University of Florida Thesis Or Dissertation

PRO-FIBROGENIC AND ESTROGEN RECEPTOR SIGNALING PATHWAYS INTERACT IN THE LUNG: IMPLICATIONS FOR FIBROTIC LUNG DISEASE

LEY CODY SMITH

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

2017

To my family and my friends

ACKNOWLEDGMENTS

I thank my advisor, Dr. Tara Sabo-Attwood, for her expert guidance, constructive advice, and patience throughout the dissertation. I would also like to thank my brilliant advisory committee for their support and guidance in the development of my project: Dr.

Nancy Denslow, Dr. Paul Cooke, and Dr. Gregory Schultz. I would like to acknowledge the members of the Center for Environmental and Human Toxicology, past and present, for all their assistance and support. I especially want to thank Dr. Candice Lavelle, Dr.

Joseph H. Bisesi, Jr., Dr. Georgia Roberts, and Dr. Erica Brockmeier for not only being excellent coworkers but also life-long friends. I thank Dr. Dale Porter of CDC NIOSH and Dr. Andrew Bryant of the UF Medical School for being enthusiastic collaborators. I thank Dr. Manjunatha Nanjappa and Terry Medrano of the UF Department of

Physiological Sciences for their assistance with immunohistochemical staining. I thank members of the UF Interdisciplinary Center for Biotechnology Research including Dr.

Alberto Riva of the Bioinformatics Core, Dr. Yanping Zhang of the Gene Expression &

Genotyping Core, and Dr. David Moraga of the NextGen DNA Sequencing Core for assistance with the RNA-Seq analysis. I thank Dr. Navid Saleh and Dipesh Dah of the

University of Texas at Austin Cockrell School of Engineering and Dr. Paul Carpinone of the UF Herbert Wertheim College of Engineering for their assistance with the nanotube experiments. I also want to thank the undergraduate researchers who have assisted me throughout the years including Santiago Moreno, Kristal Gant, Jennifer Knapp, and

Kung Shroff. Last but not least, I thank my family and friends for their unconditional love and support throughout this journey.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 9

LIST OF ABBREVIATIONS ...... 11

ABSTRACT ...... 16

CHAPTER

1 BACKGROUND ...... 18

Pulmonary Fibrosis ...... 18 TGF-β1 Signaling in Pulmonary Fibrosis ...... 20 Multi-Walled Carbon Nanotube-Induced Pulmonary Fibrosis ...... 22 Sex and Sex Hormones in Pulmonary Fibrosis ...... 24 Estrogen Signaling in the Lung ...... 26 Interactions Between TGF-β1 and E2 Signaling ...... 27 Cell-Specific Effects in Pulmonary Fibrosis ...... 29 Research Objectives ...... 30

2 MULTI-WALLED CARBON NANOTUBES INHIBIT ESTROGEN RECEPTOR ALPHA EXPRESSION IN THE LUNG VIA TGF-β1 ...... 32

Introduction ...... 32 Materials and Methods...... 34 Chemicals ...... 34 Animal Exposure ...... 35 Histopathology ...... 35 Enzyme-Linked Immunosorbent Assay (ELISA) ...... 36 Immunohistochemistry (IHC) ...... 36 Cell culture ...... 37 Nanotube Dispersion and Characterization in vitro ...... 38 Total RNA Extraction and Purification ...... 39 Quantitative Real-Time Polymerase Chain Reaction (qPCR)...... 40 Intracellular Reactive Oxygen Species (ROS) Measurement ...... 40 Statistics ...... 41 Results ...... 41 Mitsui MWCNTs Cause Histopathological Alterations in the Lungs of Mice ..... 41 Mitsui MWCNTs Increase TGF-β1 Levels in BALF ...... 42 Mitsui MWCNTs Reduce Estrogen Receptor mRNA Expression in Mouse Lungs ...... 42

ESR1 Protein is Expressed in Mouse Lung Epithelial Cells ...... 43 Mitsui MWCNTs Reduce ESR1 mRNA Expression in BEAS-2Bs ...... 43 Mitsui MWCNT-Induced Reduction of ESR1 mRNA is Not a Function of Surface Chemistry ...... 44 Mitsui MWCNTs Reduce ESR1 mRNA Expression in a TGF-β1-Dependent Manner in vitro ...... 45 Mitsui MWCNT-Induced Reduction of ESR1 mRNA Expression is Likely Not a Consequence of ROS Generation ...... 45 Discussion ...... 46

3 TRANSFORMING GROWTH FACTOR BETA1 TARGETS ESTROGEN RECEPTOR SIGNALING IN BRONCHIAL EPITHELIAL CELLS ...... 67

Introduction ...... 67 Materials and Methods...... 69 Chemicals ...... 69 Cell Culture ...... 70 Total RNA Extraction and Purification ...... 71 Quantitative Real-Time Polymerase Chain Reaction (qPCR)...... 71 Protein Extraction and Purification ...... 72 Western Blot ...... 72 RNA-Seq Library Preparation ...... 73 RNA Sequencing ...... 74 Bioinformatics ...... 74 Statistics ...... 75 Results ...... 75 TGF-β1 Induces Changes Consistent with EMT ...... 75 E2 Does Not Significantly Affect TGF-β1-Induced EMT ...... 76 TGF-β1 Reduces Estrogen Receptor mRNA and Protein Expression ...... 77 TGF-β1 and E2 Exhibit Unique Transcriptional Profiles ...... 77 TGF-β1 and E2 Differentially Regulate Gene Sets ...... 79 Discussion ...... 80

4 GENE EXPRESSION ANALYSIS IN THE LUNGS OF PATIENTS WITH IDIOPATHIC PULMONARY FIBROSIS ...... 98

Introduction ...... 98 Materials and Methods...... 100 Patient Samples ...... 100 Handling of Lung Tissue ...... 101 Total RNA Extraction and Purification ...... 101 Gene Expression Analysis in Whole Lung Tissue ...... 101 Laser Capture Microdissection ...... 102 Gene Expression Analysis in Microdissected Tissue ...... 103 Statistics ...... 103 Results ...... 103 Baseline Characteristics of Patient Samples ...... 103

Histopathology in IPF Lungs ...... 104 Gene Expression Analysis in Whole IPF Lung Tissue ...... 104 LCM Method Development ...... 105 Discussion ...... 106

5 GENERAL CONCLUSIONS ...... 120

APPENDIX

A CTGF GENE EXPRESSION IN LUNGS OF MICE EXPOSED TO MITSUI MWCNTS ...... 126

B ALIGNMENT OF MOUSE ESR1 TRANSCRIPT VARIANTS ...... 127

C ESR1 PROTEIN LEVELS IN BRONCHIAL EPITHELIAL CELLS EXPOSED TO MITSUI MWCNTS ...... 128

D RNA INTEGRITY OF HUMAN LUNG SAMPLES ...... 129

LIST OF REFERENCES ...... 130

BIOGRAPHICAL SKETCH ...... 156

LIST OF TABLES

Table page

2-1 Primer details for qPCR in Chapter 2...... 57

3-1 Primer details for qPCR in Chapter 3...... 89

3-2 Genes differentially regulated by E2...... 89

3-3 Gene set enrichment analysis of genes identified by RNA-Seq...... 90

4-1 Baseline characteristics of study groups...... 113

4-2 Fold change IPF lung mRNA expression compared to controls...... 113

4-3 Cq values of select genes in microdissected IPF lung tissue...... 113

4-4 Primer details for qPCR in Chapter 4...... 114

LIST OF FIGURES

Figure page

2-1 MWCNT characterization in cell culture media ...... 58

2-2 Histopathological changes in the lungs of mice after 4, 8, or 12 days of exposure to 10 mg/m3 Mitsui MWCNTs for 5 hours per day by whole body inhalation ...... 59

2-3 Mitsui MWCNTs upregulate TGF-β1 levels in bronchoalveolar lavage fluid (BALF) ...... 60

2-4 Mitsui MWCNTs reduce estrogen receptor expression in mouse lungs ...... 61

2-5 ESR1 protein is expressed in epithelial cells lining the bronchi and in the interstitial space in mouse lung tissue ...... 62

2-6 Mitsui MWCNTs reduce expression of ESR1 but not ESR2 or GPER1 mRNA expression in BEAS-2Bs ...... 63

2-7 Mitsui MWCNT-induced reduction of ESR1 mRNA expression is not dependent on nanotube functionalization ...... 64

2-8 Mitsui MWCNT-induced reduction of ESR1 mRNA expression is partly dependent on TGF-β1 signaling ...... 65

2-9 Mitsui MWCNT-induced reduction of ESR1 mRNA expression is likely not a consequence of ROS generation ...... 66

3-1 TGF-β1 induces gene expression changes consistent with EMT in BEAS-2B cells ...... 91

3-2 E2 does not affect TGF-β1 induced EMT ...... 92

3-3 TGF-β1 downregulates ESR1, ESR2, and GPER1 mRNA expression in BEAS-2B cells ...... 93

3-4 TGF-β1 downregulates ESR1 protein expression ...... 94

3-5 TGF-β1 and E2 exhibit distinct transcriptional profiles ...... 95

3-6 Orthogonal validation of RNA-Seq data ...... 96

3-7 TGF-β1 and E2 cause differential regulation of genes involved in extracellular matrix turnover ...... 97

4-1 Representative high resolution computerized tomography (HRCT) scans of IPF lungs ...... 115

4-2 Patients with IPF have increased deposition of collagen in the lung ...... 116

4-3 RNA isolated from control and IPF lung tissue is of high quality...... 117

4-4 Gene expression changes in lungs of patients with IPF ...... 118

4-5 LCM method development ...... 119

4-6 Optimization of gene expression analysis of microdissected tissue by qPCR .. 119

5-1 Hypothesized expression levels of TGF-β1, CTGF, and estrogen receptors throughout pulmonary fibrosis disease progression compared to expression in healthy tissue ...... 125

A-1 Mitsui MWCNTs reduce CTGF mRNA expression in mouse lung ...... 126

B-1 Alignment of Esr1 transcript variants using Multalin version 5.4.1 ...... 127

C-1 Mitsui MWCNTs do not affect ESR1 protein expression in bronchial epithelial cells ...... 128

D-1 RNA integrity of human lung samples ...... 129

LIST OF ABBREVIATIONS

ACTB Beta-actin

AECs Alveolar epithelial cells

ANOVA Analysis of variance

AP-1 Activator protein 1

AR Androgen receptor

BALF Bronchoalveolar lavage fluid

BCA Bicinchoninic acid

BEAS-2Bs Bronchial epithelial cells

BEBM Bronchial epithelial basal medium

BEGM Bronchial epithelial growth medium

BPE Bovine pituitary extract

BSA Bovine serum albumin

Ca2+ Calcium cDNA Complementary deoxyribonucleic acid

CLIC3 Chloride intracellular channel 3

CNT Carbon nanotube

Col1a1 Collagen type 1 alpha 1 chain

Col1a2 Collagen type 1 alpha 2 chain

COPD Chronic obstructive pulmonary disease

CRABP4 Cellular retinoic acid-binding protein 4

CTGF Connective tissue growth factor

CYP19 Aromatase

DCFDA 2’,7’ –dichlorofluorescin diacetate

DHT Dihydrotestosterone

DMSO Dimethyl sulfoxide

DPPC Dipalmitoylphosphatidylcholine

DUSP6 Dual specificity phosphatase 6

E2 17β-Estradiol

ECM Extracellular matrix

ELISA Enzyme-linked immunosorbent assay

EMT Epithelial to mesenchymal transition

EndoMT Endothelial to mesenchymal transition

EPM Electrophoretic mobility

ER Nuclear estrogen receptor

ERCC External RNA controls consortium

ESR Estrogen receptor

ESR1 Estrogen receptor alpha

ESR2 Estrogen receptor beta

FDR False discovery rate

Fn1 Fibronectin

FVC Forced vital capacity

FVC% Percent predicted forced vital capacity

GPER1 G-protein coupled estrogen receptor

GSEA Gene set enrichment analysis

HDR Hydrodynamic radius

HRCT High resolution computerized tomography

HRP Horseradish peroxidase

HRT Hormone replacement therapy

IHC Immunohistochemistry

IL-1 Interleukin 1

IL-18 Interleukin 18

IL-1β Interleukin 1 beta

IL-6 Interleukin 6

IPAH Idiopathic pulmonary arterial hypertension

IPF Idiopathic pulmonary fibrosis

KCl Potassium chloride

KCNQ1 Potassium voltage-gated channel subfamily Q

LAM Lymphangioleiomyomatosis

LAP Latency associated peptide

LCM Laser capture microdissection

Mg Magnesium

MIQE Minimum information for publication of qPCR experiments

MMP1 Matrix metalloproteinase 1

MMP13 Matrix metalloproteinase 13

MMP14 Matrix metalloproteinase 14

MMP2 Matrix metalloproteinase 2

MMP7 Matrix metalloproteinase 7

MMP9 Matrix metalloproteinase 9 mRNA Messenger ribonucleic acid

MT1-MMP Matrix metalloproteinase 14

MT-MMP Membrane-type matrix metalloproteinase

MWCNT Multi-walled carbon nanotube

NURF Nucleosome remodeling factor oxMWCNT Hydoxylated multi-walled carbon nanotube

PAS Periodic acid-Schiff

PBS Phosphate-buffered saline

PDGF Platelet-derived growth factor

PEN Polyethylene naphthalate

PR Progesterone receptor

PSN Penicillin-Streptomycin-Neomycin qPCR Quantitative real-time polymerase chain reaction

RASAL1 RAS protein activator like 1

RBP7 Retinol binding protein 7

RIN RNA integrity number

RNA Ribonucleic acid

RNA-Seq RNA-Sequencing

ROS Reactive oxygen species

RPS13 Ribosomal protein S13

RT-PCR Reverse transcription polymerase chain reaction

SD Standard deviation

SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis

SEM Standard error of the mean

SNAI1 Snail family transcriptional repressor 1

SPP1 Osteopontin

SPRY4 Sprouty RTK signaling antagonist 4

SWCNT Single-walled carbon nanotube

TBRI Transforming growth factor beta receptor type 1

TBRII Transforming growth factor beta receptor type 2

TGF-β Transforming growth factor beta

TGF-β1 Transforming growth factor beta1

TIMP1 Tissue inhibitor of metalloproteinases 1

TRDLS Time-resolved dynamic light scattering

UIP Usual interstitial pneumonia

Α5β6 Integrin alpha five beta six

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

PRO-FIBROGENIC AND ESTROGEN RECEPTOR SIGNALING PATHWAYS INTERACT IN THE LUNG: IMPLICATIONS FOR FIBROTIC LUNG DISEASE

Ley Cody Smith

August 2017

Chair: Tara Sabo-Attwood Cochair: Nancy Denslow Major: Veterinary Medical Sciences

Pulmonary fibrosis is characterized by progressive and irreversible scar tissue formation causing impaired function and respiratory failure in 40,000 people per year.

Pulmonary fibrosis can be caused by exposure to environmental and occupational agents although the cause is often unknown (idiopathic, IPF). Epidemiological studies suggest sex-specific trends in prevalence and mortality of IPF where it is more common in men and women have better survival rates. These data and evidence in animal models suggest sex hormones, e.g. estrogen (E2), may be involved in pathogenesis.

The overarching hypothesis of this work was that interactions exist between pro- fibrogenic and E2 signaling pathways in the lung. We tested this hypothesis by characterizing interactions between pro-fibrogenic signaling pathways, e.g. transforming growth factor beta1 (TGF-β1) and estrogen receptor alpha (ESR1), estrogen receptor beta (ESR2), and G-protein coupled estrogen receptor (GPER1) in a mouse model of multi-walled carbon nanotube (MWCNT)-induced pulmonary fibrosis and in human bronchial epithelial cells (BEAS-2Bs). We expanded these investigations by examining estrogen receptor expression in lungs from patients with IPF. We identified estrogen

receptors as transcriptional targets of pro-fibrogenic signaling in the lung where exposure to MWCNTs significantly reduced ESR1, ESR2, and GPER1 mRNA expression. We proceeded to show MWCNT-induced repression of ESR1 was mediated by TGF-β1 and TGF-β1 could independently reduce ESR1 (p<0.05), ESR2 (p>0.05), and GPER1 (p<0.05) mRNA and ESR1 (p<0.05) protein levels in BEAS-2Bs. We also report reduced expression of ESR1 (p<0.05), ESR2 (p=0.09), and GPER1 (p<0.05) in

IPF lungs. Next, we performed an RNA-Seq analysis to identify novel transcriptional targets of E2 in lung cells. Results revealed E2-specific downregulation (FDR-corrected p-value≤0.05) of chloride intracellular channel 3 (CLIC3) and retinol binding protein 7

(RBP7). Gene set enrichment analysis revealed inverse regulation of extracellular matrix turnover, airway smooth muscle cell contraction, and calcium flux regulation pathways by TGF-β1 and E2 and highlighted chromatin remodeling pathways as specific targets of E2. While these data do not support a direct role for E2, they indicate interactions between TGF-β1 and E2 signaling exist and suggest E2 modulates pathways involved in pulmonary fibrosis which may explain sex-specific trends in disease.

CHAPTER 1 BACKGROUND

Pulmonary Fibrosis

Pulmonary fibrosis is characterized by progressive and irreversible scar tissue formation in the lung that can lead to reduced lung function and ultimately increased morbidity and mortality . Prognosis is uniformly poor with limited treatment options

(194). While the cause of pulmonary fibrosis is often unknown (idiopathic), the disease can be associated with viral infection (90), exposure to environmental and occupational agents such as dusts (silica, metals) and fibers (asbestos) (138), or as a side effect of therapeutic drugs (e.g. radiation, chemotherapy) (203).

Idiopathic pulmonary fibrosis (IPF) is a particularly severe form of the disease and is essentially a fatal diagnosis (96). It is suspected that 40,000 people die of IPF every year (13). The median age at diagnosis is 66 years and the median survival is

2.5-3.5 years (96). The disease has a long (months to year) latency period, and patients with IPF do not tend to seek medical attention until the severity of the lung lesions reach a critical threshold. Patients present with chronic and progressive cough, bibasilar crackles during inspiration, and finger clubbing (96). Diagnosis is confirmed by the presence of subpleural fibrosis and honeycombing alternating with areas of less affected or normal parenchyma on high resolution computerized tomography (HRCT) scan or the presence of fibroblast foci, areas of aggregated myofibroblasts overlayed with hyperplastic reparative epithelium on lung biopsy (34, 172, 205, 223).

There exist multiple clinical phenotypes of IPF based on the progression of the disease as measured by a decline in forced vital capacity (FVC) (49). Patients with stable or slowly progressive IPF exhibit a gradual decline in lung function and worsening

dyspnea and a mean annual rate of FVC decline of 0.12 L to 0.21 L and typically die within several years (110). A subgroup of patients with IPF, predominantly male cigarette smokers, display a rapidly progressive form of the disease known as accelerated IPF (193). Patients with the accelerated variant exhibit similar lung function, chest imaging, and histological findings at time of diagnosis as patients with stable IPF but show shortened survival (193). Lastly, acute exacerbation of IPF is defined by rapidly worsening dyspnea in the absence of complications such as infection, heart failure, or pulmonary embolism (31, 202). Patients suffering from acute exacerbation

IPF exhibit worsening dyspnea within days to weeks and mortality exceeds 60% during hospital admission and greater than 90% within 6 months of discharge (31).

On a cellular level, human pulmonary fibrosis is characterized by alveolar epithelial cell injury, areas of alveolar type II cell hyperplasia, accumulation of fibroblasts and myofibroblasts, and the deposition of extracellular matrix (ECM) molecules (220).

IPF is a heterogeneous disease but the prevailing hypothesis is that it is caused by persistent lung injury and dysregulated wound repair (34). Although inflammatory mediators certainly play a role in the initiation and progression of certain types of pulmonary fibrosis (17), the relative importance of the inflammatory response to the progression of IPF has been heavily debated (243), and is punctuated by a poor response to anti-inflammatory therapy (194). Wound repair is a fundamental and intricately orchestrated process that replaces dead or damaged cells (243).

Dysregulated wound repair results in excess accumulation of ECM molecules such as hyaluronic acid, fibronectin, proteoglycans, and collagens leading to the development of permanent tissue scarring (243). The scar tissue buildup leads to organ malfunction

through reduced compliance of the lung and disruption of gas exchange, ultimately causing death from respiratory failure (243).

TGF-β1 Signaling in Pulmonary Fibrosis

Decades of research have proven a central role for transforming growth factor betas (TGF-βs) in tissue fibrosis (105). TGF-βs are pleiotropic cytokines and members of the TGF-β superfamily which includes related cytokines such as the bone morphogenetic proteins, activins, and inhibins (139). TGF-βs control a myriad of cellular functions such as cell proliferation, differentiation, and migration (40) and are thus critical for regulating embryonic development, wound healing, and other homeostatic processes (4, 105). Mammals have three different forms of TGF-βs (TGF-β1, TGF-β2, and TGF-β3) which are produced in the form of latent complexes in which the bioactive

TGF-β is covalently bonded to the latency associated protein (LAP) and linked to the

ECM (2). TGF-β can be activated by proteases such as matrix metalloproteinases (189,

252), thrombospondin-1 (192), reactive oxygen species (ROS) (8, 9), pH (115), integrins such as αvβ6 (142), or by contractile forces (241).

All activated TGF-βs signal through the same high-affinity cell surface receptors,

TGF-β type I and type II serine/threonine kinase receptors (TBRI and TBRII) of which there are five and seven members, respectively (40). Although the TGF-βs all signal through the same receptors, they exert distinct biological functions due to the diverse heterodimeric receptor combinations possible among individual type II and type I receptors which preferentially bind different ligands and induce different signaling pathways (40). Typically, TGF-β binds to the constitutively active TBRII which phosphorylates and heterodimerizes with TBR1 forming an active ligand-receptor complex (4). Phosphorylated TBRI phosphorylates the downstream effector Smads

(Smad2 and Smad3) which heterodimerize with the co-operating Smad (Smad4) to form

Smad 2/4 or Smad 3/4 complexes (185). The Smad complexes then translocate to the nucleus where they bind to Smad response elements in the promoters of TGF-β target genes (40). The third type of Smad proteins, the inhibitory Smads (Smad6 and Smad7) inhibit Smad2 and Smad3 phosphorylation or act as a competitive inhibitor of Smad2/3 binding to TBRI (145).

TGF-β1 is the isoform most closely related with the development of IPF (244), and the consensus is that the underlying mechanism of IPF is a result of abnormal behavior of alveolar epithelial cells (AECs) which produce increased amounts of TGF-

β1 (52). TGF-β1 controls several processes implicated in the pathogenesis of IPF such as the recruitment of circulating fibrocytes and bone marrow-derived progenitor cells to the lung, the induction of epithelial and endothelial to mesenchymal transition (EMT and

EndoMT, respectively), processes by which epithelial and endothelial cells transdifferentiate into a more mesenchymal phenotype, and the activation, migration, and proliferation of resident fibroblasts and their differentiation into myofibroblasts (52).

These mechanisms lead to the formation of fibroblast and myofibroblast foci which secrete ECM molecules such as collagens thereby leading to reorganization of lung architecture and scar formation (96).

Several signaling mediators enhance TGF-β-driven fibrosis including connective tissue growth factor (CTGF) (105). CTGF is a secreted and inducible matricellular protein and member of the CCN family of immediate-early genes (Cyr61 and Nov)

(106). CTGF is upregulated in bronchoalveolar lavage fluid (BALF) of patients with IPF

(1) and modulates signaling pathways by binding to ECM proteins such as fibronectin,

cell surface receptors such as integrins, and growth factors of the cysteine knot superfamily (27, 160, 161). It can act both downstream of TGF-β1 (64) but also independently (207) to activate pro-fibrotic processes such as EMT, fibroblast proliferation, and ECM production (104, 106, 208).

Multi-Walled Carbon Nanotube-Induced Pulmonary Fibrosis

There is growing concern that a class of emerging environmental contaminants termed nanomaterials, small particles that are being utilized in numerous industrial and consumer products, may also have the capacity to cause lung injury through inhalation exposures, potentially leading to pulmonary fibrosis (141, 165, 182). Carbon nanotubes

(CNTs) are a type of nanomaterial comprised of one or many layers of rolled up graphene sheets, termed single-(SWCNTs) or multi-walled carbon nanotubes

(MWCNTs), respectively, and have a diameter less than 100 nm, high aspect ratio, and unique physiochemical and electro-conductive properties (39). The unique properties of

CNTs make them appealing for a variety of applications in electronics, engineering, manufacturing and medicine (50). Their widespread application and increased use each year also increases the potential for human exposure in occupational, environmental, medical, and consumer environments (229). There is particular concern for adverse pulmonary effects through inhalation due to the high aspect ratio of CNTs similar to other particulates with known adverse effects in humans such as asbestos fibers (45).

Animal studies have indicated that exposure to CNTs through various routes of administration causes respiratory toxicity including inflammatory changes, lung remodeling and fibrosis, mesothelioma, and promotion of lung adenocarcinoma (141,

165, 181, 182, 187). The pro-fibrotic potential of MWCNTs is of particular concern given the measurement of biomarkers of fibrosis in the blood of workers in a plant

manufacturing MWCNTs (51). In animals, MWCNTs acutely cause dose-dependent fibrotic-like changes in the lung which is evidenced by excess collagen deposition in foci of inflammation, upregulation of TGF-β1, and induction of fibrotic marker gene expression such as collagen 1 (Col1a1), collagen 2 (Col1a2), and fibronectin (Fn1) (47,

167, 232). Chronic effects of MWCNT exposure are increasingly apparent as it was shown that MWCNTs increased the thickness of fibrillary collagen in the alveolar septa of mice 336-days post-exposure. This report suggests that inhaled MWCNTs are retained in the alveolar tissue and promote a progressive and persistent fibrotic response (129).

Numerous in vitro mechanistic studies indicate that MWCNTs induce changes consistent with pulmonary fibrosis through interactions with inflammatory, epithelial, and fibroblast cells [reviewed by Vietti et al. (229)]. For example, MWCNTs promote pro- fibrogenic signaling by macrophages including inflammasome activation and production of pro-inflammatory and pro-fibrogenic mediators such as platelet derived growth factor

(PDGF), IL-1β, IL-18, and TGF-β1 (69, 75, 130, 231). Direct effects of MWCNTs on fibroblasts include induction of proliferation, differentiation, and collagen production in vitro (75, 228). Mechanistic studies have identified a casual role for TGF-β1 in a number of these effects (28, 133, 231). MWCNTs also stimulate the secretion of pro-fibrogenic and pro-inflammatory mediators from epithelial cells including TGF-β1, IL-1, IL-6, IL-8, and IL-18 which promote EMT and can indirectly activate fibroblasts to differentiate and secrete ECM components (75, 78, 216, 233). The effects in epithelial cells are particularly striking as it is suspected that fibrosis is a result of persistent epithelial injury leading to altered epithelial to fibroblast signaling and dysregulated wound repair (183)

Sex and Sex Hormones in Pulmonary Fibrosis

There is increasing evidence for a role of sex hormones (estrogen, progesterone, and testosterone) in non-reproductive organs such as the lung (188). Accordingly, epidemiological studies suggest sex-specific trends in various lung diseases including

IPF, lympangeiomyomatosis (LAM), idiopathic pulmonary arterial hypertension (IPAH) chronic obstructive pulmonary disease (COPD), asthma, and lung cancer [reviewed by

Carey et al (23)]. While LAM, IPAH, and asthma seem to predominantly affect women and the rates of COPD are rising faster in women than in men (12, 68, 120, 178, 227),

IPF appears to be more prevalent in men and women diagnosed with the disease have better survival rates (63, 70, 149, 171) suggesting either an exacerbating role for androgens or a protective role for estrogens.

Evidence of a role for sex hormones in animal models of pulmonary fibrosis is largely model-dependent. In the most common animal model for IPF, the bleomycin mouse model, young and aged male mice exhibited increased deposition of collagen and reduced static compliance compared to age-matched female mice (175). Another study indicated that androgen (5-α-dihydrotestosterone, DHT) supplementation increased lung function decline in gonadectomized male and female mice treated with bleomycin and reported a negligible effect of estrogen receptor (ER) knockout (Esr1-/- or

Esr2-/-) based on measurements of quasi-static compliance (230). A recent study found that it was the presence of the Y chromosome and not necessarily sex itself that predisposed the lung to increased bleomycin induced fibrosis in male and female mice

(225). Notably, the sex-specific disparities in this common model for IPF may be due to variable activity of bleomycin hydrolase in the lungs of male and female mice (72).

Taken together, these results suggest that androgens potentiate bleomycin induced fibrosis in mice.

