growth of MCF-7 cells.

A. Colony formation of different MCF-7 cell types in soft agar. The procedure is the

same as described in Figure 5.8 except for the seeded cell number (8000 cells/well). B.

Comparison of colony efficiency (CE) among different MCF-7 cell types. CE is

determined as described in Figure 5.8. Bar, 500 µm.

The results were presented as the mean ± standard deviation (SD) for 5 individual fields

as one well. Analysis was performed using Minitab statistical software for Windows

(Minitab Inc., State College, PA, USA). Statistical differences were determined by using

one-way ANOVA for independent groups. P-values of less than 0.05 were considered

statistically significant.

Our results showed that that M7-PTPγ-800 (s) formed less and smaller colonies in soft

agar than wild type MCF-7 cells and M7- pCR®3.1-800 (s), but M7-A24-800 (s) formed more and much larger colonies in soft agar than wild type MCF-7 cells, M7- pCR®3.1-

800 (s) and M7-B19-800 (s).

137

) (s 7 0 (s) ) ) rol F- 80 0 (s (s nt C .1- 80 00 00 co M 3 γ- -8 -8 e pe CR TP 19 24 tiv ty -p -P -B -A ga ild 7 7 7 7 Ne W M M M M

60

] ]

2 2 m

m 50 *

m m

y y

3 3

. .

c c

3 3

n n /

/ 40

e e

) )

i i

c c

m m

i i

f f

µ µ

f f

5 5 E

E 30

1 1

y y

> >

( (

n n

o o

s s l

l 20 *

e e

o o

i i

C C

n n

o o

l l o

o 10

c c

[ [

0

Figure 5.9

138 Figure 5.10. Effects of PTPγ on E2’s and Z’s activities on MCF-7 cell proliferation.

4 2×10 cells were cultured in each well of 24-well plates and treated with the E2 or Z at 0,

7.5, 15, 30, 60 or 120 nM in phenol red-free high-calcium DMEM/F12 supplemented with DCC-stripped FBS (5%) for 24 hours. Cell proliferation rate was quantified by using

TM CellTiter 96 AQueous assay and optical density was read at 490nm (O.D.490nm) in 96-

well plates by an ELISA plate reader. Each point presented the mean ± standard deviation

(SD) of 4 replicate culture wells. * represents the significant difference from the control

group. A probability (P) of less than 0.05 was considered statistically significant

(P<0.05). A. Wild type MCF-7 cells; B. M7-pCR3.1-800 (s); C. M7-B19-800 (s); D.

M7-PTPγ-800 (s); E. M7-A24-800 (s)

The results were presented as the mean ± standard deviation (SD) for 4 replicate culture

wells as one group. Analysis was performed using Minitab statistical software for

Windows (Minitab Inc., State College, PA, USA). Statistical differences were determined

by using one-way ANOVA for independent groups. P-values of less than 0.05 were

considered statistically significant.

Our results showed that both Z and E2 could stimulate the cell proliferation of all tested cell types in a dose-dependent manner, and the potency of Z and E2 are very similar.

Most interestingly, the degree of Z’s or E2’s stimulation on M7-PTPγ-800 (s) cell

proliferation is much less than that on the cell proliferation of other cell types; however, the degree of Z’s or E2’s stimulation on M7-A24-800 (s) cell proliferation is much greater

than that on the cell proliferation of other cell types.

139

A.

3.00

2.75 E2 Z 2.50 N = 4 (well) 2.25 * P < 0.05 2.00 * ll 1.75

/ we 1.50 . * D. D . 1.25 * O 3.00 1.00 2.75 E2 * Z 0.75 2.50 N = 4 (well) 0.50 2.25 * P < 0.05 0.25 2.00 l 0.00 l 1.75 0.0 7.5 15.0 30.0 60.0 120.0 e

Concentrations (nM) / w 1.50 D. 1.25 O.

B. 1.00

3.00 0.75 * 2.75 E2 0.50 * * Z 2.50 0.25 N = 4 (well) 2.25 * P < 0.05 0.00 0.0 7.5 15.0 30.0 60.0 120.0 2.00 Concentrations (nM)

ll 1.75 * e *

/ w 1.50 .

D E. 1.25 O. * 3.00 1.00 * * 2.75 E2 0.75 Z * 2.50 0.50 N = 4 (well) 2.25 * P < 0.05 0.25 * 2.00 0.00

0.0 7.5 15.0 30.0 60.0 120.0 ll 1.75 e Concentrations (nM)

/ w 1.50 .

D * 1.25 O.

C. 1.00 3.00 0.75 2.75 E 2 0.50 Z 2.50 0.25 N = 4 (well) 2.25 * P < 0.05 0.00 2.00 0.0 7.5 15.0 30.0 60.0 120.0 Concentrations (nM) ll 1.75 e * *

/ w 1.50 . D 1.25 * O.

