Citation
Benzo(a)pyrene-mediated effects on cellular adhesion and signal transduction pathways in endometrial cancer

Material Information

Title:
Benzo(a)pyrene-mediated effects on cellular adhesion and signal transduction pathways in endometrial cancer
Alternate title:
Benzoapyrene mediated effects on cellular adhesion and signal transduction pathways in endometrial cancer
Alternate title:
Benzo a pyrene-mediated effects on cellular adhesion and signal transduction pathways in endometrial cancer
Creator:
McGarry, Michelle Ann, 1975-
Publication Date:
Language:
English
Physical Description:
xiv, 179 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Actins ( jstor )
Cadherins ( jstor )
Cell adhesion ( jstor )
Cell cycle ( jstor )
Cell lines ( jstor )
Cell membranes ( jstor )
Cells ( jstor )
Genes ( jstor )
Oxidative stress ( jstor )
Xenobiotics ( jstor )
Benzo(a)pyrene -- pharmacology ( mesh )
Cell Adhesion ( mesh )
Cell Cycle ( mesh )
Cytochrome P-450 CYP1A1 ( mesh )
Cytoskeleton ( mesh )
Department of Pharmacology and Therapeutics thesis Ph.D ( mesh )
Dioxins -- pharmacology ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pharmacology and Therapeutics -- UF ( mesh )
Endometrial Neoplasms ( mesh )
Research ( mesh )
Signal Transduction ( mesh )
Tumor Cells, Cultured ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2001.
Bibliography:
Bibliography: leaves 164-178.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Michelle Ann McGarry.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
027371416 ( ALEPH )
52233136 ( OCLC )

Full Text














BENZO(A)PYRENE-MEDIATED EFFECTS ON CELLULAR ADHESION
AND SIGNAL TRANSDUCTION PATHWAYS IN ENDOMETRIAL CANCER












By

MICHELLE ANN MCGARRY


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2001






















This dissertation is dedicated to my parents Joan and Donald McGarry for all of their
love, support, and encouragement in helping me to achieve my dreams.












ACKNOWLEDGEMENTS


I would like to express my sincere gratitude to the many people who have

contributed to my dissertation work over the past four years. First, and foremost, I would

like to thank my mentor, Dr. Kathleen Shiverick, for her continuous efforts to challenge,

support and guide me throughout my graduate school years. Her enthusiasm and love for

science has been an inspiration to me, and has made my graduate learning so enjoyable.

Her high expectations and encouragement in presenting my research at national and local

conferences has helped me in developing my critical thinking, problem solving, and

communication skills. She has also generously shared her time in discussing my

experiments and has provided extensive support in helping me to accomplish my

professional and personal goals. Her genuine compassion, warmth, and concern for

others have made her a role model to me in both my research and personal life, and I feel

truly fortunate to be her student.

I would also like to thank my committee members, Dr. Michael Bubb, Dr. Naser

Chegini, Dr. Stephen Roberts, and Dr. Dietmar Siemann, for their dedication and support

of my research over the past four years. Individually, and as a group, they have shared

excellent suggestions and advice that has stimulated my thoughts and have deepened my

understanding. Their enthusiasm and encouragement have contributed to my success and

enjoyment of research.

I would further like to thank Theresa Medrano from our laboratory for her

outstanding technical assistance and advice, as well as her friendship. Her patience,








kindness, and concern for others make her a pleasure to work with. I also thank all of the

past and present members of Dr. Shiverick's laboratory for their friendship and support,

including fellow graduate students Von Samedi and Renita Handayani.

In addition, I would like to thank Dr. Maria Grant for her generous assistance with

the Boyden chamber attachment assay technique, as well as Dr. Grantley Charles for his

prior experience, advice, and humor. In addition, I would like to thank Dr. Martha

Campbell-Thompson and Dr. Michael Bubb for generously sharing their time, expertise

and resources for my immunocytochemistry and actin immunostaining experiments. In

addition, I thank Melissa Chen from the flow cytometry core laboratory for her kindness

and help with the cell cycle phase distribution analysis, and Juan and Dr. Henry Baker for

assistance and suggestions on cDNA microarray data analysis. Finally, I would like to

thank Judy Adams and the entire staff in the Department of Pharmacology for their

exceptional administrative and personal support. I am additionally grateful for the

financial support of the Superfund Basic Research Program graduate fellowship award.

Finally, I would like to extend my deepest gratitude and love to my parents, who

have continuously believed in me and encouraged me to do my best. They have given

me the encouragement and support I needed to follow my dreams.













TABLE OF CONTENTS

pag.e

ACKNOWLEDGEMENTS ................................................................... i

LIST OF TABLES ........................................................................................................... viii

LIST OF FIGURES ....................................................................................................... x

ABSTRACT ..................................................................................................................... xiii

CHAPTERS

1 INTRODUCTION ........................................................................................................ 1

Study Objectives ........................................................................................................ 1
Carcinogenesis and Hum an Health .......................................................................... 3
Pathophysiology of Uterine Cancer .......................................................................... 4
Environmental Exposure to Polycyclic Aromatic Hydrocarbons and Dioxin .......... 6
Cell Culture M odels for Uterine Disease ............................................................... 10
Cell Adhesion M olecules and Uterine Disease ...................................................... 12
Oxidative Stress and Toxicology ............................................................................. 15


2. M ATERIALS AND M ETHODS ............................................................................. 19

M aterials ..................................................................................................................... 19
Chem icals and Bioreagents ............................................................................... 19
Antibodies ........................................................................................................ 20
M ethods ....................................................................................................................... 20
Cell Cultures and Chem ical Treatm ents .......................................................... 20
In Vitro Attachm ent Assay ............................................................................... 21
Imm unocytochem istry ..................................................................................... 22
Light and Confocal M icroscopy ..................................................................... 23
Cellular Protein (Membrane and Lysate) Preparations .................................... 24
W estern Im m unoblot Analysis ........................................................................ 24
Co-Imm unoprecipitation of P-catenin with a-catenin ...................................... 26
Phosphorylation Experim ents .......................................................................... 27
Flow Cytometry Analysis with Propidium Iodide Fluorescence ...................... 27
Northern Blot Analysis ..................................................................................... 28








Ultra-pure RNA Extraction and Poly A+ Purification for
cDNA Microarray Analysis ............................................................................ 29
cDNA Microarray Analysis ............................................................................ 30
cDNA Microarray Data Analysis ...................................................................... 31
Quantitation of Intracellular Oxidative Stress by DCF Assay .......................... 33
D ata A nalysis ................................................................................................... 34
Potential Hazards and Precautions .................................................................... 34


3. EFFECTS OF BAP AND TCDD ON CELL ATTACHMENT AND EXPRESSION
OF ADHERENS JUNCTION, CYTOSKELETAL, AND GROWTH FACTOR
RECEPTOR PROTEINS IN RL95-2 CELLS ........................................................ 38

Introduction ................................................................................................................. 38
R esults ......................................................................................................................... 40
Cellular Attachment on Matrigel ...................................................................... 40
Effects on Epidermal Growth Factor Receptors ............................................... 40
Effects on Cadherin and f3-catenin Cellular Adhesion Molecules .................... 41
Effects on Actin Cytoskeletal Protein ............................................................... 41
D iscussion ................................................................................................................... 42


4. SIGNAL TRANSDUCTION PATHWAYS FOR BAP AND TCDD EFFECTS ON
RL95-2 CELLS: THE AHR, CELL CYCLE, AND OXIDATIVE STRESS ........ 55

Introduction ................................................................................................................. 55
The Arylhydrocarbon Receptor ........................................................................ 55
Oxidative Stress and Cell Cycle ....................................................................... 57
Prostaglandin H -Synthase ................................................................................. 61
Oxidative Stress and Cellular Adhesion .......................................................... 62
R esu lts ......................................................................................................................... 62
Cell Cycle Phase Distribution .......................................................................... 62
Effects on CYP1A1 and CYPIBI mRNA Levels ............................................ 63
Effects on CYP1A1, PGHS-1, and PGHS-2 Protein Levels ............................ 63
cDNA Microarray Analysis for Induction of Stress-Related Genes ................. 64
Cluster and TreeView Analysis of Common Gene Expression Patterns .......... 67
Quantitation of Intracellular Oxidative Stress
by Dichlorofluorescein Assay .......................................................................... 67
D iscussion ................................................................................................................... 68


5. DIFFERENTIAL EFFECTS OF BAP ON CELL ATTACHMENT, ADHERENS
JUNCTION, AND CYTOSKELETAL COMPLEX AND PHOSPHORYLATION IN
HEC-1A AND HEC-IB CELLS ............................................................................ 95

Introduction ................................................................................................................. 95








R esu lts ......................................................................................................................... 97
Differential Localization of Actin Cytoskeletal Proteins .................................. 97
Levels of P-catenin, ax-catenin, and EGF-R Proteins ........................................ 98
Effects of BaP on HEC-1A and HEC-1B Cell Attachment and Invasion ...... 99
Effects of BaP on a-catenin, 0-catenin, and EGF-R Protein Levels .................. 100
Effects of BaP on P3-catenin and ca-catenin Protein Interactions ......................... 101
Effects of BaP on P-catenin Phosphorylation ..................................................... 101
D iscussion ................................................................................................................. 102


6. ALTERNATE SIGNAL TRANSDUCTION PATHWAYS AND CELL CYCLE
EFFECTS OF BAP ON HEC-IA AND HEC-IB CELLS ....................................... 116

Introduction ............................................................................................................... 116
Metabolic Enzymes and Cell Cycle .................................................................... 116
Alternative Signal Transduction Pathways for BaP Effects Data Mining ..... 117
R esults ....................................................................................................................... 119
Cell Cycle Phase Distribution ............................................................................. 119
Time Course Effects on HEC-1A and HEC-1B Cell Cycle Phase Distribution. 120
Effects on CYPlA1 and CYP1B1 Levels ........................................................... 121
cDNA Microarray Analysis to Profile Changes in
Toxicology Gene Expression .............................................................................. 122
Cluster and TreeView Analysis of Common Expression Patterns of Genes ...... 124
D iscussion ................................................................................................................. 125


7. C O N C LU SIO N S ....................................................................................................... 151

LIST OF REFERENCES ................................................................................................ 164

BIOGRAPHICAL SKETCH .......................................................................................... 179













LIST OF TABLES


Table Page

2-1. Antibody Sources and Dilutions for Cellular Adhesion Protein
Western Blots and Immunocytochemistry ........................................................ 36

2-2. Antibody Sources and Dilutions for Cellular Enzyme
W estern blots ................................................................................................... 37

4-1. Stress-related genes with defined ratios down-regulated by
benzo(a)pyrene in RL95-2 cells ........................................................................ 86

4-2. Stress-related genes with defined ratios up-regulated by
benzo(a)pyrene in RL95-2 cells ........................................................................ 86

4-3. Stress-related genes with defined ratios down-regulated by
TCDD in RL95-2 cells ...................................................................................... 87

4-4. Stress-related genes with defined ratios up-regulated by
TCDD in RL95-2 cells ...................................................................................... 87

4-5. Stress-related genes with defined ratios down-regulated by
t-butylhdyroperoxide in RL95-2 cells ............................................................... 88

4-6. Stress-related genes with defined ratios up-regulated by
t-butylhydroperoxide in RL95-2 cells ............................................................... 88

4-7. Biochemical Measure of Oxidative Stress in RL95-2 Cells -
DCF Fluorescence Using a Microplate Reader ............................................... 93

6-1. Human toxicology genes down-regulated by
benzo(a)pyrene in HEC-1A cells ........................................................................ 140

6-2. Human toxicology genes down-regulated by
benzo(a)pyrene in HEC-IB cells ........................................................................ 140

6-3. Human toxicology genes up-regulated by
benzo(a)pyrene in HEC-1B cells ........................................................................ 141

6-4. Human toxicology genes higher in control
HEC-1A cells than HEC-IB cells ....................................................................... 142








6-5. Human toxicology genes higher in control
HEC-IB cells than HEC-1A cells ....................................................................... 144

7-1. Cellular Morphology and Response to BaP ........................................................ 162













LIST OF FIGURES


Table Pae

1-1. Cell-cell and cell-extracellular matrix adhesions ............................................. 17

1-2. Chemical structures of xenobiotics ................................................................... 18

3-1. Effects of BaP and TCDD on RL95-2 cell attachment to membranes ............. 47

3-2. Effects of BaP and TCDD on the localization and protein levels
of EGF-R in RL95-2 cells ................................................................................. 48

3-3. Western Immunoblot analysis of the effects of BaP and TCDD
on cadherin, P-catenin, and Vinculin Levels ................................................... 50

3-4. Effects of BaP on the localization of actin filaments and protein
levels in RL95-2 cells ....................................................................................... 53

4-1. Intermediates from normal metabolism of atmospheric oxygen by
successive 1-electron reductions ...................................................................... 78

4-2. Graphical Representation of BaP and TCDD Effects on RL95-2
Cell Cycle Phase Distribution .......................................................................... 79

4-3. Differential Effects of BaP and TCDD on Cell Cycle Phase Distribution
in R L 95-2 cells ................................................................................................. 80

4-4. Induction of CYPIAl and CYP1B1 mRNA by BaP and TCDD
in R L95-2 cells ................................................................................................. 81

4-5. Western immunoblot analysis for the effects of BaP and TCDD on
CY PIA I protein levels ..................................................................................... 82

4-6. Western immunoblot analysis of the effects of BaP and TCDD
on PGH S-1 Protein Levels ............................................................................... 83

4-7. RL95-2 Microarray Experiment ScatterPlot Analysis
using AtlasImage 1.5 Globally Normalized Data ............................................. 84

4-8. Venn diagram for BaP, TCDD, and t-butylhydroperoxide effects
on R L95-2 cells ................................................................................................. 89








4-9. RL95-2 ScatterPlot for TreeView Analysis ...................................................... 90

4-10. RL95-2 cDNA Microarray analysis TreeView Diagram .................................. 91

4-11. Cellular effects of oxidative stress ................................................................... 94

5-1. Differential filamentous actin localization in HEC-1A and HEC-IB cells ........ 107

5-2. Western Immunoblot profiles of 03-catenin, ax-catenin, and EGF-R
protein levels in HEC-IA and HEC-1B cell lines .............................................. 108

5-3. Effects of BaP on HEC-IA and HEC-IB cell attachment and invasion ............. 109

5-4. Western Immunoblot analysis of the effects of BaP on c-catenin Protein
in H EC-1A and H EC-1B cells ............................................................................ 110

5-5. Western Immunoblot analysis of the effects of BaP on 0-catenin Protein
in HEC- 1A and H EC- IB cells ............................................................................ 111

5-6. Western Immunoblot analysis of the effects of BaP on EGF-R Protein
in H EC-1A and HEC-IB cells ............................................................................ 112

5-7. Schematic Representation of Immunoprecipitation Protocol ............................. 113

5-8. Protein interactions of the P-catenin / a-catenin complex in HEC-1A
and HEC- 1 B cells following BaP challenge ....................................................... 114

5-9. Tyrosine phosphorylation of immunoprecipitated 1-catenin protein in
HEC-IA and HEC-1B cells upon BaP treatment ............................................... 115

6-1. Differential Effects of BaP and TCDD on Cell Cycle Phase
Distribution in HEC-1A and HEC-1B cell lines ................................................. 133

6-2. BaP differentially regulates HEC-1A and HEC-IB cell cycle
distribution over tim e .......................................................................................... 134

6-3. Induction of CYP1A1 and CYP1B1 mRNA by BaP in HEC-1A
and H E C -1B cells ............................................................................................... 135

6-4. Western Immunoblot analysis of the effects of BaP on CYP1A1
in HEC-1A and H EC-1B cells ............................................................................ 136

6-5. Illustration representing the generalized protocol for cDNA
M icroarray analysis ............................................................................................. 137

6-6. ScatterPlot Analysis for HEC-lA and HEC-IB
M icroarray experim ent ........................................................................................ 138








6-7. Venn diagram depicting genes regulated by BaP in HEC-IA and
H E C -IB cell lines ............................................................................................... 145

6-8. HEC-IA and HEC-IB Microarray experiment ScatterPlot Analysis ................. 146

6-9. HEC-1A and HEC-IB cDNA Microarray analysis TreeView Diagram ............ 147

6-10. Selected Clusters from the HEC-IA and HEC-IB
M icroarray TreeView Diagram ........................................................................... 149














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

BENZO(A)PYRENE-MEDIATED EFFECTS ON CELLULAR ADHESION
AND SIGNAL TRANSDUCTION PATHWAYS IN ENDOMETRIAL CANCER

By

Michelle Ann McGarry

August 2001


Chair: Kathleen T. Shiverick, Ph.D.
Major Department: Pharmacology and Therapeutics

Cigarette smoking in women has been linked with a decreased incidence of

endometrial cancer and endometriosis, whereas exposure to the dioxin 2,3,7,8-

tetrachlorodibenzo-p-dioxin (TCDD) has been associated with an increased incidence of

endometriosis in animal models. This study characterized the cellular and molecular

effects of BaP, a cigarette smoke toxicant, and dioxin on three human uterine endometrial

cancer cell lines (RL95-2, HEC-lA, and HEC-1B).