While an exacerbating role for androgens in pulmonary fibrosis is consistent across studies, a precise role for estrogens is less defined. For example, bleomycin treatment of female rats caused significantly larger and more consolidated fibrotic lesions, increased expression of procollagens, and denser collagen deposition compared to male rats (59). Ovariectomy diminished bleomycin-induced lung fibrosis compared to sham-operated females and fibrosis was restored after 17β-Estradiol (E2) supplementation (59). The authors established a direct role for E2 in potentiating pro- fibrogenic signaling by exposing rat lung fibroblasts isolated from bleomycin-treated females to E2 in vitro and measuring dose-dependent increases in TGF-β1 and procollagen α1 mRNA (59). The discordant results in rats may be due to species- specific differences in expression and/or activity of bleomycin hydrolase as it was shown that pulmonary fibrotic responses to bleomycin treatment in male and female mice was driven by variable activity of bleomycin hydrolase between the sexes (72). Conversely, a protective role for estrogens against fibrosis was suggested in a study that found ovariectomy caused significant increases in total lung collagen content and airway subepithelial collagen deposition in female mice compared to sham operated females

(108).

The limited studies investigating sex-based differences in response to particulate exposure have similarly produced mixed results. For example, female mice exposed to the particulate silica had less hydroxyproline content and more inflammatory cells in the lung than silica-exposed males (16). One study found that the relative resistance of

female mice to silica-induced fibrosis compared to male mice was due to E2-mediated repression of secreted phosphoprotein 1 (SPP1, also known as osteopontin) (103).

Conversely, a study by Shvedova et al. found that female mice exhibited a greater pulmonary fibrotic response after pharyngeal aspiration of the same dose of cellulose nanocrystals compared to male mice (200). Shatkin and Oberdorster suggested that the heightened response in the latter study was due to the smaller body weight of female mice compared to male mice resulting in a greater dose per kilogram in females (198), but Shvedova et al. pointed out that although the overall body weights of male and female mice differ, the lung volume/mass between male and female mice of the same age are comparable.

Estrogen Signaling in the Lung

E2 is the major biologically active estrogen in humans and is primarily produced in the gonads from cholesterol and ultimate aromatization of testosterone by aromatase

(CYP19) (156). E2 is transported from the gonads to target tissues in the bloodstream bound to sex hormone binding globulins (201). E2 and other sex steroids are also produced locally in various non-gonadal tissues including the lung although at lower levels (102). The levels of E2 are stable in men in the pg/mL range while E2 fluctuates in different life stages in females ranging from 20 pg/mL in the follicular phase in non- pregnant and post-menopausal women to 40 ng/mL in pregnant women (188). The relative importance of locally produced E2 is suspected to depend on the levels of circulating E2 (188).

E2 primarily exerts its biological effects by classical, genomic mechanisms through binding and activation of the nuclear estrogen receptors (ERs), estrogen receptor α (ESR1) and estrogen receptor β (ESR2), which are members of the nuclear

receptor superfamily of ligand-dependent transcription factors (65). Binding of E2 to

ESR1 and ESR2 induces a conformational change that promotes hetero or homodimerization between the receptors and recruitment of various co-regulatory proteins (65). The ER transcriptional complex binds to estrogen response elements in the promoters of E2-responsive genes and regulates their transcription (65). ERs can also interact with other transcriptional factors such as c-fos/c-jun and interact with AP-1- responsive elements to indirectly influence gene transcription (188).

In addition to the classical mechanism of E2 action, there is growing recognition of a role for non-genomic mechanisms of E2 signaling. For example, E2 binds to and activates the membrane bound G-protein coupled estrogen receptor 1 (GPER1) to activate various kinase pathways and influence gene transcription (53, 169). Further, full length ESR1 (ESR1-66) and truncated isoforms (ESR1-46, ESR1-36) so named for their molecular weights, are also able to signal from the cell membrane adding another layer of biological complexity to E2 signaling (109). E2 activates rapid effects within seconds to minutes via membrane receptors such as activation of cyclic nucleotides, kinases, and phosphatases and usually modulate ion fluxes across the cell membrane

(91).

Interactions Between TGF-β1 and E2 Signaling

Few studies have investigated interactions between pro-fibrogenic signaling mediators such as TGF-β1 and E2 in the lung to explain the sex-based differences aside from one that showed increased TGF-β1 expression in rat lung fibroblasts exposed to E2 (59). Regulation of TGF-β1 by E2 has been extensively characterized in other model cell systems and the effects appear to be contextual. For example, E2 inhibited TGF-β1 signaling in breast cancer cells by reducing the expression of

activators of TGF-β1 (26) and by increasing the proteasomal-dependent degradation of signaling mediators downstream of TGF-β1 including the Smad proteins (61, 81).

Conversely, E2 increased TGF-β1 secretion in dermal fibroblasts (3), and TGF-β1 levels were reduced in the kidneys of diabetic female mice lacking ESR1 compared to wildtype mice suggesting positive regulation of TGF-β1 by ESR1 (114).

Studies aiming to decipher receptor-specific effects have also produced variable results. For example, Stope et al. found that ESR1 but not ESR2 inhibited TGF-β1 activation in gene reporter assays in breast cancer cells (213) while another group found that both ESR1 and ESR2 suppressed TGF-β1 signaling by associating with and acting as a transcriptional corepressor for Smad3 (124, 245). Other studies have shown a role for GPER1 in mediating estrogen-dependent reduction in Smad protein activation in breast cancer cells (99) and human and rat mesangial cells by suppressing Smad2/3 phosphorylation and Smad4 complex formation as well as proteasome-dependent

Smad2 degradation (112).

In addition to direct effects of E2 signaling on TGF-β1, E2 activation of ERs also targets TGF-β1-driven cellular processes. For example, a role for E2 in inhibiting EMT in humans was suggested in a study that found that reduced expression of ESR1 immunohistochemical staining was associated with increased expression of genes involved in EMT in endometrial carcinoma samples (239). Further, EMT is a target for

E2 in cell types such as breast and prostate cancer cells where E2 signaling maintained an epithelial phenotype and suppressed EMT (41, 66, 117, 249), while another found that E2 promoted a reversible EMT-like transition and collective motility in MCF-7 breast

cancer cells (163). Another study found that E2 signaled through GPER1 to inhibit TGF-

β1-induced ECM production in human and rat mesangial cells (112).

Likewise, TGF-β1-has been shown to impact ERs. For example, TGF-β1 reduced

ESR1 mRNA expression (57, 212) and ESR1 protein expression (158, 212) in breast epithelial cancer cells and ESR2 protein expression in prostate cancer cells (117).

Petrel et al. found that TGF-β1 promoted proteasomal-dependent degradation of ESR1 in breast cancer cells (158) while in human dermal fibroblasts, TGF-β1 increased ESR1 mRNA and protein levels (73).

Cell-Specific Effects in Pulmonary Fibrosis

The lung is comprised of over 40 different cell types in the epithelium, interstitial connective tissue, blood vessels, hematopoietic and lymphoid tissue, and the pleura

(131). The discrete cell types are often implicated in different pathological conditions

(56) and can respond differently to insults (242). Further, many genes are multifunctional and play different roles in different cell types (92). As such, it is important to consider cell type specific effects when assessing the molecular and cellular impacts of pathological or toxicological conditions in the lung.

Cell type specific analyses are also more sensitive measurements than whole lung endpoints. Cell type specific gene expression analyses can detect differences in gene expression between diseased and control individuals that are not apparent when the whole tissue is analyzed due to a dilution of response (92). For example, Kelly et al. measured the expression of various genes involved in pulmonary fibrosis in specific types in the lungs of patients with usual interstitial pneumonia (UIP) and healthy controls by using laser capture microdissection. They reported increased expression of TIMP1,

MMP7, and TGFB1 in discrete cell types in patients with UIP compared to healthy controls that was not apparent when whole lung tissue was analyzed (92).

Research Objectives

IPF is a devastating disease responsible for approximately 40,000 deaths annually (13), and there is growing evidence that a number of these cases can be attributed to novel occupational exposures (190). The observation of sex-based differences from epidemiology studies of IPF (63, 70, 149) and differences in fibrotic response between males and females in small animal models of experimental pulmonary fibrosis (59, 108, 175, 230) suggest sex hormones are involved. However, interactions between pro-fibrogenic and sex hormone receptors at the molecular level are poorly defined as are the molecular and cellular and transcriptional targets of sex hormones in the lung. Therefore, there is a critical need to characterize interactions between pro-fibrogenic signaling and targets of sex hormones in the lung to increase our understanding of disease pathogenesis which may help to better stratify individuals who may respond better to treatments based on sex hormone status, and to identify potential targets of therapeutic intervention.

The objective of this work is to identify interactions between pro-fibrogenic and estrogen receptor signaling pathways and to identify novel targets of E2 in the lung. The hypothesis for this work was that E2 inhibits pro-fibrogenic signaling in the lung and thus protects against the development of pulmonary fibrosis. The hypothesis was tested through three specific aims: Aim 1 identified estrogen receptors as targets of MWCNT- induced pro-fibrogenic signaling using both in vivo and in vitro models of MWCNT- induced fibrogenesis; Aim 2. identified novel transcriptional targets of E2 in lung cells that potentially influence the development of pulmonary fibrosis; Aim 3 quantified gene

expression changes in lung tissues from individuals diagnosed with IPF to identify potential sex hormone-based differences between diseased and non-diseased lesions.

CHAPTER 2 MULTI-WALLED CARBON NANOTUBES INHIBIT ESTROGEN RECEPTOR ALPHA EXPRESSION IN THE LUNG VIA TGF-β1

Introduction

Carbon nanotubes (CNTs) are rolled up graphene sheets consisting of one or concentric layers, termed single- (SWCNTs) or multi-walled carbon nanotubes

(MWCNTs), respectively, and have a diameter less than 100 nm, high aspect ratio, and unique physiochemical and electro-conductive properties (39). The unique properties of

CNTs make them appealing for a variety of applications in electronics, engineering, manufacturing and medicine (50). Their widespread application and increased use each year enhances the potential for human exposure through occupational, environmental, medical, and consumer routes (229). There is particular concern for adverse pulmonary effects occurring through inhalation due to the high aspect ratio of CNTs (45).

Animal studies have indicated that exposure to CNTs through pulmonary administration causes respiratory toxicity including inflammatory changes, lung remodeling and fibrosis, mesothelioma, and promotion of lung adenocarcinoma (141,

165, 181, 186). The pro-fibrotic potential of MWCNTs is of particular concern given the measurement of biomarkers of fibrosis in the blood of workers in MWCNT manufacturing plants (51). In animals, MWCNTs have been shown to cause acute dose-dependent fibrotic-like changes in the lung, evidenced by excess collagen deposition in foci of inflammation, upregulation of the pro-fibrogenic cytokine, transforming growth factor beta1 (TGF-β1), and induction of fibrotic marker gene expression such as collagen 1 (Col1a1), collagen 2 (Col1a2), and fibronectin (Fn1) in mice (47, 167, 232). Chronic effects of MWCNT exposure are increasingly apparent as it was shown that MWCNTs increased the thickness of fibrillary collagen in the alveolar

septa of mice 336 days post-exposure suggesting that these inhaled particulates are retained in the alveolar tissue and promote a progressive and persistent fibrotic response (129).

While evidence supports a link between MWCNTs and other particles and fibers to pulmonary fibrosis, few studies have investigated sex-specific differences in disease development and pathogenesis despite epidemiological studies indicating that pulmonary fibrosis is more common in men and women have better survival rates (63,

70). Results from studies investigating a role for sex in a common animal model for pulmonary fibrosis, the bleomcycin model, have been mixed. For example, male mice exhibited greater fibrosis after treatment with the chemotherapeutic bleomycin than female mice (175, 230), whereas estrogen (17β-Estradiol, E2) supplementation exacerbated bleomycin-induced fibrosis in rats (59). Despite the observation of sex- specific differences in fibrotic responses, few studies have explored a mechanistic role for sex hormones and sex steroid receptors in the lung to explain the sex-based differences in disease outcomes.

To begin mechanistic investigations into a role for sex hormones in particulate- induced lung disease, we sought to probe interactions between pro-fibrogenic and estrogen receptor signaling mechanisms in both in vivo and in vitro models of MWCNT exposure. Because E2 exerts its effects by binding and activating several receptors including the nuclear transcription factors estrogen receptor alpha (ESR1), estrogen receptor beta (ESR2), and several variants thereof (65), and the membrane-bound G- protein coupled estrogen receptor 1 (GPER) which potentiates rapid non-genomic effects (53, 169), we explored the effect of MWCNT exposure on estrogen receptor

expression as a surrogate for E2 signaling. Herein we report the novel observation that

MWCNTs downregulate mRNA expression of Esr1 and Esr2, and to a lesser extent mRNA expression of Gper1, in the lungs of mice exposed to MWCNTs by whole body inhalation, and ESR1 mRNA expression in human bronchial epithelial cells. We proceed to show that the MWCNT-induced reduction of ESR1 expression in vitro is largely mediated by TGF-β1 and not dependent on CNT surface chemistry. Overall, these data are the first to expose E2 signaling as a target of pro-fibrogenic pathways in the lung and lay the foundation for more directed mechanistic studies to decipher the molecular role for ESRs in sex-specific differences in lung diseases.

Materials and Methods

Chemicals

Four types of MWCNTs were used in this study. The two non-functionalized

(pristine) MWCNTs were MWCNT-XNRI-7 from Mitsui & Co., Ltd. Tokyo, Japan (Lot

#05072001K28 and Lot#061220-31) referred to here as Mitsui MWCNTs and another tube from Helix Materials Solutions, Inc., Richardson, Texas, USA referred to as Helix

MWCNTs. The functionalized CNTs used in the study included hydroxylated MWCNTs referred to as oxMWCNTs obtained from our collaborator, Dr. Navid Saleh at University of Texas at Austin, and hydroxylated SWCNTs referred here as P3 SWCNTs (Carbon

Solutions Inc.). The Selective inhibitor of TGF-β type-I receptor, LY 364947 (purity: >

99%) was purchased from Tocris Bioscience, Bristol, UK (Cat. No. 2718) and dissolved in CorningTM cellgroTM dimethyl sulfoxide (DMSO, CorningTM 25950CQC, Thermo Fisher

Scientific, Waltham, MA), final solvent concentration < 0.1%. Bovine serum albumin

(BSA) was purchased from Fisher Scientific Co LLC (BP1605, Pittsburgh, PA) and 1,2-

Dipalmitoyl-sn-glycero-3-phosphocholine (purity: ≥ 99%) (DPPC) was purchased from

Sigma (P0163, Saint Louis, MO).

Animal Exposure

The mouse inhalation exposure protocol was performed at NIOSH and has been described previously (167). Briefly, Male C57BL/6J mice (6 weeks old) obtained from

Jackson Laboratories (Bar Harbor, ME) were housed one per cage in an AAALAC- accredited, specific pathogen-free, environmentally controlled facility. For the gene expression experiments, mice were exposed to a Mitsui MWCNT aerosol (10 mg/m3, 5 h/day) for either 2, 4, 8, or 12 days, using an acoustical-based computer-controlled whole-body inhalation system designed and constructed in Dr. Porter’s lab (127), and sacrificed one day thereafter. For measurement of TGF-β1 in the bronchoalveolar lavage fluid (BALF), mice were exposed to a Mitsui MWCNT aerosol (5 mg/m3, 5 h/day) for 4 days and sacrificed one day thereafter (129). The Mitsui MWCNTs working stocks used in this study and the aerosols generated have been extensively characterized elsewhere (167). All procedures were approved by the NIOSH Institutional Animal Care and Use Committee.

Histopathology

Formalin-fixed and paraffin-embedded tissue blocks were cut at 5 μm and mounted on non-coverslipped slides. H&E staining, Masson’s Trichrome staining, and

Alcian-Blue/PAS staining was performed by Histology Tech Services (Gainesville, FL).

Qualitative analyses of histopathology were performed with the assistance of Dr. Dale

Porter and Dr. Marlene Orandle at CDC NIOSH.

Enzyme-Linked Immunosorbent Assay (ELISA)

Bronchoalveolar lavage samples (BALF) from mice exposed to clean air or Mitsui

MWCNT aerosol (5 mg/m3, 5 h/day) for 4 days were centrifuged for 20 minutes at

14,000 x g at 4°C, and 100 μL of the supernatant was removed for TGF-β1 quantification using the Quantikine® ELISA kit from R&D Systems (MB100B,

Minneapolis, MN). Latent TGF-β1 was activated by adding 20 μL of 1 N HCl to 100 μL supernatant, vortexing, and incubating for 10 minutes at room temperature. The reaction was neutralized by adding 20 μL of 1.2 N NaOH/0.5 M HEPES and vortexing.

Active TGF-β1 was quantified per manufacturer’s recommendations.

Immunohistochemistry (IHC)

Formalin-fixed and paraffin-embedded lung tissue blocks from mice exposed to clean air or 10 mg/m3 MWCNT for 5 hours per day (n=3) from each time-point were sectioned at 5 μm and mounted on non-coverslipped slides. Slides were warmed at

50°C for 30 minutes to soften paraffin and subsequently de-paraffinized in Hemo-De

(HD-150A, Scientific Safety Solvents, Keller, TX) 3 x 10 minutes. Thereafter slides were rehydrated through successive washes of 100% EtOH, 95% EtOH, 70% EtOH, 2 x 5 minutes each, then rinsed in tap water for 5 minutes. Antigen retrieval was performed by microwaving the slides in 0.1 M citrate buffer, pH 6 for 18 minutes, then allowing the slides to reach room temperature. Endogenous peroxidase activity was blocked by incubating slides in 3% hydrogen peroxide in tap water for 5 minutes, then washed 2 x 2 minutes in PBS. ESR1 was detected using Vector® M.O.M.TM Immunodetection Kit (PK-

2200, Vector Laboratories, Burlingame, CA) per manufacturer’s protocol. Sections were blocked overnight at 4°C in working solution of M.O.MTM Mouse IgG Blocking Reagent, allowed to return to room temperature for one hour the next day, then washed 2 x 2

minutes in PBS. Tissues sections were incubated in working solution of M.O.M.TM diluent prepared as described for 5 minutes after which the excess was tipped off and the sections were incubated with anti-ESR1 (sc-8002, Santa Cruz Biotechnology, Inc.,

Dallas, TX) diluted 1:50 in M.O.M.TM diluent for 30 minutes at room temperature, then washed 2 x 2 minutes in PBS. Slides were incubated with VECTASTAIN® ABC

Reagent prepared as described for 5 minutes and washed 2 x 5 minutes in PBS. Slides were developed with DAB (SK-4100, Vector Laboratories) per manufacturer’s protocol.

Slides were rinsed in running tap water for 5 minutes, counter stained with Hemotoxylin for 15 seconds, then rinsed in tap water for 5 minutes. Slides were dehydrated by successive washes of 70% EtOH, 90% EtOH, 95% EtOH, 100% EtOH x 2, Hemo-De 3 x 5 minutes each, then mounted with Fisher Chemical™ Permount™ Mounting Medium

(SP15-100, Fisher Scientific).

Cell Culture

Human bronchial epithelial cells (BEAS-2Bs, CRL-9609TM) were purchased from

ATCC and cultured per the manufacturer’s specifications. Cells were cultured in bronchial epithelial growth medium (BEGM) consisting of bronchial epithelial basal medium (BEBM, Lonza CC-3171, Walkersville, MD) and the BEGM SingleQuot Kit

Supplements & Growth Factors (Lonza CC-4175) but exchanging gentamicin for

Penicillin-Streptomycin-Neomycin (PSN) Antibiotic Mixture (Gibco 15640, ThermoFisher

Scientific Inc.). Cells were cultured in T75 Corning™ U-Shaped Cell Culture Flasks

(Corning 430641U, Fisher Scientific Co LLC) coated with a matrix (4.5 mL per 75 cm2) consisting of 0.01 mg/mL fibronectin (Akron AK8350, Boca Raton, FL), 0.03 mg/mL bovine collagen (Gibco A10644-01, ThermoFisher Scientific Inc.), and 0.01 mg/mL BSA

(Fisher BP1605, Fisher Scientific Co LLC) in BEBM. All exposures were performed in

BEGM without the supplied bovine pituitary extract (BPE) aliquot because its composition is not defined. For the gene expression experiments, BEAS-2Bs were plated at 40,000 cells/mL on matrix-coated 12-well Nunc™ Cell-Culture Treated

Multidishes (ThermoFisher Scientific Inc., 150628), allowed to adhere overnight, and subsequently exposed for 48 hours to the indicated concentrations of MWCNTs.

Nanotube Dispersion and Characterization in vitro

For in vitro experiments, stock solutions of Mitsui MWCNTs were dispersed based on the method described by Porter et al. (166). Briefly, Mitsui MWCNTs were suspended in dispersion media (Ca2+- and Mg2+-free phosphate buffered saline, pH 7.4, supplemented with 0.6 mg/mL BSA and 0.01 mg/mL DPPC) at 2 mg/mL. The MWCNT stock was sonicated in a water bath sonicator (Branson B-220) for 10 minutes, then sonicated using a Branson 450 Sonifier fitted with a 1/8-inch tip in an ice bath at 50% amplitude for 10 minutes with an 8 second pulse and 10 second rest between pulses.

Stock solutions of Helix MWCNTs were prepared based on the method described by

Wang et al. (234). Briefly, Helix MWCNTs were suspended in ultrapure water at 5 mg/mL and sonicated using the same parameters as for the Mitsui MWCNT dispersion.

Both MWCNT stock solutions were dispersed at the concentrations indicated in the manuscript in BEGM without BPE supplemented with 0.6 mg/mL BSA and 0.01 mg/mL

DPPC.

Time dependent hydrodynamic radii (HDR) of both Helix (Figure 2-1A) and Mitsui

(Figure 2-1B) MWCNTs were monitored with time resolved dynamic light scattering

(TRDLS). A 22 mW 632 nm HeNe laser incorporated ALV/CGS-3 compact goniometer system (ALV-Laser GmbH, Langen/Hessen, Germany) with QE APD detector

(photomultipliers of 1:25 sensitivity) was employed to monitor size evolution every 15

seconds for 48 hours. The TRDLS experiments were performed at 37°C for both samples with 2 mL sample in a borosilicate glass vial, which was cleaned thoroughly with 2% extran solution. These vials were vortexed for 10 seconds before inserting in the toluene-filled sample vet of the goniometer system. The scattered light was collected at 90° and analyzed using an auto cross-correlator to calculate average HDR.

The electrophoretic mobility (EPM) of both MWCNT suspensions were measured using a Möbiuζ (Wyatt Technology, Santa Barbara, CA) at 25 °C and 30 psi (Figure 2-

1C). For each measurement, 1 mL of the MWCNT suspension was introduced into a flow through cell using a 1 mL syringe. Five EPM values were collected for each of the

MWCNT suspensions. The cells were washed with deionized water and ethanol between measurements.

Total RNA Extraction and Purification

Total RNA was extracted from whole lung and cell lysates using RNA STAT-60TM

Reagent (Tel-Test, Inc. Cs-502, Friendswood, TX). Cell lysates were vortexed and whole lung tissue was mechanically disrupted. Total RNA was extracted per manufacturer’s specifications. RNA was precipitated overnight at -20°C in 100% molecular biology grade isopropanol (Fisher BioReagentsTM BP26184, ThermoFisher

Scientific Inc.) containing 0.067% GlycoBlueTM Coprecipitant (Ambion® AM9515,

ThermoFisher Scientific Inc.) and purified by washing 2X with 75% molecular biology grade absolute ethanol (Fisher BioReagentsTM BP28184, ThermoFisher Scientific Inc.).

RNA pellets were reconstituted in 15 μL RNAsecure™ (Ambion® AM7010,

ThermoFisher Scientific Inc.). RNA was quantified using a SynergyTM H1 plate reader

(BioTek Instrument, Inc., Winooski, VT) and RNA integrity was spot-checked using a

Bioanalyzer 2100 instrument (Agilent Technologies, Santa Clara, CA).

Quantitative Real-Time Polymerase Chain Reaction (qPCR)

Total RNA was DNase-treated using the PerfeCTa DNase I Kit (Quanta

BioSciences 95150-01k, VWR International LLC, Suwanee, GA) and DNase-treated

RNA was subsequently reverse transcribed using the qScript™ cDNA Synthesis Kit

(Quanta BioSciences 95047, VWR International LLC). cDNA was diluted 1:20 in

RNase-DNase free water. Each 10 μL qPCR reaction contained 1x SsoAdvanced™

Universal SYBR® Green Supermix (Bio-Rad 172-5270, Hercules, CA), 850 nM forward and reverse primers, and 3.3 μL of the cDNA dilution. Gene specific primers and cycling parameters are displayed in Table 2-1. Each qPCR reaction was followed by melt curve analysis to verify primer specificity. Cq values were determined by regression method using the CFX Manager 2.1 software and quantified using the relative ΔΔCq method

(76). Target gene expression was normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression. In the case of no amplification, a Cq value of 40 was applied.

Intracellular Reactive Oxygen Species (ROS) Measurement

Intracellular ROS was measured using the DCFDA Cellular ROS Detection

Assay kit (abcam®ab113851, Cambridge, UK). BEAS-2Bs were plated at 30,000 cells per well on matrix-coated black wall, clear bottom 96-well plates (Corning 3603, Fisher

Scientific Co LLC) and allowed to adhere overnight. Once cells reached 90% confluency, they were washed with 1X Buffer, then stained with 20 mM DCFDA dye for

45 minutes at 37°C in the dark. Thereafter, DCFDA solution was removed, and the cells were washed with 1X PBS. The washed cells were exposed to increasing concentrations of MWCNTs for 6 hours and fluorescence was measured at Ex/Em =

485/535 nm.

Statistics

Normality of experimental data was determined by D'Agostino & Pearson omnibus normality test, Shapiro-Wilk normality test, and KS normality test using

GraphPad Prism software (Version 5.01, GraphPad Software, Inc., La Jolla, CA). Data were determined to be normal by passing at least one normality test (p < 0.05). If the data were normal, significant differences (p < 0.05) in means were determined by one- way ANOVA followed by Newman-Keuls multiple comparison test or two-tailed unpaired t test using GraphPad Prism software. If the data were not normal, significant differences (p < 0.05) were determined by Kruskal-Wallis test followed by Dunn’s multiple comparison test or Mann-Whitney U test using GraphPad Prism software.

Results

Mitsui MWCNTs Cause Histopathological Alterations in the Lungs of Mice

The histopathological alterations observed in this study were reported in full detail elsewhere (167). Briefly, the lungs of exposed animals were characterized by inflammation centered at the bronchioalveolar junction, hyperplasia and hypertrophy of the bronchiolar epithelium, airway mucous metaplasia, fibrosis, and vascular changes

(167).

Bronchiolar epithelial cell hypertrophy and hyperplasia was present in all Mitsui

MWCNT-exposed mice at all time points (Figure 2-1B-D). Lung fibrosis was minimal to mild in severity as indicated by a subtle increase in collagen staining in the lungs of mice exposed to Mitsui MWCNTs for 8 and 12 days (Figure 2-1G-H). Prominent mucous metaplasia was apparent in mice exposed to Mitsui MWCNTs for 8 and 12 days and was primarily bronchiolocentric as determined by heavy magenta staining

(Figure 2-1K-L).

Mitsui MWCNTs Increase TGF-β1 Levels in BALF

The levels of TGF-β1 in the BALF of mice exposed to clean air or 5 mg/m3 Mitsui

MWCNTs for 5 hours per day for 4 days was quantified by ELISA. The mean ± SEM of

TGF-β1 levels in control and Mitsui MWCNT-exposed mice were 86.67 ± 14.92 (n = 6) and 121.9 ± 12.91 (n = 8), respectively, and the means did not quite reach statistical significance but showed an increased trend (p = 0.10) (Figure 2-3).

Mitsui MWCNTs Reduce Estrogen Receptor mRNA Expression in Mouse Lungs

To determine whether the estrogen receptors are targets for Mitsui MWCNT- induced pro-fibrogenic signaling, we evaluated mRNA expression of Esr1, Esr2, and

Gper1 by qPCR in whole lungs of mice 2, 4, 8, or 12 days after exposure to clean air or

10 mg/m3 Mitsui MWCNTs by whole-body inhalation for 5 hours per day. The expression levels Esr1, Esr2, and Gper1 in whole lungs of mice exposed to clean air for

2 days were expressed as a ratio to Esr2 based on the method described by Pfaffl et al.

(159) to determine relative baseline expression. The basal levels of expression of each receptor subtype was Gper1 > Esr1 > Esr2 (Figure 2-4A).

For the time-course analysis, fold changes for all samples were calculated relative to expression levels in lungs of mice exposed to clean air for 2 days. The relative expression of each receptor subtype varied over the time-course in both clean air- and Mitsui MWCNT-exposed mice (Figure 2-4B-D). The expression of Esr1 was significantly reduced in Mitsui MWCNT-exposed mice compared to time-point matched control mice after 4 and 12 days of exposure. Expression of Esr1 was reduced compared to controls after 2 and 8 days of exposure but the differences were not significant (p > 0.05, Figure 2-4B). The expression of Esr2 was significantly reduced compared to controls after 2 and 4 days and was reduced after 8 and 12 days, but the

differences were not significant (p > 0.05, Figure 2-4C). Gper1 expression varied the most and the relative expression levels inverted over the time-course and were only significantly different (p < 0.05) after 4 days of exposure (Figure 2-4D). We also measured CTGF mRNA expression across the time-course as an indirect measure of

TGF-β-induced pro-fibrogenic signaling. CTGF was significantly reduced in the lungs of mice exposed to MWCNTs for 2 and 4 days, but expression returned to control levels after 8 and 12 days of exposure (Figure A-1).