1.00 * 0.75

0.50

0.25

0.00 0.0 7.5 15.0 30.0 60.0 120.0 Concentrations (nM)

Figure 5.10

140 CHAPTER 6

CONCLUSIONS

PERSPECTIVES

Human health concerns: estrogenic activities of Zeranol (Z) in human breast

Epidemiological evidence of human breast cancer indicates that white American women have a 5-fold greater risk for human breast cancer than Asian women in China and Japan. Furthermore, the risk of acquiring breast cancer among Asian women immigrants in the U.S. approaches that of American women after 1 to 2 generations

(Kelsey and Berkowitz, 1988). These observations indicate a strong environmental, rather than genetic, component in the etiology of this disease (Buell, 1973). It has been speculated that dietary factors may contribute to this ethnic difference in human breast cancer incidence (Committee on Diet, Nutrition and Cancer, 1982). A recent study associated an increase in the risk for breast cancer with elevated red meat consumption, while fat, protein, dairy product, poultry or fish consumption was modestly or not at all associated with the incidence of the disease (Toniolo et al., 1994). In light of these

141 findings and the decreasing support for the association of dietary fat with breast cancer

(Toniolo et al., 1994; Lund, 1994), perhaps another component(s) within red meat may be involved in the postulated role of diet in breast cancer etiology. It is our concern that this component may be the estrogenic anabolic growth promotant used widely by the meat- producing industry.

As we have known, Zeranol (Z) (Ralgro), which is marketed as Ralgro

(Schering-Plough Corp, Kenilworth, NJ 07033), is synthesized from the mycotoxin

Zearalenone (ZL, a nonsteroidal, resorcyclic acid lactone compound produced by fungi of the genus Fusarium) and is a nonsteroidal agent with estrogenic activities that is used as a growth promoter in the U.S. beef, veal and lamb industries to accelerate weight gain, improve feed conversion efficiency and increase the lean meat-to-fat ratio. In recent years, the occurrence of compounds possessing estrogenic activities in the environment and in food products, either as natural constituents or as contaminants, has received increasing attention because of speculation concerning their ability to adversely affect human and animal endocrinology and their possible etiological role in estrogen-related carcinogenesis.

During the preparation of this dissertation, we have demonstrated that Z can induce human breast epithelial cell neoplastic transformation and has the same effects on some estrogen-targeted genes with the similar potency as E2 in our tested dose range. One possible explanation is that, since most exogenous hormone-like chemicals, including Z and the other synthetic growth promoting hormones, exhibit limited or no binding to the carrier proteins, the potential potency of Z is much larger than its actual concentrations suggested (up to 50 times) [Mastri et al., 1985; Shrimanker et al., 1985; Nagel et al.,

142 1998]. So, based on our current findings which provides a putative link between Z and

risk of human breast cancer, our concern is that the consumption of beef products,

derived from Z-implanted beef cattle may pass a potential health impact on human

consumers, particularly with respect to reproductive endocrinology and hormone-

sensitive organs, such as breasts.

PTPγ acts as an estrogen-targeted gene in human breast

Previous results, particularly our preliminary and published work as described in

this dissertation, indicate an intriguing relationship among cancer, estrogen and PTPγ.

We speculate that, in the human breast, PTPγ is an estrogen-regulated gene that possesses cancer suppressor activity, and estrogen-induced suppression of

PTPγ expression plays a role in breast tumorigenesis. Clearly, defining the mechanism(s)

of estrogenic regulation of PTPγ expression and elucidating its role in breast cancer

induction and progression is important. Such research could reveal novel mechanisms of

hormonal carcinogenesis, regulation of cancer cell growth and endocrine disruption by

environmental estrogens. As described in chapter 4 in this dissertation, we found that

downregulation of PTPγ expression by both E2 and Z is associated with ERα, which indicates this downregulation is through estrogen receptor-mediated pathway. However, there have been no reports to date of the presence of a functional estrogen response element (ERE) in the PTPγ gene, and this promotes us to hypothesize that there must be

some regulatory sequences responsible for estrogenic effects of E2 and Z on PTPγ

expression, which are not caused by the direct interaction of ligand-ER complex with an

ERE in the promoter region of the PTPγ gene in human breast. So, in the future, we need 143 to search for such regulatory sequences. Sequence analysis of the promoter of the human

PTPγ gene and evaluation of transcriptional activation through a series of truncated

PTPγ promoter-reporter constructs will allow us to identify such important regulatory domains within the 5’-flanking sequence of the PTPγ gene.

PTPγ inhibits cell proliferation and suppresses anchorage-independent growth of

human breast cancer cells

Previous results showed that alterations in PTPs activity might affect cell growth,

neoplastic processes and transformation [Gaits et al., 1995]. PTPγ has been implicated as

a tumor suppressor gene in kidney and lung cancers [Lubinski et al., 1994; van Niekerk et

al., 1999]. As described in chapter 5 in this dissertation, we examined the effect of PTPγ on the growth of human breast cancer and compare the estrogenic responses of human breast cells with different expression levels of PTPγ. We found that PTPγ overexpressing

MCF-7 cells normally show the characteristics with slow-growing and lower colony

efficiency on soft agar, and show anti-estrogenic effects on human breast cell

proliferation. In combination with our previous results, all these experimental data

support our suggested concept that PTPγ is a potential estrogen-regulated breast cancer

suppressor gene in vitro. In the future, in vivo experiments need to be done to further

support our suggested concept.

This molecular biomarker will be a powerful tool for investigating the estrogen

and/or nonsteroidal estrogenic agents in controlling the etiological process of

tumorigenesis in human breast and rodent mammary. It was claimed that the reduction of

144 PTPγ expression in human breast cancer is unlikely to be due to genetic events; therefore, epigenetic mechanisms (such as methylation) might be responsible.