Experiments show that BaP completely inhibited RL95-2 and HEC-1A cellular

attachment and decreased membrane P-catenin and EGF receptor protein levels; in

contrast, HEC-1B cells were unaffected. HEC-1A cells also showed decreased a-catenin

levels with BaP treatment, whereas HEC-1B cells remained unaffected. TCDD did not

affect RL95-2 cell attachment, 0-catenin, or EGF-R levels. BaP, but not TCDD, further

produced subcortical actin aggregates and decreased cadherin levels in RL95-2 cells,








whereas neither chemical altered overall actin and vinculin levels. Furthermore, BaP

induced an enhanced cell cycle response in HEC-1A and RL95-2, but not HEC-IB cells,

whereas TCDD had no effect.

HEC- 1 A (BaP-responsive) and HEC- lB (BaP-unresponsive) cells were compared

for key morphological differences. HEC-lA cells expressed higher levels of c-catenin

and had thick, cortical, filamentous actin, whereas HEC-1B cells exhibited nuclear-

localized actin and had non-continuous intercellular boundaries. CYP1A1 was induced

in all three cell lines upon BaP and TCDD treatment, and therefore did not account for

the cell-specific and BaP-specific effects. Microarray experiments to define profiles of

gene expression following treatment of HEC-1A and HEC-IB cells, as well as for BaP,

TCDD, and t-butylhydroperoxide treatment of RL95-2 cells, indicated that BaP does not

significantly induce stress-regulated genes by 6 hours.

Overall, results show that the Ah-Receptor ligand BaP alters a complex array of

signal transduction pathways in human endometrial adenocarcinoma cell lines, resulting

in profound alterations of cellular adhesion, cytoskeleton, and cell cycle. RL95-2, HEC-

1A, and HEC-1B cells are presented as useful models for determining alternate cellular

responsiveness to xenobiotic exposure.














CHAPTER 1
INTRODUCTION


Study Objectives

Evidence indicates that cigarette smoking in women is associated with a

decreased incidence of endometrial cancer and endometriosis. The environmental

contaminant benzo(a)pyrene (BaP), a major toxicant in cigarette smoke, has been shown

to inhibit the growth of human uterine endometrial RL95-2 cells in culture (Charles,

1997). Exposure to the dioxin 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) has been

associated with an increased incidence of endometriosis in animal models, and possibly

women. The present study explored cellular and molecular mechanisms which underlie

the growth inhibitory effects of BaP, a known carcinogen and AhR ligand, on uterine

endometrial cancer cell lines. One hypothesis herein tested is that BaP exerts growth

inhibitory effects by altering the delicate balance of cellular adhesion and cytoskeletal

proteins, thereby altering cellular attachment and invasion. An additional hypothesis

investigated was that BaP acts by disrupting the integrity of the adherens complex

proteins by altering the phosphorylation status of P3-catenin. Finally, the role of cell cycle

alterations and the induction of AhR-dependent and AhR-independent enzymatic

pathways for BaP mediated effects, including the generation of an oxidative stress

response, were investigated using BaP-responsive and BaP-unresponsive endometrial

cancer cell lines. Many experiments involved a comparison of the effects of BaP with

those of TCDD, a non-metabolized AhR ligand.








The studies presented here evaluated the usefulness of three human, uterine

endometrial cancer cell lines as in vitro models to elucidate the cellular and molecular

effects of BaP and TCDD. Experiments were conducted to characterize the effects of

BaP and TCDD treatment on key cellular adhesion molecule protein levels, localization,

and interactions in relation to cellular attachment and invasion. Levels of the major

proteins involved in cell-cell adhesion (cadherin, 03-catenin, a-catenin, vinculin, and

actin), as well as epidermal growth factor receptor (EGF-R) were quantitated in RL95-2,

HEC-lA, and HEC-IB cell lines. The localization of key cellular proteins was visualized

using light and confocal microscopy. The effects of BaP on the phosphorylation levels of

cellular adhesion proteins and the integrity of the adhesion complex were evaluated in

HEC-IA and HEC-IB cells.

Further studies were performed to characterize the effects of BaP and TCDD on

gene expression in three endometrial cancer cell lines through two separate signal

transduction pathways, the AhR and oxidative stress. Because BaP and TCDD activate

AhR-mediated genes, studies were performed to determine whether the observed

differences in gene expression between BaP and TCDD were due to the differential

activation of enzymes involved in xenobiotic metabolism and the induction of an

oxidative stress response in cells. The effects of BaP and TCDD on cellular mRNA and

protein levels of the enzymes CYP1A1, CYPIBI, PGHS-1, and PGHS-2 were

characterized in each cell line. Further comparison of major gene induction pathways of

BaP with TCDD, a non-metabolized AhR ligand, and t-butylhydroperoxide, a classic

oxidative stress inducer, were evaluated using cDNA microarray technology. A

biochemical measure for oxidative stress, the oxidation of dichlorodihydrofluorescein








diacetate, was employed to characterize the xenobiotic effects on intracellular oxidative

stress.

The temporal effects of BaP and TCDD on cell cycle phase distribution in RL95-

2, HEC-1A, and HEC-1B cells were determined using flow cytometry analysis with

propidium iodide fluorescence at 6, 12, 24, and 48 hours. Analysis of the alternate signal

transduction pathways for BaP-mediated effects on the HEC-IA and HEC-1B cell lines

was carried out using cDNA microarray technology. Gene expression data were

analyzed for fold-change induction or repression of individual genes, as well as for

common patterns of gene expression across multiple replicates and chemical treatments.

The differential response of the respective cell lines to BaP treatment serves as a

predictive model for biomarkers of effect for xenobiotic exposure on the human uterine

endometrium, as well as for differential tumor cell response to exposure to these

prototype environmental toxicants.



Carcinogenesis and Human Health

There are an estimated 200 different types of cancer known to exist in humans, all

characterized by an uncontrolled cellular growth and the capacity to metastasize to distant

sites, thereby adversely affecting the health of the patient (Thomas, 1993). Animal

studies indicate that the formation of cancer involves multiple stages, characterized as

initiation, promotion, and progression (Guyton and Kensler, 1993). The initiation stage

of carcinogenesis involves irreversible alteration of a cell, typically by mutation.

Initiation can occur through oxidative DNA modification that results in potentially

deleterious mutations. The resulting mutations may alter the function of key genes








relating to carcinogenesis, such as oncogenes or tumor suppressor genes. Promotion, the

second stage of cancer development, involves the clonal expansion of a mutated cell.

Typically, promotion will occur through the perturbation of signal transduction pathways,

including changes in methylation, accumulation of intracellular calcium, inhibition of

intercellular communication, or alterations in gene expression or apoptosis (Yoshida and

Ogawa, 2000). The ultimate stage of carcinogenesis, progression, involves the process

whereby malignant transformation occurs and cellular growth proceeds uncontrolled.

Progression typically occurs through multiple stages, as it has been estimated that 10 or

more mutational changes occur in most cancers (Barrett, 1993). Oxidative stress has

been implicated in carcinogenesis by producing alterations in signaling pathways which

lead to mutagenicity through oxidative modification of genetic material, stimulation of

the initiated cell during tumor promotion, and progression to uncontrolled growth and

malignancy (Guyton and Kensler, 1993; Yoshida and Ogawa, 2000).



Pathophysiology of Uterine Cancer

Endometrial cancer, the most common gynecologic malignancy in the United

States, remains the fourth leading cancer incidence in females and the eighth leading

cause of new cancer deaths in women (American Cancer Society, 1998). Whereas an

estimated 36,100 cases of cancer of the uterine corpus, usually of the endometrium, were

diagnosed in 1998, incidence rates have been relatively constant since the middle 1980s

at about 21 per 100,000 (American Cancer Society, 1998). The estimated deaths from

endometrial cancer of 6,300 in 1998 coincide with the relatively constant mortality rates

since 1989 of about 3 per 100,000 (American Cancer Society, 1998). Furthermore,





5


endometriosis, a painful and chronic uterine disease often associated with infertility, has

been detected in 10 to 15 percent of all premenopausal women undergoing gynecological

surgery (Haney, 1990), as well as 25 percent of all women in their thirties and forties

(Chalmers, 1980).

Endometrial adenocarcinoma occurs during the reproductive and menopausal

years, with 20 to 25 percent of cases being diagnosed before the onset of menopause

(Creasman, 1997). Initially associated with abnormal uterine bleeding or spotting,

endometrial cancer later produces pain or systemic symptoms. There are two primary

types of endometrial cancers, estrogen-dependent and estrogen-independent. The

estrogen-dependent form of endometrial cancer is associated with estrogen-related

exposures such as estrogen replacement therapy, tamoxifen, early menarche, late

menopause, nulliparity, and failure to ovulate. The estrogen-independent form of

endometrial cancer appears to lack direct hormonal influence, and could potentially arise

due to a compromised immune system with age. Additional risk factors associated with

endometrial cancer include infertility, diabetes, gallbladder disease, dietary factors,

hypertension, and obesity (American Cancer Society, 1998). Endometrial cancers are

routinely treated with surgery, radiation, hormones and/or chemotherapy, and result in a

relatively high rate of survival if diagnosed early.

It has been shown that smoking is consistently associated with an independent

reduction in risk for endometrial cancer. Epidemiological studies indicate that cigarette

smoking reduces the risk of endometrial cancer by as much as 50 percent (Meek and

Finch, 1999). It is believed that the seemingly protective effect of cigarette smoking on

endometrial cancer involves an anti-estrogenic mechanism, wherein women who smoke








appear as if estrogen deficient (Baron, 1996; Baron et al., 1990). In addition, current

smokers and postmenopausal women experience the greatest reduction in cancer risk

(Baron et al., 1990; Lesko et al., 1985). The reduction may be attributed to three primary

mechanisms: a reduction in luteal phase estrogens, early menopause, and differences in

metabolism of estrogens among smokers (Burke et al., 1996). In addition, smoking has

been shown to alter prostaglandin pathways (Burke et al., 1996). Further, there is direct

evidence for the implication of cigarette smoke in the progression of human

papillomavirus (HPV)-initiated cervical cancer (Nakao et al., 1996; Nakao et al., 1996).



Environmental Exposure to Polycyclic Aromatic Hydrocarbons and Dioxin

Polycyclic aromatic hydrocarbons (PAHs) are prevalent environmental pollutants

formed during the combustion of fossil fuels and the burning of various organic

materials. Each year, tens of thousands of PAHs are released into the environment in the

United States alone, leading to the contamination of air, water, and soil (Zedeck, 1980).

Humans are exposed to PAHs through inhalation of coal and petroleum products, wood

smoke and cigarette smoke, as well as through occupational inhalation exposures from

tars and fumes (Burchiel and Luster, 2001). Inhalation of PAHs can involve gaseous, or

volatilized components, as well as exposure to PAH-coated particles, which are

eventually ingested. Exposure to PAHs primarily occurs through ingestion, partly from

consumption of charcoal-broiled foods, through which humans are exposed to microgram

concentrations of PAHs each day (Burchiel and Luster, 2001). Additional human

exposure to PAHs occurs through dermal contact with tars or soot. Since human

exposure to PAHs often occurs in combination with exposure to other toxic compounds,








the effects associated with a single exposure to BaP may not directly reflect actual

environmental exposures.

Cigarette smoke, a common route of human exposure to PAHs, is a complex

mixture containing more than 3,500 particulate and nonparticulate components, at least

sixty of which are known toxicants. PAHs, nicotine, cadmium, nitroso compounds,

aromatic amines, protein pyrolysates, and carbon monoxide have all been associated with

the toxic effects of smoke (Shiverick and Salafia, 1999; Meek and Finch, 1999).

Benzo(a)pyrene (BaP), a major component of cigarette smoke, was selected as a target

toxicant for study due to its predominance and potency. BaP is present at levels of 20-40

ng/cigarette (Shiverick and Salafia, 1999). The average daily intake of BaP in the United

States is approximately 2.2 jig/day, the majority of exposures resulting from ingestion,

inhalation, and dermal absorption (Kim et al., 1998; Hattemer-Frey and Travis, 1991).

Further, it is estimated that the average intake of BaP in smokers of one pack of

unfiltered cigarettes is nearly 0.7 jig/day over background levels, whereas the

corresponding intake for filtered cigarettes is 0.4 jig/day over background (Miller and

Ramos, 2001b).

Whereas some PAHs are relatively innocuous, BaP is a potent chemical

carcinogen. BaP is a highly lipophilic compound that can be readily taken up by cells

through the plasma membrane. Once inside the cell, BaP is rapidly distributed

throughout the cellular compartments, including the mitochondrion and nucleus, the

Golgi, and the cytoplasmic membranes, including the plasma membrane, endoplasmic

reticulum, and nuclear envelope (Miller and Ramos, 2001b). Studies from human

placentas of women who smoke cigarettes lend support for the study of BaP-mediated








effects on uterine endometrial cancer. It has been reported that EGF-stimulated kinase

activity was markedly decreased in placental membrane proteins from women who

smoke cigarettes, whereas insulin receptor phosphorylation was unaltered or increased,

indicating that maternal smoking was associated with a selective loss of EGF receptor

autophosphorylation (Wang et al., 1988). Further studies have shown the induction of

placental aryl hydrocarbon hydroxylase (AHH) activity in placental tissue of smokers, an

observation associated with the binding of PAH adducts to DNA in vitro (Pelkonen and

Saami, 1980; Berry et al., 1977; Vaught et al., 1979). Studies in human placental cell

cultures reflect the dose-dependent decrease of BaP on EGF receptors in early gestation

cells (Guyda et al., 1990). Term placental cells, on the contrary, were shown to reflect

receptor desensitization upon 10 tM BaP exposure due to a dissociation of EGF binding

and EGF receptor protein kinase activity (Guyda et al., 1990). In addition, a dose-

dependent induction of AHH by approximately 20- to 150-fold was observed in early

placental cells exposed to BaP (Guyda et al., 1990). Furthermore, the dose of 10 gM

BaP has been shown to induce CYPIA1 and CYPIBI in the RL95-2 endometrial cancer

line (Charles and Shiverick, 1997).

In consideration of published studies, the concentration of 10 14M BaP was

selected for current experiments studying the effects of BaP on uterine cancer cell lines.

Since BaP is found in concentrations of 20-40 ng/cigarette, a concentration of 1 [tM BaP

in cell culture is equivalent to a dose of approximately 250 ng/ml of BaP, or the

approximate BaP content of 5 cigarettes. Since BaP is known to bioaccumulate, the

actual environmental relevance of the 10 jtM BaP dose is somewhat uncertain.








Additional studies investigating dose-response relationships for BaP-mediated effects on

attachment, adhesion, and cell cycle effects could prove beneficial.

A genotoxic chemical, BaP significantly alters signal transduction in human and

animal cells by binding to the intracellular cytosolic protein aryl hydrocarbon receptor

(AhR) and transcriptionally activates cytochrome P450 lAl (CYP1A1), leading to its

own metabolism (Poland et al., 1976; Whitlock, Jr. et al., 1996). The AhR protein, an N-

terminal basic helix-loop-helix protein, was found to be expressed in 43 percent of human

endometria studied by Kuchenhoff and co-workers, with apical cytoplasmic localization

in epithelial cells of endometrial glands (Kuchenhoff et al, 1999). The finding that the

PAH BaP is carcinogenic, while the structurally similar PAH benzo(e)pyrene (BeP) is

weakly or non-carcinogenic, makes BeP a useful tool in determining specific and

nonspecific effects of these chemicals on cell lines (MacLeod et al., 1982) (Figure 1-1).

The cytochrome P450 enzyme CYP1A1, known to function in the bioactivation of

procarcinogens, is induced in many tissues by BaP, 2,3,7,8-tetrachlorodibenzo-p-dioxin

(TCDD), the prototype AhR ligand, or other AhR ligands (Omiecinski et al., 1999).