ESR1 Protein is Expressed in Mouse Lung Epithelial Cells

We performed immunohistochemical staining to localize the expression of ESR1 protein in the mouse lung. Positive nuclear staining for ESR1 was apparent in all samples at every time-point. Expression was particularly apparent in epithelial cells lining the bronchi and larger airways (Figure 2-5). There was also positive staining in epithelial cells in the interstitial space (Figure 2-5). A tissue section from a control mouse that was incubated without primary antibody revealed no positive staining for

ESR1 (Data not shown).

Mitsui MWCNTs Reduce ESR1 mRNA Expression in BEAS-2Bs

To begin mechanistic evaluations regarding a role for TGF-β1 in MWCNT- induced repression of estrogen receptor mRNA expression, we characterized this response in bronchial epithelial cells (BEAS-2Bs) in vitro. First, we measured the relative baseline expression levels of ESR1, ESR2, and GPER1 in control cells as a ratio to ESR2 using the method described previously (159). The expression pattern for each receptor subtype in vitro was GPER1 > ESR1 > ESR2 which matched the relative pattern observed in mouse lungs in vivo (Figure 2-6A and Figure 2-4A).

Similar to results observed in vivo, exposure to increasing concentrations of

Mitsui MWCNTs (0.2, 2, and 20 μg/mL) for 48 hours significantly reduced ESR1 mRNA expression (Figure 2-6B) and did not significantly impact GPER1 mRNA expression

(Figure 2-6D). Contrary to the in vivo results, exposure to Mitsui MWCNTs did not affect

ESR2 expression (Figure 2-6C). Next, we sought to determine whether the reduction in

ESR1 mRNA persisted at the protein level by exposing cells to increasing concentrations of Mitsui MWCNTs (0.2, 2, and 20 μg/mL) for 96 hours and measuring

ESR1 protein expression by western blot. Results showed a decreased trend but did not appear to be dose-dependent (Figure C-1).

Mitsui MWCNT-Induced Reduction of ESR1 mRNA is Not a Function of Surface Chemistry

We next sought to determine whether the Mitsui MWCNT-induced repression of

ESR1 mRNA expression in vitro was a response to the CNT surface chemistry. For these experiments, we measured ESR1 mRNA expression in BEAS-2Bs exposed for 48 hours to increasing concentrations of Helix MWCNTs which are non-functionalized, i.e. pristine, like the Mitsui MWCNTs, and to increasing concentrations of surface functionalized (hydroxylated) MWCNTs (oxMWCNTs) and hydroxylated SWCNTs (P3

SWCNT). Expression of ESR1 mRNA was not affected by any dose of the Helix

MWCNTs (Figure 2-7A). This is despite the fact that both types of MWCNTs behaved similarly in the cell culture media as evidenced by similar HDR over time and similar

EPM values between Helix and Mitsui MWCNTs (Figure 2-1). Likewise, neither the oxMWCNTs nor P3 SWCNTs significantly affected ESR1 mRNA expression (Figure 2-

7).

Mitsui MWCNTs Reduce ESR1 mRNA Expression in a TGF-β1-Dependent Manner in vitro

To determine whether TGF-β1 mediated the Mitsui MWCNT-induced reduction in

ESR1 mRNA expression in vitro, we used a selective inhibitor of TGF-β type-I receptor

(TBRI), LY 364947 to block TGF-β1-mediated signaling. First, we performed a titration experiment to find the optimum inhibitor dose. BEAS-2Bs were pre-treated with increasing concentrations of LY 364947 for two hours, then exposed to 5 ng/mL TGF-β1 in the presence or absence of increasing concentrations of LY 364947. Concentrations of LY 364947 ≤ 0.2 μM did not rescue ESR1 mRNA expression while pre-exposure to 2

μM LY 364947 for 2 hours followed by co-exposure to 2 μM LY 364947 and 5 ng/mL

TGF-β1 partially rescued and pre-exposure followed by co-exposure to 20 μM completely rescued ESR1 mRNA expression in the presence of 5 ng/mL TGF-β1

(Figure 2-8A). The chosen dose (20 μM) did not cause any obvious cytotoxic effects.

Next, we determined whether LY 364947 could block Mitsui MWCNT-induced reduction of ESR1 mRNA. Cells were pre-treated with 20 μM LY 364947 for 2 hours, then exposed to 20 μg/mL Mitsui MWCNTs for 48 hours in the presence or absence of

20 μM LY 364947. As before, exposure of cells to Mitsui MWCNTs significantly reduced

ESR1 mRNA expression, but pre-treatment and co-exposure of cells to LY 364947 blocked Mitsui MWCNT-induced repression of ESR1, and exposure to LY 364947 individually did not affect ESR1 mRNA expression (Figure 2-8B).

Mitsui MWCNT-Induced Reduction of ESR1 mRNA Expression is Likely Not a Consequence of ROS Generation

We measured intracellular ROS generation in BEAS-2Bs exposed to the pristine

MWCNTs to determine whether differential ROS generation could explain the dichotomous effects on ESR1 mRNA expression. BEAS-2Bs were exposed to

increasing concentrations of Mitsui and Helix MWCNTs for 6 hours and ROS generation was measured by DCFDA assay. Exposure of cells to 20 μg/mL Mitsui MWCNTs caused a significant 3.147 ± 0.73 (mean ± SEM) fold increase in intracellular ROS levels compared to controls (Figure 2-9A). Exposure to 100 μg/mL Helix MWCNTs caused a comparable 2.73 ± 1.38 (mean ± SEM) fold increase in intracellular ROS levels compared to controls although the means were not significantly different (Figure

2-9B). Hydrogen peroxide was included as a positive control in each experiment to ensure experimental success (data not shown).

Discussion

MWCNT use is rapidly growing which raises concern for potential adverse effects through inhalation exposures, including the development of pulmonary fibrosis. There is extensive evidence in small animal models that MWCNTs cause fibrotic like changes in the lung (229), but limited studies have probed a role for steroid signaling despite epidemiological data indicating sex-specific trends in the incidence, prevalence, and survival rates of IPF (70) and studies in animal models indicating sex-specific differences in pulmonary fibrotic responses (59, 108, 175, 230). As such, the purpose of this work was to begin investigations into interactions between pro-fibrogenic signaling and estrogen receptors in a model of MWCNT-induced fibrosis to highlight a potential role for E2 in the pro-fibrogenic response. Through a combination of in vivo and in vitro exposure studies, we highlight Esr1 and Esr2 mRNA expression as targets of Mitsui

MWCNT-induced signaling in vivo and ESR1 mRNA expression as a target in vitro. We proceed to show that the inhibitory effects are mediated by TGF-β1 and are not likely dependent on CNT surface chemistry or ROS generation. This work is the first to

support estrogen receptors as targets of pro-fibrogenic signaling which likely leads to potential consequences on lung function influenced by E2.

Exposure to MWCNTs through various routes of exposure including aspiration, inhalation, and injection are capable of causing fibrosis like effects in the lungs of exposed mice and rats [Reviewed by Vietti et al. (229)]. In our analysis, we observed hypertrophy and hyperplasia of bronchiolar epithelial cells, airway mucous metaplasia, and increased deposition of collagen in the lungs of mice exposed to Mitsui MWCNTs

(Figure 2-2). It should be noted that based on an analysis of total lung burden of Mitsui

MWCNTs, mice exposed for four days to 10 mg/m3 MWCNTs for 5 hours per day approximates human deposition for a person performing light work for approximately

27-103 months, thus the concentrations used in this study are feasible human occupational exposures (167). The fibrosis observed in this study was minimal, yet expected for these early time-points. Furthermore, chronic effects of MWCNT exposure are increasingly apparent as Mitsui MWCNTs increased thickness of fibrillary collagen in the alveolar septa of mice 336-days post-exposure. This implies that inhaled Mitsui

MWCNTs are retained in the alveolar tissue and promote a progressive and persistent fibrotic response (129).

As TGF-β1 is a well-established driver of extracellular matrix deposition and fibrosis in the lung, particularly in MWCNT-induced fibrogenesis (133, 231), we measured TGF-β1 levels in the BALF of Mitsui MWCNT-exposed mice. A trend of increased TGF-β1 levels in the BALF of mice exposed to 5 mg/m3 aerosolized Mitsui

MWCNTs for 5 hours per day for 4 days was observed, although our results did not quite reach statistical significance (p = 0.10, Figure 2-3). Nonetheless, the 1.4-fold

increase we observed is similar to other studies reporting 1.5- to 4-fold increases in

TGF-β1 levels in the BALF of MWCNT-exposed mice (28, 47, 111, 179, 231, 233). The range of responses is likely due to differences in exposure routes, dosage, and time- points assessed. We also measured expression of connective tissue growth factor

(CTGF) because it is transcriptionally regulated by TGF-β1 (107) and is known to upregulate pathways that contribute to pulmonary fibrosis (106). Surprisingly, we observed reduced CTGF expression in the lungs of mice exposed to Mitsui MWCNTs for 2 and 4 days, with expression returning to control levels thereafter (Figure A-1).

Other groups have shown that CTGF is upregulated in the plasma and BALF during pulmonary fibrosis in humans (1), and others observed temporal heterogeneity of CTGF protein expression in lung tissue of rats exposed to bleomycin where protein expression peaked after 7 days and fell 84% thereafter (246). While we are unsure why CTGF expression is repressed in our model, our results may be a result of the temporally dynamic and compartmentalized expression patterns of CTGF.

There is a growing body of evidence suggesting that carbon-based nanoparticles are capable of interfering with endocrine signaling in reproductive tissues [reviewed by

Iavicoli et al. (79)]. For example, intratracheal administration of carbon black nanoparticles at 0.1 mg/kg body weight for ten times every week to ICR mice caused partial vacuolization of the seminiferous tubules, up to 33% reduction in daily sperm production, and increased testosterone levels (250). The same group showed that similar reproductive effects were seen in male mice exposed to carbon black in utero

(251). Another group showed that repeated intravenous injections of water-soluble

MWCNTs into male mice caused accumulation of MWCNTs in the testes, generated

oxidative stress, and decreased the thickness of the seminiferous epithelium (6). In addition, diabetic male rats exposed to C60 fullerene, another carbon-based nanoparticle, in their water exhibited increased sperm motility, epididymal sperm concentration, and decreased the abnormal sperm rate seen in diabetic rats suggesting a beneficial effect (7). Reproductive effects in females include significantly earlier onset of puberty as assessed as time of vaginal opening in C57BL/6 mice exposed in utero to carbon black nanoparticles (82). While these studies were groundbreaking, they did not investigate the molecular signaling pathways driving the phenotypic responses and none explored effects on sex hormone mediated signaling pathways in non-reproductive tissues. However, these studies strongly support that nanoparticles likely interfere with hormones and hormone signaling in some way.

Few studies have focused on effects of particulate exposure on sex hormone signaling in non-reproductive organs or effects of sex hormones on responses to particulate exposure. This is despite the growing body of evidence suggesting that sex hormones, particularly estrogens, play a role in lung physiology and disease (24, 199,

217). To increase our understanding about baseline expression and localization of estrogen receptors in the mouse lung, we quantified the relative amounts of Esr1, Esr2, and Gper1 mRNA in lung tissue isolated from control mice and performed immunohistochemistry for ESR1 protein. We found that the relative expression of estrogen receptor mRNA was Gper1 > Esr1 > Esr2 (Figure 2-4A). These results contrast with a study by Couse et al. which found that Esr2 was more abundant in the mouse lung than Esr1 (33). The discrepancy could be a result of different detection methods as we used qPCR to quantify mRNA expression while Couse et al. used an

RNase protection assay (33). While RNase protection assays are sensitive and specific, they require the use of probes that are completely homologous to the target sequences.

There exist several Esr1 transcript variants whose coding sequences vary considerably in the 3-prime end (Figure B-1). If the probes used by Couse et al. were specific to this region, then some of the transcript variants of Esr1 would not have been detected. We go on to show that ESR1 protein is expressed in epithelial cells lining the bronchi as well as in the interstitial space as evidenced by positive nuclear staining (Figure 2-5A).

This is inconsistent with a study that reported no expression of ESR1 in mouse fetal lung tissue (25), however, it is well-established that different antibodies to the same antigen exhibit variable binding affinities, which can lead to detection issues between protocols (60). It is also likely that differences in detection are a result of variable expression of ESR1 in fetal and adult lung tissue.

Next, we characterized the effect of Mitsui MWCNT exposure on Esr1, Esr2, and

Gper1 mRNA expression in the mouse lung to infer a role for Mitsui MWCNTs in modulating E2 signaling in the lung. We found that Esr1 and Esr2 mRNA expression was reduced in the lungs of Mitsui MWCNT-exposed mice compared to controls at all time points (Figure 2-4B-C). The expression of Gper1 was less affected by Mitsui

MWCNT exposure (Figure 2-4D). These data are consistent with a study by Brass et al. which found that intratracheal instillation of silica nanoparticles to male and female mice significantly reduced Esr1 and Esr2 but not Gper1 mRNA expression in the lung after 14 days (16). The reduced expression was associated with less hydroxyproline content and more inflammatory cells in the lungs of silica-exposed females than silica-exposed males (16). A recent study suggested that E2 may regulate osteopontin mRNA

expression, a target commonly induced in fibrosis which has been proposed to contribute to sex-based differences in silica-induced lung fibrosis in mice (103). We proceeded to detect ESR1 protein expression on lung tissue sections by immunohistochemistry in control and Mitsui MWCNT-exposed mice to determine whether the repression persisted at the protein level. While we detected ESR1 in both control and MWCNT treated groups we cannot at this point detect a quantifiable difference (Figure 2-5). We are currently working with a certified pathologist (Dr.

Marlene Orandle, CDC NIOSH) to delineate if notable differences in both levels and localization exist.

To begin mechanistic investigations into pathways contributing to the MWCNT- induced repression of estrogen receptor mRNA, we sought to develop an in vitro system that would allow us to more easily probe suspect signaling pathways. We chose BEAS-

2Bs because they are derived from normal, non-diseased lungs, and would represent cells that stained positive for ESR1 protein in the epithelial cells lining the bronchi in vivo

(Figure 2-5). In addition, epithelial cells are the first cells to be affected by inhalation of particulates (238), and it is suspected that pulmonary fibrosis is a result of persistent epithelial cell injury leading to altered epithelial to fibroblast signaling and dysregulated wound repair (183). We first measured the relative baseline expression of the estrogen receptors and found that the pattern in vitro matched the pattern in vivo where GPER1 >

ESR1 > ESR2 mRNA expression (Figure 2-6A). Our results are consistent with a study by Stabile et al. that found higher ESR1 than ESR2 mRNA expression in human lung adenocarcinomas and squamous cell lung tumors although a difference between ESR1 and ESR2 expression was not evident in normal lung cells (211). However, these

results are in contrast to a study by Mollerup et al. that found that ESR2 was more abundant than ESR1 in BEAS-2Bs (134), which is perhaps a result of variable detection methods as Mollerup et al. measured ESR1 and ESR2 mRNA expression by reverse transcription PCR (RT-PCR) followed by gel electrophoresis and did not report the comparable efficiency of the amplification reactions. We carefully assessed the amplification efficiencies of our primer pairs for ESR1, ESR2, and GPER1 to affirm they are similar (100.7%, 102.9%, and 97.2%, respectively, Table 2-1) and are well within the acceptable range as dictated by MIQE guidelines (219).

We then determined whether BEAS-2Bs exhibited a similar response to Mitsui

MWCNT exposure as did the mice. We exposed BEAS-2Bs to increasing concentrations of Mitsui MWCNTs for 48 hours and found that only ESR1 mRNA expression exhibited a dose-dependent reduction in expression (Figure 2-6B-D). The absence of effect on GPER1 mRNA expression is consistent with the in vivo data which also indicated a negligible effect of Mitsui MWCNT exposure on Gper1 mRNA (Figure 2-

4D). The absence of an effect on ESR2 expression is likely due to the high degree of variability among replicates and the inability to detect ESR2 mRNA in multiple samples given its minimal baseline expression level relative to the other estrogen receptors

(Figure 2-6A).

It is well known that subtle differences in nanoparticle dimension and structure can drastically influence their behavior in biological systems and their ability to alter biological responses (62, 119). As such, we questioned whether the responses observed in our study were specific to Mitsui MWCNTs and whether different physiochemical properties could influence the repression of ESR1. To answer this

question, we exposed BEAS-2Bs to a suite of different CNTs that varied in compositional structure (multi-walled vs. single-walled) and/or surface chemistry

(oxidized surface) for 48 hours and measured ESR1 expression. We chose Helix

MWCNTs because they are pristine (non-surface functionalized) like the Mitsui

MWCNTs, oxidized MWCNTs (oxMWCNT), and oxidized SWCNTs (P3 SWCNTs).

Interestingly, none of these CNTs, Helix MWCNTs, oxMWCNTs, nor P3 SWCNTs repressed ESR1 mRNA expression (Figure 2-7) suggesting that surface functionalization (oxidation) nor composition alone influences the response. We were limited to somewhat of a lower exposure dose range for both oxMWCNT and P3

SWCNT (20 μg/mL) as we discovered that the surface hydroxyl groups interfere with reverse transcription during the qPCR pipeline at doses above 2 μg/mL (Humes ST,

Hentschel S, Lavelle CM, Smith LC, Sabo-Attwood T. Biotechniques. In Press.).

We also show, through DLS analysis, that Mitsui MWCNTs and Helix MWCNTs behave similarly in cell culture media with respect to size, forming relatively stable aggregates in the range of 100 nm to 1 μm (Figure 2-1), so the behavior of the

MWCNTs in suspension is likely not a factor. The dimensions of the Helix and Mitsui

MWCNTs are mildly different where Mitsui MWCNTs have a median length of 3.86 μm and a geometric standard deviation of 1.94 and a mean width of 49 ± 13.4 nm (168) while the Helix MWCNTs exhibit lengths between 0.3 and 50 μm and a diameter between 30 and 50 nm (182). Further, the Mitsui MWCNTs have been described as having high crystallinity (143). It is possible that the slight differences in these dimensions, including rigidity and crystallinity, between the CNTs are responsible for the dichotomous effects on ESR1 mRNA expression. In support of this, a study reported

that crystalline silica caused a more robust transcriptional response than amorphous silica despite comparable effects on cell viability (157).

Because Mitsui MWCNTs are known to upregulate levels of TGF-β1 (47, 132), we questioned whether this induction could drive Mitsui MWCNT-induced inhibition of

ESR1 mRNA expression. To answer this question, we exposed BEAS-2Bs to Mitsui

MWCNTs in the presence and absence of a TBRI inhibitor for 48 hours and measured

ESR1 mRNA expression. We found that the TBR1 inhibitor rescued ESR1 expression in the presence of Mitsui MWCNTs suggesting that the inhibitory effects were at least partly mediated through TGF-β1 (Figure 2-8). The direct interactions between TGF-β1 and estrogen receptor signaling have been previously reported by a few studies, however these have all been examined in the context of reproductive cancers. For example, studies in cell culture have shown that TGF-β1 reduces ESR1 mRNA (57,

212) and ESR1 protein expression (158, 212) in breast epithelial cancer cells, and

ESR2 protein expression in prostate cancer cells (117). Only one study has examined the interactions between TGF-β1 and estrogen receptor signaling in the lung where E2 exposure increased TGF-β1 expression in lung fibroblasts isolated from rats exposed to bleomycin (59), a result that would seem to contribute to fibrosis rather than repress related signaling. However, differences in rodent models and cell type specific effects have not been delineated.

To further probe the mechanism of interaction between MWCNT-induced signaling and estrogen receptor expression, we investigated a role for ROS production as CNTs have been shown to produce oxidative species directly and upregulate ROS in biological systems (46, 119, 151), and it has also been reported that CNT-induced ROS

generation can promote pulmonary fibrosis (5). Further, high aspect ratio particulate- derived ROS can activate TGF-β1 by promoting dissociation from latency associated proteins (LAP) (164). As such we sought to determine whether ROS induction could contribute to Mitsui MWCNT-induced repression of ESR1 mRNA expression. Our finding that exposure to both Helix and Mitsui MWCNTs induced comparable intracellular ROS levels at the chosen doses (Figure 2-9), but only Mitsui MWCNTs downregulated ESR1 mRNA expression suggests that the inhibitory effects are likely independent of ROS generation. Further, we have shown in our lab that oxMWCNTs also upregulate intracellular ROS levels (data not shown) without affecting ESR1 mRNA expression. While we have not tested the ability of P3 SWCNTs to up-regulate ROS in our lab, evidence in the literature suggests that carboxylated SWCNTs are capable of producing ROS (176).

The novel observation that Mitsui MWCNT-induced reduction in Esr1 and Esr2 mRNA expression in vivo and ESR1 mRNA expression in vitro begs the question of whether there are potential adverse biological effects associated with this phenomenon.

ERs are known to influence lung development (121, 122, 155), but perhaps more relevant to this study, are the effects of E2 through ESR1 on a variety of respiratory parameters in adult mice, suggesting that E2 signaling may be a critical regulator of breathing and respiratory rhythmogenesis (22). Further, transgenic mice lacking ESR1 exhibited increased airway hyperresponsiveness to inhaled methacholine compared to wildtype mice (22) and another study using transgenic mice lacking ESR2 found that the protective effects of E2 on lung injury after trauma-hemorrhage were mediated by ESR2

(221). However, the consequence of reduced estrogen receptor expression on the

development or progression of fibrosis is less defined. One study found that bleomycin treatment of transgenic male and female mice lacking ESR1 or ESR2 did not cause a difference in fibrotic endpoints compared to wildtype mice (230). However, this study relied on a measurement of lung function that may not have been sensitive enough to discern molecular level changes. Overall, these transgenic studies highlight the adverse effects associated with the absence of estrogen receptors and suggest that the Mitsui

MWCNT-induced repression of estrogen receptor expression may contribute to similar pathology.

In conclusion, we observed a previously unreported interaction between Mitsui

MWCNT-induced signaling and estrogen receptors that was at least partly mediated through induction of the pro-fibrogenic cytokine, TGF-β1. Further, we report that the

MWCNT-induced repression was likely not a function of surface chemistry or intracellular ROS induction, but likely due to differences in rigidity or crystallinity between MWCNTs. Collectively this work suggests that E2-driven processes in the lung may be susceptible to interference via inhalation exposure to Mitsui MWCNTs and perhaps other particulates as a similar repression of ERs was also seen in response to silica. This work lays the foundation for future studies investigating the cellular consequences of impaired estrogen receptor signaling in the lung, and future studies should consider the influence of circulating E2 and its cognate receptors in both males and females.

Table 2-1. Primer details for qPCR in Chapter 2. Efficiency Gene Forward (5'-3') Reverse (5'-3') Protocol (%) Source 95C 3m; 95C 10s, 60C GAPDH GAAGGTGAAGGTCGGAGTC GAAGATGGTGATGGGATTTC 93.9 (11) 30s, x40 95C 3m; 95C 10s, 60C ESR1 CCACCAACCAGTGCACCATT GGTCTTTTCGTATCCCACCTTTC 100.7 (210) 30s, x40 95C 3m; 95C 10s, 60C ESR2 Proprietary Proprietary 102.9 Bio-Rad 30s, x40 95C 3m; 95C 10s, 60C GPER1 GCTCCCTGCAAGCAGTCTTT GAAGGTCTCCCCGAGAAAGC 97.2 (248) 30s, x40 95C 3m; 95C 10s, 58C Gapdh AGGTCATCCCAGAGCTGAACG CACCCTGTTGCTGTAGCCGTAT 91.9 (67) 10s, 72C 30S, x40 95C 3m; 95C 10s, 62C Esr1 GGAAGCTCCTGTTTGCTCCT AACCGACTTGACGTAGCCAG 95.1 (248) 30s, x40 95C 3m; 95C 10s, 60C Esr2 CGCAGACGAAGAGTGCTGT AGCCAAGGGGTACATACTGG 98.7 (248) 30s, x40 95C 3m; 95C 10s, 60C Gper1 CAGTCTTTCCGTCACGCCTA GCTCGTCTTCTGCTCCACAT 91.8 (248) 30s, x40

Figure 2-1. MWCNT characterization in cell culture media. (A-B) Measurements of hydrodynamic radii of A) Helix MWCNTs and B) Mitsui MWCNTs in cell culture media for 24 hours. C) Electrophoretic mobility (EPM) values of Helix MWCNTs and Mitsui MWCNTs in cell culture media. All experiments were performed at 37°C.

Figure 2-2. Histopathological changes in the lungs of mice after 4, 8, or 12 days of exposure to 10 mg/m3 Mitsui MWCNTs for 5 hours per day by whole body inhalation. (A, E, I) Mice exposed to clean air for 12 days had normal tissue architecture as determined by A) H&E, E) Masson’s Trichrome, and I) Alcian- Blue/PAS-stained lung tissue sections. (B, C, D) Mice exposed to MWCNTs for B) 4 days, C) 8 days, and D) 12 days had hypertrophy and hyperplasia of the bronchiolar epithelium (arrows). (F, G, H) Mice exposed to MWCNTs for F) 4 days, G) 8 days, and H) 12 days exhibited subtle increases in fibrosis as indicated by blue staining of collagen with Masson’s Trichrome (arrows). (J, K, L) Mice exposed to MWCNTs for 4 days exhibited minimal mucous metaplasia while mice exposed for K) 8 and L) 12 days exhibited prominent mucous metaplasia, predominantly in the epithelium as indicated by magenta cytoplasmic staining of mucosubstances on slides stained with Alcian- Blue/PAS (arrows). All images were taken at 40X magnification.

Figure 2-3. Mitsui MWCNTs upregulate TGF-β1 levels in bronchoalveolar lavage fluid (BALF). Mice were exposed to clean air or 5 mg/m3 aerosolized MWCNTs for 5 hours per day for 4 days and TGF- β1 levels were determined in BALF by ELISA. Each point represents TGF- β1 level in one mouse. Lines indicate mean ± SEM. Means were not significantly different (p = 0.10) as determined by two-tailed, unpaired t test.

Figure 2-4. Mitsui MWCNTs reduce estrogen receptor expression in mouse lungs. A) Relative baseline expression levels of each estrogen receptor in whole lung tissue of mice exposed to clean air for two days (n = 8) was Gper1 > Esr1 > Esr2. Mice were exposed by whole body inhalation to clean air for 2, 4, 8, or 12 days (n = 8, 8, 6, 8, respectively) or 10 mg/m3 aerosolized Mitsui MWCNTs for 5 hours per day for 2, 4, 8, or 12 days (n = 9, 7, 5, 6, respectively) and expression of B) Esr1, C) Esr2, and D) Gper1 was determined by qPCR. Data are mean ± SEM of mRNA expression relative to day 2 control mice. Asterisks (*) indicate statistically significant (p < 0.05) differences from control at each time-point as determined by two-tailed, unpaired t test.

Figure 2-5. ESR1 protein is expressed in epithelial cells lining the bronchi and in the interstitial space in mouse lung tissue. (A-D) Immunohistochemical staining of ESR1 protein in mice exposed to clean air for A) 12 days or to 10 mg/m3 Mitsui MWCNTs for 5 hours per day for B) 4, C) 8, or D) 12 days by whole body inhalation. Images correspond to histopathological images in Figure 1. All images were taken at 40X magnification.

Figure 2-6. Mitsui MWCNTs reduce expression of ESR1 but not ESR2 or GPER1 mRNA expression in BEAS-2Bs. A) Relative baseline expression levels of each receptor in control cells was GPER1 > ESR1 > ESR2. (B-D) BEAS-2Bs were exposed to increasing concentrations of Mitsui MWCNTs for 48 hours and expression of B) ESR1, C) ESR2, and D) GPER1 was measured by qPCR. Data are mean ± SEM of three independent experiments. Letters indicate statistically significant (p < 0.05) differences as determined by one way ANOVA followed by Newman-Keuls multiple comparison test.

Figure 2-7. Mitsui MWCNT-induced reduction of ESR1 mRNA expression is not dependent on nanotube functionalization. A) BEAS-2Bs were exposed to increasing concentrations of Helix MWCNTs for 48 hours and expression of ESR1 was measured by qPCR. Data are mean ± SEM of three independent experiments (n = 3). Different letters indicate significant differences (p < 0.05) as determined by Newman-Keuls Multiple Comparison Test. (B-C) BEAS-2Bs were exposed to increasing concentrations of B) hydroxylated MWCNTs (oxMWCNT) or C) hydroxylated SWCNTs (P3 SWCNTs) for 48 hours and expression of ESR1 was measured by qPCR. Data are mean ± SEM of one experiment (n = 1). Different letters indicate significant differences (p < 0.05) as determined by Kruskal-Wallis test followed by Dunn’s multiple comparison test.