FUTURE WORKS AND CLINICAL IMPLICATIONS

Mechanism(s) of estrogenic regulation of PTPγ expression in human breast –

Characterization and analysis of the human PTPγ promoter region

The work in this dissertation has shown that PTPγ can be downregulated by

estrogenically active agents E2 and Z, which is associated with ERα. These results suggested that this regulation is through estrogen receptor-mediated pathway. However, we couldn’t find any EREs in PTPγ promoter region, and it is not likely that E2 and Z

downregulated PTPγ expression by direct interaction of ligand-ER complex and EREs in

the human PTPγ gene. Future goals will focus on further characterization and analysis of

the human PTPγ promoter region by scanning of the promoter region with different

constructs of PTPγ: plasmid and look for the important elements in the

promoter region responsible for the estrogenic regulation of PTPγ expression. This work

will allow us to identify the possible molecules and/pathways involved in the estrogenic

regulation of PTPγ by E2 and Z in human breast and define the mechanisms for this

regulation.

145 Investigations of the anti-tumorigenicity of PTPγ in nude mice

During this dissertation, in combination with our previous results, we have

showed that there is an intriguing relationship among cancer, estrogen and PTPγ. All

these experimental data suggested that PTPγ is a potential estrogen-regulated breast

cancer suppressor gene, at least in vitro. If similar results can be reproduced in nude

mouse testing model and we can show that PTPγ overexpression can inhibit mammary

gland tumor growth in nude mice, then, taken together, these results indicate that PTPγ

overexpression can suppress the development of human breast cancer both in vitro and in

vivo, and we can further support our concept that PTPγ is a potential estrogen-regulated

breast cancer suppressor gene and the presence of this potential cancer suppressor gene.

Clinical relevance of PTPγ in human breast cancer patients

Based on these results, we conclude that there will be some clinical relevance for

PTPγ. PTPγ might be able to serve as a molecular biomarker for diagnosis or prognosis for breast cancer patient treatment. We expect that PTPγ protein will significantly prolong the survivability of human breast cancer patients by inhibiting or slow down the proliferative ability of human breast cancer cells. The signaling pathway with PTPγ involvement may provide a new target for cancer prevention and treatment. In the future,

PTPγ might be able to be applied in clinical trials for cancer prevention and cancer therapy.

146 CONCLUDING REMARKS

The studies in this dissertation were carried out with the intention to claim the estrogenic activities of Zeranol in human breast neoplastic transformation and identify a new potential tumor suppressor gene in human breast, which help us to understand some basic mechanisms regarding to the effects of environmental disruptors on human breast and contribute to our overall understanding of the tumor suppressing activities of PTPγ in human breast. But, there are still many questions to be answered with more investigations in the future. The underlying goal of this research was to provide rationale for the prevention and treatment of human breast cancer with some novel target sites. Thus, with this knowledge, we hope that it should be possible to design new therapies targeted to this novel tumor suppressor and lead to the improved treatment of human breast cancer, and hope that these studies can be used as a basis for the similar investigations in other human cancers.

147 LIST OF REFERENCES

Aaronson SA, Bottaro DP, Miki T, Ron D, Finch PW, Fleming TP, Ahn J, Taylor WG, and Rubin JS, 1991. Keratinocyte growth factor, a fibroblast growth factor family member with unusual target cell specificity. Ann. NY Acad. Sci., 638, 62-77.

Akiyama T, Ishida J, Nakagawa, S, Ogawara H, Watanabe S, Itoh N, Shibuya M, and Fukami Y, 1987. Genistein, a specific inhibitor of tyrosine-specific protein kinases. Journal of Biological Chemistry, 262, 5592-5595.

American Cancer Society, 2001. Breast Cancer Facts and Figures, 2001-2002. 01-50M- No. 8610.01-R.

Ashburn SP, Han X, Liehr JG, 1993. Microsomal hydroxylation of 2- and 4- fluoroestradiol to catechol metabolites and their conversion to methyl ethers: catechol estrogens as possible mediators of hormonal carcinogenesis. Mol Pharmacol, 43, 534- 541.

Ball P, Knuppen R, 1980. Catecholestrogens (2- and 4-hydroxy-oestrogens). Chemistry, biosynthesis, metabolism, occurrence and physiological significance, Acta. Endocrinol. (Copenh), 232 (Suppl. 1), 127.

Barford D, Jia Z, and Tonks NK, 1995. Protein tyrosine phosphatases take off. Nature Struct. Biol., 2, 1043–1053.

Barnea G, Silvennoinen O, Shaanan B, Honegger AM, Canoll PD, D'Eustachio P, Morse B, Levy JB, LaForgia S, Huebner K, Musacchio JM, Sap J and Schlessinger J. 1993. Identification of a carbonic anhydrase-like domain in the extracellular region of RPTPγ defines a new subfamily of receptor tyrosine phosphatase. Mol Cell Biol, 13, 1497-1506.

Bernstein L, Rose Rk, 1993. Endogenous hormones and breast cancer risk. Epidemiol Rev, 15, 48-65.

Berry M, Nunez AM, and Chambon P, 1989. Estrogen-responsive element of the human pS2 gene is an imperfectly palindromic sequence. Proceedings of the National Academy of Sciences of the United States of America, 86, 1218-1222.