Whereas cigarette smoke has been associated with a decreased incidence of endometrial

cancer and endometriosis (Baron, 1996), TCDD has been associated with the promotion

of uterine disease (Cummings and Metcalf, 1995; Koninckx et al, 1994; Mayani et al,

1997; Rier et al, 1993). In addition, dioxin exposure has been associated with toxic

effects such as atrophy, reduced sperm counts, chloracne, teratogenicity, and

carcinogenicity (Koninckx et al., 1994; Mayani et al., 1997). Recent data analyzing the

twenty year effects on residents from Seveso, Italy after the 1976 massive dioxin

exposure indicate a link of dioxin with an overall increased rate of carcinogenesis relative








to unexposed neighboring populations (Bertazzi et al., 2001). However, uterine cancer

was not shown among the cancers increased by dioxin exposure after twenty years

(Bertazzi et al., 2001). In addition, TCDD responsiveness has been recently been shown

in a human, uterine endometrial explant culture model as determined by increased

induction of CYP1A1 and CYP1B1 mRNA, protein, and enzymatic activities (Bofinger

et al., 2001).

Although the AhR mediates both BaP and dioxin actions, unlike BaP, none of the

symptoms from TCDD exposure are believed to result from the metabolism of the parent

compound (Figure 1-1) (Kolluri et al., 1999). Aside from the AhR pathway, TCDD

further exerts effects through estrogen/estrogen receptor mechanisms and immunologic

mechanisms (Hazan and Norton, 1998; Umbreit et al., 1989; Safe et al., 1998).



Cell Culture Models for Uterine Disease

Female reproduction is characterized by cyclic transformation of the endometrium

mediated by steroids and steroid receptors. During each menstrual cycle, the upper layers

of the endometrium, comprising glandular ducts and stroma, are routinely shed, forming

a uterine environment in continual flux. Evidence indicates that altered hormonal states

are key to the etiology of certain uterine disorders (Hughes and Pfaff, 1998). However,

due to the continued flux of the uterine endometrium, careful study of the direct role of

cellular adhesion and cytoskeletal proteins proves more variable on primary cultures than

on established cell lines.

The cadherins, integrins, and cytoskeletal proteins are key candidates for studies

of cell-cell and cell-extracellular matrix adhesion processes involved in endometrial








cancer. Careful determination of their roles as mediators of signal transduction pathways

leading to altered cellular attachment requires the use of appropriate cell culture models

to control for inherent variability found in primary cultures. Established human

endometrial cell lines offer the advantage of providing a homogeneous cell type and a

controllable condition for studying the action of environmental chemicals in vitro.

RL95-2 cells have been shown to have a smooth surface structure, a thin

glycocalyx, and a non-polarized phenotype which allow them to effectively adhere to

trophoblast cells (Thie et al., 1998; Thie et al., 1996). RL95-2 cells express E-cadherin

and integrins over their entire surface and represent late stage endometrial cancer (Thie et

al., 1996). Further, RL95-2 cells expresses both cytoplasmic and nuclear estrogen

receptors at early passage (subcultured < 30 times), but not at high passage (subcultured

> 200 times), thereby providing a useful model for both primary malignancy and

metastatic tumor formation (Sundareshan and Hendrix, 1992; Way et al., 1983). The

RL95-2 cell line is distinct from others in the adhesiveness of its apical pole for

trophoblast cells, thereby serving as an in vitro model of the human uterine epithelium

receptive for implantation (Thie et al., 1997). Preliminary data using RL95-2 cell

cultures indicate that BaP markedly decreases cellular invasion and attachment (Charles,

1997). The observed alterations in cellular attachment and invasion were hypothesized to

result from BaP-mediated altered protein levels or localization of cellular adhesion or

cytoskeletal proteins.

HEC-IA and HEC-1B cells, by comparison, have been shown to have a rough

surface structure and a thick glycocalyx, and to represent early stage endometrial cancer

(Thie et al., 1998). HEC-IA and HEC-1B cells are well established in vitro models often








used to study the effects of hormones and/or growth factors on uterine endometrial cell

growth. Both cell lines are characterized by similar proliferative patterns, including

doubling time and cell cycle kinetics (Borri et aL, 1998). HEC-1A and HEC-IB cells are

highly polarized and consequently are non-adhesive upon contact with trophoblast cells

(Thie et al., 1998). Further, HEC-1A and HEC-1B cells express E-cadherin and integrins

at their lateral membranes. Several reports indicate that HEC-1A cells express an

estrogen responsive phenotype with ER-a expression and estrogen-induced cellular

proliferation (Castro-Rivera and Safe, 1998), whereas the HEC-lB cell subtype fails to

show expression of the ER through RT-PCR analysis (Holsapple et al., 1996). Given

recent discoveries of ER subtypes, however, evidence is uncertain over the actual ER

status of the HEC-IA and HEC-1B cellular substrains. The studies herein described

detail the differences in cellular morphology among the three cell lines likely contributing

to their differential cellular responses to chemical insult. The RL95-2, HEC-1A, and

HEC-1B cell lines are proposed as an excellent model system for investigations on

xenobiotic-mediated cellular adhesion alterations. The three cell lines serve as useful

models for the development of biomarkers for the aryl hydrocarbon receptor- (AhR) and

oxidative stress-mediated signal transduction pathways for chemical effects. The BaP-

mediated dysregulation of cell adhesion molecules has beene explored for alterations in

uterine epithelial cell polarization that may impact upon cellular attachment and invasion.



Cell Adhesion Molecules and Uterine Disease

Cell adhesion molecules play a fundamental role in the determination of tissue

architecture and functions of cell assembly and connection to the internal cytoskeleton








and, consequently, are of great significance in the pathophysiology of endometrial cancer.

The differential regulation of cadherin, P3-catenin, a-catenin, and actin are central to

investigations of altered cellular attachment to other cells (Figure 1-2). Cadherins are

integral membrane glycoproteins functioning in epithelial cells to form calcium-

dependent linkages between cells (Potter et al., 1999; Peralta et al., 1997). Cadherins

play a key role in mediating the formation and breakage of cell-to-cell contacts and in

maintaining strength in cellular adhesion through homophilic interaction of their

extracellular domains, allowing for cellular aggregation to occur (Potter et al., 1999).

Classical cadherins, such as E-cadherin found largely in epithelial cells, consist of an

extracellular domain, followed by a transmembrane region and highly conserved

cytoplasmic domain where cadherins critically interact with 3-catenin (Potter et aL,

1999). A key membrane-associated protein, 1-catenin is responsible for the

colocalization of cadherins to sites of cell-cell contact with the actin cytoskeleton. Actin

works in conjunction with cytoskeletal microtubules and intermediate filaments in

performing essential functions in locomotion and cytokinesis (Hulka and Brinton, 1995).

Actin bundles are attached to integral membrane cadherin and integrin proteins, adapter

proteins and the contractile bundle. P-catenin forms a signal transduction pathway with

actin, the key cytoskeletal player, through oa-catenin (Yamada and Geiger, 1997). Both

3- and ax-catenins are known to associate early with E-cadherin at the endoplasmic

reticulum during its synthesis; yet oa-catenin is integrated into the adherens complex at

the later stage of plasma membrane insertion (Potter et al., 1999). Activation of

epidermal growth factor receptor (EGF-R), as well as other tyrosine kinases, has been








shown to directly affect the adhesive function of E-cadherin through catenin alterations

(Hayes et al., 1996).

Tumor cell invasion has been highly linked to changes in the integrin family of

receptors, as well as to cadherins. Integrins mediate stable adhesion to extracellular

matrix (ECM) components; therefore, alterations in integrin expression contribute to

tumor cell invasiveness and metastasis (Sanders et al., 1998). Integrins are integral

membrane proteins that bind to fibronectin or laminin molecules and are of considerable

interest since they not only mediate cell-matrix and cell-cell adhesion, but also function

in plasma membrane signaling (Thie et al., 1997). Integrins are composed of two distinct

subunits, a and 03, which are non-covalently associated with each other (Thie et al.,

1997). Unlike the cadherins which produce strong adherence between cells, integrins

exhibit relatively low affinities for their ligands, yet allow cells to remain firmly anchored

to the matrix due to multiple weak interactions generated by numerous integrins binding

to ECM proteins (Hopfer et at., 1996). For cells to successfully migrate, they must be

able to make and break numerous, weak integrin contacts. In particular, the disregulation

of integrins a5p1, a6P4, ot5P3, and a6131 likely play a significant role in uterine disease

pathogenesis, as well as adherence of the trophoblast during implantation (Beliard et al.,

1997; Thie et al., 1997). The a5pl integrin fibronectin receptor has been shown to be

lower in endometriotic tissues, suggesting a potential role of the fibronectin receptor in

the persistent attachment of endometriotic cells during menstruation (Beliard et al.,

1997). Immunolocalized a5P11 function at focal contacts of cells with the substratum

keeps the cells tightly bound to the matrix, likely through indirect binding to the actin

cytoskeleton. The ot5Pl protein has been shown to be reduced in tumor cells, and when








overexpressed, it has been linked with a suppression of cellular growth and tumorigenesis

(Plantefaber and Hynes, 1989; Giancotti and Ruoslahti, 1990).



Oxidative Stress and Toxicology

Oxidative stress occurs when there is an imbalance in the generation and removal

of reactive oxygen species in an organism. Reactive oxygen species (ROS) have the

potential to damage tissues and cellular components, including the cell membrane, DNA,

and proteins (Guyton and Kensler, 1993). ROS are produced in a cell as a result of

normal, physiologic processes, as well as from oxidizing enzymes and xenobiotic

metabolism (Guyton and Kensler, 1993; Dalton et al., 1999). Free radicals play a variety

of normal physiological roles, including functioning in immune responses,

neurotransmission, and muscle relaxation, as well as control of certain transcription

factors (Ramos, 1999; Morel and Barouki, 1999). In addition, ROS production can elicit

toxic effects in cells, and thereby adversely effect human health (Guyton and Kensler,

1993).

A number of xenobiotics, including BaP and TCDD, can increase ROS

production within cells. Benzo(a)pyrene is a carcinogenic polycyclic aromatic

hydrocarbon (PAH) known to exert toxic effects on cells through multiple mechanisms.

BaP may act directly, initiating tumors through cytochrome P450-mediated metabolic

pathways (Miller, 1970; Heidelberger, 1975; Dipple, 1994). BaP may also act indirectly,

affecting cell signaling pathways through oxidative and electrophilic signaling pathways,

including redox cycling and quinone formation (Bulun et al., 2000). Redox cycling

involves the univalent reduction of the xenobiotic to radical intermediates by enzymes








such as NADPH-cytochrome P450 reductase (Kelly et aL, 1998). In the process, a

radical intermediate transfers an electron to oxygen, producing oxygen radicals (.O2),

and regenerating the parent compound. Multiple oxygen radicals are produced in the

process, and NADPH is depleted (Kelly et aL, 1998).

Excessive production of oxygen radicals can have toxic effects on cells. ROS can

damage lipids, DNA, and proteins. Lipid damage occurs via a chain reaction including

initiation, propagation, and termination, thereby altering biological membranes (Guyton

and Kensler, 1993). ROS can additionally be damaging to DNA, resulting from

hydroxide (HO.) attack on nitrogenous bases, or the DNA backbone itself. The damage

produced can include hydroxylation, ring opening, and DNA fragmentation (Kelly et al.,

1998). In particular, BaP exposure is associated with 8-hydroxy-2'-deoxyguanosine (8-

OhdG) formation. Typically, DNA damage is repaired by excision repair and

postreplication repair enzymes, a process often associated with a prolonged S phase of

the cell cycle. In severe cases, inhibition of cell cycle progression occurs to prevent

transmission of mutations to daughter cells. ROS can further damage cells by producing

protein-DNA crosslinks.

In summary, the present dissertation studies were undertaken to investigate the

role of BaP on the initiation of complex AhR- and oxidative stress-mediated signal

transduction pathways in three human endometrial cancer cell lines. The differential

effects of BaP and TCDD on cellular adhesion, cytoskeleton, and cell cycle checkpoint

responses were determined specific to each cell line, thereby providing a model for

biomarkers of effect for xenobiotic exposure on the human uterine endometrium, as well

as for differential tumor cell response to xenobiotic exposure.














B.


A.















C.


Figure 1-1. Chemical structures of xenobiotics.
The chemical structures of the test compound benzo(a)pyrene (BaP) (A), the structure
activity control Benzo(e)pyrene (BeP) (B), and the prototype dioxin 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) (C).














































Figure 1-2. Cell-cell and cell-extracellular matrix adhesions.
Schematic diagram depicting key cell-cell connections by adherens junctions and cell-extracellular
matrix connections by hemidesmosomes and adhesion plaques. Key cellular adhesion molecules
involved are depicted, along with their relationship to the actin cytoskeleton.














CHAPTER 2
MATERIALS AND METHODS


Materials

Chemicals and Bioreagents

TCDD was obtained from Midwest Research Institute (Kansas City, MO) through

the National Cancer Institute Chemical Carcinogen Reference Repository and

benzo(a)pyrene (BaP) was purchased from the Sigma Chemical Co. (St. Louis, MO).

Cell culture media was purchased from GIBCO BRL/Life Technologies (Gaithersburg,

MD), fetal bovine serum (FBS) from Hyclone Laboratories (Logan, UT), and Matrigel

from Collaborative Biomedical Products (Bedford, MA). Penicillin and ampicillin were

purchased from the Sigma Chemical Company (St. Louis, MO). [t-33P] dATP was

purchased from ICN Biomedicals Inc. (Irvine, CA). The Fisher LeukostatTM stain was

from Fisher Scientific (Lexington, MA). LabTec by Nunc four well chamber slides

were purchased from Fisher Scientific (Lexington, MA). 5-(and-6)-chloromethyl-2'7'-

dichlorodihydrofluorescein diacetate (CM-H2DCFDA) was purchased from Molecular

Probes (Eugene, OR). The Human Stress Toxicology Array and the Atlas Human

Toxicology 1.2 Array were purchased from Clontech (cat#7747-1). All other chemicals

were reagent or molecular biology grade and were obtained from standard commercial

sources.








Antibodies

Polyclonal sheep anti-human EGF receptor antiserum was purchased from

Upstate Biotechnology Inc. (Lake Placid, NY). Polyclonal goat anti-rat CYP1AI and

CYPIBI antisea were purchased from Gentest (Woburn, MA). Monoclonal mouse anti-

human 13-catenin, a-catenin, a5 integrin, and 1 integrin antisera were purchased from

Transduction Laboratories (Lexington, KY). Monoclonal mouse anti-chicken vinculin

antiserum was purchased from Sigma (St. Louis, MO). Monoclonal mouse anti-chicken

actin antiserum was purchased from Amersham Pharmacia Biotech (Piscataway, NJ).

Polyclonal rabbit anti-human PGHS-1 and PGHS-2 antisera were purchased from Oxford

Biochemical (Oxford, Michigan) (Tables 2-1 and 2-2).



Methods

Cell Cultures and Chemical Treatments

The human endometrial adenocarcinoma cell lines RL95-2 (passage 127-147),

HEC-lA (passage 115-135), and HEC-IB (passage 119-139) were obtained from

American Tissue Culture Collection (ATCC). They were maintained in DMEM:HAMS

F-12 (1:1), McCoy's Medium, or Minimum Essential Medium, respectively,

supplemented with 10% (w/v) fetal bovine serum (FBS) in a humidified atmosphere

containing 5% CO2 at 37C. All media contained penicillin and streptomycin at 100

jLg/ml. Media were changed every 2-3 days and cells were maintained at a pH of 7.15-

7.2. For routine cell culture, cells were rinsed with Hanks Balanced Salt Solution

(HBSS) prior to detachment.








All experiments were initiated when cells reached 50-75% confluence. Stock

solutions of BaP and TCDD were initially prepared in DMSO and were added to cultures

with a final concentration in DMSO of 0.1% (v/v). Unless otherwise indicated, cells

were treated with 10 [iM BaP and 10 nM TCDD. Tert-butyl hydroperoxide was added to

cultures with a final concentration of 200 utM in DMSO. DMSO treated cultures served

as vehicle controls for all experiments.