A 2.0 B

2.0

) )

ESR1 1.5 ESR1 1.5

1.0 1.0 * * *

0.5 0.5

Fold Change to Control ( to Control Change Fold Fold Change Fold to Control (

0.0 0.0 TGF-1 - + + + + + MWCNT - + + - LY 364947 (M) - - 0.02 0.2 2 20 LY 364947 - - + +

Figure 2-8. Mitsui MWCNT-induced reduction of ESR1 mRNA expression is partly dependent on TGF-β1 signaling. A) BEAS-2B cells were exposed to the TGF beta receptor 1 inhibitor (LY 364947) for two hours, then exposed to 5 ng/mL TGF-β1 in the presence or absence of increasing concentrations of LY 364947 for 48 hours, and ESR1 mRNA expression was measured by qPCR. Data are mean ± SEM of one experiment. B) BEAS-2B cells were exposed to 20 μM LY 364947 for two hours, then exposed to 20 μg/mL Mitsui MWCNTs in the presence or absence of 20 μM LY 364947 for 48 hours. ESR1 mRNA expression was measured by qPCR. Data are mean ± SEM of three independent experiments. Asterisks (*) indicate significant differences (p < 0.05) compared to controls as determined by a two-tailed, unpaired t test.

Figure 2-9. Mitsui MWCNT-induced reduction of ESR1 mRNA expression is likely not a consequence of ROS generation. (A-B) BEAS-2Bs were exposed to increasing concentrations of A) Mitsui or B) Helix MWCNTs for 6 hours and intracellular ROS levels were measured by DCFDA assay. Data are mean ± SEM of 3 independent experiments (n = 3) for Mitsui (A) and 2 independent experiments (n = 2) for Helix MWCNTs. Different letters indicate significant differences (p < 0.05) as determined by Kruskal-Wallis test followed by Dunn’s multiple comparison test.

CHAPTER 3 TRANSFORMING GROWTH FACTOR BETA1 TARGETS ESTROGEN RECEPTOR SIGNALING IN BRONCHIAL EPITHELIAL CELLS

Introduction

Epidemiological studies have associated the variable of sex with the prevalence and incidence of idiopathic pulmonary fibrosis (IPF) where males are more negatively impacted and females have better survival rates (63, 70, 149, 171, 173). Based on these observations, several hypotheses have been put forward that suggest a protective role for estrogens and/or an exacerbating role for androgens. In addition, a study by our group found a statistical interaction between sex and severity of IPF based on gene expression in diseased lung tissue for select targets (126).

In attempts to explain sex-based differences, a number of studies have investigated a role for hormones in IPF using small animal models, but results have been mixed. Several reports support the notion that estrogen (17β-Estradiol, E2) is protective and androgens exacerbate fibrotic responses. For example, in a chemically induced model of fibrosis (bleomycin) male mice exhibited an enhanced fibrotic response compared to females (175, 230). Interestingly, a recent study found that it was the presence of the Y chromosome and not necessarily sex itself that predisposed the lung to increased bleomycin induced fibrosis in male and female mice (225).

Conversely, a study in rats indicated that females exhibited increased pulmonary fibrosis in response to bleomycin treatment compared to males (59). It has been noted that variable responses to bleomycin may be due to differential activity of bleomycin hydrolase between males and females (72). A protective role for E2 has also been supported by data produced from ovariectomized mice where a significant increase in

total lung collagen content and airway subepithelial collagen deposition was observed

(108).

Current reports have not probed the molecular mechanisms that may be impacted by hormones in the lung that contribute to the sex-influenced differences in fibrotic disease. Only one study has specifically probed the interactions between pro- fibrogenic signaling mediators such as transforming growth factor beta 1 (TGF-β1) and

E2 in the lung which showed increased TGF-β1 expression in rat fibroblasts exposed to

E2 (59). Regulation of TGF-β1 by E2 has been extensively characterized in other model cell systems and the effects appear to be contextual. For example, E2 inhibited

TGF-β1 signaling in breast cancer cells by reducing the expression of activators of TGF-

β1 (26) and by increasing the proteasomal-dependent degradation of signaling mediators downstream of TGF-β1 including Smad proteins (61, 81). Conversely, E2 increased TGF-β1 secretion in dermal fibroblasts (3), and TGF-β1 levels were reduced in the kidneys of diabetic female mice lacking estrogen receptor alpha (ESR1) compared to wildtype mice suggesting positive regulation of TGF-β1 by E2 (114).

E2 exerts its effects by binding and activating several receptors including the nuclear transcription factors ESR1, estrogen receptor beta (ESR2), and several variants thereof, and the membrane-bound G-protein coupled estrogen receptor 1 (GPER1) which potentiates rapid non-genomic effects (65, 169). Studies aiming to decipher receptor-specific effects on TGF-β1 signaling have been limited to non-lung cells and tissues (i.e. breast). For example, Stope et al. found that ESR1 but not ESR2 inhibited

TGF-β1 activation in gene reporter assays in breast cancer cells (213) while others found that both ESR1 and ESR2 suppressed TGF-β1 signaling by associating with and

acting as a transcriptional corepressor for Smad3 (124, 245). Other studies have shown a role for GPER1 in mediating estrogen-dependent reduction in Smad protein activation in breast cancer cells (99) and TGF-β1-induced extracellular matrix production in human and rat mesangial cells (112).

Given the preponderance of opposing actions of E2 on TGF-β1 in in vitro systems other than the lung and the epidemiological evidence suggesting a male sex- bias in incidence, prevalence, and severity of fibrosis, we hypothesized that E2 would inhibit TGF-β1-induced signaling in lung epithelial cells. We investigated the impact of

E2 on the classic pro-fibrotic and TGF-β1-driven cellular event of epithelial to mesenchymal transition (EMT), a process by which epithelial cells transdifferentiate into a more mesenchymal phenotype (10, 85, 89, 95, 240). Although our studies did not support a role for E2 in directly modulating EMT, we do report the novel observation that

TGF-β1 inhibited ESR1, ESR2, and GPER1 mRNA expression, with the greatest reduction observed for ESR1. We go on to further identify specific targets of E2 signaling likely involved in fibrogenesis that include chromatin remodeling and opposing regulation of processes targeted by TGF-β1 such as extracellular matrix turnover, airway smooth muscle cell contraction, and calcium flux regulation.

Materials and Methods

Chemicals

Recombinant human transforming growth factor beta1 (TGF-β1) was purchased from R&D Systems (>97% purity, Catalog No. 240-B002, Minneapolis, MN) and 17β-

Estradiol (E2) was purchased from Sigma (≥98% purity, Catalog No. E2758, Saint

Louis, MO).

Cell Culture

Human bronchial epithelial cells (BEAS-2Bs, CRL-9609TM) were purchased from

ATCC and cultured per manufacturer’s specifications. STR analysis and mycoplasma contamination testing was not performed. Cells were cultured in bronchial epithelial growth medium (BEGM) consisting of bronchial epithelial basal medium (BEBM, Lonza

CC-3171, Walkersville, MD) and the BEGM SingleQuot Kit Supplements & Growth

Factors (Lonza CC-4175) but exchanging gentamicin for Penicillin-Streptomycin-

Neomycin (PSN) Antibiotic Mixture (Gibco 15640, ThermoFisher Scientific Inc.,

Waltham, MA). Cells were cultured in T75 Corning™ U-Shaped Cell Culture Flasks

(Corning 430641U, Fisher Scientific Co LLC, Pittsburgh, PA) coated with a matrix (4.5 mL per 75 cm2) consisting of 0.01 mg/mL fibronectin (Akron AK8350, Boca Raton, FL),

0.03 mg/mL bovine collagen (Gibco A10644-01), and 0.01 mg/mL BSA (Fisher BP1605) in BEBM. Cells were subcultured up to 11 times before use in exposure studies. All exposures were performed in BEGM without the supplied bovine pituitary extract (BPE) aliquot because its composition is not defined. For the gene expression experiments,

BEAS-2Bs were plated at 40,000 cells/mL on matrix-coated 12-well Nunc™ Cell-Culture

Treated Multidishes (ThermoFisher Scientific Inc.), allowed to adhere overnight, and subsequently exposed for 48 hours to the indicated concentrations of TGF-β1 dissolved in 0.1% BSA (Fisher BP1605) and 4 mM HCl or 10 nM E2 dissolved in DMSO

(Mediatech Inc. MT25950CQC, Manassas, VA). For the protein expression experiments, BEAS-2Bs were plated at 60,000 cells/mL on matrix-coated 6-well

Corning® Costar® cell culture plates (Costar 3516), allowed to adhere overnight, and subsequently exposed for 48 hours to the indicated concentrations of TGF-β1 in BEBM without BPE. All chemical solvent concentrations were maintained below 0.1%.

Total RNA Extraction and Purification

Cell lysates were collected in RNA STAT-60TM Reagent (Tel-Test, Inc. Cs-502,

Friendswood, TX) and RNA was extracted per manufacturer’s specifications. RNA was precipitated overnight at -20°C in 100% molecular biology grade isopropanol (Fisher

BioReagentsTM BP26184) containing 0.067% v/v GlycoBlueTM Coprecipitant (Ambion®

AM9515) and purified by washing 2X with 75% molecular biology grade absolute ethanol (Fisher BioReagentsTM BP28184). RNA pellets were reconstituted in 15 μL

RNAsecure™ (Ambion® AM7010). RNA was quantified using a SynergyTM H1 plate reader (BioTek Instrument, Inc., Winooski, VT) and RNA integrity was analyzed using a

Bioanalyzer 2100 instrument (Agilent Technologies, Santa Clara, CA).

Quantitative Real-Time Polymerase Chain Reaction (qPCR)

A total of 1 μg total RNA was DNase-treated using the PerfeCTa DNase I Kit

(Quanta BioSciences 95150-01k, VWR International LLC, Suwanee, GA) and subsequently reverse transcribed using the qScript™ cDNA Synthesis Kit (Quanta

BioSciences 95047). cDNA was diluted 1:20 in RNase-DNase free water. Each 10 μL qPCR reaction contained 1x SsoAdvanced™ Universal SYBR® Green Supermix (Bio-

Rad 172-5270, Hercules, CA), 850 nM forward and reverse primers, and 3.3 μL of the cDNA dilution. Gene specific primers and cycling parameters are displayed in Table 3-1.

ESR2 primers were purchased from Bio-Rad (Unique Assay ID: qHsaCID0013184).

Each qPCR reaction was followed by melt curve analysis to verify primer specificity. Cq values were determined by regression method using the CFX Manager 2.1 software and quantified using the relative ΔΔCq method (76) or the ratio method (159) when indicated. Target gene expression was normalized to glyceraldehyde 3-phosphate

dehydrogenase (GAPDH) expression. In the case of no amplification, a Cq value of 40 was applied.

Protein Extraction and Purification

After the exposure, media was removed, and the cells were washed 3X with ice cold PBS and subsequently collected in 200 μL RIPA Lysis and Extraction Buffer

(Pierce Biotechnology 89900, Waltham, MA) containing PierceTM Protease Inhibitor

Tablets (Pierce Biotechnology 88265) using a cell scraper. The lysates were passed 5X through a 25-gauge needle and incubated on ice for 30 minutes with intermittent mixing.

Thereafter, the lysates were centrifuged at 14,000 x g for 20 minutes at 4°C, supernatants were removed, and total protein quantified using the Pierce BCA Protein

Assay Kit (Pierce Biotechnology 23225).

Western Blot

For SDS-PAGE, 15 μg of total protein was diluted in NovexTM Tris-Glycine SDS

Sample Buffer (2X) (Invitrogen LC2676) and loaded onto a Novex™ WedgeWell™ 4-

12% Tris-Glycine Mini Gel (Invitrogen XP04120BOX). Electrophoresis was performed for 30 minutes at 225 volts and electrophoresed proteins were subsequently transferred to nitrocellulose membrane (GVS Life Sciences EP4HY00010) by semi-dry transfer method under 15 volts for 30 minutes. Blots were blocked with 5% dehydrated milk dissolved in TBST for 1 hour at room temperature, then incubated with mouse monoclonal antibody specific for estrogen receptor α (anti-ESR1, Santa Cruz

Biotechnology, Inc. SC-514857, Dallas, TX) diluted 1:100 in blocking solution overnight at 4°C. Then, blots were washed 3X in TBST for 10 minutes and incubated with HRP- linked Rabbit anti-Mouse IgG (H+L) Secondary Antibody (Pierce 31450) diluted 1: 4,000 in TBST for 1 hour at room temperature. Blots were washed 3X with TBST for 10

minutes at room temperature, then 1X with TBS at room temperature. Thereafter, blots were incubated with 1:1 solution of Clarity™ Western ECL Blotting Substrates (Bio-Rad

170-5060, Hercules, CA) and imaged using the auto-exposure option on a Bio-Rad

ChemiDocTM MP system (Bio-Rad 17001402). After probing with anti- ESR1, blots were incubated 2X for 10 minutes at room temperature in mild stripping buffer (0.1 M Glycine,

20 mM MgAcetate, 50 mM KCl, pH 2.2), then washed 3X with TBST for 5 minutes at room temperature. To verify antibody stripping, the blot was probed with HRP linked secondary antibody and re-imaged as before. After verification of anti-ESR1 removal, the blot was incubated in blocking solution for 1 hour at room temperature, then incubated with mouse monoclonal antibody specific for β-Actin (anti-β-Actin, Sigma

A5441) diluted 1:5000 in blocking solution overnight at 4°C. The blot was subsequently washed as before, incubated in HRP-linked Rabbit anti-Mouse IgG (H+L) Secondary

Antibody (Pierce 31450) diluted 1:5000 in TBST for 1 hour at room temperature, and imaged as before. Densitometry was performed in ImageJ (191) using the Gel Analysis method outlined in the ImageJ documentation.

RNA-Seq Library Preparation

RNA library construction was performed at the Interdisciplinary Center for

Biotechnology Research (ICBR) Gene Expression Core, University of Florida (UF). RNA concentration was determined on Qubit® 2.0 Fluorometer (ThermoFisher/Invitrogen,

Grand Island, NY), RNA quality was assessed using the Agilent 2100 Bioanalyzer. Total

RNA with RNA integrity number (RIN) ≥ 7 was used for RNA-Seq library construction.

The RINs of all the total RNA was 9.7-10. 2 μl of 1:2000 diluted External RNA Controls

Consortium (ERCC) RNA spike-in (half of amount suggested in the ERCC user guide:

Cat# 4456740) spike to 1 μg of high quality total RNA followed by mRNA isolation using

NEBNext Poly(A) mRNA Magnetic Isolation module (New England Biolabs E7490,

Ipswich, MA) and RNA library construction with NEBNext Ultra RNA Library Prep Kit for

Illumina (New England Biolabs E7530) according to the manufacturer's user guide.

Briefly, 100 ng of total RNA together with 2 μl of 1:2000 diluted ERCC was extracted with 15 μl of NEBNext Magnetic Oligo d(T)25 and fragmented in NEBNext First Strand

Synthesis Buffer by heating at 94 °C for 8 min, followed by first strand cDNA synthesis using reverse transcriptase and random primers. Synthesis of cDNA was performed using the 2nd strand master mix provided in the kit. The resulting double-stranded cDNA was end-repaired, dA-tailing and ligated with NEBNext adaptors. Finally, library was enriched by 13 cycles of amplification, and purified by Meg-Bind RxnPure Plus beads (Omega Biotek M1386, Norcross, GA). Barcoded libraries were sized on the bioanalyzer, quantitated by QUBIT and qPCR (Kapa Biosystems KK4824, Wilmington,

MA). 17 individual libraries were pooled at equal molar concentration of 20 nM.

RNA Sequencing

Barcoded cDNA was sequenced using the 2x100 configuration in 2 lanes of a

HiSeq 3000 instrument (Illumina, San Diego, CA). The yield for the run was in the expected range, the quality was good with Q30% > 96.25%, and the pool was well balanced (in terms of number of reads per samples). ERCC Spike-In Mix (AmbionTM

4456740, ThermoFisher Scientific Inc.) for RNA-Seq projects was used as an internal control by adding 2 μL of 1:2000 diluted spike-in to 100 ng of total RNA input.

Bioinformatics

Short reads were trimmed and filtered to remove low-quality reads, using trimmomatic version 0.36. Quality control was assessed using the FastQC tool, version

0.11.4. Short reads were aligned to the transcriptome (hg38) using STAR version

2.5.2a. Transcript quantification and differential analysis were performed using RSEM version 1.2.31. Differential analysis was performed at the level of coding genes, all transcript, and all splicing isoforms. Coding genes, transcripts, and splicing isoforms were considered statistically significant if FDR-corrected p-value ≤ 5% and fold change

> 1.5 in either direction. Clustering analysis was performed using the ‘gplots’ package in

R version 3.3.2 (2016-10-31). Gene set enrichment analysis was performed using

Pathway Studio® Version 11.4.0.8 operating on the ResNet Mammalian database

(Elsevier). Statistically significant enrichment (p ≤ 0.05) of predefined gene sets was determined by Mann-Whitney U-test.

Statistics

Normality of experimental data was determined by D'Agostino & Pearson omnibus normality test, Shapiro-Wilk normality test, or KS normality test using

GraphPad Prism software (Version 5.01, GraphPad Software, Inc., La Jolla, CA). Data were determined to be normal by passing at least one normality test (p < 0.05). For qPCR and western blot analyses, if data were normal, statistically significant differences

(p < 0.05) in mean fold changes between experimental groups were determined by one- way ANOVA followed by Newman-Keuls multiple comparison test using GraphPad

Prism software.

Results

TGF-β1 Induces Changes Consistent with EMT

To begin our investigations, we first optimized a well-characterized model of

EMT. BEAS-2Bs were exposed to increasing concentrations of TGF-β1 for 48 hours and the mRNA expression of molecular markers for EMT were assayed by qPCR.

Results revealed a dose-dependent response where exposure of cells for 48 hours to

0.1, 1, and 5 ng/mL TGF-β1 caused a 0.867, 0.527 (p < 0.05), and 0.559 (p < 0.05) fold reduction in expression of the epithelial cell type marker, E-cadherin (CDH1), respectively, compared to control cells (Figure 3-1A). Conversely, exposure of cells to

0.1, 1, and 5 ng/mL TGF-β1 caused a 1.38- (p < 0.05), 2.06- (p < 0.05), and 2.21- (p <

0.05) fold increase in expression of the mesenchymal cell type marker, vimentin (VIM), respectively, compared to control cells (Figure 3-1B). Exposure of cells to TGF-β1 did not affect expression of the myofibroblast cell type marker, alpha smooth muscle actin

(ACTA2) at any of the doses tested (Figure 3-1C).

E2 Does Not Significantly Affect TGF-β1-Induced EMT

To probe a role for E2 in modulating TGF-β1-induced EMT, BEAS-2B cells were exposed to 5 ng/mL TGF-β1 in the presence and absence of 10 nM E2 for 48 hours.

Thereafter, expression of the EMT marker genes was assayed by qPCR. As expected, exposure of cells to 5 ng/mL TGF-β1 caused a reduction (0.686-fold) in expression of epithelial cell type marker, CDH1, (Figure 3-2A). However, the addition of 10 nM E2 to the cells did not significantly affect the TGF-β1-induced response, nor affected CDH1 expression individually (Figure 3-2A). Exposure of cells to 5 ng/mL TGF-β1 caused a

3.02-fold increase in expression of the mesenchymal marker, VIM, (Figure 3-2B) and similar to CDH1, co-exposure of cells to 5 ng/mL TGF-β1 in the presence of 10 nM E2 did not result in a statistically significant difference from cells exposed to TGF-β1 individually (Figure 3-2B). As before, there were no statistically significant differences in expression of ACTA2 compared to control cells in any exposure group, however there was a 1.74-fold trend of increased expression in the co-exposure group (Figure 3-2C).

TGF-β1 Reduces Estrogen Receptor mRNA and Protein Expression

To highlight estrogen receptors as targets for pro-fibrogenic signaling mediated by TGF-β1, we characterized the effect of TGF-β1 on estrogen receptor mRNA expression. The baseline expression levels of ESR1, ESR2, and GPER1 mRNA in control cells were expressed as a ratio to ESR1 based on the method described by

Pfaffl et al. (159) to determine relative baseline expression levels. The relative expression of each receptor subtype was GPER1 > ESR1 > ESR2 (Figure 3-3A).

Exposure of BEAS-2Bs to increasing concentrations of TGF-β1 (0.1, 1, and 5 ng/mL) for

48 hours caused a 1.81-, 3.11-, and 2.87-fold significant (p < 0.05) decrease in ESR1 mRNA expression compared to controls (Figure 3-3B). Similar trends were observed for

ESR2 mRNA expression compared to controls (Figure 3-3C), and a 1.44-, 1.72-, and

1.78-fold significant (p < 0.05) decrease in GPER1 mRNA expression compared to controls (Figure 3-3D), was observed for the three doses, respectively.

To determine if TGF-β1 reduced the expression of ESR1 protein levels, we exposed BEAS-2Bs to increasing concentrations of TGF-β1 (0.1, 1, and 5 ng/mL) for 48 hours and detected ESR1 by western blot (Figure 3-4A). Signal intensity was quantified by densitometry using ImageJ (Figure 3-4B). Similar to the mRNA results, exposure of

BEAS-2Bs to 0.1, 1, and 5 ng/mL TGF-β1 caused a 2.09-, 2.77-, and 3.76-fold significant decrease in ESR1 protein levels, respectively (Figure 3-4B).

TGF-β1 and E2 Exhibit Unique Transcriptional Profiles

We performed RNA-Seq analysis to identify transcriptional targets of E2 in lung cells and cellular processes that may be affected by the observed downregulation of

ESR expression by TGF-β1. For this experiment, BEAS-2Bs were exposed to either vehicle control, 5 ng/mL TGF-β1 for 48 hours, 10 nM E2 for 24 hours, or pre-exposed to

5 ng/mL TGF-β1 for 24 hours and subsequently co-exposed to both 5 ng/mL TGF-β1 and 10 nM E2 for 24 hours (Figure 3-5A). Differential expression analysis resulted in

2,182 coding genes with FDR-corrected p-value ≤ 0.05 and Log2(Fold Change) > |0.6| compared to controls in the TGF-β1 group. The expression of 2,119 coding genes was altered in the group co-exposed to TGF-β1 and E2, and 10 in the group exposed to E2.

In sum, 379, 316, and 6 genes were specifically altered in the TGF-β1, TGF-β1 + E2, and E2 groups, respectively, while 1,798 genes were differentially regulated in both the

TGF-β1 and TGF-β1 + E2 groups, and 4 were differentially regulated in all groups

(Figure 3-5B). Many of the genes significantly upregulated by TGF-β1, such as connective tissue growth factor (CTGF), matrix metalloproteinase 2 (MMP2) and VIM, are well known targets of this pathway. Other genes significantly upregulated in all treatment groups included sprouty RTK signaling antagonist 4 (SPRY4) and dual specificity phosphatase 6 (DUSP6), and significantly downregulated genes included potassium voltage-gated channel subfamily Q member 1 (KCNQ1) and RAS protein activator like 1 (RASAL1). Genes that were selectively regulated by E2 only included retinol binding protein 7 [RBP7, Log2(Fold Change) = -1.65] and chloride intracellular channel 3 [CLIC3, Log2(Fold Change) = -0.73] (Table 3-2). A clustering analysis in R of genes differentially regulated in at least one exposure group showed that the expression profiles of the TGF-β1 and TGF-β1 + E2 group were more similar to each other than to the expression profile of E2 (Figure 3-5C).

The expression of select genes relevant to pulmonary fibrosis and this current work was validated by qPCR in an independent experiment (Figure 3-6). As expected, exposure to TGF-β1 caused a significant reduction in ESR1 mRNA expression (Figure

3-6A) and increased the expression of known targets of TGF-β1 such as CTGF (Figure

3-6B), VIM (Figure 3-6C), and MMP2 (Figure 3-6D). The presence of E2 in the co- exposure group did not have a clear effect on the expression of these select genes compared to the TGF-β1 group.

TGF-β1 and E2 Differentially Regulate Gene Sets

We performed a gene set enrichment analysis (GSEA) (214) to identify statistical enrichment in the RNA-Seq data of curated pathways using Pathway Studio® Version

11.4.0.8 (Elsevier). The GSEA resulted in differential enrichment of biological function and disease pathways among the exposure groups. As expected, exposure to TGF-β1 resulted in enrichment of pathways such as extracellular matrix turnover and skin fibrosis (Table 3-3). Exposure to TGF-β1 also resulted in statistical enrichment of pathways including alveolar epithelial cell dysfunction, Ca2+ flux regulation, classical and alternative complement pathways, and neutrophil chemotaxis (Table 3-3). In most cases, similar enrichment and median changes were observed in the co-exposure group (TGF-β1 + E2) and the TGF-β1 only exposure group except for alveolar epithelial cell dysfunction which was not statistically enriched in the co-exposure group (Table 3-

3).

Exposure to E2 also resulted in enrichment of classical and alternative complement pathways, airway smooth muscle cell contraction, Ca2+ flux regulation, and extracellular matrix turnover, however, the median change of the latter two pathways was inverse (downregulated) compared to the median change observed in the TGF-β1 and co-exposure groups (Table 3-3). The extracellular matrix pathway is presented graphically to highlight the inverse regulation of genes in the pathway by

TGF-β1 (Figure 3-7A) and E2 (Figure 3-7B). Exposure to E2 caused specific statistical

enrichment of pathways including histone acetylation and phosphorylation pathways,

NURF in chromatin remodeling, and vasodilation activation (Table 3-3).

Discussion

The premise of this work was motivated by several lines of evidence that suggest hormones may influence gene regulation in the lung which may contribute to sex differences in pulmonary diseases, such as fibrosis. Using a well-established model lung epithelial cell line, we focused on modulation of EMT, a known TGF-β1-driven cellular process important in fibrogenesis. We report that although TGF-β1-driven EMT was not significantly affected by E2, this may be due to the novel observation of TGF-

β1-induced repression of estrogen receptor mRNA expression, most notably ESR1. We extended this observation to identify novel targets of E2 by RNA-Seq that may also be susceptible to TGF-β1-induced repression of estrogen receptor mRNA and protein expression such as chromatin remodeling processes, extracellular matrix turnover, and airway smooth muscle cell contraction.

We first sought to characterize the relative mRNA expression levels of the estrogen receptors in our model cell line. We found that GPER1 was the most abundant followed by ESR1 while ESR2 was least expressed (Figure 3-3A). Our results are similar to a study by Stabile et al. that found higher expression of ESR1 than ESR2 in human lung adenocarcinomas and squamous cell lung tumors although a difference between ESR1 and ESR2 expression was not evident in normal lung cells (211). These results are in contrast to a study by Mollerup et al. that found that ESR2 was more abundantly expressed than ESR1 (134) and another study by Couse et al. that found greater expression of ESR2 in mouse lung (33). The discrepancies may be a result of variable detection methods as we used qPCR to measure the relative expression while

Mollerup et al. and Stabile et al. measured ESR1 and ESR2 mRNA expression by RT-

PCR followed by gel electrophoresis (134, 211), and Couse et al. used an RNase protection assay (33). Further, the study by Mollerup et al. and Stabile et al. did not report the efficiency of the amplification reactions whereas we show that the amplification efficiencies of our primer pairs for ESR1, ESR2, and GPER1 are similar

(100.7%, 102.9%, and 97.2%, respectively, Table 3-1) and are well within the acceptable range as dictated by MIQE guidelines (219).

To begin investigations into a role for E2 in modulating pro-fibrogenic signaling, we focused on EMT as this process is well understood to be an important contributor to fibrogenesis (85, 89, 95, 240). Using TGF-β1 as a trigger, we were able to cause a significant reduction in expression of CDH1 mRNA and a significant increase in expression of VIM mRNA (Figure 3-1B-C) similar to other studies (44, 77, 87). Unlike the study by Doermer et al., we did not measure a significant increase in expression of

ACTA2 mRNA (Figure 3-1C) which may be a result of the duration of exposure as this particular marker tends to be more highly induced at later time-points (5 days) (44) and is indicative of further differentiation of fibroblasts into the contractile myofibroblast (98).

Using this model system, we determined whether E2 affected TGF-β1-induced

EMT. A role for E2 in inhibiting EMT in humans was suggested in a study which found that reduced expression of ESR1 was associated with increased expression of genes involved in EMT in endometrial carcinoma samples (239). Further, EMT is a target for sex hormones in cell types such as breast and prostate cancer cells where E2 signaling maintains an epithelial phenotype and suppresses EMT (41, 66, 117, 249). In our analysis, exposure to E2 did not significantly affect EMT marker gene expression

individually nor did it impact the normal TGF-β1 response (Figure 3-2A-C). Co-exposure of the cells to TGF-β1 and E2 resulted in a trend of increased expression of VIM and

ACTA2 compared to TGF-β1 alone which is consistent with one study that reported E2 promoted reversible EMT-like transition and collective motility in breast cancer cells

(163). The lack of a robust response by E2 in our study may be due to ligand- independent functions of estrogen receptors, particularly ESR1, in EMT as reported by

Cardamone et al. who showed that un-liganded ESR1 constitutively maintained CDH1 expression through direct binding to a half estrogen response element in the CDH1 promoter and that E2 binding facilitated dissociation of co-activator proteins and recruitment of co-repressor proteins (21).