148 Bishop JM, 1985. The search for genetics damage in neoplastic cells has become a central theme of contemporary cancer research. Trends. Genet., 1, 245-249.

Brady-Kalnay SM, and Tonks NK, 1995. Protein tyrosine phosphatases as adhesion receptors. Curr Opin Cell Biol, 7, 650–657.

Buell P, 1973. Changing incidence of breast cancer in Japanese-American women. J. Natl. Cancer Inst, 51, 1479.

Carroll KK, 1991. Review of clinical studies on cholesterol-lowering response to soy protein. Journal of the American Dietetic Association, 91, 820-827.

Clark R, Dickson RB, Lippman ME, 1992. Hormonal aspects of breast cancer. Growth factors, drugs and stromal interactions. Crit Rev Oncol Hematol, 12, 1-23.

Code of Federal Regulations. 1991. Title 21, Sections 522 and 556, U.S. Government Printing Office, Washington, D.C., pp. 224 and 452.

Coe JE, Ishak KG, Ward JM, and Ross MJ, 1992. Tamoxifen prevents induction of hepatic neoplasia by zeranol, an estrogenic food contaminant. Proceedings of the National Academy of Sciences of the United States of America, 89, 1085-1089.

Colditz GA, 1996. The benefits of hormone replacement therapy do not outweigh the increased risk of breast cancer. J NIH Res, 8, 41-44.

Collins BM, McLachlan JA, Arnold SF, 1997. The estrogenic and antiestrogenic activities of phytochemicals with the human estrogen receptor expressed in yeast. , 62 (4): 365-372.

Committee on Diet, Nutrition and Cancer, Assembly of Sciences, National Research Council, 1982. Diet, Nutrition and Cancer, Washington DC, National Academy Press.

Couse JF, Hewitt SC, Bunch DO, Sar M, Walker VR, Davis BJ, Korach KS, 1999. Postnatal sex reversal of the ovaries in mice lacking estrogen receptors alpha and beta. Science, 286, 2328-2331.

Cowley SM, Hoare S, Mosselman S, Parker MG, 1997. Estrogen receptors alpha and beta form heterodimers on DNA. J Biol Chem, 272, 19858-19862.

Cullen KJ, Lippman ME. 1991. Stromal-epithelial interactions in breast cancer. In: Dickson RB and Lippman ME (eds.), Gene, oncogenes, and hormones: Advances in cellular and molecular biology of breast cancer. Kluwer Academic Publisher, Boston. pp. 413-431.

Cunha GR, 1994. Role of mesenchymal-epithelial interactions in normal and abnormal development of the mammary gland and prostate. Cancer (suppl.), 74, 1030-1044. 149

Dodge JA, Glasebrook AL, Magee DE, Phillips DL, Sato M, Short LL, and Bryant HU, 1996. Environmental estrogens: effects on cholesterol lowering and bone in the ovariectomized rat. Journal of Steroid Biochemistry & Molecular Biology, 59, 155-161.

Dong-LeBourhis X, Berthois Y, Millot G, Degeorges A, Sylv M, Martin P, Calvo F, 1997. Effect of stromal and epithelial cells derived from normal and tumorous breast tissue on the proliferation of human breast cancer cell lines in co-culture. Int J Cancer, 71, 42-48.

Donjacour AA, Cunha GR, 1991. Stromal regulation of epithelial function. Kluwer Academic Publishers, Boston, 335-364.

Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M, 2000. Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes. Development, 127, 4277-4291.

Fotsis T, Pepper M, Adlercreutz H, Fleischmann G, Hase T, Montesano R, and Schweigerer L, 1993. Genistein, a dietary-derived inhibitor of in vitro angiogenesis. Proceedings of the National Academy of Sciences of the United States of America, 90, 2690-2694.

Freiss G, Vignon F, 1994. increase protein tyrosine phosphatase activity in human breast cancer cells. Mol Endocrinol, 8, 1389-96.

Gaits F, Li RY, Ragab A, Ragab-Thomas JMF, and Chap H, 1995. Increase in receptor- like protein tyrosine phophatase activity and expression level on density-dependent growth arrest of endothelial cells. Biochem J, 311, 97-103.

Gaub MP, Bellard M, Scheuer I, Chambon P, Sassone-Corse P, 1990. Activation of the gene by the estrogen receptor involves the fos-jun complex. Cell, 63, 1267- 1276.

Ghoussoub RA, Dillon DA, D’Aquila T, Rimm EB, Fearon ER, Rimm DL, 1998. Expression of c-met is a strong independent prognostic factor in breast carcinoma. Cancer, 82, 1513-1520.

Gupta RR, Sen S, Diepenhorst LL, Rudick CN, Maren S, 2001. Estrogen modulates sexually dimorphic contextual fear conditioning and hippocampal long-term potentiation (LTP) in rats(1). Brain Res, 12, 356-365.

Henderson BE, Rose R, Bernstein L, 1988. Estrogen as a cause of human cancer. Cancer Res, 48, 246-253.

Hennipman A, van Oirschot BA, Smits J, Rijksen G, Staal GEJ, 1989. Tyrosine kinase 150 activity in breast cancer, benign breast disease, and normal breast tissue. Cancer Res, 49, 516-521.