In Vitro Attachment Assay

The Matrigel invasion assay was performed using a modified Boyden Chamber

apparatus (Grant et al., 1987). RL95-2, HEC-1A, and HEC-1B cell cultures were

incubated with 10 iM BaP and 10 nM TCDD for 48 hours, after which cells were

trypsinized and collected by centrifugation at 500 x g for 5 minutes. Cells were

resuspended in Hank's buffer, washed twice, and counted using either a hemocytometer

or coulter counter. Approximately 30,000 cells treated with respective chemicals were

resuspended in 27 g1 of serum-free media and were aliquoted into the lower wells of the

Boyden chamber. The wells were overlaid with a Matrigel-coated polyvinyl-pyrrolidone-

free polycarbonate membrane (Nucleopore, 8 micron diameter pore size for RL95-2 cell

experiments, 10 micron diameter pore size for HEC-1A and HEC-1B experiments) and

the gasket seal and upper chamber attached. The apparatus was inverted at 371C/5% CO2

for 2 hours to allow for cell attachment. The apparatus was then disassembled and

attached cells stained with LeukostatTM (Fisher Scientific, Lexington, MA) and

quantitated by light microscopy. Cells were counted in quadruplicate wells for each

treatment regimen. Images of stained cells were digitally captured using a Zeiss








fluorescence microscope under low power magnification and cells quantitated manually

using Microsoft Paint software. All data were expressed as the mean standard error

measurement (SEM) of the number of migrating cells from three separate experiments.



Immunocytochemistry

Cells were plated on poly-L lysine-coated four-well chamber slides. Cultures

were treated for 48 hours with 10 jtM BaP, 10 nM TCDD, or DMSO vehicle for 48

hours. Cells were fixed in freshly prepared 3.7% paraformaldehyde in modified HBSS

for 10 minutes at room temperature followed by phosphate buffered saline (PBS) washes

(10 mM KPO4, 150 mM NaCl, pH 7.5). Cells were then blocked in 10% normal serum

for 10 minutes at 37*C followed by anti-human EGF-R antibody in PBS/1% BSA (10

jtg/ml) overnight at 4'C or for 1 hour at 37C. In control experiments, the primary

antibody was replaced with PBS alone. Slides were washed in PBS and endogenous

peroxidase activity blocked by incubation with 0.3% H202 in PBS for 10 minutes. Cells

were incubated with rabbit anti-sheep peroxidase-labeled secondary antibody (1:50

dilution; Southern Biotechnology Associates, Birmingham, AL) for 10 minutes followed

by PBS washes. Diaminobenzidine tetrahydrochloride with COC12 enhancement was

used as substrate for visualization. Cells were counterstained with hematoxylin and

mounted with Fluoromount-G prior to visualization.

For phalloidin immunostaining of actin filaments, fixed and permeabilized cells

were incubated with 25 jtl of fluorescein-labelled phalloidin (Molecular Probes) in

PBS/well. No secondary antibody was necessary for visualization.








Light and Confocal Microscopy

Fluorescent light microscopy was utilized to collect images of altered adhesion

molecule and cytoskeletal protein alterations, followed by confocal imaging to determine

cellular protein variations in alternate focal planes. Confocal microscopy is a specialized

technique used to view polarization of cell surface and cytoskeletal proteins in RL95-2

cells by permitting fluorescent molecule visualization in a single plane of focus, thereby

creating an immensely sharper image. Through confocal imaging, optical sections (serial

sections of fluorescent images at different depths of the sample) are pooled into one

three-dimensional image providing information on the distribution of particular protein(s)

within a cell. As a result, confocal imaging allows one to create three-dimensional

images, see surface contour in minute detail, and accurately measure critical cellular

dimensions. Likewise, as compared with standard fluorescence light microscopy,

confocal microscopy allows for the acquisition of bright, three-dimensional, high

resolution and high contrast images, while overcoming the obstacle of depth penetration

which has continually proven a challenge in working with RL95-2 cells.

Fluorescence microscopy was performed on a Ziess Axiophot microscope

equipped with epi-illumination. All confocal microscopy was performed on a Zeiss

Inverted Axiovert 100 M BP microscope using a LSM 510 confocal module and Zeiss

2.01 Proprietary software. FITC imaging was perfomed using a 488 nm excitation argon

laser and rhodamine imaging using a 543 nm helium laser. Images were collected

through the 40x and 100x power lenses and all image analysis was performed using a

Personal Computer with Windows NT software.








Cellular Protein (Membrane and Lysate) Preparations

Cells were rinsed three times in ice cold phosphate buffered saline (PBS) solution

and collected with a cell scraper in 1 ml PBS containing protease inhibitors (1 mM

phenyl-methyl sulfonyl flouride (PMSF), 1 mM Na3VO4, 1 jig/ml aprotinin, 1 gtg/ml

leupeptin, and I tl/ml pepstatin). Cells were then lysed by three freeze-thaw cycles in

liquid nitrogen. The final membrane fractions were obtained by centrifugation at 10,000

x g for 10 minutes at 4C. Pellets containing cellular membranes were resuspended in

300 jad PBS and protease inhibitors.

Alternately, cells were rinsed three times in ice cold PBS and collected in 1 ml

RIPA buffer containing 50 mM Tris, 150 mM NaC1, 1 mM ethylene glycol bis-

N,N,N,Tetraacetic acid (EGTA), 1 % Nonidet P-40 (NP-40), deoxycholic acid Na+ salt

(0.25%), and sodium fluoride (1 mM), pH 7.4, and protease inhibitors as listed above.

Cell solutions were transferred to a 2.0 ml microfuge tube using a 25 gage needle and 1

ml syringe to further lyse cells. Upon 30 minute incubation on ice to ensure lysis, cells

were vortexed, and centrifuged at 10,000 x g for 15 minutes at 4C to separate detergent-

extracted cellular lysates. Protein concentrations in supernatants were determined using

the BCA protein assay (Pierce #23223).



Western Immunoblot Analysis

Cell membrane or detergent-extracted cell lysate protein samples (40 rtg) were

separated by 7.5% or 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis

(SDS-PAGE). Samples were then transferred overnight at 40C to nitrocellulose in 25

mM Tris, 192 mM glycine buffer at pH 8.2, with 20 % methanol according to the method








of Towbin et al. (Towbin et al., 1979). Membranes were blocked in 10% dried milk, tris-

buffered saline-Tween (TBS-T) or phosphate buffered saline-Tween (PBS-T) solution for

2 hours with agitation. Membranes were washed, then incubated in primary antibody

[See Tables 1 and 2 for respective dilutions] in 5% dried milk and TBS or PBS overnight

at 4'C. Following washing, the membranes were incubated in appropriate horseradish

peroxidase (HRP)-conjugated or biotinylated secondary antibody [See Tables 1 and 2] in

5% dried milk and TBS or PBS for 2 hours at 37C with agitation. For experiments

using a biotinylated secondary antibody, the membranes were incubated in the diluted

biotinylated HRP streptavidin complex subsequent to washing for 45-60 minutes at 37C.

Immunoreactive protein levels were determined using the ECL detection method

according to the protocol of Amersham Life Science company.

For EGF-R and CYP1A1 Western immunoblots, immunostaining was performed

according to published protocols (Wang et al., 1988). Membranes were incubated in

sheep anti-human EGF receptor or goat anti-rat CYP1A1, followed by horseradish-

peroxidase conjugated anti-sheep IgG for EGF-R blots and anti-goat IgG for the CYP1A1

blots. Bands were visualized by incubation with 3-amino-9-ethylcarbazole.

For all Western immunoblot experiments, immunoreactive bands were quantitated

by scanning on a Microtek ScanMaker II scanner and quantitation was performed using

NIH image software. Negative controls were run using preabsorption of antibody or

omission of primary antibody to ensure specificity.









Co-Immunoprecipitation of 0-catenin with a-catenin

Immunoprecipitation experiments were performed according to the protocols of

Dr. Allan Parrish (Parrish et aL, 1999), as based on the procedures of Transduction

Laboratories (Lexington, KY). Experiments were performed to co-immunoprecipitate

proteins under native, non-denaturing, conditions. Given the reactivity of antibodies

from the same species, immunoprecipitation experiments were carried using a polyclonal

antibody (goat anti-human P3-catenin from Santa Cruz, CA), followed by Western

immunoblot analysis with the corresponding monoclonal primary antibody for the

complexed protein (mouse anti-human a-catenin from Transduction Laboratories,

Lexington, KY).

Briefly, cells were rinsed with ice cold PBS (pH 7.4), followed by cell lysis with

500 p1 of IX immunoprecipitation buffer (1% Triton X-100, 150 mM NaCl, 10 mM Tris,

pH 7.4, 1 mM EDTA, 1 mM EGTA, pH 8.0, 0.2 mM sodium orthovanadate, 0.2 mM

PMSF, 0.5% NP-40). Samples were mixed by inversion for 15 minutes at 4C, followed

by centrifugation for 20 minutes at 12,000 rpm. Supernatants were collected and protein

concentration determined using the BCA protein assay (Pierce #23223). Cell lysates

were then immunoprecipitated by rocking with appropriate soluble antibody (1 jIg

antibody/100 Vtg protein) in a total volume of 500 ill immunprecipitation buffer for I

hour at 4C. Twenty pi1 of Gamma-Bind Plus-Sepharose or Agarose was added to each

tube and the incubation continued overnight at 4'C. The samples were centrifuged for 15

minutes at 12,000 rpm and the supernatant removed. The immunoprecipitate (pellet) was

washed 3 times in immunoprecipitation buffer, followed by the addition of 30 pA of 2X

sample buffer with 10% P-mercaptoethanol, and samples were then boiled for 5 minutes.








Samples were analyzed for co-immunoprecipitation by SDS-PAGE, followed by Western

immunoblot analysis for the co-immunoprecipitated proteins.



Phosphorylation Experiments

The effects of chemical treatment on P-catenin phosphorylation status were

evaluated by immunoprecipitation for P-catenin protein using the polyclonal goat anti-

human P3-catenin antibody (Santa Cruz, CA), followed by subsequent Western

immunoblot analysis for phosphotyrosine residues using a monoclonal a-human

phosphotyrosine antibody (cat#P 11120, Transduction Laboratories, Lexington, KY).

Experiments were carried out according to the immunoprecipitation protocol previously

described.



Flow Cytometry Analysis with Propidium Iodide Fluorescence

Cells were plated in 10 ml culture dishes and allowed to reach approximately 75%

confluency prior to treatment in regular media. Upon 6, 12, 24, and 48 hours treatment,

cells were collected and washed in ice-cold PBS and final concentration of cells adjusted

to 1 x 106 cells/ml. Cells were then stained with the fluorochrome solution propidium

iodide (PI) in sodium citrate buffer using the CycleTest kit (Becton Dickinson, Mountain

View, CA). The PI fluorescence of individual nuclei of control and treated samples were

measured using a FACSort cell sorter (Becton Dickinson Immunocytometry Systems,

Mountain View, CA) equipped with an ionized argon laser operating at 15 mW output at

488-nm. Red fluorescence due to PI staining of DNA was collected by a 58521 nm

band-pass filter and at least 30,000 cells were analyzed for each sample. The percentages








of cells in GO/G1, S, and G2/M phases were determined by a graphical curve fitting

method using ModFit LT 2.0 software (Verity SoftwareHouse, Inc., Topshan, ME).



Northern Blot Analysis

Total cellular RNA was isolated from cultured cells by acid guanidium

thiocyanate phenol-chloroform extraction according to a published protocol (Xie and

Rothblum, 1991). Poly (A)+ RNA was prepared according to the methods of Celano et

al. (Celano et aL, 1993). For Northern blot analysis, 40 tg samples of total cellular RNA

or 10 pg samples of poly (A)+ RNA were denatured and fractioned on a 1.2%

formaldehyde RNA gel and transferred to MSI Magnacharge membrane by overnight

capillary transfer in 20x SSC solution at room temperature. Upon transfer, the membrane

was dried and auto-crosslinked at 1200 P./s.

Membrane hybridization with a cDNA probe was accomplished by labeling

probes with the Stratagene Prime-it II Random Primer Kit reagents. Upon washing, the

membranes were dried and incubated in Clontech ExpressHyb solution in a

Kapak/Scotchpak Heat sealable pouch at 68'C. Meanwhile, 20 pt1 ddH20, 10 p1l random

primers, and 4 ptl of the cDNA probe were boiled for 5 minutes, then cooled to room

temperature, followed by the addition of 10 p.1 of the 5x dCTP buffer, 5 gt1 32P and 1 l1

Klenow reagent at 37C for 10-20 minutes. Upon incubation, Stop solution was added to

quench the reaction. The purified probe eluents were collected upon passage through a

Stratagene Nuctrap column and 100 p1 sonicated salmon sperm DNA was added.

Following subsequent boiling and cooling, the probes were incubated with the

membranes for 1.5 to 4 hours at 68C. After hybridization, the membranes were washed








with agitation and the presence of specific RNA was detected on each by

autoradiography. All resulting mRNA bands were quantitated by densitometry through

the use of NIH image software.




Ultra-pure RNA Extraction and Poly A+ Purification for cDNA Microarray

Analysis

High purity RNA samples free from DNA contamination were required for cDNA

microarray analysis. High purity RNA extraction from confluent cells was carried using

the AtlasTM Pure Total RNA Labeling System kit (Clontech laboratories, Palo Alto, CA).

All volumes which follow are based on a starting cell number of 1-3 x 107 cells (actual

volumes used were adjusted accordingly), and all steps were carried out on ice or at 40C

to prevent RNA degradation. Following chemical treatment, cultured cells were scraped

in 3 ml of denaturing solution, followed by thorough mixing and centrifugation at 12,000

rpm for 5 minutes. Cellular supernatants underwent two rounds of separate washes in

saturated phenol and chloroform, followed by precipitation of RNA from the aqueous

phase in 6 ml of isopropanol. Samples were then spun at 12,000 rpm for 15 minutes,

pellets washed in 80% ethanol, air dried, and resuspended in RNAse-free water to yield a

final concentration of 1-2 pig/pl. Purified samples were then stored at -70C prior to use.

Subsequently, total RNA samples (0.5 mg) were treated with DNAse I and

samples extracted in saturated phenol/chloroform prior to purification of poly-A+

(Clontech laboratories, Palo Alto, CA). The overall yield and purity of the total RNA

was assessed by measuring absorbance readings at 260 and 280, as well as by comparing

intensities of the 28S and 18S ribosomal RNA bands (4.5 and 1.9 kb) after agarose gel








electrophoresis. The ratio of 260 to 280 intensities was determined for RNA purity, with

acceptable values for samples in the range of 2:1. Total RNA samples were enriched for

Poly A+ RNA using Streptavidin magnetic beads according to the protocol of the AtlasTM

Pure Total RNA Labeling System (Clontech laboratories, Palo Alto, CA).



cDNA Microarray Analyses

The RL95-2 microarray experiments were performed using the Clontech AtlasTM

Human Stress Array (cat#7747-1) and the HEC-IA and HEC-1B experiments using the

Clontech AtlasTM Human Toxicology 1.2 Array (cat# 7859-1). For all experiments, Poly

A+ enriched RNA free from DNA contamination was isolated from treated cells at

selected time points. 33P-labelled cDNA probes were generated by reverse transcribing

each RNA population to cDNA in the presence of [a-33P] dATP using a gene-specific

primer mix as described in the Clontech protocol. Subsequent purification of the

labeled cDNA from unincorporated 33p-labeled nucleotides and small (<0.1 kb) cDNA

fragments was accomplished using column chromatography with NucleoSpin Extraction

Spin columns. The resulting activity of purified probes was determined by scintillation

counting. Poly A+ RNA-derived probes resulting in the range of 5-20 x 106 cpm were

used for analysis.

Each cDNA probe was hybridized overnight to the respective Atlas Array in the

presence of Cot-1 DNA according to instructions outlined in the Clontech User Manual.

After a high stringency wash, the hybridized membranes were exposed to a

phosphorimager screen for up to two weeks, prior to development. Hybridized arrays

were visualized with a Storm Phosphorimager (Molecular Dynamics Inc., Sunnyvale,








CA) at a pixel resolution of 150 microns. Hybridization results were globally normalized

against all genes on the membrane. A set of nine housekeeping controls, including

ubiquitin and P-actin, as well as negative controls, such as pUC 18 and lambda DNA,

were present on each membrane. Simultaneous comparison of the expression levels of

hundreds of genes was made through side-by-side comparisons of hybridizations from

treated and vehicle control cell populations.



cDNA Microarray Data Analysis

Digital images of hybridized arrays were quanitated using Atlaslmage 1.5

software from Clontech. The data was processed by global normalization, using the

value of signal over background for all genes on the arrays for normalization. The

Atlaslmage 1.5 program was then utilized to generate composite arrays for the three

replicate experiments for each designated treatment condition and cell line. The adjusted

intensities of genes on each array were normalized when the composite arrays were

generated for each treatment group, and then while determining the fold change between

treatment groups. Comparison of adjusted average intensity values for each gene to a

diagonal line of "identity" using ScatterPlot analysis in the Prism 2.01 software program

resulted in determination of the fold change cut-off for altered genes in each experiment.