An additional explanation for the lack of effect of E2 on TGF-β1-driven EMT is related to the direct actions of TGF-β1 on estrogen receptors themselves. Interestingly, we found that exposure to increasing concentrations of TGF-β1 caused a dose- dependent and significant reduction in ESR1, ESR2, and GPER1 mRNA expression

(Figure 3-3B-C). We extended this to show that TGF-β1-induced repression of ESR1 persisted at the protein level (Figure 3-4A-B). Other studies have shown that TGF-β1 reduces ESR1 mRNA expression (57, 212) and ESR1 protein expression (158, 212) in breast epithelial cancer cells and ESR2 protein expression in prostate cancer cells

(117), however this is the first study to show that TGF-β1 reduced ESR2 and GPER1 mRNA expression and certainly the first to report any interaction between TGF-β1 and estrogen receptors in lung cells. Future studies should investigate which signaling mediators downstream of TGF-β1, e.g. Smads, CTGF, or SNAI1, among others, are responsible for the observed repression.

Few studies to date have revealed a functional role for E2 in lung cells as measured by genes and pathways modulated downstream. Using a global transcriptome RNA-Seq approach we developed profiles and searched for enriched pathways in BEAS-2Bs exposed to E2 and TGF-β1 individually, and in combination, to both identify points of convergence of E2 and TGF-β1 signaling and to highlight novel

E2 targets that may be susceptible to TGF-β1-induced repression of estrogen receptor expression. Data from the RNA-Seq analysis indicated greater regulation of genes in response to TGF-β1 exposure in comparison to E2 exposure, perhaps consistent with the well-recognized strong pro-fibrotic response associated with TGF-β1. Although some genes were differentially regulated between the TGF-β1 and TGF-β1 + E2 exposure groups, the majority of genes modulated by each were shared suggesting that the presence of E2 had a minimal effect on the TGF-β1-induced transcriptome (Figure

3-5B) potentially due to TGF-β1-induced repression of ESRs. We confirmed the expression of select genes relevant to this work and/or known to be targets of TGF-β1

(Figure 3-6, ESR1, VIM, CTGF, MMP2) which was consistent with our previous results and those reported in the literature. For example, CTGF is a prototypic member of the

CCN protein family and a well-established downstream signaling mediator of TGF-β1 and is involved in EMT (208) and the activation of extracellular matrix production and myofibroblast differentiation (104). MMP2 is another well-established target of TGF-β1 in other cell types including keratinocytes (118), breast epithelial cells (94), and lens epithelial cells (197). MMP2 is a member of a family of zinc-dependent endopeptidases that regulate extracellular matrix organization in organogenesis and wound repair (93,

174) and is upregulated in the lungs of patients with IPF (58) and suspected to promote

EMT and fibrogenesis [reviewed by Craig et al., (38)].

Exposure to E2 did not induce as robust a transcriptional response in BEAS-2Bs compared to TGF-β1. However, we identified statistically significant regulation of 10 genes by E2 that have not been previously reported (Table 3-2). Two genes that were specifically downregulated by E2 included chloride intracellular channel 3 (CLIC3) and retinol binding protein 7 (RBP7) (Table 3-2). CLIC3 promotes migration and invasion of cancer cells by facilitating the functions of MT1-MMP (MMP14) (48, 116). MT1-MMP is the most highly expressed MMP in IPF lungs (58) and may protect against PF by degrading collagen (100) and promoting lung repair (88). Another study indicated that

MT1-MMP promoted pulmonary fibrosis by activating latent TGF-β1 (140). Our results suggest E2 may repress MT1-MMP function by downregulating CLIC3 mRNA expression. RBP7, also known as CRABP4, is a retinol binding protein thought to play an important role in retinol uptake, storage, and metabolism (55). RBP7 has been shown to be upregulated in IPF lung tissue (136) and in wound tissue in the normal chicken chorioallantoic wound model (209) although its role in fibrosis is unclear. RBP7 mRNA is positively regulated by E2 in breast cancer cells (97) and mouse mammary gland (20). The discrepancy in regulation in our study may be a result of variable exposure dynamics as the study by Calvo et al. exposed mice to one dose of E2 and sacrificed the animals three hours later (20) while we exposed cells in vitro for 24 hours.

Nonetheless, E2 appears to regulate RBP7 which may exhibit an unexplored effect on fibrogenic signaling.

Using gene set enrichment analysis (GSEA) we identified curated pathways statistically enhanced in the RNA-Seq data set. This approach is a powerful analysis option because it considers the rank of each gene [based on experimental data values i.e. Log2(Fold Change)] when identifying enrichment which allows the entire dataset to be used without applying a statistical threshold (i.e. p-value) (214). This is important because it considers small changes in a large number of biologically related genes that would be missed if the list of genes had been filtered to only those that are statistically significant from controls (214).

As expected, exposure to TGF-β1 resulted in significant enrichment of the extracellular matrix turnover (Figure 3-7A), alveolar epithelial cell dysfunction, and skin fibrosis pathways (Table 3-3). It is well known that TGF-β1 is involved in organization of the extracellular matrix (14) and neutrophil chemotaxis (154), and one of the prevailing hypotheses in IPF research is that it is a result of dysfunctional behavior of alveolar epithelial cells (52). In this case, skin fibrosis serves as a surrogate for pulmonary fibrosis because the underlying mechanisms are similar and largely regulated by TGF-

β1 (15), and pulmonary fibrosis does not exist as a curated, predefined pathway in

Pathway Studio. In most cases, pathways enriched in the TGF-β1 individual exposure group were also enriched in the TGF-β1 + E2 co-exposure group, and the overall directionality as indicated by the median change was similar which suggests that E2 had a limited effect on TGF-β1 once that pathways were in motion and/or was a result of TGF-β1-induced repression of estrogen receptors thus mirroring the results seen at the gene level (Figure 3-5B-C).

Interestingly, exposure to E2 alone resulted in specific enrichment of multiple pathways involved in epigenetic regulation of chromatin structure and organization including histone acetylation, histone phosphorylation, and NURF in chromatin remodeling (Table 3-3). Indirect epigenetic regulation by ESR1 and ESR2 is well known as they are nuclear receptors which upon activation recruit various co-regulatory proteins with chromatin modifying and remodeling capabilities to gain access to promoters and influence gene transcription (42). However, there is less known about a role for estrogen in transcriptionally regulating the expression of genes involved in chromatin remodeling, particularly in the lung. This is important because evidence for the importance of epigenetics and chromatin organization in lung disease is growing, particularly in the context of pulmonary fibrosis (34-36, 148, 170, 184, 247). For example, histone deacetylases are involved in activation of lung fibroblasts to myofibroblasts (235) and accumulation of extracellular matrix components and EMT in the diabetic kidney (147). Notably, exposure to TGF-β1 individually or in the presence of

E2 did not result in enrichment of chromatin remodeling gene sets. This suggests that the absence of enrichment in the co-exposure group, despite the presence of E2, was likely a result of TGF-β1-induced repression of estrogen receptor expression and not through direct regulation of genes by TGF-β1. Collectively, these data imply that E2 may influence pro-fibrogenic signaling through regulation of epigenetic modifications and chromatin remodeling and that TGF-β1-induced repression of estrogen receptor expression impairs E2-induced effects.

Similar to TGF-β1, exposure to E2 resulted in statistical enrichment of the extracellular matrix turnover, airway smooth muscle cell contraction, and calcium flux

regulation pathways (Table 3-3). There is evidence linking E2 to airway smooth muscle cell contraction and airway responsiveness in asthma (22, 123) and these effects seem to be driven by E2-induced regulation of calcium signaling (43, 222). Future studies should investigate the contribution of non-genomic signaling by E2 in modulating calcium flux regulation and airway smooth muscle cell contraction because it is known that changes in intracellular calcium levels are often mediated by membrane-bound receptors including GPER1 (137, 177).

E2 is also known to influence the extracellular matrix in the uterus and vaginal tissues (37), in osteoblasts (101), and in the skin (206). Interestingly, the overall directionality of the pathway as indicated by the median change, was opposite

(negative) compared to the directionality of the pathway in the TGF-β1 and TGF-β1 +

E2 groups (positive, Table 3-3). This is consistent with a study that found that E2 inhibited TGF-β1-induced extracellular matrix production in human and rat mesangial cells through GPER1 activation (112). The difference in directionality is most apparent in the graphical representation of the pathway which highlights upregulation of genes in the pathway by TGF-β1 but downregulation of the same genes by E2 (Figure 3-7). Of note is the repression of MMP14 and MMP2 as another study showed that E2 decreased MMP2, MMP13, and MMP14 mediated tissue matrix destruction (162).

These results are consistent with the significant reduction of CLIC3 mRNA expression by E2 (Table 3-2) which is known to regulate MMP14 (48, 116). Future studies should delineate the precise role of each ESR in regulating extracellular matrix turnover.

In conclusion, we were not able to decipher an effect of E2 directly on TGF-β1- induced EMT in our study but we do report the novel observation that TGF-β1 inhibited

ESR1, ESR2, and GPER1 mRNA expression and ESR1 protein expression in BEAS-

2Bs. We also report that E2 specifically downregulated the expression of select genes such as CLIC3 and RBP7 which are associated with pulmonary fibrosis. Most strikingly, we highlight cellular pathways involved in chromatin remodeling as novel and specific targets of E2 in lung cells and opposing actions of TGF-β1 and E2 signaling on extracellular matrix turnover, airway smooth muscle cell contraction, and calcium flux regulation. Although these data do not explicitly indicate a protective role for E2 in pulmonary fibrosis, these results suggest that E2 likely influences pro-fibrogenic signaling independently and through modulation of pathways that are targets of TGF-β1 and highlights potential roles for E2 in the lung that may contribute to sex-specific differences in IPF.

Table 3-1. Primer details for qPCR in Chapter 3. Efficiency Gene Forward (5'-3') Reverse (5'-3') Protocol (%) Source GAPDH GAAGGTGAAGGTCGGAGTC GAAGATGGTGATGGGATTTC 95C 3m; 95C 10s, 60C 30s, x40 93.9 (11) ESR1 CCACCAACCAGTGCACCATT GGTCTTTTCGTATCCCACCTTTC 95C 3m; 95C 10s, 60C 30s, x40 100.7 (210) ESR2 Proprietary Proprietary 95C 3m; 95C 10s, 60C 30s, x40 102.9 Bio-Rad GPER GCTCCCTGCAAGCAGTCTTT GAAGGTCTCCCCGAGAAAGC 95C 3m; 95C 10s, 60C 30s, x40 97.2 (248) CDH1 GAAAGCGGCTGATACTGACC CTCAGACTAGCAGCTTCGGA 95C 3m; 95C 10s, 58C 30s, x40 104.9 (248) ACTA2 CATCATGCGTCTGGATCTGG GGACAATCTCACGCTCAGCA 95C 3m; 95C 10s, 60C 30s, x40 94.8 (218) MMP2 TGTGTTCTTTGCAGGGAATG TCCAGAATTTGTCTCCAGCA 95C 3m; 95C 10s, 58C 30s, x40 93.6 (224) CTGF AATGCTGCGAGGAGTGGGT CGGCTCTAATCATAGTTGGGTCT 95C 3m; 95C 10s, 60C 30s, x40 96.4 (144) VIM GCGTGAAATGGAAGAGAACT GGTATCAACCAGAGGGAGTG 95C 3m; 95C 10s, 56C 10s, 72C 30S, x40 104.0 (77) MUC15 CCATCGGCGACTTTATGACG TCTTCACTTTCTGGCATGGCT 95C 3m; 95C 10s, 60C 30s, x40 92.1 (248)

Table 3-2. Genes differentially regulated by E2. ENSEMBL gene ID Gene symbol Gene name Log2(fold change) P-value ENSG00000258588 TRIM6-TRIM34 Tripartite motif-containing 6 and tripartite motif-containing 34 5.36 1.33E-06 ENSG00000256966 RP11-613M10.8 AL513165.2 3.61 2.98E-02 ENSG00000274944 RP5-864K19.6 AL139260.3 2.93 1.34E-14 ENSG00000255439 RP11-196G11.1 AC135050.2 1.43 9.12E-03 ENSG00000187678 SPRY4 Sprouty RTK signaling antagonist 4 0.76 3.20E-05 ENSG00000139318 DUSP6 Dual specificity phosphatase 6 0.75 2.25E-08 ENSG00000169583 CLIC3 Chloride intracellular channel 3 -0.73 1.40E-06 ENSG00000053918 KCNQ1 Potassium voltage-gated channel subfamily Q member 1 -0.86 4.87E-03 ENSG00000111344 RASAL1 RAS protein activator like 1 -1.34 1.79E-03 ENSG00000162444 RBP7 Retinol binding protein 7 -1.65 3.08E-02

Table 3-3. Gene set enrichment analysis of genes identified by RNA-Seq.

TGF-β1 TGF-β1 + E2 E2 Name Median Median Median Pathway type P-value P-value P-value change Change change Airway Smooth Muscle Cell Contraction Diseases 1.14 4.72E-04 1.10 4.34E-03 -1.01 3.96E-02 Alveolar Epithelial Cell Dysfunction Diseases 1.01 4.98E-02 - - - - Ca2+ Flux Regulation Biological Function 1.02 2.12E-07 1.06 1.40E-05 -1.00 5.48E-09 Complement Alternative Pathway Biological Function -1.63 3.17E-04 -1.49 2.64E-05 -1.11 1.36E-03 Complement Classical Pathway Biological Function -1.34 2.27E-04 -1.49 3.09E-05 -1.15 4.87E-04 Extracellular Matrix Turnover Biological Function 1.55 2.26E-08 1.53 3.98E-07 -1.04 6.86E-04 Histone Acetylation Biological Function - - - - 1.04 3.68E-02 Histone Phosphorylation Biological Function - - - - 1.05 2.99E-05 Neutrophil Chemotaxis Biological Function 1.09 1.27E-03 1.07 7.25E-03 - - NURF in Chromatin Remodeling Biological Function - - - - 1.06 2.48E-02 Skin Fibrosis Diseases 1.07 2.15E-02 1.06 1.64E-02 - - Vasodilation Activation Biological Function - - - - -1.00 3.34E-02 (-) indicates gene set not significantly enriched in treatment group.

Figure 3-1. TGF-β1 induces gene expression changes consistent with EMT in BEAS- 2B cells. (A-C) BEAS-2Bs were exposed to increasing concentrations of TGF- β1 for 48 hours and mRNA expression of A) CDH1, B) VIM, and C) ACTA2 was measured by qPCR. Data are mean ± SEM of three independent experiments. Letters indicate statistically significant (p < 0.05) differences as determined by one-way ANOVA and Newman-Keuls multiple comparison Test.

Figure 3-2. E2 does not affect TGF-β1 induced EMT. (A-C) BEAS-2Bs were exposed to 5 ng/mL TGF-β1 in the presence or absence of 10 nM E2 for 48 hours and mRNA expression of A) CDH1, B) VIM, and C) ACTA2 was measured by qPCR. Data are mean ± SEM of four independent experiments. Letters indicate statistically significant (p < 0.05) differences as determined by one- way ANOVA and Newman-Keuls multiple comparison test.

Figure 3-3. TGF-β1 downregulates ESR1, ESR2, and GPER1 mRNA expression in BEAS-2B cells. A) Relative expression of estrogen receptors in control cells was GPER1 > ESR1 > ESR2. (B-D) BEAS-2Bs were exposed to increasing concentrations of TGF-β1 for 48 hours and mRNA expression of B) ESR1 (n = 3), C) ESR2 (n = 2), and D) GPER1 (n = 3) was measured by qPCR. Data are mean ± SEM and letters indicate statistically significant (p < 0.05) differences as determined by one-way ANOVA and Newman-Keuls multiple comparison test.

Figure 3-4. TGF-β1 downregulates ESR1 protein expression. A) BEAS-2Bs were exposed to increasing concentrations of TGF-β1 for 48 hours and ESR1 protein expression was measured by western blot. B) Densitometric analysis was performed using ImageJ. ESR1 expression was normalized to Beta-actin (ACTB). Data are mean ± SEM normalized arbitrary density units of duplicate measurements per blot of three independent experiments (n = 3). Letters indicate statistically significant (p < 0.05) differences as determined by one- way ANOVA and Newman-Keuls multiple comparison test.

Figure 3-5. TGF-β1 and E2 exhibit distinct transcriptional profiles. (A-C) BEAS-2Bs were exposed to 5 ng/mL TGF-β1 and 10 nM E2 individually and in combination. A) Exposure design schematic. B) Venn diagram highlighting distribution of differentially expressed genes [Log2(Fold Change) ≥ |0.6| and FDR-corrected p-value < 0.05] among treatment groups. C) Heat map showing the clustering and relative expression levels [Log2(Fold Change) compared to controls] of genes that were differentially expressed in at least one treatment group. Red coloring indicates upregulation compared to controls and green coloring indicates downregulation compared to controls, (T, TGF-β1; T+E, TGF-β1 + E2; E, E2).

Figure 3-6. Orthogonal validation of RNA-Seq data. Expression of A) ESR1, B) CTGF, C) VIM, and D) MMP2 was verified in an identical and independent experiment by qPCR. Bars represent expression [Log2(Fold Change)] of each gene in the RNA-Seq analysis, and black dots represent expression [Log2(Fold Change)] in each sample unit (n = 6) in the orthogonal experiment as determined by qPCR. Asterisks (*) indicate differential expression [Log2(Fold Change) ≥ |0.6| and FDR-corrected p-value < 0.05] in the RNA- Seq analysis.

Figure 3-7. TGF-β1 and E2 cause differential regulation of genes involved in extracellular matrix turnover. A GSEA using Pathway Studio of genes identified by RNA-Seq revealed that exposure to A) 5 ng/mL TGF-β1 and B) 10 nM E2 caused statistically significant (p < 0.05) enrichment of the extracellular matrix turnover pathway. Gray boxes denote cellular processes involved in the extracellular matrix turnover pathway. Red proteins indicate upregulation and blue proteins indicate downregulation as determined by RNA-Seq.

CHAPTER 4 GENE EXPRESSION ANALYSIS IN THE LUNGS OF PATIENTS WITH IDIOPATHIC PULMONARY FIBROSIS

Introduction

Idiopathic pulmonary fibrosis (IPF) is a disease characterized by progressive and irreversible scar tissue formation in the lung resulting in increased morbidity and mortality (243). The median age at diagnosis is 66 years and the median survival is only

2.5-3.5 years (96). It is suspected that 40,000 people die of IPF every year (13). While the cause of IPF is unknown, the current consensus is that it is a result of persistent epithelial injury and dysregulated wound repair (34) resulting in excess accumulation of extracellular matrix (ECM) molecules and the development of permanent scarring (243).

Scar tissue accumulation leads to organ malfunction due to reduced compliance of the lung and disruption of gas exchange, ultimately causing death from respiratory failure

(243).

Data from epidemiological studies indicate sex-specific trends exist for both the incidence and prevalence of IPF where prevalence is higher in males and females have better survival rates (70). This is in contrast to other lung diseases such as chronic obstructive pulmonary disease (COPD) where women show greater incidence and severity (71). To explain these sex biases, sex hormones have been postulated to play a role in lung biology and are increasingly contemplated in studies investigating COPD, asthma, and fibrosis (188). Sex hormones signal through sex steroid receptors which are members of the nuclear hormone receptor superfamily of ligand activated transcription factors (65). Sex hormones can also signal through membrane-bound receptors to initiate rapid, non-genomic signaling mechanisms that can also influence gene transcription (113). It is possible that cellular and molecular mechanisms

contributing to IPF progression may vary between sexes due to differential actions of sex steroid receptors on transcriptional targets in the lung.

We have identified estrogen receptor alpha (ESR1), estrogen receptor beta

(ESR2), and G-protein coupled estrogen receptor (GPER1) as transcriptional targets of pro-fibrogenic signaling in the lung in vivo and in vitro (Chapter 2 and Chapter 3).

However, there has been minimal effort to characterize the expression of sex steroid receptors or membrane-bound receptors in the lungs of patients with IPF to infer a potential role in disease pathogenesis. Our group previously analyzed mRNA expression of sex steroid receptors including androgen receptor (AR), ESR1, ESR2, progesterone receptor (PR), and GPER1 in whole lung tissue from normal and IPF patients, but did not detect significant differences in expression (126). Additionally, another study found positive expression of PR but no expression of AR or ESR1 protein by immunohistochemistry (IHC) in normal and usual interstitial pneumonia (UIP) lungs

(128). However, several limitations of the aforementioned studies, including antibody specificity and the lack of analysis of cell type-specific mRNA expression, may mask true differences. Expression of sex steroid receptors may vary in different cell types rendering gene expression analyses in whole tissue inadequate for detecting changes in specific cell types that are diluted by expression levels in neighboring cells (54, 92).

In this study, we first followed up our previous investigations by measuring ESR1,

ESR2, and GPER1 mRNA expression in whole lung tissue from control and IPF lungs, in addition to other genes suspected to be involved in IPF, such as connective tissue growth factor (CTGF), bone morphogenetic protein 4 (BMP4), matrix metalloproteinase

1 (MMP1), matrix metalloproteinase 2 (MMP2), and matrix metalloproteinase 7 (MMP7).

As expected, we observed increased expression of MMP1 and MMP7 in IPF lungs, but report reduced expression of CTGF, BMP4, and MMP2. Most strikingly, we observed reduced ESR1, ESR2, and GPER1 mRNA expression in IPF lungs indicating that estrogen receptor expression is associated with disease and suggesting that estrogen receptors may play a functional role in normal lung physiology or IPF pathogenesis. To probe deeper into a role for estrogen receptors in IPF, we describe the development of a method to quantify gene expression changes in specific cell types in the lung using laser capture microdissection (LCM). We expect that identification of the specific cell types in which estrogen receptors are differentially expressed will more precisely inform potential transcriptional and cellular targets of estrogen receptors allowing us to better infer a role for estrogens in IPF.

Materials and Methods

Patient Samples

The human patient samples in this study were a kind gift from Dr. Andrew Bryant.

The explanted lung tissue was obtained from subjects undergoing lung transplant for

IPF and from lungs rejected for transplant from normal controls per the National

Institutes of Health Lung Tissue Research Consortium (protocol no. 14-99-0011). This study consisted of fifteen patients, eight with IPF (n = 8) and seven controls (n= 7). A pattern consistent with UIP and exclusion of fibrotic nonspecific UIP and other forms of idiopathic interstitial pneumonias and interstitial lung disease associated with occupational or environmental exposure, systemic disease, or drugs was used for diagnosis of IPF (204, 223). The protocol for collection of lung tissue samples, and subsequent studies, were approved by the institutional review board at Vanderbilt

University and the University of Florida (19).

100

Handling of Lung Tissue

Immediately after lung biopsy or resection, a portion of the lung was removed and flash frozen in liquid nitrogen or dry ice and stored at -80°C. For LCM, a portion of the frozen lung tissue was chiseled off and embedded in a mold containing Tissue-Tek®

O.C.T. Compound (Sakura® Finetek 4583, VWR International LLC, Suwanee, GA), frozen slowly in liquid nitrogen-cooled isopentane (Electron Microscopy Sciences

18550, ThermoFisher Scientific, Waltham, MA), and stored at -80°C. The lung tissue remained frozen throughout the embedding process to maintain RNA integrity.

Total RNA Extraction and Purification

Frozen whole lung samples were pulverized over liquid nitrogen using a mortar and pestle then mechanically disrupted in RNA STAT-60TM Reagent (Tel-Test, Inc. Cs-

502, Friendswood, TX) using a handheld homogenizer. RNA was extracted per the manufacturer’s specifications followed by overnight precipitation at -20°C in 100% molecular biology grade isopropanol (Fisher BioReagentsTM BP26184) containing

0.067% v/v GlycoBlueTM Coprecipitant (Ambion® AM9515). Precipitates were collected by centrifugation at 14,000 RPM at 4°C for 45 minutes and purified by washing 2X with

75% molecular biology grade absolute ethanol (Fisher BioReagentsTM BP28184). RNA pellets were reconstituted in 15 μL RNAsecure™ (Ambion® AM7010). RNA was quantified using a SynergyTM H1 plate reader (BioTek Instrument, Inc., Winooski, VT) and RNA integrity was analyzed using a Bioanalyzer 2100 instrument (Agilent

Technologies, Santa Clara, CA).

Gene Expression Analysis in Whole Lung Tissue

A total of 2 μg total RNA was DNase-treated using the PerfeCTa DNase I Kit

(Quanta BioSciences 95150-01k, VWR International LLC) and 1 μg was subsequently

101

reverse transcribed using the qScript™ cDNA Synthesis Kit (Quanta BioSciences

95047). cDNA was diluted 1:20 in RNase-DNase free water. Each 10 μL qPCR reaction contained 1x SsoAdvanced™ Universal SYBR® Green Supermix (Bio-Rad 172-5270,

Hercules, CA), 850 nM forward and reverse primers, and 3.3 μL of the cDNA dilution.

Gene specific primers and cycling parameters are displayed in Table 4-4. ESR2 primers were purchased from Bio-Rad (Unique Assay ID: qHsaCID0013184). Each qPCR reaction was followed by melt curve analysis to verify primer specificity. Cq values were determined by regression method using the CFX Manager 2.1 software and quantified using the relative ΔΔCq method (76). Target gene expression was normalized to ribosomal protein S13 (RPS13) expression. In the case of no amplification, a Cq value of 40 was applied.

Laser Capture Microdissection

An IPF lung from a different consortium was analyzed by LCM to optimize the protocol. The frozen lung tissue was prepared as above and processed for LCM. The frozen tissue block was sectioned at 8 μm under RNase free conditions and four serial sections were mounted onto two PEN (polyethylene naphthalate) Membrane Frame

Slides (ThermoFisher Scientific LCM0521), two sections per slide. The PEN slides containing mounted tissue were stained with H&E to localize structures. LCM was performed using a Leica 7000 Laser Microdissection Microscope. Alveoli were captured from the section and collected in 30 μL ArcturusTM PicoPureTM RNA Extraction Buffer in

ExtracSure HS LCM caps (Arcturus). In total, 211 punches were captured with a total surface area of 975,313 μm2 from four sections (two sections on two PEN slides). The time interval from sectioning the block to collecting the captured cells was less than two hours.

102

Gene Expression Analysis in Microdissected Tissue

Total RNA was extracted and purified using the ArcturusTM PicoPureTM RNA

Isolation kit (ThermoFisher Scientific KIT0204, Waltham, MA) per the manufacturer’s protocol with on-column DNase treatment using the RNase-Free DNase Set (Qiagen

79254, Hilden Germany). Purified RNA was eluted in 11 μL ARCTURUS® PicoPureTM

RNA Elution Buffer and stored at -80°C. The entire volume of purified RNA was amplified through two rounds of linear amplification using the ArcturusTM RiboAmpTM HS

PLUS kit (ThermoFisher Scientific KIT0525) per the manufacturer’s protocol and maximizing the in vitro transcription incubation steps to maximize yield. 10 μL of control

RNA was amplified in parallel to ensure protocol performance. A total of 250 ng of amplified mRNA was reverse transcribed using the qScript™ cDNA Synthesis Kit

(Quanta BioSciences 95047) and diluted 1:20 in RNase-DNase free water. qPCR was performed as previously described.

Statistics

Statistically significant differences (p < 0.05) in mean fold changes between control and IPF lungs were determined by Mann Whitney U test using GraphPad Prism software (Version 5.01, GraphPad Software, Inc., La Jolla, CA).

Results

Baseline Characteristics of Patient Samples

The IPF cohort consisted of eight male patients ranging in age from 47 to 69 years (61.5 ± 7.11). Some IPF patients also exhibited pulmonary hypertension, respiratory bronchiolitis, cryptogenic organizing pneumonia, desquamative interstitial pneumonia, or emphysema. The percent predicted forced vital capacity (FVC%) ranged

103

from 24 to 70 (42.9 ± 14.39) indicating severe disease. Six IPF patients never smoked cigarettes while two had smoked cigarettes at one point in their lives (Table 4-1).

The control group was comprised of three females ranging in age from 45 to 69 years (56 ± 12.12) and two males ranging in age from 72 to 82 years (77 ± 7.07). Some patients exhibited emphysema from COPD, granulomatous inflammation, or non-small cell carcinoma. The FVC% of the female controls ranged from 100 to 115 (107.67 ±

7.51) and 95 to 108 (101.5 ± 9.19) in the male controls. Four control patients never smoked while one had smoked at one point in their life (Table 4-1).

Histopathology in IPF Lungs

The IPF lungs exhibited characteristic honeycombing, scar tissue accumulation, and traction bronchiectasis on a high resolution computerized tomography (HRCT) scan

(Figure 4-1A), and alternating areas of fibrosis and less affected or normal parenchyma highlighting the heterogeneous disposition of the disease (Figure 4-1B). Excessive collagen deposition evidenced by increased red staining was apparent on an IPF lung section stained with Sirius Red compared to a normal lung (Figure 4-2).