Hilakivi-Clarke L, 2000. Estrogens, BRCA1, and breast cancer. Cancer Res, 60, 4993- 5001.

Hofland LJ, Van der Burg B, Van Eijck CH, Sprij DM, Van Koetsveld PM, and Lamberts SW, 1995. Role of tumor-derived fibroblasts in the growth of primary cultures of human breast-cancer cells: effects of epidermal growth factor and the somatostatin analogue octreotide. Int J Cancer, 60, 93-99.

Hu Y-F, 1995. Human breast cancer and DES-inducible hamster kidney tumor: cellular and molecular analysis of hormonal carcinogenesis and chemoprevention. Ph.D dissertation, The Ohio State University. pp.76-108.

Hu YF, Lau KM, Ho SM, Russo J, 1998. Increased expression of estrogen receptor β in chemically transformed human breast epithelial cells. International Journal of Oncology, 12, 1225-1228.

Hunter T, Cooper JA, 1980. Protein tyrosine kinase. Proc Natl Acad Sci USA , 77(3), 1311-1315.

Irshaid F, Kulp SK, Sugimoto Y, Lee K, Lin YC, 1999. Zeranol stimulates estrogen- regulated gene expression on MCF-7 human breast cancer cells and normal human breast epithelial cells. Biol Reprod, 60 (Suppl 1), 234-235.

Kaplan R, Morse B, Huebner K, Croce C, Ravera M, Ricca G, Jaye M, and Schlessinger J, 1990. Cloning of three novel human tyrosine phosphatases reveals the existence of a multigene family of receptor-linked protein tyrosine phosphatases. Proc Natl Acad Sci USA, 87, 7000-7004.

Katzenellenbogen BS, Tsai TS, Tatee T, and Katzenellenbogen JA, 1979. Estrogen and antiestrogen action: studies in reproductive target tissues and tumors. Advances in Experimental Medicine & Biology, 117, 111-132.

Katzenellenbogen BS, Kendra KL, Norman MJ, Berthois Y, 1987. Proliferation, hormonal responsiveness, and estrogen receptor content of MCF-7 human breast cancer cells grown in the short-term and long-term absence of estrogens. Cancer Res., 47, 4355-4360.

Katzenellenbogen BS, Korach KS, 1997. A new actor in the estrogen receptor drama- enter Erb. Endocrinology, 138, 861-862.

Kellis JT, Jr, and Vickery LE, 1984. Inhibition of human estrogen synthetase (aromatase) by flavones. Science, 225, 1032-1034.

151 Kelsey JL, Berkowitz GS, 1988. Breast cancer epidemiology. Cancer Res, 48, 5615.

Kelsey JL, Gammon MD, John EM, 1993. Reproductive factors and breast cancer. Epidemiol Rev, 15 (1), 36-47.

Klarlund JK, 1985. Transformation of cells by an inhibitor of phosphatases acting on phosphotyrosine in proteins. Cell, 41(3), 707-17.

Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, Sar M, Korach KS, Gustafsson JA, Smithies O, 1998. Generation and reproductive phenotypes of mice lacking . Proc Natl Acad Sci USA, 95, 15677-15682.

Krop IE, Sgroi D, Porter DA, Lunetta KL, LeVangie R, Seth P, Kaelin CM, Rhei E, Bosenberg M, Schnitt S, Marks JR, Pagon Z, Belina D, Razumovic J, and Polyak K, 2001. HIN-1, a putative highly expressed in normal but not cancerous mammary epithelial cells. Proc Natl Acad Sci USA, 98, 9796-9801.

Krueger NX, Streuli M, and Saito H. 1990. Structural diversity and of human receptor-like protein tyrosine phosphatases. EMBO J, 9, 3241-3252.

Krueger NX and Saito H, 1992. A human transmembrane protein tyrosine phosphatase, PTPζ, is expressed in brain and has N-terminal receptor domain homologous to carbonic anhydrases. Proc Natl Acad Sci USA, 8, 7417-7421.

Kulp SK, Liu S, Sugimoto Y, Brueggemeier RW and Lin YC, 2000. Localization of protein tyrosine phosphatase γ (PTPγ) to the mammary epithelium of ACI rats. Biol Reprod, 62(suppl 1), 181-182.

Kumar V, Chambon P, 1988. The estrogen receptor binds tightly to its responsive element as a ligand-induced homodimer. Cell, 55, 145-156.

Kuiper GG, Enmark E, Pelto-HM, Nilsson S, Gustafsson JA, 1996. Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA, 93, 5925- 5930.

Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA. 1997. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptor α and β. Endocrinology, 138 (3), 863-870.

Kuiper GG, Lemmen JG, Carlsson B, Corton JC, Safe SH, van der Saag PT, van der Burg B, and Gustafsson JA, 1998. Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology, 139, 4252-4263.

Laborda J, 1991. 36B4 cDNA used as an estradiol-independent mRNA control is the cDNA fro human acidic ribosomal phosphoprotein PO. Nucl Acids Res, 19, 3998.

152 LaForgia S, Morse B, Levy J, Barnea G, Cannizzaro LA, Li F, Nowell PC, Boghosian- sell L, Glick J, Weston A, Harris CC, Drabkin H, Patterson D, Crose CM, Schlessinger J, and Huebner K, 1991. Receptor protein-tyrosine phosphatase gamma is a candidate tumor suppressor gene at human chromosome region 3p21. Proc. Natl Acad Sci USA, 88 (11), 5036-5040.