Also, in HEC-1A and HEC-1B experiments, average intensity values were

compared for each gene in control cell lines prior to treatment to determine basic

differences in cellular metabolism enzymes prior to chemical challenge. Genes which

showed an average fold change of 2-fold or greater in their expression were identified

along with functional categories. Subsequently, GenBank and SwissProt, genomic








information public databases, were used to determine function of alternately expressed

genes (Bassett, Jr. et al., 1999).

Since knowledge of where and when a gene is expressed can provide insight into

gene function, the genes analyzed using microarrays were organized on the basis of

similarities in their expression profiles (Bassett, Jr. et aL, 1999). The adjusted average

intensity values prior to normalization for each gene were analyzed for functional

significance by Cluster and TreeView Software programs developed by Dr. Michael

Eisen (http://RANA.stanford.edu/software, Stanford University) (Eisen et al., 1998). The

means and standard deviations were calculated across experiments for each individual

gene on each array, and the half of the total population of genes showing the highest

standard deviations, or greatest deviation from the mean, were selected for analysis.

Variance normalizations were then calculated for the selected half of array genes

according to the following equation:



Variance Normalization = (Normalized Intensity for each gene Mean value)

Standard Deviation



The resulting variance normalization values were then analyzed by binary,

agglomerative, heirarchical clustering using the Cluster and TreeView software programs

(Eisen et al., 1998). Genes were clustered according to common expression patterns

across treatments and cell lines using Average Linkage Clustering, and the clusters were

visualized using interactive TreeView Software. The expression profiles for the top 50

percent of genes were plotted in a hierarchical tree graph according to common








expression patterns, with genes expressing strong similarity being joined by very short

branches, and increasingly dissimilar genes joined by longer branches. Genes with a log

ratio of 0 (unchanged genes) were colored black, genes with increasingly positive log

ratios red (increase) and genes with decreasing log ratios green (decrease) (Eisen et al.,

1998). Through organizing data using inherent, orderly features and graphically

representing the data in a naturalistic manner, the resulting patterns of gene expression

provided indication of the activities of signal transduction pathways within the cells

(Brown and Botstein, 1999). Human toxicology and stress array analysis generated a

unique fingerprint of genetic alterations due to chemical exposure, providing a useful tool

in classifying toxicants based upon their characteristic expression profiles and mehanisms

of action (Bartosiewicz et al., 2001 a; Bartosiewicz et al., 2001 a).



Quantitation of Intracellular Oxidative Stress by Dichlorofluorescein Assay

RL95-2 cells were plated into 96-well plates (Coming) 1 day prior to

experiments. The following day, cells at approximately 70% confluence were treated

with respective chemicals (10 jtM and 20 jiM BaP, 10 jtM and 20 j.M BeP, 10 nM

TCDD, 1 jtM, 10 jiM, and 200 jiM T-butyl hydroperoxide, and 0.1 % v/v DMSO) in

regular, 10% serum-containing medium. Following 6, 24, and 48 hours treatment, the

cells were washed twice in PBS and were incubated with the identical chemical

treatments in PBS in the presence of 10 jtM 5-(and-6)-chloromethyl-2'7'-

dichlorodihydrofluorescein diacetate (CM-H2DCFDA) for 30 minutes at 370C.








The fluorescence of the cells from each well was measured and recorded using a

FL600 Microplate fluorescence reader (Bio-Tek Instruments) with constant temperature

at 370C. The excitation filter was set at 485 10 nm and the emission filter was set at

530 12.5 nm. The fluorescence from each well was captured, digitized, and stored on a

computer using KC4 software (Bio-Tek Instruments). Data points were taken every 5

minutes for 30 minutes and the data were exported to Excel spreadsheet software

(Microsoft, Seattle, WA) for analysis. The net fluorescence was calculated for each

treatment relative to background controls (chemicals in the presence or absence of CM-

H2DCFDA). Data were expressed as the net change in fluorescence for each treatment.




Data Analysis

All quantitative experiments were performed in triplicate for each experiment.

For scanned image analysis, control lanes were standardized to 100% and treatments

were assessed relative to controls for each. The paired Student's t-test was used to

analyze numerical data averaged over several experiments. All statistical analyses were

performed using Microsoft Excel software, with the exception of cDNA microarray

experiments, which were analyzed using Clontech Atlaslmage software, as well as

Cluster and Treeview interactive software.



Potential Hazards and Precautions

The experiments herein described contain numerous potential hazards which

could pose a risk for personnel involved. Such hazards include the use of dangerous

chemicals, cancer cell lines, and radioactivity. All chemicals used were handled with





35


extreme care and disposed of as described by the Chemical Waste Management Guide of

the University of Florida. Cancer cell lines were treated as biohazards and were handled

with extreme precaution and disposed in bleach/water solution. Radioactivity was

similarly handled according to the methods outlined by the University of Florida Division

of Environmental Health and Safety Department of Radiation Control and Radiological

Services. All laboratory personnel were certified in the use of hazardous chemicals and

radioactivity by the University of Florida.









TABLE 2-1: Antibody Sources and Dilutions for Cellular Adhesion Protein Western Blots and Immunocytochemistry

Target Protein/source Molecular Weight Western Blot: Immunocytochemistry:
TargetProtein/sour _e Molecular _Weight Primary/Secondary Antibody Primary/Secondary Antibody


Cadherin (W. blot)
Santa Cruz, #sc- 1499
1-Catenin
Transduction Labs #c 19220
a-Catenin
Transduction Labs #c21620
Actin
Amersharn # N-350
Actin (Phalloidin)
Molecular Probes
Vinculin
Sigma #Vin 11-5
a5 Integrin
Transduction Labs #i55220
31 Integrin
Transduction Labs #i41720
EGF-R
Upstate Biotech. #06-129


120 kDa

92 kDa

120 kDa

42 kDa

N/A

130 kDa

150 kDa

140 kDa

170 kDa


1:160 / 1:2000

1:500 / 1:2000

1:250 / 1:2000

1:1000 / 1:2000

N/A

1:1000 / 1:2000

1:5000/ 1:2000

1:2500/ 1:2000

1:1000 / 1:2000


___________________ 1 ______________ 1 ______________________ 1~


N/A

10 lpg/ml / 1:2000

10 pg/ml / 1:2000

N/A

25 pl/well / no secondary

N/A

N/A

N/A

1:100 / 1:100











TABLE 2-2: Antibody Sources and Dilutions for Cellular Enzyme Western Blots


Western Blot: Western Blot:
Target Protein/source Molecular Weight Primary Antibody Secondary Antibody
CYP1A1
Gentest # 210105 (MlIlR 55kDa 1:500 1:1000
ctrl)
CYP1BI 50 kDa 1:500 1:500
Gentest #A21 1 (M120R ctrl)
PGHS-1 70 kDa 1:100 1:1000
Oxford Biochem. #PG 16
PGHS-2 72 kDa 1:1000 1:1000
Oxford Biochern















CHAPTER 3
EFFECTS OF BAP AND TCDD ON CELL ATTACHMENT, INVASION,
AND EXPRESSION OF ADHERENS JUNCTION, CYTOSKELETAL, AND
GROWTH FACTOR RECEPTOR PROTEINS IN RL95-2 CELLS


Introduction

Evidence from epidemiological and laboratory studies support a role for several

environmental toxicants in the etiology of endometrial cancer and endometriosis. The

environmental chemicals benzo(a)pyrene (BaP) and 2,3,7,8-tetrachlorodibenzo-p-dioxin

(TCDD) have been shown to produce adverse reproductive effects in animals. Studies in

human populations indicate that women who smoke cigarettes have less than half the risk

of developing endometrial cancer of non-smokers, and maintain similar decreased

incidence rates for endometrial hyperplasia, uterine fibroids and endometriosis (Baron et

al., 1990; Cramer et aL, 1986; Matorras et al., 1995). In contrast, recent studies in rhesus

monkeys and rodents indicate that TCDD may promote endometriosis (Cummings et al.,

1996; Cummings and Metcalf, 1995; Rier et al., 1993). In addition, human

epidemiological studies provide some evidence for an association between dioxin

exposure and the promotion of uterine disease, including carcinogenesis (Koninckx et al.,

1994; Bertazzi et al., 2001; Mayani et al., 1997). Uterine cancer, however, was not

shown to be increased by dioxin exposure in human epidemiological studies up to twenty

years from the time of exposure (Bertazzi et al., 2001).








The present studies were conducted to test the hypothesis that BaP, and not

TCDD, decreases cellular attachment and invasion through alterations in membrane

adherens junction and cytoskeletal proteins, thereby producing an anti-adhesive

phenotype in RL95-2 cells. Benzo(a)pyrene, a major toxicant of cigarette smoke, was

selected as a target for investigation due to its predominance and potency. A genotoxic

chemical, BaP is recognized to significantly alter signal transduction in human and

animal cells by binding to the intracellular cytosolic protein aryl hydrocarbon receptor

(AhR). BaP transcriptionally activates cytochrome P450 IA1 (CYP1AI), leading to its

own metabolism (Poland et aL, 1976; Whitlock, Jr. et al., 1996). TCDD, a classical, non-

genotoxic carcinogen, binds to the AhR in affecting cellular proliferation and division.

Current studies were undertaken to evaluate the effects of the AhR-ligands BaP

and TCDD on the ability of RL95-2 cells to attach to matrigel-coated membranes, and on

key adherens junction, cytoskeletal, and growth factor proteins involved in these

processes. Cell adhesion mechanisms play a fundamental role in the determination of

tissue architecture and the functions of cell assembly and connection to the internal

cytoskeleton and, consequently, are of great significance to the pathophysiology of

endometrial cancer. The well-studied human endometrial RL95-2 cell line was selected

as a model for these studies because of its relevance as an in vitro model for endometrial

cancer (Way et al., 1983). In addition, the RL95-2 cell line has been shown to exhibit

adhesiveness of its apical pole for the trophoblast, thereby also serving as a model for

human uterine epithelium receptive for implantation (Thie et al., 1997).








Results

Cellular Attachment on Matrigel

Cells were cultured in the presence of 10 RtM BaP, 10 nM TCDD, or DMSO

vehicle for 48 hours, then trypsinized and applied to Matrigel-coated porous membranes.

All treated cells exhibited greater than 95% viability as determined by tryphan blue dye

exclusion. Pretreated cells were applied to the Matrigel surface for 2 hours, after which

the membranes were stained for adherent nuclei. Figure 3-1 shows that BaP-pretreated

cells exhibit minimal attachment, whereas TCDD-exposed cells show a high level of

attachment comparable to DMSO controls.



Effects on Epidermal Growth Factor Receptors (EGF-R)

EGF-R immunoreactivity was detected as brown immunostaining primarily

localized in intercellular regions of adjacent cells and associated with the plasma

membrane, with a lesser amount detected in the cytoplasm (Figure 3-2A). Treatment

with 10 RtM BaP for 48 hours resulted in a marked decrease in EGF-R staining along cell

membranes, whereas cytoplasmic staining was still present (Figure 3-2B). In contrast, 10

nM TCDD treatment resulted in no apparent effect in EGF-R localization, relative to

control (Figure 3-2C). Replacement of the primary antibodies with PBS showed a

complete lack of EGF-R immunostaining (Figure 3-2D).

Western immunoblot data showed the expression of EGF-R as a 170 kDa protein

in RL95-2 cells in both cell membrane and cell lysate fractions. Data further indicated a

selective loss of plasma membrane EGF-R following 10 jiM BaP treatment, with no

change in EGF-R protein levels in detergent extracted cell lysates (Figure 3-2E).








Treatment with 10 nM TCDD failed to show an effect on EGF-R expression levels. In

data not shown, 10 ptM BaP caused a substantial loss of membrane EGF-R protein levels

by 7 hours treatment.



Effects on Cadherin and fB-catenin Cellular Adhesion Molecules

Studies have been conducted on the cell adherens junction proteins cadherin, B-

catenin, and vinculin. Data in Figure 3-3A show that cadherin is expressed in RL95-2

cells as a 120 kDa protein. Treatment with 10 gM BaP, but not 10 nM TCDD, produced

a significant 38% decrease in cadherin levels in RL95-2 cell lysates (p < 0.005). 13-

Catenin, a second adherens junction protein, is detected as a 92 kDa protein present in

both cell membrane and detergent-extracted cell lysate preparations of RL95-2 cells.

Treatment with 10 IiM BaP is associated with a selective loss of B-catenin in cell

membrane fractions, a significant 80% decrease as compared to control, whereas cell

lysate protein levels remain unchanged. In contrast, treatment with 10 nM TCDD had no

effect on B-catenin levels in either preparation (Figure 3-3B). In comparison, 10 p.M BaP

and 10 nM TCDD treatments had no effect on vinculin protein levels in cell membrane

and lysate fractions as determined by Western immunoblot analysis (Figure 3-3C).



Effects on Actin Cytoskeletal Protein

Subsequent immunocytochemical studies were performed investigating BaP-

mediated effects on actin cytoskeletal localization in RL95-2 cells. Experiments utilized

the fluorescently-labelled peptide phalloidin to selectively visualize filamentous actin

using fluorescence light microscopy. Results indicate that filamentous actin is localized








in a subcortical layer in control cells (Figure 3-4A). In contrast, 10 jiM BaP treatment is

associated with the formation of subcortical actin aggregates in RL95-2 cells (Figure 3-

4B). At the same time, however, overall actin levels in cell membrane and triton X-100

fractions remain unaltered as demonstrated by Western immunoblot analysis of the 42

kDa actin protein in RL95-2 cells (Figure 3-4C).



Discussion

Endometrial cancer is the most common gynecologic malignancy in the United

States. Due to recent evidence implicating environmental agents like TCDD and

cigarette smoke in the etiology of endometrial cancer and endometriosis (Matorras et aL,

1995; Cummings et aL, 1996; Rier et aL, 1993; Mayani et aL, 1997), current studies

investigated roles of these environmental agents in altering cellular attachment of

endometrial cells.

The present study demonstrates that BaP produced a significant decrease in the

cell attachment of the RL95-2 endometrial cell line. In contrast, TCDD exposure did not

significantly affect the attachment of RL95-2 cells. The observation that BaP inhibits

endometrial cancer cell attachment is potentially indicative of a conversion to a more

metastatic phenotype. Since increased metastasis is associated with a poorer prognosis in

endometrial cancers (Wronski et al., 1993), the decreased cellular attachment associated

with BaP exposure cannot account for the seemingly "protective" effect of cigarette

smoke (Cramer et al., 1986; Matorras et al., 1995). The reported effects of BaP on

inhibiting cellular invasion and proliferation could, however, potentially account for the

seemingly protective effects against endometrial cancer (Charles, 1997). These data are








strengthened by reports of BaP having independent effects on growth-related gene

expression and signaling compared with effects on acute liver cytotoxicity (Parrish et aL,

1999).

Studies indicate that BaP might act to alter cellular attachment through a change

in the expression of cell adhesion molecules. The BaP-mediated alterations in

endometrial cancer cell attachment appear to be linked with a loss of cell surface

expression of adherens junction and cytoskeletal proteins, such as cadherin, B-catenin,

and actin. Cadherins are integral membrane glycoproteins which function in epithelial

cells to form calcium-dependent linkages between cells (Potter et al., 1999; Knudsen and

Soler, 2000). Cadherins play a key role in mediating the formation and breakage of cell-

to-cell contacts and in maintaining strength in cellular adhesion through homophilic

interaction of their extracellular domains, allowing for cellular aggregation to occur

(Potter et al., 1999). Cadherins connect to B-catenin, a key membrane-associated protein

responsible for the colocalization of cadherins to sites of cell-cell contact with the actin

cytoskeleton (Potter et al., 1999). The BaP-mediated alterations in the membrane

associated adherens junction proteins indicate the likely role of BaP in decreasing cellular

attachment in RL95-2 cells. A phenotype with decreased attachment has been associated

with increased cellular metastasis (Wronski et al., 1993).

Although the present studies provide useful information about BaP-mediated

membrane P3-catenin effects, they do not address whether there is a true

compartmentalization of membrane-associated f3-catenin after BaP treatment since cell

lysate preparations currently described have centrifuged out detergent insoluble proteins.