Gene Expression Analysis in Whole IPF Lung Tissue

The mRNA expression of the estrogen receptors and select genes involved in fibrosis was measured in control and IPF whole lung tissue by qPCR. The RNA from the human samples was generally of high quality as evidenced by high RNA integrity numbers (RINs) (80). The RINs of the samples ranged from 2.3 to 9.3 (7.62 ± 1.92,

Figure D-1). The two samples with RINs < 6.7 were excluded from the analysis. Two representative electropherograms of RNA from a control and IPF lung are presented in

Figure 4-3.

104

The mRNA expression of ESR1 and GPER1 was significantly reduced in IPF lungs compared to control lungs (Table 4-2, Figure 4-4A and 4-4C) while there was a trend of reduced ESR2 mRNA expression in IPF lungs (p = 0.09, Table 4-2, Figure 4-

4B). The mRNA expression of BMP4 and MMP2 was also significantly reduced in IPF lungs (Table 4-2, Figure 4-4E and 4-4G). The mRNA expression of MMP1 was significantly increased in IPF lungs and exhibited the largest increase in expression

(Table 4-2, Figure 4-4F), and there was a trend of increased expression of MMP7 although the difference did not quite reach statistical significance (p = 0.13, Table 4-2,

Figure 4-4H).

LCM Method Development

We optimized a LCM-based method to isolate individual cell types for gene expression analysis by qPCR with the goal of comparing cell type-specific gene expression patterns in epithelial cells immediately adjacent to fibrotic lesions compared to distant epithelial cells in IPF lungs and epithelial cells in control lungs.

First, we captured cells from an IPF lung from a different patient cohort as proof of concept. Cells were successfully captured as evidenced by their presence in the lung section prior to capture (Figure 4-5A), absence in the lung section after capture (Figure

4-5B), and presence in the collecting tube after capture (Figure 4-5C). In total, we collected 211 punches with a total surface area of 975,313 μm2 from four serial sections. RNA was not detected after purification from captured cells, however, we obtained 100.43 ng of mRNA after one round of linear amplification, and 591.89 ng after two rounds of linear amplification.

Next, we quantified the mRNA expression of two housekeeping genes, RPS13 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in the microdissected IPF

105

lung tissue to determine the quality of the captured mRNA. The average Cq values of two technical replicate qPCR amplifications for GAPDH and RPS13 as measured in the microdissected IPF lung sample were 26.28 and 18.4, respectively (Table 4-3). The amplification curves and melt peaks for RPS13 and GAPDH generated from the microdissected IPF lung sample are presented in Figure 4-6. Importantly, there is only one melt peak per gene (Figure 4-6B). We also measured the expression of ESR1,

ESR2, GPER1, and CTGF in microdissected lung tissue and all showed minimal expression (Table 4-3).

Discussion

Few studies have probed a functional role for sex steroid receptors in IPF despite the existence of sex-specific trends in the incidence, prevalence, and survival rates where IPF is more common in males and females have better survival rates (70). The epidemiological data suggest that either androgens promote fibrosis or estrogens are protective. To begin investigations into a role for sex hormones and sex steroid receptors in IPF, we sought to characterize mRNA expression of ESR1, ESR2, and

GPER1 in IPF lungs and report the novel observation that ESR1, ESR2, and GPER1 mRNA expression is reduced in IPF lungs. We also measured the mRNA expression of genes suspected to be involved in fibrosis including CTGF, BMP4, MMP1, MMP2, and

MMP7. Lastly, we optimized a method to quantify cell type-specific gene expression by

LCM with the goal of comparing gene expression patterns in epithelial cells immediately adjacent to fibrotic lesions compared to distant epithelial cells in IPF lungs and epithelial cells in control lungs. Using this method, we will precisely determine in which cell types the observed changes in ESR1, ESR2, and GPER1 mRNA expression occur which will better inform a role of estrogen receptors in the fibrotic lung.

106

The IPF lung tissue used in this the study was all derived from male donors. This is a common limitation in IPF research given the preponderance of disease in males compared to females (70). The IPF patients exhibited severe disease as determined by their FVC% which is a reliable, valid, and responsive measure of clinical status in patients with IPF (49). Both IPF and control groups contained patients with a history of smoking and several comorbidities. These are common covariates in human research that if carefully considered and controlled for do not confound results or conclusions

(18). The control group contained three female patients whose average age was 56 years. Two women are above the age that is typically associated with hormone replacement therapy (HRT) while one is within the age. This could be a confounding variable for our study but unfortunately, we do not have information regarding HRT status of any of the female patients in this study.

The IPF patients exhibited characteristic scar tissue accumulation, honeycombing, and traction bronchiectasis on HRCT scans (Figure 4-1). Scar tissue accumulation is observed as opacity on HRCT scans (84) and honeycombing is the hallmark of UIP seen in IPF (146) and was first described in 1949 (150). On HRCT scans, honeycombing appears as clustered cystic airspaces that are typically the same diameter, usually 3-10 mm, with well-defined walls (84). The presence of honeycombing indicates advanced ‘end-stage’ disease (236). Traction bronchiectasis is the irreversible enlargement of bronchi and bronchioles within areas of pulmonary fibrosis (237), and was recently found to be highly associated with IPF (odds ratio, 3.64) (215). The IPF lungs also exhibited increased Sirius Red staining compared to the control lungs (Figure

107

4-2) indicating enhanced collagen deposition which is a major contributor to IPF pathogenesis (52).

To begin investigations into a role for E2 signaling in the fibrotic lung, we characterized the expression of ESR1, ESR2, and GPER1 in whole IPF lung tissue. We hypothesized that if estrogen receptor mRNA was differentially expressed in diseased individuals, then it would suggest, by extension, that E2 signaling has a functional role in normal lung physiology and/or IPF pathogenesis. We found that mRNA expression of each receptor was reduced in IPF lungs compared to controls (Table 4-2, Figure 4-4A-

C). A previous study by our group reported a slight reduction in ESR1 mRNA in male

IPF lungs compared to male controls, however the difference was not significant (126).

The discrepancy could be a factor of disease progression as the patients in the previous study exhibited mild or medium severity IPF based on FVC% whereas the patients analyzed in the present study had severe IPF. The same study also reported a reduction in ESR2 mRNA expression in lungs from male patients with mild IPF compared to male controls (p > 0.05) but no difference between males with medium severity IPF and control males (126). Similar to the study by McGee et al., we were not able to detect GPER1 mRNA in most samples, the majority of which were IPF lungs. It should be noted that differences in magnitude of fold changes between studies may be a result of sampling bias as the study by McGee et al. stratified individuals based on sex, whereas we did not have enough male and female samples to properly stratify our experimental groups. Both studies are in contrast with recent data from Mehrad et al. which indicate no expression of ESR1 in UIP or control lungs by immunohistochemistry

(IHC) (128). The discrepancy may be a result of variability in disease severity among

108

patients as described previously or a result of variable staining protocols and antibody binding affinities (60).

We also probed the expression of genes suspected to be involved in IPF including CTGF, BMP4, MMP1, MMP2, and MMP7. As expected, MMP1 and MMP7 mRNA expression was increased in IPF lungs (Figure 4-4F and 4-4H). These MMPs are intricately involved in IPF pathogenesis (153) and have been proposed as specific biomarkers of IPF given their increased expression in the blood, lung tissue, and BALF of IPF patients (152, 180). CTGF is a matricellular protein that promotes tissue fibrosis

(207), thus we expected it to be upregulated in IPF lungs, yet our results indicate it was significantly reduced. This is in contrast to other studies that have reported increased

CTGF in IPF patients, however, these studies measured CTGF in the plasma and BALF

(1). It is possible that CTGF is not transcriptionally induced in lungs of patients with severe IPF or perhaps it is only expressed in the lungs in earlier stages of disease.

Likewise, BMP4 also produced an interesting result as we expected it to be upregulated in IPF lungs given its involvement in pro-fibrogenic processes such as EMT in vitro

(125, 135). Furthermore, a study by Selman et al. also measured increased expression of BMP4 in IPF lungs by microarray analysis (195). The discrepancy may be a result of variable expression across disease severity classifications as the patients in the study by Selman et al. exhibited FVC% of 56.7 ± 13.5 indicating more patients had medium severity IPF whereas the majority of patients in our study had severe IPF. Further, the relative contribution of EMT to fibrogenesis throughout disease progression is heavily debated (10).

109

MMP2 is suspected to play a role in fibrogenesis in the lung (38), and it has been reported that MMP2 enzymatic activity is increased in BALF from IPF lungs (58), however, we observed reduced MMP2 mRNA expression in IPF lungs (Figure 4-4G).

Other studies investigating MMP2 expression in IPF and control lungs have produced conflicting results. One study reported positive MMP2 IHC staining in IPF but not in control lungs (196) while another reported positive staining in both control and IPF lungs

(74). The variability among studies is potentially due to complex regulatory mechanisms as MMP2 is transcriptionally inducible by TGF-β1 (Figure 3-6D) but also requires proteolytic activation by membrane type-MMPs (MT-MMPs) (38). Discrepancies between staining protocols and antibody affinities could also be a factor (60).

The ultimate goal of this work was to measure gene expression in specific cellular compartments in the lung. This is important because the lung is comprised of over 40 different cell types (131) that are often implicated in discrete pathological conditions (56) and can respond variably to insults (242). In addition, toxicologically or pathologically-associated changes in mRNA or protein expression in one cell type can be masked by changes in expression in neighboring cells when analysis is performed in whole tissue (54, 92). Further, many genes are multifunctional and play different roles in different cell types (92). As such, we intend to measure mRNA expression of estrogen receptors in specific cell types as a next step to better inform a role for E2 signaling in normal and fibrotic lung. We expect this method to allow us to use epithelial cells located in normal parenchyma as a control for epithelial cells adjacent to fibrotic lesions within each lung. By including each subject as their own control, we can circumvent

110

problems associated with our heterogenous sample populations and covariates such as sex, comorbidities, and smoking status.

To begin such investigations, we first optimized a method to quantify gene expression from laser capture microdissected tissue. LCM is emerging as a powerful tool to precisely isolate specific cells for downstream transcriptomic or proteomic analyses (54). We captured cells from an IPF lung (Figure 4-5), isolated RNA, performed two rounds of linear amplification, and measured mRNA expression of a suite of genes (Table 4-3, Figure 4-6). GAPDH and RPS13 were the most abundantly expressed as evidenced by their lower Cq values while ESR1, ESR2, and GPER1 mRNA were minimally expressed, which was expected given their limited expression in whole IPF lung tissue (Table 4-2, Figure 4-4).

In conclusion, we observed increased expression of genes known to be upregulated and involved in IPF such as MMP1 and MMP7, and reported reduced expression of genes including CTGF, BMP4, and MMP2 which are known to contribute to fibrosis although their temporal expression dynamics throughout IPF pathogenesis and across disease severity is less well understood. Most strikingly, we report reduced mRNA expression of the nuclear estrogen receptors, ESR1 and ESR2, and the membrane-bound G-protein coupled receptor for E2, GPER1, in IPF lungs suggesting that transcriptional regulation of estrogen receptors is associated with severe IPF.

Notably, we only measured expression of the receptors and not activity. Future studies should investigate binding affinities and activation of the estrogen receptors in our fibrotic models as activation is a better representation of receptor function. Overall, this work suggests that estrogen receptors play a functional role in the lung and begs the

111

question of whether there are adverse or protective effects associated with reduced expression. Future studies delineating cell type-specific changes in expression by LCM will better inform a role for estrogen receptors and transcriptional and cellular targets of estrogen receptor signaling in IPF.

112

Table 4-1. Baseline characteristics of study groups. Group Variable Male Female Control Number of subjects (n) 2 3 Age 77 ± 7.07 56 ± 12.12 Smoking Status Never 1 3 Ever 1 0 FVC, % predicted 101.5 ± 9.19 107.67 ± 7.51 IPF Number of subjects (n) 8 0 Age 61.5 ± 7.11 - Smoking Status Never 6 - Ever 2 - FVC, % predicted 42.9 ± 14.39 - Data are mean ± SD; (-) indicates absence of sample in study group.

Table 4-2. Fold change IPF lung mRNA expression compared to controls. Gene Gene name Control IPF P-value ESR1 Estrogen receptor alpha 1.21 ± 0.46 0.04 ± 0.02 3.10E-03 ESR2 Estrogen receptor beta 1.65 ± 0.77 0.26 ± 0.12 9.00E-02 GPER1 G-protein coupled estrogen receptor 1 17.17 ± 9.20 0.16 ± 0.06 6.20E-03 CTGF Connective tissue growth factor 1.51 ± 0.93 0.15 ± 0.03 6.20E-03 BMP4 Bone morphogenetic protein 4 0.99 ± 0.24 0.35 ± 0.06 1.09E-02 MMP1 Matrix metalloproteinase 1 4.09 ± 2.17 111.2 ± 42.79 2.95E-02 MMP2 Matrix metalloproteinase 2 2.99 ± 2.22 0.30 ± 0.18 4.51E-02 MMP7 Matrix metalloproteinase 7 1.09 ± 0.33 4.52 ± 1.98 1.27E-01 Data are mean ± SEM.

Table 4-3. Cq values of select genes in microdissected IPF lung tissue. Gene LCM lung Cq GAPDH 26.28 ± 0.16 RPS13 18.4 ± 0.11 ESR1 N/A ESR2 32.38 GPER1 33.2 CTGF 31.67 Data are mean ± SD; mean Cq without SD indicates that expression was only detected in one sample; N/A indicates that expression was not measured in any sample.

113

Table 4-4. Primer details for qPCR in Chapter 4. Gene Forward (5'-3') Reverse (5'-3') Protocol Source RPS13 CGAAAGCATCTTGAGAGGAACA TCGAGCCAAACGGTGAATC 95C 3m; 95C 10s, 58C 30s, x40 (83) GAPDH GAAGGTGAAGGTCGGAGTC GAAGATGGTGATGGGATTTC 95C 3m; 95C 10s, 60C 30s, x40 (11) ESR1 CCACCAACCAGTGCACCATT GGTCTTTTCGTATCCCACCTTTC 95C 3m; 95C 10s, 60C 30s, x40 (210) ESR2 Proprietary Proprietary 95C 3m; 95C 10s, 60C 30s, x40 Bio-Rad GPER GCTCCCTGCAAGCAGTCTTT GAAGGTCTCCCCGAGAAAGC 95C 3m; 95C 10s, 60C 30s, x40 (248) CTGF AATGCTGCGAGGAGTGGGT CGGCTCTAATCATAGTTGGGTCT 95C 3m; 95C 10s, 60C 30s, x40 (144) BMP4 TCCAGTATCCCCAAAGCCTG ACTCATCCAGGTACAGCATGG 95C 3m; 95C 10s, 60C 30s, x40 (248) MMP1 AGCTGCTTACGAATTTGCCG TGTCCTTGGGGTATCCGTGT 95C 3m; 95C 10s, 60C 30s, x40 (248) MMP2 TGTGTTCTTTGCAGGGAATG TCCAGAATTTGTCTCCAGCA 95C 3m; 95C 10s, 58C 30s, x40 (224) MMP7 TCGATGAGGATGAACGCTGG TCAGAGGAATGTCCCATACCCA 95C 3m; 95C 10s, 60C 30s, x40 (248) MMP9 CGTGAACATCTTCGACGCCA AAAGACCGAGTCCAGCTTGC 95C 3m; 95C 10s, 60C 30s, x40 (248)

114

Figure 4-1. Representative high resolution computerized tomography (HRCT) scans of IPF lungs. A) Supine HRCT scan of IPF lung highlight characteristic honeycombing associated with IPF (red arrow), scar tissue accumulation (green arrow), and traction bronchiectasis (blue arrow). B) Coronal HRCT scan of IPF lung highlight heterogeneous disposition of fibrotic areas (green arrow) alternating with less affected or normal parenchyma (red arrow).

115

Figure 4-2. Patients with IPF have increased deposition of collagen in the lung. (A-B) Representative lung sections from A) control and B) IPF patients stained with Sirius Red to identify collagen fibrils. Arrows indicated increased deposition of collagen.

116

Figure 4-3. RNA isolated from control and IPF lung tissue is of high quality. (A-B) Electropherograms of RNA from a representative A) control and B) IPF sample obtained by capillary gel electrophoresis using an Agilent Bioanalyzer 2100 instrument.

117

Figure 4-4. Gene expression changes in lungs of patients with IPF. (A-H) The expression of genes related to E2 signaling and fibrosis was measured in whole lung tissue samples in individuals with and without IPF by qPCR. Each point represents a different subject, n = 5 for controls n = 8 for IPF lungs. Lines indicate mean ± SEM. Asterisks (*) indicate significant different differences from controls as determined by two-tailed Mann Whitney U test (p < 0.05).

118

Figure 4-5. LCM method development. (A-C) Cells from an IPF lung were collected by LCM. A) Brightfield image of H&E-stained IPF lung section before LCM. B) Brightfield image of the same tissue section after LCM. C) Brightfield image of captured cells in collecting tube. Arrows indicate representative microdissected areas.

Figure 4-6. Optimization of gene expression analysis of microdissected tissue by qPCR. A) Amplification curves of the housekeeping genes ribosomal protein S13 (RPS13) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in amplified RNA isolated from microdissected cells. B) Melt peaks of RPS13 and GAPDH amplicons indicating absence of non-specific amplification (one peak).

119

CHAPTER 5 GENERAL CONCLUSIONS

Pulmonary fibrosis is a devastating disease characterized by progressive and irreversible scar tissue formation in the lung that can lead to impaired lung function and increased morbidity and mortality (243). Pulmonary fibrosis can be caused by exposure to environmental and occupational agents (138), viral infection (90), as a side effect of chemotherapy (203), and potentially exposure to a particulate of emerging concern, multi-walled carbon nanotubes (MWCNTs) (51, 229), however, the cause is often unknown (idiopathic, IPF). IPF is a particularly severe form of the disease with limited treatment options and is suspected to cause 40,000 deaths per year (13).

Epidemiological studies report sex-specific trends in IPF where the disease is more prevalent in men and females have better survival rates (70) suggesting a role for sex hormones in pathogenesis. Studies in animal models have produced mixed results with some suggesting that androgens promote pulmonary fibrosis (175, 230) and that estrogen (E2) is protective (108), while others suggest an exacerbating role for estrogen

(59). These studies rarely probe the molecular mechanisms contributing to sex-specific differences. As such, the purpose of this work was to characterize interactions between pro-fibrogenic and sex hormone signaling in the lung, focusing on a protective role for

E2, using a complement of in vivo and in vitro models including a MWCNT-induced mouse model (167), bronchial epithelial cells in vitro (44), and human lung tissue samples from patients with IPF (19).

Most notably, we identified estrogen receptors as transcriptional targets of pro- fibrogenic signaling in the lung. To our knowledge this is the first study to report transcriptional repression of estrogen receptor alpha (ESR1), estrogen receptor beta

120

(ESR2), and G-protein coupled estrogen receptor (GPER1) in the lungs of mice exposed to MWCNTs. We proceeded to show that the MWCNT-induced repression of

ESR1 was mediated by transforming growth factor beta1 (TGF-β1) in bronchial epithelial cells. In support of this interaction, we verified that TGF-β1 independently reduced ESR1, ESR2, and GPER1 mRNA expression and ESR1 protein levels in vitro.

Interactions between TGF-β1 and estrogen receptors have been characterized in other cell systems (26, 61, 81, 99, 112, 124, 245), however, they are typically described in the context of cancers. It is important that these interactions be considered in fibrosis because cancer and IPF share many fundamental pathogenic hallmarks (226).

Collectively, these data suggest that estrogen receptors exert a functional role in the lung and that their repression influences physiological and pathological processes.

We began investigations into a functional role for E2 and estrogen receptors in the lung by focusing on epithelial to mesenchymal transition (EMT), a process by which epithelial cells transdifferentiate into a more mesenchymal phenotype. EMT is suspected to play a role in IPF pathogenesis (86, 240) and is a target of E2 signaling mediated by ESR1, ESR2, and GPER1 in cancer cells (29, 117, 249) where E2 signaling typically inhibits EMT. As such, we hypothesized that E2 would inhibit TGF-β1- induced EMT in bronchial epithelial cells, however, we were not able to discern a clear effect. Nonetheless, we proceeded to identify novel transcriptional targets of E2 by

RNA-Seq. We detected E2-specific transcriptional repression of chloride intracellular channel 3 (CLIC3) and retinol binding protein 7 (RBP7). CLIC3 regulates proteins such as membrane-type matrix metalloproteinases which are suspected to be involved in

121

pulmonary fibrosis (48, 58, 116, 140). Similarly, RBP7 is upregulated in IPF although its precise role in fibrosis is unclear (136).

Focusing on regulation of specific genes is not always the most informative approach. Therefore, we performed gene set enrichment analysis to identify cellular processes enriched in the RNA-Seq data that may represent targets of E2 signaling in lung cells. We observed inverse regulation of pathways including extracellular matrix

(ECM) turnover, airway smooth muscle cell contraction, and calcium flux regulation by

E2 and TGF-β1. These data are intriguing considering regulation of the ECM is intimately associated with pulmonary fibrosis (30). In addition, we identified E2-specific enrichment of chromatin remodeling pathways which is particularly striking given the increased evidence of a role for histone modifications and other epigenetic marks in pulmonary fibrosis (148, 247).

We then measured ESR1, ESR2, and GPER1 mRNA expression in human IPF lungs as well as other genes suspected to be involved in pulmonary fibrosis. Similar to our results obtained in the MWCNT-induced fibrosis mouse model and in bronchial epithelial cells in vitro, we observed reduced expression of ESR1, ESR2, and GPER1.

At this point, it is unclear whether the repression is a direct effect of pro-fibrotic signaling mechanisms in IPF lung or a byproduct of a severely diseased parenchyma. Current work is investigating the cell type-specific regulation of estrogen receptor expression and E2 targets identified in the RNA-Seq analysis by laser capture microdissection

(LCM). Through these analyses we expect to better infer a role for estrogen receptors in the lung and in pulmonary fibrosis by determining the precise cell types in which ESR1,

122

ESR2, and GPER1 mRNA expression is changing because discrete cell types differentially contribute to disease pathogenesis (56).

Other important contributions of this work include the observation that MWCNT- induced repression of ESR1 mRNA expression was specific to one type of MWCNT and not a function of CNT surface chemistry or dimension. This is important because it substantiates the notion of ‘safety by design’ where nanomaterials can be designed in a certain way to minimize adverse effects. It still remains to be determined whether estrogen receptor repression in the lung is indeed adverse, but it is reassuring to know that other CNTs do not elicit the same response and may serve as suitable alternatives.

Furthermore, the formation of ROS by CNTs is heavily cited in the literature as a driving mechanism behind cellular responses yet we failed to observe a direct connection with

ROS production and repressed ESR1 mRNA expression. These results highlight that some mechanisms may be a result of ROS production, but repression of ESR1 does not seem to be a downstream target and other cellular triggers not yet identified that induce

TGF-β1 pathways are responsible.

A strength of this work is the use of multiple models of fibrosis. Through this integrated analysis, we identified estrogen receptors as transcriptional targets of both earlier pro-fibrogenic signaling mechanisms and in end-stage lungs overwhelmed with fibrotic lesions. Taken together, we can infer that estrogen receptor expression is altered throughout all stages of fibrosis but it remains to be seen whether the functional role of such receptors is consistent throughout disease. While it is too preliminary based on the data generated so far to suggest that E2 is protective against fibrosis, this work lays a solid foundation for future mechanistic probing of functional roles for E2 and

123

estrogen receptors, in particular ESR1, in modulating pro-fibrogenic signaling mechanisms. A hypothesized role for estrogen receptors is proposed in Figure 5-1 which indicates that any protective role for E2 during pulmonary fibrosis is precluded by repression of estrogen receptors by TGF-β1 and/or CTGF throughout the disease progression. We suspect that E2 may be more effective in preventing pro-fibrogenic signaling in the initiation of the disease and that by targeting the TGF-β1-induced repression of estrogen receptors, we may be able to repress pro-fibrogenic signaling in the future.

124

Figure 5-1. Hypothesized expression levels of TGF-β1, CTGF, and estrogen receptors throughout pulmonary fibrosis disease progression compared to expression in healthy tissue. After an insult in the lung (red lightning bolt) TGF-β1 levels are increased followed by an up-regulation of CTGF in the initiation of pulmonary fibrosis where processes such as EMT, fibroblast proliferation and activation, and fibrocyte recruitment occur. Up-regulation of TGF-β1 and CTGF contribute to reduced expression of estrogen receptors which persists throughout disease progression precluding any anti-fibrotic effects of estrogen. TGF-β1 and CTGF levels return to baseline levels during maturation of the disease where activated fibroblasts secrete collagens and extracellular matrix components. Expression of TGF-β1 and CTGF is similarly reduced in the terminal or end-stage period of pulmonary fibrosis and estrogen receptor repression persists.

125

APPENDIX A CTGF GENE EXPRESSION IN LUNGS OF MICE EXPOSED TO MITSUI MWCNTS

Ctgf 2.0

1.5

1.0 * 0.5 *

to 2 ControlDay Control

Fold Change Relative Change Fold Mitsui MWCNT 0.0 2 4 8 12 Days

Figure A-1. Mitsui MWCNTs reduce CTGF mRNA expression in mouse lung. Mice were exposed by whole body inhalation to clean air for 2, 4, 8, or 12 days (n = 8, 8, 6, 8, respectively) or 10 mg/m3 aerosolized Mitsui MWCNTs for 5 hours per day for 2, 4, 8, or 12 days (n = 9, 7, 5, 6, respectively). Data are mean ± SEM of mRNA expression relative to day 2 control mice. Asterisks (*) indicate statistically significant (p < 0.05) differences from controls at each time-point as determined by two-tailed, unpaired t test.

126

APPENDIX B ALIGNMENT OF MOUSE ESR1 TRANSCRIPT VARIANTS

Figure B-1. Alignment of Esr1 transcript variants using Multalin version 5.4.1 (32). Red bases indicate homology while blue bases indicate nonhomology.

127

APPENDIX C ESR1 PROTEIN LEVELS IN BRONCHIAL EPITHELIAL CELLS EXPOSED TO MITSUI MWCNTS

ESR1 1.5

1.0

0.5 Fold Change to ControlChange Fold 0.0 0 0.2 2 20 Mitsui MWCNTs (g/mL)

Figure C-1. Mitsui MWCNTs do not affect ESR1 protein expression in bronchial epithelial cells. BEAS-2Bs were exposed to increasing concentrations of Mitsui MWCNTs for 96 hours and expression of ESR1 protein was measured by western blot followed by densitometric analysis in ImageJ. ESR1 was normalized to beta-actin (ACTB) and expressed as fold change to control. Data are mean ± SEM normalized arbitrary density units of duplicate measurements per blot of three independent experiments.

128

APPENDIX D RNA INTEGRITY OF HUMAN LUNG SAMPLES

Figure D-1. RNA integrity of human lung samples. The RNA integrity of the human lungs samples was determined using an Agilent Bioanalyzer 2100 instrument.

129

LIST OF REFERENCES

1. Allen JT, Knight RA, Bloor CA, Spiteri MA. Enhanced insulin-like growth factor binding protein–related protein 2 (connective tissue growth factor) expression in patients with idiopathic pulmonary fibrosis and pulmonary sarcoidosis. Am J Respir Cell Mol Biol 21: 693-700, 1999.

2. Annes JP, Munger JS, Rifkin DB. Making sense of latent TGFβ activation. J Cell Sci 116: 217-224, 2003.

3. Ashcroft GS, Dodsworth J, Van Boxtel E, Tarnuzzer RW, Horan MA, Schultz GS, Ferguson MWJ. Estrogen accelerates cutaneous wound healing associated with an increase in TGF-1 levels. Nat Med 3: 1209-1215, 1997.

4. Attisano L and Wrana JL. Signal transduction by the TGF-β superfamily. Science 296: 1646-1647, 2002.

5. Azad N, Iyer AKV, Wang L, Liu Y, Lu Y, Rojanasakul Y. Reactive oxygen species-mediated p38 MAPK regulates carbon nanotube-induced fibrogenic and angiogenic responses. Nanotoxicology 7: 157-168, 2013.

6. Bai Y, Zhang Y, Zhang J, Mu Q, Zhang W, Butch ER, Snyder SE, Yan B. Repeated administrations of carbon nanotubes in male mice cause reversible testis damage without affecting fertility. Nat Nanotechnol 5: 683-689, 2010.

7. Bal R, Türk G, Tuzcu M, Yilmaz O, Ozercan I, Kuloglu T, Gür S, Nedzvetsky VS, Tykhomyrov AA, Andrievsky GV. Protective effects of nanostructures of hydrated C 60 fullerene on reproductive function in streptozotocin-diabetic male rats. Toxicology 282: 69-81, 2011.

8. Barcellos-Hoff M, Derynck R, Tsang M, Weatherbee J. Transforming growth factor-beta activation in irradiated murine mammary gland. J Clin Invest 93: 892, 1994.

9. Barcellos-Hoff M and Dix TA. Redox-mediated activation of latent transforming growth factor-beta 1. Mol Endocrinol 10: 1077-1083, 1996.