LaForgia S, Lasota J, Latif F, Boghosian-sell L, Kastury K, Ohta M, Druck T, Atchison L, Cannizzaro L, Barnea G, Schlessinger J, Modi W, Kuzmin I, Tory K, Zbar B, Croce CM, Lerman M, and Huebner K, 1993. Detailed genetic and physical map of 3p chromosome region surrounding the familial RCC chromosome translocation, t(3;8)(p14.2;q24.1). Cancer Res., 53, 3118-3124.

Lazennec G, Bresson D, Lucas A, Chauveau C, and Vignon F, 2001. ERß Inhibits proliferation and invasion of breast cancer cells. Endocrinology, 142 (9), 4120-4130.

Lee HP, Gourley L, Duffy SW, Esteve J, Lee J, and Day NE, 1991. Dietary effects on breast-cancer risk in Singapore. [comment]. Lancet, 337, 1197-1200.

Leffers H, Nasby M, Vendelbo B, Skakkebak NE, and Jorgensen M, 2001. Oestrogenic potencies of Zeranol, oestradiol, diethylstilboestrol, bisphenol-A and genistein: implications for exposure assessment of potential endocrine disrupters. Human reproduction, 16 (5), 1037-1045.

Levy JB, Canoll PD, Silvennoinen O, Barnea G, Morse B, Honegger AM, Huang J-T, Cannizzaro LA, Park S-H, Druck T, Huebner K, Sap J, Ehrlich M, Musacchio JM, and Schlessinger J, 1993. The cloning of a receptor-type protein tyrosine phosphatase expressed in the central nervous system. J Biol Chem, 268,10573-10581.

Li JJ, Li SA, 1987. Estrogen carcinogenesis in syrian hamster tissues: Role of metabolism. Fed Proc, 46, 1858-1863.

Liehr JG, Ulubelen AA, Strobel HW, 1986. Cytochrome P-450-mediated redox cycling of estrogens. J Biol Chem, 261, 16865-16870.

Lin YC, Chang CJG, Sugimoto Y, Chen R, Canatan H, Brueggemeier RW, and Dayton MA, 1994. Detection of protein tyrosine phosphatase γ gene in hamster kidney. Proceedings of the American Association for Cancer Research Eighty-fifth Annual Meeting, 35, 607a (Abstract #3619).

Lin YC, Mulla Z, Kulp SK, Sugimoto Y, Farrar WB, Brueggemeier RW, 1996. Biological activity in serum and meat of Zeranol-implanted beef cattle: Regulation of proliferation and estrogen-induced gene expression in normal breast cells and MCF-7 cells. Proceedings of 10th international congress of endocrinology, P2-798.

Lin YC, Kulp SK, Sugimoto Y, Brueggemeier RW, 2000. Potential risk of growth promoter in beef for breast cancer growth. Era of Hope, Department of Defense Breast 153 Cancer Research Program Meeting Proceedings, II, 480.

Lippman ME, Dickson RB, 1989. Mechanisms of growth control in normal and malignant breast epithelium. Rec Prog Horm Res, 45, 383-440.

Liu S, Kulp SK, Sugimoto Y, Jiang J, Chang HL, Lin YC, 2002a. Involvement of breast epithelial-stromal interactions in the regulation of protein tyrosine phosphatase-γ (PTPγ) mRNA expression by estrogenically active agents. Breast Can Res and Treatment 71: 21- 35.

Liu S, Sugimoto Y, Kulp SK, Jiang J, Chang HL, Park KY, Kashida Y, Lin YC, 2002b. Estrogenic downregulation of Protein Tyrosine Phosphatase γ (PTPγ) in human breast is associated with Estrogen Receptor α. Anticancer Res, 22, 3917-3924.

Lubahn DB, Moyer JS, Golding TS, Couse JF, Korach KS, Smithies O, 1993. Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc Natl Acad Sci USA, 90, 11162-11166.

Lubinski J, Hadaczek P, Podolski J, Toloczko A, Sikorski A, McCue P, Druck T and Huebner K, 1994. Common regions of deletion in chromosome regions 3p12 and 3p14.2 in primary clear cell renal carcinomas. Cancer Res, 54, 3710-3713.

Lund E, 1994. Editorial: The research tide ebbs for the dietary fat hypothesis in breast cancer. Epidemiology, 5, 387.

Luttrell LM, and Lefkowitz RJ, 1995. and G beta gamma MAP kinase activation by a common signaling pathway. Nature, 376, 781-784.

Macgregor JI, and Jordan VC, 1998. Basic guide to the mechanisms of antiestrogen action. Pharmacological Reviews, 50 (2), 151-196.

Makela S, Poutanen M, Lehtimaki J, Kostian ML, Santti R, and Vihko R, 1995. Estrogen-specific 17beta-hydroxysteroid oxidoreductase type 1 (E.C. 1.1.1.62) as a possible target for the action of phytoestrogens. Proceedings of the Society for Experimental Biology & Medicine, 208, 51-59.