Future experiments should address BaP-mediated effects on cell membrane proteins from








whole cell lysate preparations prepared from lysed cells without centrifugation to

determine whether the loss of membrane-associated P-catenin is compensated for by

increases in cytoplasmic P3-catenin. It can be speculated that if BaP treatment is causing a

redistribution of 1P-catenin to the cytoplasmic and nuclear fractions, where 13-catenin can

initiate signal transduction pathways leading to carcinogenesis, BaP exposure is thereby

not protective against endometrial cancer.

The actin cytoskeleton was additionally investigated for BaP effects due to its

fundamental role in cellular adhesion. Actin works in conjunction with cytoskeletal

microtubules and intermediate filaments in performing essential functions in locomotion

and cytokinesis (Hulka and Brinton, 1995). Actin bundles are attached to integral

membrane cadherin and integrin proteins, adapter proteins and the contractile bundle.

Recent studies have shown that treatment of MCF-1OA nontransformed human mammary

epithelial cells with BaP and UV light resulted in the reorganization of actin filaments

into substrate-associated aggregates (Seagrave and Burchiel, 2000). Similar alterations in

filamentous actin conformation following BaP, yet not TCDD treatment, were observed

in RL95-2 cells, reflecting another fundamental structural alteration with BaP exposure.

Activation of epidermal growth factor receptor (EGF-R), as well as other receptor

tyrosine kinases, has been shown to directly affect the adhesive function of cells (Hazan

and Norton, 1998). In particular, EGF-R has been shown to directly regulate cell-cell

adhesion through effects on E-cadherin interactions with actin in a human breast cancer

cell line (Hazan and Norton, 1998). Cigarette smoking has been consistently associated

with a decrease in placental EGF receptors (Shiverick and Salafia, 1999). It has also

been reported that both BaP and TCDD mimic growth factor signaling pathways in








human mammary epithelial cells, as shown by increased tyrosine phosphorylation of

insulin-like growth factor (IGF-I) receptor beta, IRS-I and Shc (Tannheimer et aL, 1998).

Studies in three endometrial cancer cell lines indicate an inverse relationship between

cellular EGF-R levels and the grade of the tumor (Lelle et aL, 1993). The marked loss of

cell surface-EGF-R at the cell-cell interface of RL95-2 cells in the present study

represents the dramatic alterations in the adherens complex upon BaP treatment. The

BaP-specific loss of expression of EGF-Rs in RL95-2 cells is in agreement with

previously published reports of BaP-, yet not TCDD-mediated loss of EGF-Rs in human

placental choriocarcinoma JEG-2 cells (Zhang and Shiverick, 1997).

Published studies indicate that the first step of invasion and metastasis is the

detachment of cancer cells from the primary tumor, a process mainly controlled by the

adherens junction proteins, consisting of E-cadherin, a- and P-catenins, vinculin, and

actin. Consequently, it can be inferred that the decreased cellular attachment of the

RL95-2 cells upon BaP exposure is consistent with promotion of a more metastatic

phenotype which would actually promote, rather than protect against, endometrial cancer.

This view is supported by studies from human breast cancer patients which show that

tumors negative for either cc- or P-catenin expression demonstrate a higher incidence of

distant metastasis than those expressing both catenins (Yoshida et al., 2001). Further, a

reduction or loss of E-cadherin expression has been associated with lymph node

metastases and poor prognosis of invasive breast cancers (Yoshida et aL, 2001).

Additional studies from colorectal cancer patients indicate an association between

decreased ot-catenin expression and increased metastasis (Gofuku et aL, 1999).

However, a conflicting report from pancreatic cancer patients found an association








between an intact E-cadherin/catenin complex and increased liver metastasis (Gunji et

al., 1998).

The RL95-2 endometrial carcinoma cell line, which has characteristics distinct

from other endometrial cell lines, exhibits an adhesiveness of its apical pole for tropho-

blast cells and thereby serves as an in vitro model for the human uterine epithelium

receptive for implantation (Thie et al., 1997). BaP exposure therefore likely inhibits

attachment in benign uterine disorders, while additionally proving unfavorable for

implantation and the establishment of pregnancy.





















Control
*i...!


10 gM
BaP


10 nM
TCDD

40 0


Figure 3-1. Effects of BaP and TCDD on RL95-2 cell attachment to membranes.
Cells were pre-treated for 48 hours with 10 gtM BaP, 10 nM TCDD, or 0.1% DMSO
vehicle (control). Cells (30,000 per well) in serum-free media were aliquoted into the
lower wells of the Boyden Chamber apparatus and were allowed to attach to a Matrigel-
coated membrane for 2 hours. Upon attachment, cells were stained with Leukostat and
counted by light microscopy.






















Figure 3-2. Effects of BaP and TCDD on the localization and protein levels
of EGF-R in RL95-2 cells.
A-D: Immunocytochemical localization of EGF-R in cells following 48 hours
treatment with DMSO vehicle control (A and D), 10 jiM BaP (B), and 10 nM
TCDD (C). Cells were incubated with sheep anti-human EGF-R antiserum (A-
C) or PBS (D). E: Western Immunoblot analysis of cell membrane preparations
compared with detergent-extracted cell lysates of cells following BaP and TCDD
treatment. Samples containing 100 jig protein were electrophoresed, transferred
to nitrocellulose, and immunostained with anti-EGF-R antibody.

















DMSO CONTROL


NEGATIVE CONTROL


CELL MEMBRANES


CELL LYSATES


CONTROL TCDD BaP


CONTROL TCDD BaP


ICDD


doainrto I


BaP


I loolow,,von
















Cadherin


BaP


RL95-2 Cell Lysates


Cadherin
120 kDa


DMSO


BaP TCDD


Figure 3-3. Western Immunoblot analysis of the effects of BaP and TCDD on
Cadherin, f-Catenin, and Vinculin Levels.
RL95-2 cells were treated with 10 pM BaP, 10 nM TCDD, or 1% v/v DMSO control for
48 h. Membrane and detergent-extracted cell lysate preparations were electrophoresed,
transferred to nitrocellulose, and immunostained with antibodies to cadherin, B-catenin,
or vinculin, followed by horseradish peroxidase conjugated IgG. A. Results from
cadherin Western immunoblot analysis represents the mean SE of three experiments
(p

*1~


120
100
80
60
40
20
0


DMSO


TCDD


















120-
100-
80-
60-
40-
20-
0-


DNESO


B-Catenin











B(a)P


P-Catenin
92 kDa


Membranes Lysates
DMSO BaP TCDD DMSO BaP TCDD


amw


Figure 3-3-continued. B. Data for P3-catenin Western Immunoblot analysis represents
the mean SE of three experiments (p<0.05).


TCDD
















Vinculin


DMSO B(a)P


Membranes


DMSO BaP


TCDD DMSO BaP


Figure 3-3-continued. C. Data for vinculin Western immunoblot analysis represents
the mean SE of three experiments.


250 -
200 -
150 -
100 -
50 -


0 -


TCDD


Lvsates


Vinculin
130 kDa


TCDD






















Figure 3-4. Effects of BaP on the localization of actin filaments and protein levels
in RL95-2 cells.
A-B: Immunocytochemical localization of actin in RL95-2 cells following 48 hours
treatment with DMSO vehicle control (A) and 10 gtM BaP (B). Cells were incubated
with fluorescently labeled phalloidin to visualize altered actin structure upon BaP
treatment. C: Western Immunoblot analysis of cell membrane and cell extract
preparations following DMSO, BaP, and TCDD treatment. Samples containing 40 g
of protein were electrophoresed, transferred to nitrocellulose, and immunostained with
anti-actin antibody.









































Membranes Triton-X 100 Extracts
DMSO BaP TCDD DMSO BaP TCDD
Actin
42 kD a o .. .... .. moo














CHAPTER 4
SIGNAL TRANSDUCTION PATHWAYS FOR BAP AND TCDD EFFECTS
ON RL95-2 CELLS: THE AHR, CELL CYCLE, AND OXIDATIVE STRESS


Introduction

BaP and its metabolites are known to modulate mammalian gene expression

through alterations in cell cycle and epigenetic mechanisms involving the AhR signaling

pathway, oxidative stress, and altered mitogenic signaling. In particular, oxidative

metabolism of BaP is associated with most adverse health effects of BaP. Further, DNA

adducts can result in deleterious genetic mutations, leading to disruption of gene

expression.



The Arylhydrocarbon Receptor

Benzo(a)pyrene (BaP) is a ubiquitous environmental pollutant heavily implicated

in human cancer etiology. Whereas the parent compound BaP is not directly biologically

active, BaP is readily metabolized to mutagenic and carcinogenic metabolites (Kim et al.,

1998). Consequently, BaP is considered a procarcinogen because it requires metabolic

activation to reactive intermediates to elicit toxic effects. The enzymes involved in BaP

metabolism include Phase I enzymes, such as cytochrome P450s, epoxide reductases, and

epoxide hydrolases, as well as Phase II enzymes, such as glutathione transferases, UDP-

glucuronyl transferases, and suflotransferases (Miller and Ramos, 2001 a). The primary








products of BaP metabolites include epoxides, dihydrodiols, phenols, and quinones

(Miller and Ramos, 2001 a).

Xenobiotics such as BaP and the dioxin TCDD bind to the cytosolic

arylhydrocarbon receptor (AhR) protein in mediating gene transcription. The AhR is a

member of the bHLH/PAS transcription factor family characterized by the basic helix-

loop-helix DNA binding domain and a Per/AhR/Amt/Sim homology region (Burbach et

al., 1992). In the process of binding to xenobiotics, the AhR dissociates two heat-shock

90 proteins and the BaP-AhR complex translocates to the nucleus where it binds to the

Ah receptor nuclear translocator (Amt) protein. Subsequent binding of the xenobiotic-

AhR-Arnt complex to the xenobiotic response elements (XRE) on the CYPIA1 gene

promoter locally modifies the chromatin structure and activates gene transcription

(Whitlock, Jr. et aL, 1996).

In vivo studies of AhR gene expression in the human uterine endometrium

indicate that AhR mRNA levels were significantly lower in women who smoke cigarettes

than in non-smokers (Igarashi et al., 1999). However, studies indicate that no significant

difference was observed in the AhR mRNA levels in human uterine endometrium in

women with or without endometriosis (Igarashi et aL, 1999). Further, the levels of

CYP1A1 transcripts have been shown to be a marked 8.7 times higher in endometriotic

tissues than in the eutopic endometrium (Bulun et al., 2000).

Cytochrome P450 (CYP) enzymes including CYP1A1 catalyze the reactions

leading to the oxidation of endogenous or xenobiotic substrates. Among the CYP

enzymes, numerous genes are highly inducible by xenobiotics, including CYPlA1,

CYP1A2, and CYP1B1. CYPIA1 is expressed in numerous tissues after induction by








polycyclic aromatic hydrocarbons (PAHs) such as BaP, and polyhalogenated

hydrocarbons such as TCDD (Barouki and Morel, 2001; Omiecinski et al., 1999).

CYP1A2 is expressed constitutively in the liver, and is induced through the AhR pathway

(Omiecinski et al., 1999). The enzyme CYP1B1 is constitutively expressed in many

tissues and is inducible through the AhR pathway (Omiecinski et al., 1999). CYPIBI

has been shown to be an important activator of PAHs in tissues such as the mammary

gland when environmental chemical exposures minimally induce CYP I Al (Larsen et al.,

1998). Further, CYPIBI has been shown to carry out the metabolism of BaP to BPDE at

higher rates than CYP1A2, but slower than CYP1A1 (Kim et al., 1998).

Whereas both BaP and the dioxin TCDD are potent AhR ligands known to induce

CYPlA1 and CYPIBI, TCDD, yet not BaP, exerts toxic effects on cells in the absence of

prior metabolism of the parent compound. Recent knock-out studies indicate that AhR

activation is essential to TCDD-mediated toxicity. Data show that in the absence of a

functional AhR, most toxicity associated with TCDD is not observed (Fernandez-

Salguero et al., 1996). BaP, on the contrary, can exert toxic effects through both AhR-

dependent and AhR-independent pathways (Kim et al., 1997; Mamett, 1990).



Oxidative Stress and Cell Cycle

In 1953, Howard and Pelc demonstrated that DNA synthesis occurs at a precise

period during the cell cycle, designated the S phase (Howard and Pelc, 1953). The period

before the S phase became known as the Gap 1 (GI) phase, whereas the period

subsequent to the S phase and before mitosis became known as the Gap 2 (G2) phase.

The modem concept of the cell cycle emerged with identification of distinct GI, S, G2,








and M (mitosis) phases. Subsequently, it became known that dividing cells contained

factors which ensure orderly progression and timing throughout the cell cycle, and that a

lack of genomic integrity can result in cell death or the accumulation of deleterious

mutations.

While cells typically proceed throughout the cell cycle without interruption, DNA

damaged cells have the ability to pause in the GI, S, or G2 phase to allow time for DNA

repair. In cases of severe DNA damage, cells may undergo apoptosis or enter into an

irreversible GO state (Shackelford et aL, 2000). Cell cycle checkpoints are crucial to

allow cells more time to pause and repair DNA damage before progression through the

more critical phases of DNA replication and mitosis. Although cells are highly sensitive

to DNA damage throughout the cell cycle, once they pass the restriction point, their

checkpoint pause is delayed until the subsequent phase of the cell cycle (Shackelford et

aL, 2000). Defects in cellular checkpoints have been observed in hereditary cancer

syndromes and in early stages of cellular transformation (Kaufinann and Paules, 1996).

Regulation of cell cycle progression is accomplished through complex pathways

involving cyclin accumulation and degradation, phosphorylation of cyclin dependent

kinases (Cdks), cyclins, and other proteins, regulation of cyclin/Cdk dimerization, and the

binding of Cdk inhibitory proteins (Shackelford et al., 2000). Unrepaired DNA damage

from the GI phase, as well as S phase damage, can cause a checkpoint response which

lowers the rate of DNA synthesis in cells. The S phase checkpoint response is known to

be a rapid biological response, similar to the G2 checkpoint, and to be far more sensitive

to DNA damage than the GI checkpoint (Kaufmann and Paules, 1996). Cyclin A / Cdk2

activity is known to be a requirement for S phase progression and DNA synthesis, and








consequently functions in protecting cells against DNA replication errors (Shackelford et

aL, 2000).

Certain DNA damaging agents have been shown to induce the S-phase checkpoint

response in cells, including infrared (IR), ultraviolet (UV-B and UV-C), methylmethane

sulfonate, and benzo(a)pyrene diolepoxide (BPDE) (Shackelford et al., 2000). Evidence

demonstrates that the BPDE metabolite of BaP forms a DNA adduct with the N2 nitrogen

on a guanine base in DNA, causing cells to accumulate in the S phase of the cell cycle

(Gezer et al., 1988). Whereas some adducts can be removed during transcription,

unpaired adducts in the coding strand of transcriptionally active genes can block RNA

polymerase II and may interfere with gene synthesis (Kauflnann and Paules, 1996). The

toxicity of adducts progresses as they persist throughout a cell's progression through the

cell cycle. For example, adducts formed during the GI phase typically lack permanent

genotoxicity provided they are repaired before the S phase, whereas S phase adducts have

been shown to block DNA synthesis, leaving gaps in daughter DNA (Kaufnann and

Paules, 1996). As a result, base substitution errors can occur during postreplication repair

and damage can proceed to a double strand break in the DNA. Incorrect rejoining of the

double strand break can produce genetic deletion, translocation, or amplification through

chromosomal aberrations (Kauflnann and Paules, 1996). In brief, errors in repair of DNA

damage at any stage can produce far more permanent damage than errors at a preceding

stage.

Studies indicate that increases in oxygen pressure or reactive oxygen species

(ROS) can cause mutations, chromosomal and DNA damage, inhibition of cellular

division, and tumor promotion (Figure 4-1) (Shackelford et al., 2000). Further, the AhR








battery of genes have been shown to be capable of both producing and preventing

oxidative stress through stress-detoxifying enzymes, as well as to assist cells in choosing

between apoptosis and continuation throughout the cell cycle (Nebert et al., 2000). It is

believed that dioxin, the most potent AhR ligand, activates oxidative stress genes

primarily through the AhR-mediated aromatic hydrocarbon response element (AhRE)

DNA motif, whereas electrophiles and certain BaP metabolites additionally activate

genes through the electrophile response element (EpRE) (Nebert et al., 2000).

Although BaP, an AhR ligand, is a relatively unreactive 5-ring polycyclic planer

hydrocarbon, this compound is readily metabolized to toxic or tumorigenic metabolites.