10. Bartis D, Mise N, Mahida RY, Eickelberg O, Thickett DR. Epithelial- mesenchymal transition in lung development and disease: does it exist and is it important? Thorax 69: 760-765, 2014.

130

11. Beck LA, Tancowny B, Brummet ME, Asaki SY, Curry SL, Penno MB, Foster M, Bahl A, Stellato C. Functional analysis of the chemokine receptor CCR3 on airway epithelial cells. J Immunol 177: 3344-3354, 2006.

12. Birring SS, Brightling CE, Bradding P, Entwisle JJ, Vara DD, Grigg J, Wardlaw AJ, Pavord ID. Clinical, radiologic, and induced sputum features of chronic obstructive pulmonary disease in nonsmokers: a descriptive study. Am J Respir Crit Care Med 166: 1078-1083, 2002.

13. Blackwell TS, Tager AM, Borok Z, Moore BB, Schwartz DA, Anstrom KJ, Bar-Joseph Z, Bitterman P, Blackburn MR, Bradford W. Future directions in idiopathic pulmonary fibrosis research. An NHLBI workshop report. Am J Respir Crit Care Med 189: 214-222, 2014.

14. Blobe GC, Schiemann WP, Lodish HF. Role of transforming growth factor β in human disease. New Engl J Med 342: 1350-1358, 2000.

15. Border WA and Noble NA. Transforming growth factor β in tissue fibrosis. New Engl J Med 331: 1286-1292, 1994.

16. Brass DM, McGee SP, Dunkel MK, Reilly SM, Tobolewski JM, Sabo-Attwood T, Fattman CL. Gender influences the response to experimental silica-induced lung fibrosis in mice. Am J Physiol Lung Cell Mol Physiol 299: L664-671, 2010.

17. Bringardner BD, Baran CP, Eubank TD, Marsh CB. The role of inflammation in the pathogenesis of idiopathic pulmonary fibrosis. Antioxid Redox Signal 10: 287-302, 2008.

18. Brookhart MA, Stürmer T, Glynn RJ, Rassen J, Schneeweiss S. Confounding control in healthcare database research: challenges and potential approaches. Med Care 48: S114, 2010.

19. Bryant AJ, Carrick RP, McConaha ME, Jones BR, Shay SD, Moore CS, Blackwell TR, Gladson S, Penner NL, Burman A. Endothelial HIF signaling regulates pulmonary fibrosis-associated pulmonary hypertension. Am J Physiol Lung Cell Mol Physiol 310: L249-L262, 2016.

20. Calvo E, Belleau P, Martel C, Labrie F. Specific transcriptional response of four blockers of estrogen receptors on estradiol-modulated genes in the mouse mammary gland. Breast Cancer Res Treat 134: 625-647, 2012.

131

21. Cardamone MD, Bardella C, Gutierrez A, Di Croce L, Rosenfeld MG, Di Renzo MF, De Bortoli M. ER as ligand-independent activator of CDH-1 regulates determination and maintenance of epithelial morphology in breast cancer cells. Proc Natl Acad Sci USA 106: 7420-7425, 2009.

22. Carey MA, Card JW, Bradbury JA, Moorman MP, Haykal-Coates N, Gavett SH, Graves JP, Walker VR, Flake GP, Voltz JW. Spontaneous airway hyperresponsiveness in estrogen receptor-α–deficient mice. Am J Respir Crit Care Med 175: 126-135, 2007.

23. Carey MA, Card JW, Voltz JW, Arbes SJ, Germolec DR, Korach KS, Zeldin DC. It's all about sex: gender, lung development and lung disease. Trends Endocrinol Metab 18: 308-313, 2007.

24. Carey MA, Card JW, Voltz JW, Germolec DR, Korach KS, Zeldin DC. The impact of sex and sex hormones on lung physiology and disease: lessons from animal studies. Am J Physiol Lung Cell Mol Physiol 293: L272-278, 2007.

25. Carvalho O and Gonçalves C. Expression of oestrogen receptors in foetal lung tissue of mice. Anat Histol Embryol 41: 1-6, 2012.

26. Chang EC, Frasor J, Komm B, Katzenellenbogen BS. Impact of estrogen receptor β on gene networks regulated by estrogen receptor α in breast cancer cells. Endocrinology 147: 4831-4842, 2006.

27. Chen CC and Lau LF. Functions and mechanisms of action of CCN matricellular proteins. Int J Biochem Cell Biol 41: 771-783, 2009.

28. Chen T, Nie H, Gao X, Yang J, Pu J, Chen Z, Cui X, Wang Y, Wang H, Jia G. Epithelial–mesenchymal transition involved in pulmonary fibrosis induced by multi- walled carbon nanotubes via TGF-beta/Smad signaling pathway. Toxicol Lett 226: 150- 162, 2014.

29. Chen ZJ, Wei W, Jiang GM, Liu H, Wei WD, Yang X, Wu YM, Liu H, Wong CK, Du J. Activation of GPER suppresses epithelial mesenchymal transition of triple negative breast cancer cells via NF-κB signals. Mol Oncol 10: 775-788, 2016.

30. Clarke DL, Carruthers AM, Mustelin T, Murray LA. Matrix regulation of idiopathic pulmonary fibrosis: the role of enzymes. Fibrogenesis Tissue Repair 6: 20, 2013.

132

31. Collard HR, Moore BB, Flaherty KR, Brown KK, Kaner RJ, King Jr TE, Lasky JA, Loyd JE, Noth I, Olman MA. Acute exacerbations of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 176: 636-643, 2007.

32. Corpet F. Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res 16: 10881-10890, 1988.

33. Couse JF, Lindzey J, Grandien K, Gustafsson J, Korach KS. Tissue distribution and quantitative analysis of estrogen receptor-α (ERα) and estrogen receptor-β (ERβ) messenger ribonucleic acid in the wild-type and ERα-knockout mouse. Endocrinology 138: 4613-4621, 1997.

34. Coward WR, Saini G, Jenkins G. The pathogenesis of idiopathic pulmonary fibrosis. Ther Adv Respir Dis 4: 367-388, 2010.

35. Coward WR, Watts K, Feghali-Bostwick CA, Jenkins G, Pang L. Repression of IP-10 by interactions between histone deacetylation and hypermethylation in idiopathic pulmonary fibrosis. Mol Cell Biol 30: 2874-2886, 2010.

36. Coward WR, Watts K, Feghali-Bostwick CA, Knox A, Pang L. Defective histone acetylation is responsible for the diminished expression of cyclooxygenase 2 in idiopathic pulmonary fibrosis. Mol Cell Biol 29: 4325-4339, 2009.

37. Cox DA and Helvering LM. Extracellular matrix integrity: a possible mechanism for differential clinical effects among selective estrogen receptor modulators and estrogens? Mol Cell Endocrinol 247: 53-59, 2006.

38. Craig VJ, Zhang L, Hagood JS, Owen CA. Matrix metalloproteinases as therapeutic targets for idiopathic pulmonary fibrosis. Am J Respir Cell Mol Biol 53: 585- 600, 2015.

39. De Volder MF, Tawfick SH, Baughman RH, Hart AJ. Carbon nanotubes: present and future commercial applications. Science 339: 535-539, 2013.

40. Derynck R and Zhang YE. Smad-dependent and Smad-independent pathways in TGF-β family signalling. Nature 425: 577-584, 2003.

133

41. Dhasarathy A, Kajita M, Wade PA. The transcription factor snail mediates epithelial to mesenchymal transitions by repression of estrogen receptor-α. Mol Endocrinol 21: 2907-2918, 2007.

42. Dilworth FJ and Chambon P. Nuclear receptors coordinate the activities of chromatin remodeling complexes and coactivators to facilitate initiation of transcription. Oncogene 20: 3047, 2001.

43. Dimitropoulou C, White RE, Ownby DR, Catravas JD. Estrogen reduces carbachol-induced constriction of asthmatic airways by stimulating large-conductance voltage and calcium-dependent potassium channels. Am J Respir Cell Mol Biol 32: 239- 247, 2005.

44. Doerner AM and Zuraw BL. TGF-β1 induced epithelial to mesenchymal transition (EMT) in human bronchial epithelial cells is enhanced by IL-1β but not abrogated by corticosteroids. Respir Res 10: 100, 2009.

45. Donaldson K, Poland CA, Murphy FA, MacFarlane M, Chernova T, Schinwald A. Pulmonary toxicity of carbon nanotubes and asbestos—similarities and differences. Adv Drug Deliv Rev 65: 2078-2086, 2013.

46. Dong J and Ma Q. Suppression of basal and carbon nanotube-induced oxidative stress, inflammation and fibrosis in mouse lungs by Nrf2. Nanotoxicology 10: 699-709, 2016.

47. Dong J, Porter DW, Batteli LA, Wolfarth MG, Richardson DL, Ma Q. Pathologic and molecular profiling of rapid-onset fibrosis and inflammation induced by multi-walled carbon nanotubes. Arch Toxicol 89: 621-633, 2015.

48. Dozynkiewicz MA, Jamieson NB, MacPherson I, Grindlay J, van den Berghe PV, von Thun A, Morton JP, Gourley C, Timpson P, Nixon C. Rab25 and CLIC3 collaborate to promote integrin recycling from late endosomes/lysosomes and drive cancer progression. Dev Cell 22: 131-145, 2012.

49. du Bois RM, Weycker D, Albera C, Bradford WZ, Costabel U, Kartashov A, King Jr TE, Lancaster L, Noble PW, Sahn SA. Forced vital capacity in patients with idiopathic pulmonary fibrosis: test properties and minimal clinically important difference. Am J Respir Crit Care Med 184: 1382-1389, 2011.

134

50. Endo M, Strano MS, Ajayan PM. Potential Applications of Carbon Nanotubes. In: Carbon Nanotubes: Advanced Topics in the Synthesis, Structure, Properties and Applications, edited by Jorio A, Dresselhaus G, and Dresselhaus MS. Berlin, Heidelberg: Springer Berlin Heidelberg, 2008, p. 13-62.

51. Fatkhutdinova LM, Khaliullin TO, Vasil'yeva OL, Zalyalov RR, Mustafin IG, Kisin ER, Birch ME, Yanamala N, Shvedova AA. Fibrosis biomarkers in workers exposed to MWCNTs. Toxicol Appl Pharmacol 299: 125-131, 2016.

52. Fernandez IE and Eickelberg O. The impact of TGF-β on lung fibrosis: from targeting to biomarkers. Proc Am Thorac Soc 9: 111-116, 2012.

53. Filardo EJ, Quinn JA, Frackelton Jr AR, Bland KI. Estrogen action via the G protein-coupled receptor, GPR30: stimulation of adenylyl cyclase and cAMP-mediated attenuation of the epidermal growth factor receptor-to-MAPK signaling axis. Mol Endocrinol 16: 70-84, 2002.

54. Fink L and Bohle RM. Laser Microdissection and RNA Analysis. In: Laser Capture Microdissection: Methods and Protocols, edited by Murray GI, and Curran S. Totowa, NJ: Humana Press, 2005, p. 167-185.

55. Folli C, Calderone V, Ramazzina I, Zanotti G, Berni R. Ligand binding and structural analysis of a human putative cellular retinol-binding protein. J Biol Chem 277: 41970-41977, 2002.

56. Franks TJ, Colby TV, Travis WD, Tuder RM, Reynolds HY, Brody AR, Cardoso WV, Crystal RG, Drake CJ, Engelhardt J. Resident cellular components of the human lung: current knowledge and goals for research on cell phenotyping and function. Proc Am Thorac Soc 5: 763-766, 2008.

57. Fridriksdottir AJ, Kim J, Villadsen R, Klitgaard MC, Hopkinson BM, Petersen OW, Rønnov-Jessen L. Propagation of oestrogen receptor-positive and oestrogen-responsive normal human breast cells in culture. Nat Commun 6: 8786, 2015.

58. García-Alvarez J, Ramirez R, Sampieri CL, Nuttall RK, Edwards DR, Selman M, Pardo A. Membrane type-matrix metalloproteinases in idiopathic pulmonary fibrosis. Sarcoidosis Vasc Diffuse Lung Dis 23: 13-21, 2006.

135

59. Gharaee-Kermani M, Hatano K, Nozaki Y, Phan SH. Gender-based differences in bleomycin-induced pulmonary fibrosis. Am J Pathol 166: 1593-1606, 2005.

60. Gomez-Fernandez C, Mejias A, Walker G, Nadji M. Immunohistochemical expression of estrogen receptor in adenocarcinomas of the lung: the antibody factor. Appl Immunohistochem Mol Morphol 18: 137-141, 2010.

61. Goto N, Hiyoshi H, Ito I, Tsuchiya M, Nakajima Y, Yanagisawa J. Estrogen and antiestrogens alter breast cancer invasiveness by modulating the transforming growth factor-β signaling pathway. Cancer Sci 102: 1501-1508, 2011.

62. Grabinski C, Hussain S, Lafdi K, Braydich-Stolle L, Schlager J. Effect of particle dimension on biocompatibility of carbon nanomaterials. Carbon 45: 2828-2835, 2007.

63. Gribbin J, Hubbard RB, Le Jeune I, Smith CJ, West J, Tata LJ. Incidence and mortality of idiopathic pulmonary fibrosis and sarcoidosis in the UK. Thorax 61: 980-985, 2006.

64. Grotendorst GR. Connective tissue growth factor: a mediator of TGF-β action on fibroblasts. Cytokine Growth Factor Rev 8: 171-179, 1997.

65. Gruber CJ, Tschugguel W, Schneeberger C, Huber JC. Production and actions of estrogens. New Engl J Med 346: 340-352, 2002.

66. Guttilla IK, Adams BD, White BA. ERα, microRNAs, and the epithelial– mesenchymal transition in breast cancer. Trends Endocrinol Metab 23: 73-82, 2012.

67. Ha J, Choi HS, Lee Y, Kwon HJ, Song YW, Kim HH. CXC chemokine ligand 2 induced by receptor activator of NF-κB ligand enhances osteoclastogenesis. J Immunol 184: 4717-4724, 2010.

68. Hackman AM and Lackner TE. Pharmacotherapy for idiopathic pulmonary arterial hypertension during the past 25 years. Pharmacother 26: 68-94, 2006.

69. Hamilton Jr RF, Xiang C, Li M, Ka I, Yang F, Ma D, Porter DW, Wu N Holian A. Purification and sidewall functionalization of multiwalled carbon nanotubes and resulting bioactivity in two macrophage models. Inhal Toxicol 25: 199-210, 2013.

136

70. Han MK, Murray S, Fell CD, Flaherty KR, Toews GB, Myers J, Colby TV, Travis WD, Kazerooni EA, Gross BH, Martinez FJ. Sex differences in physiological progression of idiopathic pulmonary fibrosis. Eur Respir J 31: 1183-1188, 2008.

71. Han MK, Postma D, Mannino DM, Giardino ND, Buist S, Curtis JL, Martinez FJ. Gender and chronic obstructive pulmonary disease: why it matters. Am J Respir Crit Care Med 176: 1179-1184, 2007.

72. Haston CK, Wang M, Dejournett RE, Zhou X, Ni D, Gu X, King TM, Weil MM, Newman RA, Amos CI, Travis EL. Bleomycin hydrolase and a genetic locus within the MHC affect risk for pulmonary fibrosis in mice. Hum Mol Gen 11: 1855-1863, 2002.

73. Hattori T, Stawski L, Nakerakanti SS, Trojanowska M. Fli1 is a negative regulator of estrogen receptor α in dermal fibroblasts. J Invest Dermatol 131: 1469- 1476, 2011.

74. Hayashi T, Stetler-Stevenson WG, Fleming MV, Fishback N, Koss MN, Liotta LA, Ferrans VJ, Travis WD. Immunohistochemical study of metalloproteinases and their tissue inhibitors in the lungs of patients with diffuse alveolar damage and idiopathic pulmonary fibrosis. Am J Pathol 149: 1241-1256, 1996.

75. He X, Young S-H, Schwegler-Berry D, Chisholm WP, Fernback JE, Ma Q. Multiwalled carbon nanotubes induce a fibrogenic response by stimulating reactive oxygen species production, activating NF-κB signaling, and promoting fibroblast-to- myofibroblast transformation. Chem Res Toxicol 24: 2237-2248, 2011.

76. Hellemans J, Mortier G, De Paepe A, Speleman F, Vandesompele J. qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data. Genome Biol 8: R19, 2007.

77. Hosper NA, van den Berg PP, de Rond S, Popa ER, Wilmer MJ, Masereeuw R, Bank RA. Epithelial-to-mesenchymal transition in fibrosis: collagen type I expression is highly upregulated after EMT, but does not contribute to collagen deposition. Exp Cell Res 319: 3000-3009, 2013.

78. Hussain S, Sangtian S, Anderson SM, Snyder RJ, Marshburn JD, Rice AB, Bonner JC, Garantziotis S. Inflammasome activation in airway epithelial cells after multi-walled carbon nanotube exposure mediates a profibrotic response in lung fibroblasts. Part Fibre Toxicol 11: 28, 2014.

137

79. Iavicoli I, Fontana L, Leso V, Bergamaschi A. The effects of nanomaterials as endocrine disruptors. Int J Mol Sci 14: 16732-16801, 2013.

80. Imbeaud S, Graudens E, Boulanger V, Barlet X, Zaborski P, Eveno E, Mueller O, Schroeder A, Auffray C. Towards standardization of RNA quality assessment using user-independent classifiers of microcapillary electrophoresis traces. Nucleic Acids Res 33: e56-e56, 2005.

81. Ito I, Hanyu A, Wayama M, Goto N, Katsuno Y, Kawasaki S, Nakajima Y, Kajiro M, Komatsu Y, Fujimura A. Estrogen inhibits transforming growth factor β signaling by promoting Smad2/3 degradation. J Biol Chem 285: 14747-14755, 2010.

82. Jackson P, Vogel U, Wallin H, Hougaard KS. Prenatal exposure to carbon black (Printex 90): Effects on sexual development and neurofunction. Basic Clin Pharmacol Toxicol 109: 434-437, 2011.

83. Jacob F, Guertler R, Naim S, Nixdorf S, Fedier A, Hacker NF, Heinzelmann- Schwarz V. Careful selection of reference genes is required for reliable performance of RT-qPCR in human normal and cancer cell lines. PLoS One 8: e59180, 2013.

84. Johkoh T, Sakai F, Noma S, Akira M, Fujimoto K, Watadani T, Sugiyama Y. Honeycombing on CT; its definition, pathologic correlation, and future direction of its diagnosis. Eur J Radiol 83: 27-31, 2014.

85. Kage H and Borok Z. EMT and interstitial lung disease: a mysterious relationship. Curr Opin Pulm Med 18: 517-523, 2012.

86. Kalluri R, and Weinberg RA. The basics of epithelial-mesenchymal transition. J Clin Invest 119: 1420-1428, 2009.

87. Kamitani S, Yamauchi Y, Kawasaki S, Takami K, Takizawa H, Nagase T, Kohyama T. Simultaneous stimulation with TGF-β1 and TNF-α induces epithelial mesenchymal transition in bronchial epithelial cells. Int Arch Allergy Immunol 155: 119- 128, 2011.

88. Karow M, Popp T, Egea V, Ries C, Jochum M, Neth P. Wnt signalling in mouse mesenchymal stem cells: impact on proliferation, invasion and MMP expression. J Cell Mol Med 13: 2506-2520, 2009.

138

89. Kasai H, Allen JT, Mason RM, Kamimura T, Zhang Z. TGF-1 induces human alveolar epithelial to mesenchymal cell transition (EMT). Respir Res 6: 56, 2005.

90. Kelly BG, Lok SS, Hasleton PS, Egan JJ, Stewart JP. A rearranged form of Epstein–Barr virus DNA is associated with idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 166: 510-513, 2002.

91. Kelly MJ and Levin ER. Rapid actions of plasma membrane estrogen receptors. Trends Endocrinol Metab 12: 152-156, 2001.

92. Kelly MM, Leigh R, Gilpin SE, Cheng E, Martin G, Radford K, Cox G, Gauldie J. Cell-specific gene expression in patients with usual interstitial pneumonia. Am J Respir Crit Care Med 174: 557-565, 2006.

93. Kheradmand F, Rishi K, Werb Z. Signaling through the EGF receptor controls lung morphogenesis in part by regulating MT1-MMP-mediated activation of gelatinase A/MMP2. J Cell Sci 115: 839-848, 2002.

94. Kim ES, Sohn YW, Moon A. TGF-β-induced transcriptional activation of MMP-2 is mediated by activating transcription factor (ATF) 2 in human breast epithelial cells. Cancer Lett 252: 147-156, 2007.

95. Kim KK, Kugler MC, Wolters PJ, Robillard L, Galvez MG, Brumwell AN, Sheppard D, Chapman HA. Alveolar epithelial cell mesenchymal transition develops in vivo during pulmonary fibrosis and is regulated by the extracellular matrix. Proc Natl Acad Sci USA 103: 13180-13185, 2006.

96. King TE, Pardo A, Selman M. Idiopathic pulmonary fibrosis. Lancet 378: 1949- 1961, 2011.

97. Kinyamu HK, Collins JB, Grissom SF, Hebbar PB, Archer TK. Genome wide transcriptional profiling in breast cancer cells reveals distinct changes in hormone receptor target genes and chromatin modifying enzymes after proteasome inhibition. Mol Carcinog 47: 845-885, 2008.

98. Kis K, Liu X, Hagood JS. Myofibroblast differentiation and survival in fibrotic disease. Expert Rev Mol Med 13: e27, 2011.

139

99. Kleuser B, Malek D, Gust R, Pertz HH, Potteck H. 17-β-estradiol inhibits transforming growth factor-β signaling and function in breast cancer cells via activation of extracellular signal-regulated kinase through the G protein-coupled receptor 30. Mol Pharmacol 74: 1533-1543, 2008.

100. Koenig GC, Rowe RG, Day SM, Sabeh F, Atkinson JJ, Cooke KR, Weiss SJ. MT1-MMP–dependent remodeling of cardiac extracellular matrix structure and function following myocardial infarction. Am J Pathol 180: 1863-1878, 2012.

101. Komm BS, Terpening CM, Benz DJ, Graeme KA, Gallegos A. Estrogen binding, receptor mRNA, and biologic response in osteoblast-like osteosarcoma cells. Science 241: 81, 1988.

102. Labrie F. Extragonadal synthesis of sex steroids: intracrinology. Ann Endocrinol 64: 95-107, 2003.

103. Latoche JD, Ufelle AC, Fazzi F, Ganguly K, Leikauf GD, Fattman CL. Secreted phosphoprotein 1 and sex-specific differences in silica-induced pulmonary fibrosis in mice. Environ Health Perspect 124: 1199-1207, 2016.

104. Leask A and Abraham DJ. The role of connective tissue growth factor, a multifunctional matricellular protein, in fibroblast biology. Biochem Cell Biol 81: 355-363, 2003.

105. Leask A and Abraham DJ. TGF-beta signaling and the fibrotic response. FASEB J 18: 816-827, 2004.

106. Leask A, Holmes A, Abraham DJ. Connective tissue growth factor: a new and important player in the pathogenesis of fibrosis. Curr Rheumatol Rep 4: 136-142, 2002.

107. Leask A, Holmes A, Black CM, Abraham DJ. Connective tissue growth factor gene regulation. J Biol Chem 278: 13008-13015, 2003.

108. Lekgabe ED, Royce SG, Hewitson TD, Tang MLK, Zhao C, Moore XL, Tregear GW, Bathgate RA, Du XJ, Samuel CS. The effects of relaxin and estrogen deficiency on collagen deposition and hypertrophy of nonreproductive organs. Endocrinology 147: 5575-5583, 2006.

140

109. Levin ER. Extranuclear estrogen receptor's roles in physiology: lessons from mouse models. Am J Physiol Endocrinol Metab 307: E133-E140, 2014.

110. Ley B, Collard HR, King Jr TE. Clinical course and prediction of survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 183: 431-440, 2011.

111. Li R, Wang X, Ji Z, Sun B, Zhang H, Chang CH, Lin S, Meng H, Liao YP, Wang M. Surface charge and cellular processing of covalently functionalized multiwall carbon nanotubes determine pulmonary toxicity. ACS Nano 7: 2352-2368, 2013.

112. Li YC, Ding XS, Li HM, Zhang Y, Bao J. Role of G protein-coupled estrogen receptor 1 in modulating transforming growth factor-β stimulated mesangial cell extracellular matrix synthesis and migration. Mol Cell Endocrinol 391: 50-59, 2014.

113. Lösel R and Wehling M. Nongenomic actions of steroid hormones. Nat Rev Mol Cell Biol 4: 46-55, 2003.

114. Lovegrove AS, Sun J, Gould KA, Lubahn DB, Korach KS, Lane PH. Estrogen receptor α-mediated events promote sex-specific diabetic glomerular hypertrophy. Am J Physiol Renal Physiol 287: F586-F591, 2004.

115. Lyons RM, Keski-Oja J, Moses HL. Proteolytic activation of latent transforming growth factor-beta from fibroblast-conditioned medium. J Cell Biol 106: 1659-1665, 1988.

116. Macpherson IR, Rainero E, Mitchell LE, Van Den Berghe PV, Speirs C, Dozynkiewicz MA, Chaudhary S, Kalna G, Edwards J, Timpson P. CLIC3 controls recycling of late endosomal MT1-MMP and dictates invasion and metastasis in breast cancer. J Cell Sci 127: 3893-3901, 2014.

117. Mak P, Leav I, Pursell B, Bae D, Yang X, Taglienti CA, Gouvin LM, Sharma VM, Mercurio AM. ERβ impedes prostate cancer EMT by destabilizing HIF-1α and inhibiting VEGF-mediated snail nuclear localization: implications for Gleason grading. Cancer Cell 17: 319-332, 2010.

118. Mäkelä M, Larjava H, Pirilä E, Maisi P, Salo T, Sorsa T, Uitto V-J. Matrix metalloproteinase 2 (gelatinase A) is related to migration of keratinocytes. Exp Cell Res 251: 67-78, 1999.

141

119. Manke A, Wang L, Rojanasakul Y. Mechanisms of nanoparticle-induced oxidative stress and toxicity. BioMed Res Int 2013: 15, 2013.

120. Mason R, Schwarz M, Hunninghake G, Musson R. NHLBI workshop summary. Pharmacological therapy for idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 160: 1771-1777, 1999.

121. Massaro D and Massaro GD. Estrogen receptor regulation of pulmonary alveolar dimensions: alveolar sexual dimorphism in mice. Am J Physiol Lung Cell Mol Physiol 290: L866-L870, 2006.

122. Massaro D and Massaro GD. Estrogen regulates pulmonary alveolar formation, loss, and regeneration in mice. Am J Physiol Lung Cell Mol Physiol 287: L1154-L1159, 2004.

123. Matsubara S, Swasey CH, Loader JE, Dakhama A, Joetham A, Ohnishi H, Balhorn A, Miyahara N, Takeda K, Gelfand EW. Estrogen determines sex differences in airway responsiveness after allergen exposure. Am J Respir Cell Mol Biol 38: 501- 508, 2008.

124. Matsuda T, Yamamoto T, Muraguchi A, Saatcioglu F. Cross-talk between transforming growth factor-β and estrogen receptor signaling through Smad3. J Biol Chem 276: 42908-42914, 2001.

125. McCormack N, Molloy EL, O’Dea S. Bone morphogenetic proteins enhance an epithelial-mesenchymal transition in normal airway epithelial cells during restitution of a disrupted epithelium. Respir Res 14: 36, 2013.

126. McGee SP, Zhang H, Karmaus W, Sabo-Attwood T. Influence of sex and disease severity on gene expression profiles in individuals with idiopathic pulmonary fibrosis. Int J Mol Epidemiol Genet 5: 71-86, 2014.

127. McKinney W, Chen B, Frazer D. Computer controlled multi-walled carbon nanotube inhalation exposure system. Inhal Toxicol 21: 1053-1061, 2009.

128. Mehrad M, Trejo Bittar HE, Yousem SA. Sex steroid receptor expression in idiopathic pulmonary fibrosis. Human Pathol 2017.

142

129. Mercer RR, Scabilloni JF, Hubbs AF, Battelli LA, McKinney W, Friend S, Wolfarth MG, Andrew M, Castranova V, Porter DW. Distribution and fibrotic response following inhalation exposure to multi-walled carbon nanotubes. Part Fibre Toxicol 10: 33, 2013.

130. Meunier E, Coste A, Olagnier D, Authier H, Lefèvre L, Dardenne C, Bernad J, Béraud M, Flahaut E, Pipy B. Double-walled carbon nanotubes trigger IL-1β release in human monocytes through Nlrp3 inflammasome activation. Nanomedicine 8: 987- 995, 2012.

131. Mills SE. Histology for pathologists. Philadelphia, Pa: Lippincott Williams & Wilkins, 2007, p. 1272.

132. Mishra A, Rojanasakul Y, Chen BT, Castranova V, Mercer RR, Wang L. Assessment of pulmonary fibrogenic potential of multiwalled carbon nanotubes in human lung cells. J Nanomater 2012: 4, 2012.