Masiakowski P, Breathnach R, Bloch J, Gannon F, Krust A, Chambon P, 1982. Cloning of cDNA sequences of hormone-regulated genes from the MCF-7 human breast cancer cell lines. Nucl Acids Res, 10, 7895-7903.

Mastri C, Mistry P, and Lucier GW, 1985. In vivo oestrogenicity and binding characteristics of alpha-zearalanol (P-1496) to different classes of oestrogen binding proteins in rat liver. Journal of Steroid Biochemistry, 23, 279-289.

Miksicek RJ, 1995. Estrogenic flavonoids: structural requirements for biological activity. Proceedings of the Society for Experimental Biology & Medicine, 208, 44-50. 154

Mishra S, Hamburger AW, 1993. O-phospho-L-tyrosine inhibits cellular growth by activating protein tyrosine phosphatases. Cancer Res, 53, 557-563.

Moran C, Quirke JF, Prendiville DJ, Bourke S, and Roche JF, 1991. The effect of estradiol, trenbolone acetate, or zeranol on growth rate, mammary development, carcass traits, and plasma estradiol concentrations of beef heifers. Journal of Animal Science, 69, 4249-4258.

Nagel SC, vom Saal FS, and Welshons WV, 1998. The effective free fraction of estradiol and xenoestrogens in human serum measured by whole cell uptake assays: physiology of delivery modifies estrogenic activity. Proceedings of the Society for Experimental Biology & Medicine, 217, 300-309.

Ogawa S, Inoue S, Watanabe T, Hiroi H, Orimo A, Hosoi T, Ouchi Y, Muramatsu M, 1998. The complete primary structure of human estrogen receptor beta (hER beta) and its heterodimerization with ER alpha in vivo and in vitro. Biochem Biophys Res Commun, 243, 122-126.

Ogawa, S, Chester, AE, Hewitt SC, Walker VR, Gustafsson JA, Smithies O, Korach KS, Pfaff DW, 2000. From the cover: abolition of male sexual behaviors in mice lacking estrogen receptors alpha and beta (alpha beta ERKO). Proc Natl Acad Sci USA, 97, 14737-41471.

Pace P, Taylor J, Suntharalingam S, Coombes RC, Ali S, 1997. Human estrogen receptor beta binds DNA in a manner similar to and dimerizes with . J Biol Chem, 272, 25832-25838.

Paech K, Webb P, Kuiper GGJM, Nilsson S, Gustafsson JA, Kushner PJ, Scanlan TS, 1997. Differential ligand activation of estrogen receptors ERα and ERβ at AP1 sites. Science, 277, 1508-1510.

Peterson G, 1995. Evaluation of the biochemical targets of genistein in tumor cells. Journal of Nutrition, 125, 784S-789S.

Pham TA, Elliston JF, Nawaz Z, McDonnell DP, Tsai MJ, O’Malley BW, 1991. Antiestrogens can establish nonproductive receptor complexes and alter chromatin structure at target enhancers. Proc Natl Acad Sci USA, 88, 3125-3129.

Pink JJ, and Jordan VC, 1996. Models of estrogen receptor regulation by estrogens and antiestrogens in breast cancer cell lines. Cancer Res, 56, 2321-2330.

Poliseno L, Mariani L, Collecchi P, Piras A, Zaccaro L, Rainaldi G, 2002. Bcl2-negative MCF7 cells overexpress p53: implications for the cell cycle and sensitivity to cytotoxic drugs. Cancer Chemotherapy and Pharmacology, 50 (2), 127-130.

155 Ross JS, and Fletcher JA, 1999. The HER-2/neu oncogene: prognostic factor, predictive factor and target for therapy. Semin Cancer Biol, 9 (2), 125-138.

Roy D, and Liehr JG, 1988. Temporary decrease in renal quinone and activity induced by chronic administration of estradiol to male Syrian hamsters-increased superoxide formation by redox cycling of estrogen. J Biol Chem, 263, 3646-3651.

Russo J, Hasan Lareef M, Tahin Q, Hu YF, Slater C, Ao X, Russo IH, 2002. 17β- Estradiol is carcinogenic in human breast epithelial cells. The Journal of Steroid Biochemistry and Molecular Biology, 80 (2), 149-162.

Schaapveld R, Wieringa B, Hendriks W, 1997. Receptor-like protein tyrosine phosphatases: alike and yet so different. Molecular Biology Reports, 24, 247-262.

Schumann G, Fiebich BL, Menzel D, Hull M, Butcher R, Nielsen P, Bauer J. 1998. Cytokine-induced transcription of protein-tyrosine-phosphatases in human astrocytoma cells. Molecular brain research, 62, 56-64.

Sheffield LG, and Welsch CW, 1985. Zeranol (beta-resorcylic acid lactone), a common residous component of natural foodstuffs, stimulates developmental growth of the mouse mammary gland. Cancer Letters, 28, 77-83.

Shock LP, Bare DJ, Klinz PF, and Maness PF, 1995. Protein tyrosine phosphotases expressed in developing brain and retinal Muller glia. Mol Brain Res, 28, 110-116.

Shrimanker K, Salter LJ, and Patterson RL, 1985. Binding of steroid hormones and anabolic agents to bovine sex-hormone binding globulin. Hormone & Metabolic Research, 17, 454-457.