BaP contains the structural motif known as a "bay region" which is responsible for steric

hindrance in the molecule, thereby increasing metabolism through oxidation and radical

formation, while decreasing detoxification and conjugation. Through metabolic

transformation, BaP is converted to a variety of products which can exert cellular effects

by directly binding DNA, perhaps the most noted being benzo(a)pyrene 7,8-diol-9,10-

epoxide (BPDE) (Miller and Ramos, 2001a), as well as the various quinone and

semiquinone metabolites formed during redox cycling (Joseph and Jaiswal, 1994).

Studies in a non-human primate model indicate high levels of BaP-induced adduct

formation in maternal and fetal organs throughout gestation (Lu et al., 1993).

Furthermore, human epidemiological studies indicate the presence of BaP-adducted bases

in the urine of three of seven cigarette smokers and three of seven women exposed to coal

smoke, whereas no BaP-adducted bases were observed in non-smoking women (Casale et

al., 2001). TCDD, on the contrary, has been associated with cell cycle arrest in the Gi








phase of the cell cycle in human hepatoma cells through a mechanism known to involve

the AhR (Ge and Elferink, 1998).



Prostaglandin H-Synthase

Prostaglandin H-synthase (PGHS) is a membrane protein localized in the

endoplasmic reticulum, the nuclear membrane, and the plasma membrane which

functions in the oxidation of xenobiotics in the presence of arachadonic acid or lipid

peroxides (Degen, 1993). PGHS consists of two primary components, a cyclooxygenase

which generates the peroxide PGG2 and a peroxidase that reduces PGG2 to PGH2. The

PGHS-mediated oxidations can occur by three mechanisms: direct oxidation of the

carcinogen by the PGHS peroxidase, oxidation from peroxyl radicals generated during

prostaglandin biosynthesis, and bioactivation through a second oxidant species formed

from the metabolite derived by the peroxidase (Eling and Curtis, 1992). PGHS plays a

crucial role in cellular metabolism, complementing the cytochrome P450 isoenzymes in

metabolic bioactivation reactions, as well as providing an alternate metabolic pathway for

xenobiotic metabolism from the P450-mediated pathways (Degen, 1993). Specifically,

PGHS has been shown to contribute to the bioactivation of various procarcinogens,

including benzo(a)pyrene-7,8-diol, in extrahepatic tissues where the P-450

monooxygenases have low activity (Marnett, 1990; Kelley et al., 1997). Consequently,

PGHS likely does not play a role in systemic drug metabolism, but is substantially

involved in organ and cell specific toxicities (Degen, 1993).








Oxidative Stress and Cellular Adhesion

It has been shown that oxidative injury in cultured cells is capable of markedly

impairing cellular adhesion, in part through alterations in cellular membranes,

mitochondria, and the cytoskeleton (Bellomo et aL, 1990). The mechanism for oxidative

stress-mediated cellular alterations is believed to involve oxidative modification of

cytoskeletal protein sulfhydryl groups, thereby impairing the cytoskeletal network

(Bellomo and Mirabelli, 1992). In particular, studies have shown that cells treated with

the oxidative-stress-inducing compounds have markedly disrupted cytoskeletal structures

(Bellomo et al., 1990). Since the cytoskeleton is crucial to the structural and metabolic

function of the cell, disruptions in the cytoskelton can have a profound inpact on cellular

response to toxicants. Cytoskeletal structures such as actin microfilaments, microtubules,

and intermediate-size filaments are considered significant targets in quinone-induced

oxidative stress (Bellomo et al., 1990). Current studies were undertaken to determine the

role of the various BaP metabolism enzymes on the observed cellular adhesion and

cytoskeletal effects of BaP.



Results

Cell Cycle Phase Distribution

Cells were cultured in the presence of 10 gtM BaP, 10 nM TCDD, or 0.1% v/v

DMSO vehicle control for 48 hours and then trypsinized, washed, and stained with

propidium iodide prior to analysis for alterations in cell cycle phase distribution. Data in

Figure 4-2 indicate that control RL95-2 cells have a large percentage of cells in the GO-

GI phases of the cell cycle (74 %), with considerably less cells in S (20 %) and G2-M








phases (6 %). BaP treatment produces a 2 fold increase in the percentage of cells in the S

phase, as compared to DMSO vehicle controls, with a marked 50% decrease in the

percentage of cells in the GO-Gi phases (Figure 4-2). In contrast, TCDD-treated cell

cultures appeared similar to controls after 48 hour treatment (Figure 4-2). Data from a

representative experiment indicating BaP and TCDD effects on cell cycle phase

distribution are shown in Figure 4-3. Results indicate that treatment with BaP, but not

TCDD, altered cell cycle phase distribution in the direction of increased S and G2M

phase accumulation, with a corresponding marked decrease in the GO-G1 phases.



Effects on CYP1A1 and CYP1B1 mRNA Levels

Since both BaP and TCDD have been shown to act through the aryl hydrocarbon

(Ah) receptor, initial studies were performed investigating the induction of CYP1A1 and

CYP1B1 mRNA. Studies evaluated the effects of 48 hour treatment with 10 jtM BaP and

10 nM TCDD on cytochrome P450 CYPlA1 and CYPIBI mRNA levels. Northern blot

data indicate that control RL95-2 cells express low levels of a 3.0 kb CYP1A1 mRNA

transcript (Figure 4-4A). BaP and TCDD treatment of RL95-2 cells results in a strong

induction of CYPlA1 mRNA. Further, data indicate that RL95-2 cells express a

constitutively low level of a 5.2 kb CYPIBI mRNA transcript, which is markedly

induced by both BaP and TCDD (Figure 4-4 B).



Effects on CYPlAl, PGHS-1, and PGHS-2 Protein Levels

Experiments investigating the effects of BaP- and TCDD-treatment on CYP1A1,

PGHS-1 and PGHS-2 protein levels were performed in RL95-2 cells after 48 hour








treatment. Western immunoblot analysis indicates that CYP1Al protein is not detected

in control cells, but is expressed as a 55 kDa protein with 10 pM BaP and 10 nM TCDD

treatment (Figure 4-5). It is noteworthy that TCDD produces a considerably higher level

of CYP1A1 protein expression than BaP treatment.

In addition to cytochrome P450 proteins, it is proposed that the enzyme

prostaglandin H-synthase (PGHS-1) may be involved in the biotransformation of BaP in

RL95-2 cells. Data in Figure 4-6 show the constitutive expression of PGHS-1 in RL95-2

cells as a 70 kDa protein by Western immunoblot analysis. Data further indicate that

PGHS-1 is significantly increased by 2.2 fold (p<0.005) following 10 utM BaP exposure.

On the contrary, treatment with 10 nM TCDD did not alter PGHS-1 levels (Figure 4-6).

In data not shown, PGHS-2 protein was not detected in control RL95-2 cells, as

determined by comparison with positive PGHS-2 control (Cayman), and PGHS-2

expression was not induced by either BaP or TCDD treatment at 48 hours.



cDNA Microarray Analysis for Induction of Stress-related Genes

Further studies were performed to determine whether BaP or TCDD was causing

transcriptional alterations in stress-related genes, as compared to an overall cellular

response as described above. Experiments were conducted using Clontech Human Stress

microarrays to determine changes in 236 genes known to function in stress response

regulation, including genes involved in DNA damage response, repair, and

recombination, base excision repair, nucleotide excision repair, mismatch repair, and

drug and xenobiotic metabolism. Functionally, the cDNA microarray procedure involves

reverse transcription of mRNA to radiolabelled cDNAs, which are then hybridized to








nylon membrane arrays containing 236 gene sequences. The radioactive signals bound to

each cDNA on the array represent relative expression levels of each gene as detected by

phosphorimaging. Each array contains 600 to 2400 base pairs of cDNA per gene, spotted

in duplicate for each gene represented, thereby providing information about differential

gene induction profiles.

The rationale for the present gene array experiments was to profile mRNA

expression after a 6 hour exposure to BaP, TCDD, or t-butylhydroperoxide. It was

expected that the non-metabolized AhR ligand TCDD would activate primarily the AhR-

mediated battery of genes, whereas the classic oxidative stress inducer t-butyl peroxide

would induce genes characteristic of a cellular oxidative stress response. It was further

expected that BaP would activate oxidative stress-response genes at an earlier time point

than AhR-regulated genes; however, since BaP has been shown to exert effects through

both pathways, it was expected that both batteries of genes would likely be activated by

BaP.

The pattern of gene induction by BaP at 6 hours was compared with patterns

induced by TCDD and t-butyl peroxide to provide some insight into which signal

transduction pathways were being activated by cellular exposure to BaP. Atlaslmage 1.5

software with global normalization was utilized to generate composite arrays for each

experiment. The subsequent comparison of composite arrays provided overall fold

change inductions for each gene. ScatterPlot analysis was used to determine the

distribution of all globally normalized genes relative to a diagonal line of identity (Figure

4-7). Genes distributed outside the two diagonal lines have a greater than two-fold

change in intensity relative to control. Genes from composite arrays which showed an








average fold-change of two-fold or greater between control and treated samples or

between the two untreated control cell lines were included in the final report only if their

adjusted intensities after global normalization had a defined ratio (a ratio not equal to

zero). However, exclusive consideration of genes with Clontech defined ratios of two-

fold or greater may neglect to acknowledge a number of additional genes which are

significantly altered and thus should be open for future consideration.

Results indicate four genes had a two-fold or greater down-regulation in RL95-2

cells following 6 hour treatment with 10 ,M BaP (Table 4-1), whereas 10 genes were up-

regulated two-fold or more (Table 4-2). The genes altered by BaP include stress-

response regulators, DNA damage, repair, and recombination, and xenobiotic metabolism

genes. Similarly, 6 hour treatment of RL95-2 cells with 10 nM TCDD resulted in the

down-regulation of 10 genes (Table 4-3), whereas 5 genes were up-regulated with TCDD

relative to control (Table 4-4). TCDD-altered genes include primarily stress-response

regulators and xenobiotic metabolism enzymes, with some effect on DNA damage

response and one housekeeping gene.

Treatment of RL95-2 cells with 200 p.M t-butylhydroperoxide for 6 hours, a

concentration known to produce a strong oxidative stress response in cell cultures,

resulted in the two-fold or greater down-regulation of 5 genes (Table 4-5), whereas 10

genes were up-regulated (Table 4-6). Tert-butylhydroperoxide treatment was largely

responsible for alterations in genes associated with stress response, with some effect on

genes for base and nucleotide excision repair, xenobiotic metabolism, as well as two

housekeeping genes.








Comparison of differential gene induction across BaP, TCDD, and t-

butylhydroperoxide treatments is shown in the Venn diagram (Figure 4-8). As illustrated,

no genes were similarly regulated by the three chemical treatments. However, BaP and

TCDD treatments affected the same two genes, BaP and t-butylhydroperoxide the same

one gene, and TCDD and t-butylhydroperoxide the same four genes.



Cluster and TreeView Analysis of Common Gene Expression Patterns

Unlike the array analysis previously described, current experiments utilized the

raw, non-globally normalized adjusted intensities for all genes on the membrane in the

analysis. All intensities were individually normalized, including the intensities which fell

below the background level for individual experiments. The net intensities from each

gene across all three experiments were transformed using variance normalization

calculations. The data were normalized manually using Microsoft Excel software and

were analyzed using ScatterPlot analysis to represent the distribution of genes around a

diagonal line of identity (Figure 4-9). As compared to the ScatterPlot analysis in Figure

4-7, the actual values for all genes were included in the analysis. The data were then

analyzed using Cluster and Treeview software for common expression patterns across

three separate experiments and differential chemical treatments (Figure 4-10).



Quantitation of Intracellular Oxidative Stress by Dichlorofluorescein Assay

Xenobiotic-mediated induction of intercellular oxidative stress in RL95-2 cells

was determined by monitoring the oxidation of dichlorofluorescin using a microplate

reader. 5-(and-6)-chloromethyl-2',7'-dichlorodihydrofluorescein diacetate (CM-








H2DCFDA) (Molecular Probes, Cat. #C-6827), a chloromethyl derivative of H2DCFDA

that is believed to exhibit much better retention in live cells, was added to buffer

containing xenobiotics. It is understood that CM-H2DCFDA passively diffuses into

cells, where its acetate groups are cleaved by intracellular esterases and a thiol-reactive

chloromethyl group reacts with intracellular glutathione and other thiols. Subsequent

oxidation yields a fluorescent adduct that is trapped inside the cell, thus facilitating the

monitoring of intracellular stress over time.

Results shown in Table 4-7 indicate that no definite conclusions may be drawn

regarding the relative levels of oxidative stress in RL95-2 cells upon differential chemical

treatment. Tert-butylhydroperoxide (TBH), the positive control for oxidative stress, did

not display a dose-response relationship for oxidative stress as measured by the DCF

assay. Significance was shown for 20 pM BeP, 10 nM TCDD, and 10 .tM TBH

treatments at 6 hours, with all three treatments producing a decreased level of

intracellular oxidative stress relative to DMSO vehicle control (Figure 4-7). Results are

speculative, at best, due to the inconsistencies in establishing a dose-response relationship

for TBH using the DCF fluorescence assay.



Discussion

Cancerous neoplastic progression has been strongly associated with increasing

genetic instability (Shackelford et aL, 1999). Since DNA mutations have been associated

with alterations in cell cycle progression, the present experiments were carried out to

determine the effect of the xenobiotics BaP and TCDD on cell cycle phase distribution.

Flow cytometry analysis with propidium iodide fluorescence was used to measure








relative percentages of cells in each phase of the cell cycle at specified time periods.

RL95-2 cells, which are reported to have a doubling time of 22-34 hours, have

presumably undergone a minimum of one cell cycle by 48 hours (Way et al., 1983). Data

in Figures 4-2 and 4-3 show a prolonged S phase response of RL95-2 cells following 48

hour BaP treatment. In contrast, TCDD, a non-genotoxic AhR ligand, did not elicit an

increased percentage of cells in the S phase, indicating that ligand activation of the AhR

alone is not sufficient to block cell cycle progression.

Present findings of a specific enhanced S phase response in RL95-2 cells upon

BaP treatment complement prior studies of human MCF-7 breast cancer cells treated with

BaP (Khan and Dipple, 2000). This study found that that cells treated with low, non-

cytotoxic doses of the genotoxic BaP metabolite BPDE result in an accumulation of cells

in the S phase of the cell cycle (Khan and Dipple, 2000). Further, MCF-7 cells failed to

show an accumulation in the GI phase, as would be expected from an early GI

checkpoint response (Khan and Dipple, 2000). The investigators therefore hypothesized

that the "stealth characteristic" of BaP enables cells to evade GI arrest and likely

increases the probability of malignant change because DNA replication with BaP

treatment subsequently occurs on a damaged template (Khan and Dipple, 2000). Similar

studies in human lymphoblast cells further reflect the prolonged S phase response of cells

upon BPDE treatment (Black et al., 1989). Alterations in RL95-2 cell cycle phase

distribution upon BaP treatment are therefore likely the result of a mechanism preventing

the replication of BaP-adducted, damaged DNA.

Many carcinogens exert their effects on cellular growth and differentiation by

perturbing the signal transduction pathways involved in normal cell cycle control. BaP








specifically is known to modulate mammalian gene expression through epigenetic

mechanisms involving the AhR signaling pathway, oxidative stress, and altered

mitogenic signaling.

BaP has been demonstrated to undergo differrent multi-step pathways in its

conversion to toxic or tumorigenic metabolites (Kim et al., 1997; Wells et al, 1997;

Ramos, 1999). The predominant pathway utilizes the cytochrome P450 isozyme

CYP1AI as an initial step in converting BaP to oxide-intermediates, followed by the

enzyme epoxide hydrolase, which results in dihydrodiol formation. The present study

found that BaP and TCDD treatment induced CYP1A1 mRNA and protein levels in

RL95-2 cells, as well as increased CYPIB1 mRNA levels (Figures 4-4 and 4-5). The

observed effects of BaP and TCDD on CYP1A1, and additionally on CYP1A2 mRNA

levels were further validated by cDNA microarray data, indicating enhanced mRNA of

both enzymes upon 6 hour treatment (Tables 4-2 and 4-4). CYPIBI was not included on

the Human Stress array, and therefore could not be validated by this method. Since

CYPIA1 and CYPIBI are both induced by BaP and TCDD, whereas cellular adhesion,

cytoskeleton, and cell cycle effects are specific to BaP treatment alone, factors beyond

AhR control likely account for the specific BaP effects on cellular attachment and

morphology.