133. Mishra A, Stueckle TA, Mercer RR, Derk R, Rojanasakul Y, Castranova V, Wang L. Identification of TGF- receptor-1 as a key regulator of carbon nanotube- induced fibrogenesis. Am J Physiol Lung Cell Mol Physiol 309: L821-833, 2015.

134. Mollerup S, Jørgensen K, Berge G, Haugen A. Expression of estrogen receptors α and β in human lung tissue and cell lines. Lung Cancer 37: 153-159, 2002.

135. Molloy EL, Adams A, Moore JB, Masterson JC, Madrigal-Estebas L, Mahon BP, O'Dea S. BMP4 induces an epithelial–mesenchymal transition-like response in adult airway epithelial cells. Growth Factors 26: 12-22, 2008.

136. Molyneaux PL, Willis-Owen SAG, Cox MJ, James P, Cowman S, Loebinger M, Blanchard A, Edwards LM, Stock C, Daccord C, Renzoni EA, Wells AU, Moffatt MF, Cookson WOC, Maher TM. Host–microbial interactions in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 195: 1640-1650, 2017.

137. Morley P, Whitfield JF, Vanderhyden BC, Tsang BK, Schwartz JL. A new, nongenomic estrogen action: the rapid release of intracellular calcium. Endocrinology 131: 1305-1312, 1992.

138. Mossman BT and Churg A. Mechanisms in the pathogenesis of asbestosis and silicosis. Am J Respir Crit Care Med 157: 1666-1680, 1998.

143

139. Moustakas A and Heldin C-H. The regulation of TGFβ signal transduction. Development 136: 3699-3714, 2009.

140. Mu D, Cambier S, Fjellbirkeland L, Baron JL, Munger JS, Kawakatsu H, Sheppard D, Broaddus VC, Nishimura SL. The integrin αvβ8 mediates epithelial homeostasis through MT1-MMP–dependent activation of TGF-β1. J Cell Biol 157: 493- 507, 2002.

141. Muller J, Huaux F, Moreau N, Misson P, Heilier JF, Delos M, Arras M, Fonseca A, Nagy JB, Lison D. Respiratory toxicity of multi-wall carbon nanotubes. Toxicol Appl Pharmacol 207: 221-231, 2005.

142. Munger JS, Huang X, Kawakatsu H, Griffiths MJD, Dalton SL, Wu J, Pittet JF, Kaminski N, Garat C, Matthay MA, Rifkin DB, Sheppard D. A mechanism for regulating pulmonary inflammation and fibrosis: the integrin αvβ6 binds and activates latent TGFβ1. Cell 96: 319-328, 1999.

143. Nagai H, Okazaki Y, Chew SH, Misawa N, Yamashita Y, Akatsuka S, Ishihara T, Yamashita K, Yoshikawa Y, Yasui H. Diameter and rigidity of multiwalled carbon nanotubes are critical factors in mesothelial injury and carcinogenesis. Proc Natl Acad Sci USA 108: E1330-E1338, 2011.

144. Nagaraja T, Chen L, Balasubramanian A, Groopman JE, Ghoshal K, Jacob ST, Leask A, Brigstock DR, Anand AR, Ganju RK. Activation of the connective tissue growth factor (CTGF)-transforming growth factor β 1 (TGF-β 1) axis in hepatitis C virus- expressing hepatocytes. PLoS One 7: e46526, 2012.

145. Nakao A, Afrakhte M, Morn A, Nakayama T, Christian JL, Heuchel R, Itoh S, Kawabata M, Heldin NE, Heldin CH. Identification of Smad7, a TGFβ-inducible antagonist of TGF-β signalling. Nature 389: 631-635, 1997.

146. Nishimura K, Kitaichi M, Izumi T, Nagai S, Kanaoka M, Itoh H. Usual interstitial pneumonia: histologic correlation with high-resolution CT. Radiology 182: 337-342, 1992.

147. Noh H, Oh EY, Seo JY, Yu MR, Kim YO, Ha H, Lee HB. Histone deacetylase-2 is a key regulator of diabetes-and transforming growth factor-β1-induced renal injury. Am J Physiol Renal Physiol 297: F729-F739, 2009.

148. O’Reilly S. Epigenetics in fibrosis. Mol Aspects Med 54: 89-102, 2017.

144

149. Olson AL, Swigris JJ, Lezotte DC, Norris JM, Wilson CG, Brown KK. Mortality from pulmonary fibrosis increased in the United States from 1992 to 2003. Am J Respir Crit Care Med 176: 277-284, 2007.

150. Oswald N and Parkinson T. Honeycomb lungs. QJM Int J Med 18: 1-20, 1949.

151. Pacurari M, Yin XJ, Ding M, Leonard SS, Schwegler-berry D, Ducatman BS, Chirila M, Endo M, Castranova V, Vallyathan V. Oxidative and molecular interactions of multi-wall carbon nanotubes (MWCNT) in normal and malignant human mesothelial cells. Nanotoxicology 2: 155-170, 2008.

152. Pan L, Yamauchi K, Uzuki M, Nakanishi T, Takigawa M, Inoue H, Sawai T. Type II alveolar epithelial cells and interstitial fibroblasts express connective tissue growth factor in IPF. Eur Respir J 17: 1220-1227, 2001.

153. Pardo A and Selman M. Role of matrix metaloproteases in idiopathic pulmonary fibrosis. Fibrogenesis Tissue Repair 5: S9, 2012.

154. Parekh T, Saxena B, Reibman J, Cronstein BN, Gold LI. Neutrophil chemotaxis in response to TGF-beta isoforms (TGF-beta 1, TGF-beta 2, TGF-beta 3) is mediated by fibronectin. J Immunol 152: 2456-2466, 1994.

155. Patrone C, Cassel TN, Pettersson K, Piao YS, Cheng G, Ciana P, Maggi A, Warner M, Gustafsson JÅ, Nord M. Regulation of postnatal lung development and homeostasis by estrogen receptor β. Mol Cell Biol 23: 8542-8552, 2003.

156. Payne AH and Hales DB. Overview of steroidogenic enzymes in the pathway from cholesterol to active steroid hormones. Endocr Rev 25: 947-970, 2004.

157. Perkins TN, Shukla A, Peeters PM, Steinbacher JL, Landry CC, Lathrop SA, Steele C, Reynaert NL, Wouters EF, Mossman BT. Differences in gene expression and cytokine production by crystalline vs. amorphous silica in human lung epithelial cells. Part Fibre Toxicol 9: 6, 2012.

158. Petrel TA and Brueggemeier RW. Increased proteasome-dependent degradation of estrogen receptor-alpha by TGF-β1 in breast cancer cell lines. J Cell Biochem 88: 181-190, 2003.

145

159. Pfaffl MW. A new mathematical model for relative quantification in real-time RT– PCR. Nucleic Acids Res 29: e45-e45, 2001.

160. Pi L, Ding X, Jorgensen M, Pan JJ, Oh SH, Pintilie D, Brown A, Song WY, Petersen BE. Connective tissue growth factor with a novel fibronectin binding site promotes cell adhesion and migration during rat oval cell activation. Hepatology 47: 996-1004, 2008.

161. Pi L, Shenoy AK, Liu J, Kim S, Nelson N, Xia H, Hauswirth WW, Petersen BE, Schultz GS, Scott EW. CCN2/CTGF regulates neovessel formation via targeting structurally conserved cystine knot motifs in multiple angiogenic regulators. FASEB J 26: 3365-3379, 2012.

162. Pincus DJ, Kassira N, Gombosh M, Berho M, Glassberg M, Karl M, Elliot SJ, Thaller S. 17β-Estradiol modifies diabetic wound healing by decreasing matrix metalloproteinase activity. Wounds 22: 171-178, 2010.

163. Planas-Silva MD and Waltz PK. Estrogen promotes reversible epithelial-to- mesenchymal-like transition and collective motility in MCF-7 breast cancer cells. J Steroid Biochem Mol Biol 104: 11-21, 2007.

164. Pociask DA, Sime PJ, Brody AR. Asbestos-derived reactive oxygen species activate TGF-β1. Lab Invest 84: 1013-1023, 2004.

165. Poland CA, Duffin R, Kinloch I, Maynard A, Wallace WAH, Seaton A, Stone V, Brown S, MacNee W, Donaldson K. Carbon nanotubes introduced into the abdominal cavity of mice show asbestos-like pathogenicity in a pilot study. Nat Nanotechnol 3: 423-428, 2008.

166. Porter D, Sriram K, Wolfarth M, Jefferson A, Schwegler-Berry D, Andrew ME, Castranova V. A biocompatible medium for nanoparticle dispersion. Nanotoxicology 2: 144-154, 2008.

167. Porter DW, Hubbs AF, Chen BT, McKinney W, Mercer RR, Wolfarth MG, Battelli L, Wu N, Sriram K, Leonard S. Acute pulmonary dose–responses to inhaled multi-walled carbon nanotubes. Nanotoxicology 7: 1179-1194, 2012.

168. Porter DW, Hubbs AF, Mercer RR, Wu N, Wolfarth MG, Sriram K, Leonard S, Battelli L, Schwegler-Berry D, Friend S. Mouse pulmonary dose-and time course-

146

responses induced by exposure to multi-walled carbon nanotubes. Toxicology 269: 136- 147, 2010.

169. Prossnitz ER, Arterburn JB, Smith HO, Oprea TI, Sklar LA, Hathaway HJ. Estrogen signaling through the transmembrane G protein–coupled receptor GPR30. Annu Rev Physiol 70: 165-190, 2008.

170. Rabinovich EI, Kapetanaki MG, Steinfeld I, Gibson KF, Pandit KV, Yu G, Yakhini Z, Kaminski N. Global methylation patterns in idiopathic pulmonary fibrosis. PLoS One 7: e33770, 2012.

171. Raghu G, Chen SY, Hou Q, Yeh WS, Collard HR. Incidence and prevalence of idiopathic pulmonary fibrosis in US adults 18–64 years old. Eur Respir J 48: 179-186, 2016.

172. Raghu G, Collard HR, Egan JJ, Martinez FJ, Behr J, Brown KK, Colby TV, Cordier JF, Flaherty KR, Lasky JA. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med 183: 788-824, 2011.

173. Raghu G, Weycker D, Edelsberg J, Bradford WZ, Oster G. Incidence and prevalence of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 174: 810-816, 2006.

174. Ravanti L and Kahari V. Matrix metalloproteinases in wound repair. Int J Mol Med 6: 391-407, 2000.

175. Redente EF, Jacobsen KM, Solomon JJ, Lara AR, Faubel S, Keith RC, Henson PM, Downey GP, Riches DW. Age and sex dimorphisms contribute to the severity of bleomycin-induced lung injury and fibrosis. Am J Physiol Lung Cell Mol Physiol 301: L510-L518, 2011.

176. Ren L and Zhong W. Oxidation reactions mediated by single-walled carbon nanotubes in aqueous solution. Environ Sci Technol 44: 6954-6958, 2010.

177. Revankar CM, Cimino DF, Sklar LA, Arterburn JB, Prossnitz ER. A transmembrane intracellular estrogen receptor mediates rapid cell signaling. Science 307: 1625-1630, 2005.

147

178. Rhodes L, Moorman JE, Redd SC. Sex differences in asthma prevalence and other disease characteristics in eight states. J Asthma 42: 777-782, 2005.

179. Ronzani C, Spiegelhalter C, Vonesch J-L, Lebeau L, Pons F. Lung deposition and toxicological responses evoked by multi-walled carbon nanotubes dispersed in a synthetic lung surfactant in the mouse. Arch Toxicol 86: 137-149, 2012.

180. Rosas IO, Richards TJ, Konishi K, Zhang Y, Gibson K, Lokshin AE, Lindell KO, Cisneros J, MacDonald SD, Pardo A. MMP1 and MMP7 as potential peripheral blood biomarkers in idiopathic pulmonary fibrosis. PLoS Med 5: e93, 2008.

181. Ryman-Rasmussen JP, Cesta MF, Brody AR, Shipley-Phillips JK, Everitt JI, Tewksbury EW, Moss OR, Wong BA, Dodd DE, Andersen ME. Inhaled carbon nanotubes reach the subpleural tissue in mice. Nat Nanotechnol 4: 747-751, 2009.

182. Ryman-Rasmussen JP, Tewksbury EW, Moss OR, Cesta MF, Wong BA, Bonner JC. Inhaled multiwalled carbon nanotubes potentiate airway fibrosis in murine allergic asthma. Am J Respir Cell Mol Biol 40: 349-358, 2009.

183. Sakai N and Tager AM. Fibrosis of two: epithelial cell-fibroblast interactions in pulmonary fibrosis. BBA Mol Basis Dis 1832: 911-921, 2013.

184. Sanders YY, Ambalavanan N, Halloran B, Zhang X, Liu H, Crossman DK, Bray M, Zhang K, Thannickal VJ, Hagood JS. Altered DNA methylation profile in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 186: 525-535, 2012.

185. Santibañez JF, Quintanilla M, Bernabeu C. TGF-β/TGF-β receptor system and its role in physiological and pathological conditions. Clinical science 121: 233-251, 2011.

186. Sargent LM, Porter DW, Staska LM, Hubbs AF, Lowry DT, Battelli L. Promotion of lung adenocarcinoma following inhalation exposure to multi-walled carbon nanotubes. Part Fibre Toxicol 11: 2014.

187. Sargent LM, Porter DW, Staska LM, Hubbs AF, Lowry DT, Battelli L, Siegrist KJ, Kashon ML, Mercer RR, Bauer AK. Promotion of lung adenocarcinoma following inhalation exposure to multi-walled carbon nanotubes. Part Fibre Toxicol 11: 3, 2014.

148

188. Sathish V, Martin YN, Prakash YS. Sex steroid signaling: implications for lung diseases. Pharmacol Ther 150: 94-108, 2015.

189. Sato Y and Rifkin DB. Inhibition of endothelial cell movement by pericytes and smooth muscle cells: activation of a latent transforming growth factor-beta 1-like molecule by plasmin during co-culture. J Cell Biol 109: 309-315, 1989.

190. Sauler M and Gulati M. Newly recognized occupational and environmental causes of chronic terminal airways and parenchymal lung disease. Clin Chest Med 33: 667-680, 2012.

191. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nature Methods 9: 671, 2012.

192. Schultz-Cherry S and Murphy-Ullrich JE. Thrombospondin causes activation of latent transforming growth factor- secreted by endothelial cells by a novel mechanism. J Cell Biol 122: 923-932, 1993.

193. Selman M, Carrillo G, Estrada A, Mejia M, Becerril C, Cisneros J, Gaxiola M, Pérez-Padilla R, Navarro C, Richards T. Accelerated variant of idiopathic pulmonary fibrosis: clinical behavior and gene expression pattern. PLoS One 2: e482, 2007.

194. Selman M, King TE, Pardo A. Idiopathic pulmonary fibrosis: prevailing and evolving hypotheses about its pathogenesis and implications for therapy. Ann Intern Med 134: 136-151, 2001.

195. Selman M, Pardo A, Barrera L, Estrada A, Watson SR, Wilson K, Aziz N, Kaminski N, Zlotnik A. Gene expression profiles distinguish idiopathic pulmonary fibrosis from hypersensitivity pneumonitis. Am J Respir Crit Care Med 173: 188-198, 2006.

196. Selman M, Ruiz V, Cabrera S, Segura L, Ramírez R, Barrios R, Pardo A. TIMP-1, -2, -3, and -4 in idiopathic pulmonary fibrosis. A prevailing nondegradative lung microenvironment? Am J Physiol Lung Cell Mol Physiol 279: L562-L574, 2000.

197. Seomun Y, Kim J, Choun-Ki J. Overexpression of matrix metalloproteinase-2 mediates phenotypic transformation of lens epithelial cells. Biochem J 358: 41-48, 2001.

149

198. Shatkin JA and Oberdörster G. Comment on Shvedova et al. (2016), “gender differences in murine pulmonary responses elicited by cellulose nanocrystals”. Part Fibre Toxicol 13: 59, 2016.

199. Shim B, Pacheco-Rodriguez G, Kato J, Darling TN, Vaughan M, Moss J. Sex-specific lung diseases: effect of oestrogen on cultured cells and in animal models. Eur Respir Rev 22: 302-311, 2013.

200. Shvedova AA, Kisin ER, Yanamala N, Farcas MT, Menas AL, Williams A, Fournier PM, Reynolds JS, Gutkin DW, Star A. Gender differences in murine pulmonary responses elicited by cellulose nanocrystals. Part Fibre Toxicol 13: 28, 2016.

201. Siiteri PK, Murai JT, Hammond G, Nisker JA, Raymoure W, Kuhn RW. The serum transport of steroid hormones. Recent Prog Horm Res 38: 457-510, 1982.

202. Silva CIS, Müller NL, Fujimoto K, Kato S, Ichikado K, Taniguchi H, Kondoh Y, Johkoh T, Churg A. Acute exacerbation of chronic interstitial pneumonia: high- resolution computed tomography and pathologic findings. J Thorac Imaging 22: 221- 229, 2007.

203. Sleijfer S. Bleomycin-induced pneumonitis. Chest 120: 617-624, 2001.

204. Society AT. Idiopathic pulmonary fibrosis, diagnosis and treatment. (International consensus statement). Am J Respir Crit Care Med 161: 646-664, 2000.

205. Society AT. Idiopathic pulmonary fibrosis: diagnosis and treatment. International consensus statement. American Thoracic Society (ATS), and the European Respiratory Society (ERS). Am J Respir Crit Care Med 161: 646-664, 2000.

206. Son ED, Lee JY, Lee S, Kim MS, Lee BG, Chang IS, Chung JH. Topical application of 17β-estradiol increases extracellular matrix protein synthesis by stimulating TGF-β signaling in aged human skin in vivo. J Invest Dermatol 124: 1149- 1161, 2005.

207. Sonnylal S, Shi-Wen X, Leoni P, Naff K, Van Pelt CS, Nakamura H, Leask A, Abraham D, Bou-Gharios G, de Crombrugghe B. Selective expression of connective tissue growth factor in fibroblasts in vivo promotes systemic tissue fibrosis. Arthritis Rheum 62: 1523-1532, 2010.

150

208. Sonnylal S, Xu S, Jones H, Tam A, Sreeram VR, Ponticos M, Norman J, Agrawal P, Abraham D, de Crombrugghe B. Connective tissue growth factor causes EMT-like cell fate changes in vivo and in vitro. J Cell Sci 126: 2164-2175, 2013.

209. Soulet F, Kilarski WW, Antczak P, Herbert J, Bicknell R, Falciani F, Bikfalvi A. Gene signatures in wound tissue as evidenced by molecular profiling in the chick embryo model. BMC Genomics 11: 495, 2010.

210. Spizzo R, Nicoloso M, Lupini L, Lu Y, Fogarty J, Rossi S, Zagatti B, Fabbri M, Veronese A, Liu X. miR-145 participates with TP53 in a death-promoting regulatory loop and targets estrogen receptor-α in human breast cancer cells. Cell Death Differ 17: 246-254, 2010.

211. Stabile LP, Davis ALG, Gubish CT, Hopkins TM, Luketich JD, Christie N, Finkelstein S, Siegfried JM. Human non-small cell lung tumors and cells derived from normal lung express both estrogen receptor α and β and show biological responses to estrogen. Cancer Res 62: 2141-2150, 2002.

212. Stoica A, Saceda M, Fakhro A, Solomon HB, Fenster BD, Martin MB. The role of transforming growth factor-β in the regulation of estrogen receptor expression in the MCF-7 breast cancer cell line. Endocrinology 138: 1498-1505, 1997.

213. Stope MB, Popp SL, Knabbe C, Buck MB. Estrogen receptor α attenuates transforming growth factor-β signaling in breast cancer cells independent from agonistic and antagonistic ligands. Breast Cancer Res Treat 120: 357-367, 2010.

214. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 102: 15545-15550, 2005.

215. Sumikawa H, Johkoh T, Ichikado K, Taniguchi H, Kondoh Y, Fujimoto K, Tateishi U, Hiramatsu T, Inoue A, Natsag J. Usual interstitial pneumonia and chronic idiopathic interstitial pneumonia: analysis of CT appearance in 92 patients. Radiology 241: 258-266, 2006.

216. Sweeney S, Berhanu D, Misra SK, Thorley AJ, Valsami-Jones E, Tetley TD. Multi-walled carbon nanotube length as a critical determinant of bioreactivity with primary human pulmonary alveolar cells. Carbon 78: 26-37, 2014.

151

217. Tam A, Morrish D, Wadsworth S, Dorscheid D, Man SP, Sin DD. The role of female hormones on lung function in chronic lung diseases. BMC Womens Health 11: 24, 2011.

218. Tanaka KI, Tanaka Y, Namba T, Azuma A, Mizushima T. Heat shock protein 70 protects against bleomycin-induced pulmonary fibrosis in mice. Biochem Pharmacol 80: 920-931, 2010.

219. Taylor S, Wakem M, Dijkman G, Alsarraj M, Nguyen M. A practical approach to RT-qPCR—publishing data that conform to the MIQE guidelines. Methods 50: S1-S5, 2010.

220. Todd NW, Luzina IG, Atamas SP. Molecular and cellular mechanisms of pulmonary fibrosis. Fibrogenesis Tissue Repair 5: 11, 2012.

221. Toth B, Schwacha MG, Kuebler JF, Bland KI, Wang P, Chaudrya IH. Gender dimorphism in neutrophil priming and activation following trauma-hemorrhagic shock. Int J Mol Med 11: 357-364, 2003.

222. Townsend EA, Thompson MA, Pabelick CM, Prakash Y. Rapid effects of estrogen on intracellular Ca2+ regulation in human airway smooth muscle. Am J Physiol Lung Cell Mol Physiol 298: L521-L530, 2010.

223. Travis WD, King TE, Bateman ED, Lynch DA, Capron F, Center D, Colby TV, Cordier JF, DuBois RM, Galvin J. American Thoracic Society/European Respiratory Society international multidisciplinary consensus classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med 165: 277-304, 2002.

224. Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M, Rozen SG. Primer3—new capabilities and interfaces. Nucleic Acids Res 40: e115- e115, 2012.

225. Valera IC, Kafaja S, Pham B, Fishbein M, Singh RR. Investigating mechanisms of sex bias in lung fibrosis using mice with sex chromosome complement. J Immunol 196: 197-198, 2016.

226. Vancheri C, Failla M, Crimi N, Raghu G. Idiopathic pulmonary fibrosis: a disease with similarities and links to cancer biology. Eur Respir J 35: 496-504, 2010.

152

227. Varkey AB. Chronic obstructive pulmonary disease in women: exploring gender differences. Curr Opin Pulm Med 10: 98-103, 2004.

228. Vietti G, Ibouraadaten S, Palmai-Pallag M, Yakoub Y, Bailly C, Fenoglio I, Marbaix E, Lison D, van den Brule S. Towards predicting the lung fibrogenic activity of nanomaterials: experimental validation of an in vitro fibroblast proliferation assay. Part Fibre Toxicol 10: 52, 2013.

229. Vietti G, Lison D and van den Brule S. Mechanisms of lung fibrosis induced by carbon nanotubes: towards an Adverse Outcome Pathway (AOP). Part Fibre Toxicol 13: 11, 2016.

230. Voltz JW, Card JW, Carey MA, DeGraff LM, Ferguson CD, Flake GP, Bonner JC, Korach KS, Zeldin DC. Male sex hormones exacerbate lung function impairment after bleomycin-induced pulmonary fibrosis. Am J Respir Cell Mol Biol 39: 45-52, 2008.

231. Wang P, Nie X, Wang Y, Li Y, Ge C, Zhang L, Wang L, Bai R, Chen Z, Zhao Y. Multiwall carbon nanotubes mediate macrophage activation and promote pulmonary fibrosis through TGF-β/Smad signaling pathway. Small 9: 3799-3811, 2013.

232. Wang X, Duch MC, Mansukhani N, Ji Z, Liao Y-P, Wang M, Zhang H, Sun B, Chang CH, Li R. Use of a pro-fibrogenic mechanism-based predictive toxicological approach for tiered testing and decision analysis of carbonaceous nanomaterials. ACS Nano 9: 3032-3043, 2015.

233. Wang X, Xia T, Addo Ntim S, Ji Z, Lin S, Meng H, Chung CH, George S, Zhang H, Wang M. Dispersal state of multiwalled carbon nanotubes elicits profibrogenic cellular responses that correlate with fibrogenesis biomarkers and fibrosis in the murine lung. ACS Nano 5: 9772-9787, 2011.

234. Wang X, Xia T, Ntim SA, Ji Z, George S, Meng H, Zhang H, Castranova V, Mitra S, Nel AE. Quantitative techniques for assessing and controlling the dispersion and biological effects of multiwalled carbon nanotubes in mammalian tissue culture cells. ACS Nano 4: 7241-7252, 2010.

235. Wang Z, Chen C, Finger S, Jung M, Schwarz H, Swanson N, Lareu R, Raghunath M. Suberoylanilide hydroxamic acid: a potential epigenetic therapeutic agent for lung fibrosis? Eur Respir J 34: 145-155, 2009.

153

236. Webb WR, Muller NL, Naidich DP. High-resolution CT of the lung. Philadephia, Pa: Lippincott Williams & Wilkins, 2014.

237. Westcott JL and Cole SR. Traction bronchiectasis in end-stage pulmonary fibrosis. Radiology 161: 665-669, 1986.

238. Whitsett JA. Intrinsic and innate defenses in the lung: intersection of pathways regulating lung morphogenesis, host defense, and repair. J Clin Invest 109: 565-569, 2002.

239. Wik E, Ræder MB, Krakstad C, Trovik J, Birkeland E, Hoivik EA, Mjos S, Werner HM, Mannelqvist M, Stefansson IM. Lack of estrogen receptor-α is associated with epithelial–mesenchymal transition and PI3K alterations in endometrial carcinoma. Clin Cancer Res 19: 1094-1105, 2013.

240. Willis BC and Borok Z. TGF--induced EMT: mechanisms and implications for fibrotic lung disease. Am J Physiol Lung Cell Mol Physiol 293: L525-534, 2007.

241. Wipff PJ, Rifkin DB, Meister JJ, Hinz B. Myofibroblast contraction activates latent TGF-β1 from the extracellular matrix. J Cell Biol 179: 1311-1323, 2007.

242. Wong MH and Johnson MD. Differential response of primary alveolar type I and type II cells to LPS stimulation. PLoS One 8: e55545, 2013.

243. Wynn TA. Integrating mechanisms of pulmonary fibrosis. J Exp Med 208: 1339- 1350, 2011.

244. Xaubet A, Marin-Arguedas A, Lario S, Ancochea J, Morell F, Ruiz-Manzano J, Rodriguez-Becerra E, Rodriguez-Arias JM, Iñigo P, Sanz S. Transforming growth factor-β1 gene polymorphisms are associated with disease progression in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 168: 431-435, 2003.

245. Yamamoto T, Saatcioglu F, Matsuda T. Cross-talk between bone morphogenic proteins and estrogen receptor signaling. Endocrinology 143: 2635-2642, 2002.

246. Yan LF, Wei YN, Nan HY, Yin Q, Qin Y, Zhao X, Chen BY, Zhao G, Wei JG, Cui GB. Proliferative phenotype of pulmonary microvascular endothelial cells plays a critical role in the overexpression of CTGF in the bleomycin-injured rat. Exp Toxicol Pathol 66: 61-71, 2014.

154

247. Yang IV, and Schwartz DA. Epigenetics of idiopathic pulmonary fibrosis. Transl Res 165: 48-60, 2015.

248. Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, and Madden TL. Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinformatics 13: 134, 2012.

249. Ye Y, Xiao Y, Wang W, Yearsley K, Gao J, Shetuni B, and Barsky S. ERα signaling through slug regulates E-cadherin and EMT. Oncogene 29: 1451-1462, 2010.

250. Yoshida S, Hiyoshi K, Ichinose T, Takano H, Oshio S, Sugawara I, Takeda K, and Shibamoto T. Effect of nanoparticles on the male reproductive system of mice. Int J Androl 32: 337-342, 2009.

251. Yoshida S, Hiyoshi K, Oshio S, Takano H, Takeda K, and Ichinose T. Effects of fetal exposure to carbon nanoparticles on reproductive function in male offspring. Fertil Steril 93: 1695-1699, 2010.

252. Yu Q, and Stamenkovic I. Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-β and promotes tumor invasion and angiogenesis. Genes Dev 14: 163-176, 2000.

155

BIOGRAPHICAL SKETCH

Dr. Ley Cody Smith was born in 1988 in Orlando, FL. After graduating from

William R. Boone High School in 2006, he enrolled at the University of Florida where he received his Bachelor of Science in integrative biology in 2010. Dr. Smith then worked as a research associate in the lab of Dr. David Barber in the Center for Environmental and Human Toxicology for the summer after completing his bachelor’s. Thereafter, he enrolled in the University of Florida graduate school in August 2011 and received his

Master of Science in veterinary medical sciences the spring of 2013 under the direction of Dr. Tara Sabo-Attwood. He completed his Doctor of Philosophy in veterinary medical sciences with a concentration in toxicology in the summer of 2017.

156