Sorio C, Mendrola Jeannine, Lou Z, Laforgia S, Croce CM, and Huebner K, 1995. Characterization of the receptor protein tyrosine phosphatase gene product PTPg: binding and activation by triphosphorylated necleosides. Cancer Res, 55, 4855-4864.

Sorio C, Melotti P, D'Arcangelo D, Mendrola J, Calabretta B, Croce CM, and Huebner K 1997. Receptor protein tyrosine phosphatase gamma, Ptpg, regulates hematopoietic differentiation. Blood, 90(1), 49-57.

Stamenkovic I, Sgroi D, Aruffo A, Sy MS, Anderson T, 1991. The B lymphocyte adhesion molecule CD22 interacts with leukocyte common antigen CD45RO on T cells and α2-6 sialotransferase, CD75, on B cells. Cell, 66, 1133-1144.

Sun H, and Tonks NK, 1994. The coordinated action of protein tyrosine phosphatases and kinases in cell signaling. Trends Biochem Sci, 19, 480–485.

Tennant JR, 1964. Evaluation of the trypan blue technique for determination of cell viability. Transplantation, 2, 685-694. 156

Tonks NK, Diltz CD, and Fischer EH, 1988. Purification of the major protein-tyrosine- phosphatases of human placenta. J Biol Chem, 263, 6722–6730.

Toniolo P, Riboli E, Shore RE, and Pastemack BS, 1994. Consumption of meat, animal products, protein, A fat and risk of breast cancer: a prospective cohort study in New York. Epidemiology, 5, 391.

Tremblay GB, Tremblay A, Copeland NG, Gilbert DJ, Jenkins NA, Labrie F, and Giguere V, 1997. Cloning, chromosomal localization, and functional analysis of the murine estrogen receptor beta. Mol Endocrinol, 11, 353-365.

Tsukamoto T, Takahashi T, Ueda R, Hibi K, Saito H and Takahashi T, 1992. Molecular analysis of the protein tyrosine phosphatase g gene in human lung cancer cell lines. Cancer Res, 51, 3506-3509.

Ushiro H, and Cohen S, 1980. Identification of phosphotyrosine as a product of epidermal growth factor-activated in A-431 cell membranes. J Biol Chem, 255 (18), 8363-8365. van Biesen T, Hawes BE, Luttrell DK, Krueger KM, Touhara K, Porfiri E, Sakaue M, van Niekerk CC, and Poels LG, 1999. Reduced expression of protein tyrosine phosphatase gamma in lung and ovarian tumors. Cancer letters, 137, 61-73.

Van Roozendaal CE, Van Ooijen B, Klijn JG, Claassen C, Eggermont AM, Henzen- Logmans SC, Foekens JA, 1992. Stromal influences on breast cancer cell growth. Brit J Cancer, 65, 77-81.

Van Roozendaal CE, Klijn JG, Van Ooijen B, Claassen C, Eggermont AM, Henzen- Logmans SC, Foekens JA, 1996. Differential regulation of breast tumor cell proliferation by stromal fibroblasts of various breat tissue sources. Int J Cancer, 65,120-125.

Warner M, Nilsson S, Gustafsson JA, 1999. The estrogen receptor family. Curr Opin Obstet Gynecol, 11, 249-254.

Welshons WV, Rottinghaus GE, Nonneman DJ, Dolan-Timpe M, and Ross PFA, 1990. Sensitive bioassay for detection of dietary estrogens in animal feeds. Journal of Veterinary Diagnostic Investigation, 2, 268-273.

Yan ZJ, and Roy D, 1997. Mutations in DNA polymerase P mRNA of stilbene estrogen- induced kidney tumors in Syrian hamster. Biochem Mol Biol Int, 37, 175-183.

157 Zava DT, Blen M, and Duwe G, 1997. Estrogenic activity of natural and synthetic estrogens in human breast cancer cells in culture. Environmental Health Perspectives, 105 Suppl 3, 637-645.

Zhang Y. 1999. Cell-cell interactions and gossypol’s effects on cell functions of primary cultured human breast epithelial, stromal and adipose stromal cells. Ph.D dissertation, The Ohio State University. pp.1-34.

Zheng J, Kulp SK, Zhang Y, Sugimoto Y, Dayton MA, Govindan MV, Brueggemeier RW, and Lin YC. 2000. 17ß-estradiol-regulated expression of protein tyrosine phosphatase gamma gene in cultured human normal breast and breast cancer cells. Anticancer Res, 20, 11-20.

Zhu BT, Bui QD, Weisz J, Liehr JG, 1994. Conversion of estrone to 2- and 4- hydroxyestrone by hamster kidney and liver microsomes: implications for the mechanism of estrogen-induced carcinogenesis. Endocrinology, 135, 1772-1779.

Ziegler RG, Hoover RN, Pike MC, Hildesheim A, Nomura AM, West DW, Wu-Williams AH, Kolonel LN, Horn-Ross PL, and Rosenthal JF, 1993. Migration patterns and breast cancer risk in Asian-American women. Journal of the National Cancer Institute, 85,1819- 27.

Zondag GCM, Koningstein GM, Jiang Y-P, Sap J, Moolwnaar WH, and Gebbink MFBG, 1995. Homophilic interactions mediated by receptor tyrosine phosphatases µ and κ: a critical role for the novel extracellular MAM domain. J Biol Chem, 270, 14247-14250.

158