In addition to AhR-mediated effects, BaP has been shown to be metabolized in

AhR-independent pathways. The pathway leading to the well-characterized toxic

metabolite of BaP, (+) 7,8-dihydrodiol-9,10 epoxide (BPDE), utilizes CYP1A1 and

prostaglandin H-synthase (PGHS) in the conversion of the dihydrodiol to its bioactive

form (Degen, 1993). PGHS is an enzyme that co-oxidizes some xenobiotics and








carcinogens in vitro in the presence of acrachadonic acid or lipid peroxides (Degen,

1993). PGHS probably does not play a major role in systemic drug metabolism, but may

be involved in organ and cell specific toxicities. PGHS has been localized in the uterus

primarily in the surface endometrial epithelium, although some localization was evident

in glandular epithelium (Van Voorhis et al., 1990).

PGHS has been shown to occur in two isoforms, namely PGHS-1 and PGHS-2.

PGHS-1 is the constitutive form of PGHS, which supplies prostanoids crucial for basal

cellular functions and appears to be regulated developmentally (Smith and DeWitt,

1996). PGHS-2, on the contrary, is expressed in extremely low levels in most cells, but is

rapidly induced by stimuli such as mitogens, tumor promoters, and cytokines,

maintaining important roles in cellular inflammation and proliferation (Smith and

DeWitt, 1996; Kulmacz, 1998). The present study shows that constitutive PGHS-1

protein levels are increased 2.2 fold by BaP treatment, with no significant change in

PGHS-1 levels with TCDD treatment (Figure 4-6). PGHS-2 expression, in contrast, was

not detected in control RL95-2 cells, and was not induced upon BaP or TCDD treatment,

evidence that PGHS-2 is not involved in the specific BaP effects. The lack of expression

of PGHS-2 in uterine RL95-2 cells is in contrast to published results from oral epithelial

cells which indicate that BaP induces PGHS-2 (Kelley et al., 1997). Interestingly,

benzo(e)pyrene, a non-carcinogenic congener of BaP, did not effect PGHS-2 levels

(Kelley et al., 1997). The sensitivity of PGHS levels to transcriptional and translational

stimulation by growth factors such as epidermal growth factor (EGF) can potentially

impact on BaP effects (Mamett, 1990). Overall, it is likely that PGHS-1 plays a key role








in the pathogenesis of BaP-mediated effects on uterine endometrial cells through its dual

role in carcinogen activation and catalysis of prostaglandin biosynthesis.

Whereas CYPIA1 and PGHS have been classically implicated in toxic metabolite

formation, a few recent studies have provided evidence that BaP can be metabolized to

potentially toxic quinone metabolites through PGHS alone (Degen, 1993; Sivarajah et aL,

1981). In particular, BaP can be metabolized to 1,6-, 3,6-, and 6,12- quinones (BPQs),

which are able to undergo redox cycling and thereby produce reactive oxygen species

(ROS) such as 02--, OH., OH-, and H202 (Liu et aL, 1998; Mauthe et al., 1995).

Although quinones themselves are not considered mutagenic or carcinogenic in animal

models, quinone redox cycling leading to the generation of reactive oxygen species

(ROS) is responsible for marked cellular toxicity and mutagenicity (Miller and Ramos,

2001a). Moreover, BaP metabolism has been associated with oxidative stress by the

depletion of reduced cellular glutathione (GSH) and ATP levels (Romero et al., 1997).

Consequently, BaP may have effects due to metabolites generated at the cellular level

through cytochrome P450, prostaglandin H-synthase (PGHS), lipoxygenase, and/or lipid

peroxidation.

TCDD, as well as BaP, has been reported to be associated with oxidative stress in

cells, although to a lesser extent than BaP (Morel and Barouki, 1999; Yoshida and

Ogawa, 2000; Radjendirane and Jaiswal, 1999). In particular, dioxin exposure has been

associated with a sustained oxidative stress response in the mouse (Shertzer et al., 1998).

It has been reported that mice exposed to 5 gg TCDD/kg for 3 consecutive days had

oxidized hepatic glutathione levels and elevated urinary 8-hydroxydeoxyguanosine

levels, a product of DNA base oxidation (Shertzer et al., 1998).








Due to the instability and high reactivity of ROS and their low steady state levels,

quantitation of ROS is a difficult task. In the current study, two main approaches were

taken to evaluate the oxidative stress response of RL95-2 cells to BaP and TCDD

treatments. First, experiments utilizing cDNA microarrays were performed to determine

changes in the transcriptional expression of the genes for enzymes involved in the

production and removal of ROS at 6 hours. Subsequently, a biochemical assay based on

oxidation of dichlorofluorescin was employed to estimate the overall levels of

intracellular ROS production.

Microarray technology has emerged as a cutting-edge tool for investigating

normal biological and disease processes, profiling differential gene expression, and

discovering potential therapeutic and diagnostic drug targets. Data obtained using the

Human Stress/Toxicology Array indicate that a total of 4 genes were down-regulated and

10 genes were up-regulated by BaP relative to DMSO vehicle control by a two-fold or

greater magnitude. All of the genes had defined ratios, meaning the intensities of both

genes were above background. Further, two of the genes regulated by BaP were similarly

regulated by TCDD, namely CYP1A1 and CYP1A2, thereby serving as a control for

AhR induction.

Interestingly, cyclic AMP response element binding protein 1 (CREB 1), a

transcription factor for cAMP-regulated genes, was shown to be down-regulated by BaP

relative to control. Cyclic AMP second messenger pathways are important to cellular

growth, differentiation, and function through the regulation of extracellular signals. It

could be speculated that a down-regulation of CREB 1 could prevent key signal

transduction pathways involved in cellular growth. Further, results indicate a down-








regulation of cyclin-dependent kinase inhibitor 1 (or p21), known to associate with

cyclins A, D, and E in the control of the GI to S phase transition. P21 is known to

inhibit cyclin dependent kinase proteins (cdks), as well as to inhibit the phosphorylation

of the retinoblastoma protein.

Whereas 2 stress response-regulator genes were down-regulated by BaP in RL95-

2 cells, 7 stress-response regulator genes were up regulated (Tables 4-1 and 4-2). Results

are potentially indicative of BaP producing a low level of oxidative stress in cells.

Among the genes up-regulated by BaP relative to control was the 150 kDa oxygen-

regulated protein, shown to play an important role in protein folding and secretion in the

endoplasmic reticulum, likely acting as a molecular chaperone protein to cope with

environmental stress. In addition, the UDP-glucuronosyltransferase 1-6 precursor protein

was up-regulated in RL95-2 cells with BaP treatment, indicating the induction of

secondary metabolism pathways for BaP. As expected, results indicate the induction of

CYP1A1 and CYP1A2 mRNA by BaP.

TCDD treatment, by contrast, was shown to effect the regulation of 15 genes, two

of which were similarly regulated by Ba? and 4 other genes similarly regulated by t-

butylhydroperoxide (Tables 4-3 and 4-4). The genes for heat shock cognate 71-kDa

protein and peptidyl-propyl cis-trans isomerase were down-regulated by both TCDD and

t-butylhydroperoxide, whereas the gene heat shock 70-kDa protein 4 was up-regulated by

both chemicals. In contrast, the gene for glutathione S-transferase theta 1 was down-

regulated by TCDD treatment, whereas it was up-regulated upon treatment with t-

butylhydroperoxide. Interestingly, 5 stress-response regulator genes were down-

regulated in response to TCDD treatment, whereas only 2 were up-regulated, a trend








opposite to that of BaP which predominantly up-regulated stress response genes.

Interestingly, the gene NADH-cytochrome B5 reductase (DT diaphorase) was shown to

be down-regulated by TCDD in RL95-2 cells. Diaphorase is known to function in the

disassociation and elongation of fatty acids, cholesterol biosynthesis, and drug

metabolism.

Finally, a total of 15 genes were perturbed by t-butylhydroperoxide treatment,

including 5 stress-response regulator genes which were down-regulated and 3 stress

response regulators which were up-regulated (Tables 4-5 and 4-6). Since TBH is known

to elicit a strong oxidative stress response in cells, the seeming contradiction of a

predominantly down-regulated stress response set of genes is believed to result from the

cells experiencing severe DNA damage at the 200 ,tM concentration. Evidence for DNA

damage is shown in the up-regulation of various base-excision and nucleotide-excision

repair genes upon TBH treatment. Interestingly, endonuclease II homolog 1, which

excises damaged pyramidines, was shown to be up-regulated with t-butylhydroperoxide

treatment, as well as the genes for UV excision repair protein RAD23A and DNA

excision repair protein ERCC2. Glutathione S-transferase theta 1, which functions in

secondary detoxification metabolism, was similarly up-regulated, as well as glutathione

peroxidase-related protein 2, which catalyzes the reduction of H202, organic and lipid

peroxides by reduced glutathione.

Further studies were employed to estimate the total levels of intracellular ROS

production upon BaP, TCDD, and TBH treatment. It has been reported that increased 5-

(and-6)-chloromethyl-2'7'-dichlorodihydrofluorescein diacetate (CM-H2DCFDA)

oxidation is a useful assay for oxidative stress (Jakubowski and Bartosz, 2000; Xie et al.,








1999; Wang and Joseph, 1999). CM-H2DCFDA has been used successfully to measure

intracellular reactive oxygen species in cardiac myocytes (Xie et al, 1999). Studies in

cultured monocytes indicate a significantly enhanced production of superoxide ions upon

treatment with BaP, whereas treatment with benzo(e)pyrene (BeP), a relatively

noncarcinogenic congener of BaP, failed to detect superoxide at significant levels (Xie et

aL, 1999).

The present study used the DCF assay failed to demonstrate a dose-response

relationship for intracellular oxidative stress using the positive control t-

butylhydroperoxide. Consequently, current results from the DCF assay do not adequately

assess the effects of BaP or TCDD on intracellular oxidation. Likely flaws in the

experimental design include unequal detachment of cells from the 96-well plates after

repeated PBS washes and difficulties in uniformly removing all traces of phenol red from

samples prior to absorbance readings. It is speculated that pre-treatment of the wells with

collagen may alleviate the unequal cellular attachment, and enable the acquisition of

more reproducible data. Further development of a biochemical measure for oxidative

stress will need to be made in future studies.

In conclusion, the development of an increased understanding of the role of signal

transduction pathways involved in cell-cycle checkpoint responses to environmental

exposures holds great promise in disease prevention, as well as in development of

efficacious therapeutic strategies for the treatment of environmentally-linked cancers.

Current data provide unique molecular fingerprints for xenobiotic effects on RL95-2 cells

and indicate that BaP may produce a low level of oxidative stress in cells. Further, data

suggest that inducible CYPIA1, CYP1B1 and/or PGHS-1 may mediate a pathway for





77


BaP bioactivation and toxicity. It is hoped that such information will prove useful in the

assessment of cancer risk for the whole population, as well as subpopulations of

individuals with particular genetic susceptibilities.














e e e
-4 H202- .OH H20


Figure 4-1. Intermediates from normal metabolism of atmospheric oxygen by successive
1-electron reductions.


e
02" .O









RL95-2 Ce Cycle Fhise IstnIhn

90
80
~70 *

~50-

Cw30-

10-
10 -

DNW BaP 1UD "T T a 1UC IIO a TCDD

GO-G1 Phase S Phase G2-M Phase



Figure 4-2. Graphical Representation of BaP and TCDD Effects on Cell Cycle Phase Distribution.
Graphical representation of BaP and TCDD effects on RL95-2 cell cycle phase distribution at 48 hours. Data
represent the mean SEM for four separate experiments (p<0.05).























RL95-2 DMSO


GO-GI: 82.39%
S: 12.01 %
G2-M: 5.61%



I



o,,,!. O p.t ... .


GO-GI: 36.27 %
S: 46.28 %
G2-M: 17.45 %


it:




o1 A C 'mt "


0-G 1:78.79 %
S: 13.98 %
G2-M: 7.23 %






L


Figure 4-3. Representative Experiment Showing the Differential Effects of BaP and
TCDD on RL95-2 Cell Cycle Phase Distribution.
Flow cytometry analysis with propidium iodide fluorescence was performed to evaluate the
effects of 10 pM BaP and 10 nM TCDD on cell cycle phase distribution in RL95-2 cells.


RL95-2 BaP?


RL95-2 TCDD

















RL95-2 cells
DMSO BaP TCDD









RL95-2 cells


DMSO


BaP TCDD


Figure 4-4. Induction of CYPlAI and CYP1B1 mRNA by BaP and TCDD in RL95-2 cells.
Northern blot analysis was performed to analyze CYP1A1 and CYPIBI mRNA in RL95-2 cells.
Results are representative blots for two separate experiments.


A.



CYPlA1
3.0 kb





B.


CYP1B1
5.2 kb




















RL95-2 cells


DMSO


BaP TCDD


CYPIA1
55 kDa


Figure 4-5. Western immunoblot analysis for the effects of BaP and TCDD on CYP1AI
protein level.
RL95-2 cells were treated with 10 giM BaP or 10 nM TCDD for 48 hours, and the detergent-
extracted cell lysates were electrophoresed, transferred to a nitrocellulose membrane, and
immunostained with anti-human CYP1A1 followed by horseradish peroxidase conjugated IgG.
Results were consistent among four replicates.
















NCHSA1


300-
250-
200-
150-
100-
50-
0-


I'm
DNSO BaP TCDD



RL95-2 cells


DMSO


PGHS-1


BaP


TCDD


70 kDa




Figure 4-6. Western immunoblot analysis of the effects of BaP and TCDD on PGHS-1
protein levels.
RL95-2 cells were treated with 10 giM BaP or 10 nM TCDD for 48 hours, and detergent-
extracted cell lysates were electrophoresed, transferred to a nitrocellulose membrane, and
immunostained with anti-human PGHS-1 followed by horseradish peroxidase conjugated IgG.
Data represent the mean SE for 5 experiments (p<0.005).























Figure 4-7. RL95-2 Microarray experiment ScatterPlot Analysis.
The Clontech Atlas Human Stress array with AtlasImage 1.5 software was used to
analyze overall intensities of gene expression for the 236 genes known to modulate
toxic cellular responses. Total RNA was recovered from control, BaP, TCDD, and
t-butylhydroperoxide-treated RL95-2 cells and a cDNA target prepared with
incorporation of 33p. Values for gene expression were obtained for composite
arrays from three separate experiments per treatment condition. Intensities for gene
expression were globally normalized to all genes on the membrane using
AtlasImage 1.5 and Excel software. Globally normalized (AtlasImage 1.5) values
were calculated for xenobiotic-treated samples relative to control, and were
analyzed using ScatterPlot analysis for each gene relative to a two-fold or greater
change. Genes lying close to the diagonal line of identity show similar gene
expression, whereas those lying farther away have greater variation.














A

CL


3m













0-
.. C


.2

S



.Z .







w E


Clontech Globally Normalized
Intensities In Untreated
RL95-2 ces


Clontech Globally Normalized
Intensities In Untreated
RL95-2 cels


Clontech Globally Normalized
Intensities In Untreated
RL95-2 cells








TABLE 4-1. Stress-related genes with defined ratios down-regulated by benzo(a)pyrene in RL95-2 cells.

Gene Name Access. # BaP BaP:DMSO Function
CAMP-response element binding protein M34356 0.38 Stress-response regulator
C-jun N-terminal kinase 3 alpha 2 U34819 0.42
Purine-rich single-stranded DNA-binding protein alpha M96684 0.47
Cyclin-dependent kinase inhibitor 1 U09579 0.49





TABLE 4-2. Stress-related genes with defined ratios up-regulated by benzo(a)pyrene in RL95-2 cells.

Gene Name Access. # BaP BaP:DMSO Function
Mitochondrial I0-kDa heat shock protein U07550 + 2.03
Protein disulfide isomerase-related protein ERP72 precursor J05016 + 2.79
51 -kDa fK506-binding protein U42031 + 2.42
94-kDa glucose-regulated protein X15187 + 5.00 Stress-response regulator
NADP-regulated thyroid hormone-binding protein L02950 + 4.11
Probable protein disulfide isomerase P5 precursor D49489 + 3.00

150-kDa oxygen-regulated protein U65785 + 2.47
Cytochrome P450 1 A2 Z00036 + 5.38
Cytochrome P450 I Al K03191 + 4.60 Xenobiotic metabolism
Microsomal UDP-glucoronosyltransferase 1-6 precursor J04093 + 2.52




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EPY6SX9JQ_FWXMJ8 INGEST_TIME 2014-06-17T17:46:06Z PACKAGE AA00022176_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES