THE ROLE OF HYPOXIA INDUCIBLE FACTORS 1 AND 2 IN COBALT-INDUCED LUNG
INFLAMMATION AND DEVELOPMENT OF LUNG IMMUNITY
By
Steven Paul Proper

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Biochemistry & Molecular Biology - Environmental Toxicology - Doctor of
Philosophy
2013

ABSTRACT
THE ROLE OF HYPOXIA INDUCIBLE FACTORS 1 AND 2 IN COBALT-INDUCED LUNG
INFLAMMATION AND DEVELOPMENT OF LUNG IMMUNITY
By
Steven Paul Proper

Cobalt is a transition metal utilized frequently in industry for the production of alloys,
such as tungsten carbide. Human exposures to cobalt result mostly from workers inhaling
cobalt-containing dusts. Many of these workers develop hard metal lung disease (HMLD or
“cobalt lung”). Cobalt has been shown to be the necessary component in tungsten carbide dust
to cause HMLD, though the mechanism is largely unknown. Interestingly, cobalt is a wellknown hypoxia mimic and activates hypoxia-inducible factor (HIF)-mediated signaling. Our
previous studies using a doxycycline inducible, lung epithelial-specific knockout of HIF1α
showed a switch from neutrophilic to eosinophilic recruitment during cobalt-induced
inflammation, suggesting a role for epithelial HIF1α in modulating the immune response of the
lung. Little is known, however, about the role of HIF2α in cobalt-induced lung injury or the
possible overlap of HIF1α and HIF2α in this context. With this in mind, several experiments
have been performed to elucidate the mechanism of HIF1α-deficient induced eosinophilia and
the role of HIF1α and HIF2α in cobalt-induced lung inflammation.
To characterize the mechanism of change in cobalt-induced inflammatory response in
the HIF1α-deficient mice, several experiments were performed, including flow cytometry of

resident lung immune cells, different temporal HIF1α deletions, use of specific pharmacologic
inhibitors of the suspected effector pathway of NF-κB and the protective pathway of adenosine
receptor A2B (Adora2b). These studies revealed that cobalt-treated HIF1α-deficient mice
produced more GATA3+ T-helper cells, confirming that the eosinophilia involved a T-helper type
2 (TH2) response. Also, inducing the HIF1α deletion in the early postnatal time period (P4-14),
and not the adult (P32 onward) was required in establishing a predisposition for eosinophilic
inflammation. Adora2b exerts protective effects on inflamed tissues, and inhibition of this
receptor prior to cobalt dosing decreased the eosinophilia seen in HIF1α-deficient mice to
control levels, suggesting a role for adenosine signaling in promoting eosinophilic inflammation.
Use of the proteasome inhibitor TCH013, thought to selectively inhibit NF-κB, immediately prior
to cobalt dosing, was ineffective in altering the expected eosinophilia. Overall these data imply
that HIF1α in alveolar epithelial type II and Club (Clara) cells plays a vital role in establishing the
lung’s immune environment early in postnatal lung development.
HIF2α-deficient mice treated with cobalt display eosinophilia which peaks at the 14 day
time point, later than eosinophilia observed at 2-5 days in HIF1α-deficient mice. When mice
lose both HIF1α and HIF2α by recombination, results are similar to the HIF1α-deficient mice at
5 days, suggesting that the effects of HIF1α loss are more important in driving earlier
eosinophilia. Gene expression, histopathological analyses and cytokine profiling of lungs
showed increases in classic TH2 markers correlated with eosinophilic inflammation. All
together these data show that both HIF1α and HIF2α are likely involved in post-natal
development of proper immune responses to cobalt-induced lung inflammation.

Copyright by
STEVEN PAUL PROPER
2013

ACKNOWLEDGEMENTS
I would first like to thank my mentor, Dr. John J. LaPres. When I arrived at MSU in 2006 I
was very interested in his lab’s research, and after a rotation I was relieved that he was willing
to take me as his first DO-PhD student. Over the years he has been incredibly understanding
when medical school, clinical rotations, commuting or major life events demanded my time.
John has provided a supportive learning environment while challenging me to expand my skill
set and has taught me to think more clearly about the literature, my experiments, writing and
presentations. His door was always open, and I have enjoyed our many chats over a quick
lunch in his office during an incubation step. Sometimes it was research, other times music and
movie quotes. I am eternally grateful for the guidance he has provided me.
I am immensely appreciative for my committee members Dr. Jack R. Harkema, Dr.
Timothy Zacharewski, Dr. Eric Hegg, and Dr. William (Bill) Henry who have all played a
significant role in my research by providing valuable guidance and insightful comments during
our meetings and discussions. I would especially like to thank Jack Harkema for providing a
wealth of knowledge and collaboration regarding toxicologic pathology and the art of
histopathology, his incredibly helpful manuscript reviews and also for providing additional
mentorship to me during conferences.
A huge debt of gratitude is owed to the DO-PhD Program, especially Dr. J. Justin
McCormick and Bethany Heinlen for their unwavering support, advocacy and guidance since I
joined the program in 2006. I have enjoyed watching them advance the profile of the DO-PhD
Program at MSU by enhanced programming, training, and communication among the many

v

involved departments at MSU, all for the benefit of the students. I am grateful and proud to
have been given an opportunity to be in this program.
I would like to thank the Biochemistry & Molecular Biology (BMB) Program, especially
Dr. Tom Sharkey (Dept. Chair), Dr. John L. Wang (Graduate Recruitment Director) and Dr. Jon
M. Kaguni (Graduate Student Director) for accepting me into their program and their years of
support and guidance afterwards. They have been especially helpful for me as I navigated the
unique schedule of the DO-PhD Program through the BMB Department. Sincere thanks go out
to all BMB Department staff for their assistance over the years, especially Jessica Lawrence,
Teresa Vollmer, Melinda Kochenderfer, Pappan (Kaillathe) Padmanabhan, Laurie Secord, Billy
Yang, Carol Vanderjagt, Ron Norris and Joyce Robinson. I must also thank Joe Leykam and all
the past members of the RTSF for their assistance.
Furthermore, I would like to thank the Center for Integrative Toxicology and the
Environmental and Integrative Toxicological Sciences (EITS) Program, especially Dr. Robert Roth
and Dr. Norbert Kaminski for allowing me the opportunity to gain expertise in the area of
Toxicology as a second PhD major while at MSU. Their support has been instrumental in my
training in toxicology, and has definitively set the course of my research interests in the future.
I am also grateful for the help of Amy Swagart, Kasey Baldwin and Lauren St John during my
time in the EITS Program.
Importantly, I owe a huge debt of gratitude to all the members of the LaPres Lab who
have helped directly with my research, especially Krista K. Greenwood for her immense amount
of time spent on animal work, as well as Nathan Downing, Corey Cook and Christian Merrill. I

vi

also must thank Dr. Yogesh Saini for creating the mouse strains that I have used and for his
excellent training in animal techniques and collaboration since his graduation from MSU. I also
want to thank current and past LaPres lab members for their friendship and creating a fun
atmosphere in the lab, including Hye Jin Hwang, Dr. Dorothy Tappenden, Dr. Scott G. Lynn, Dr.
Kyunghee Burkitt, Dr. Ajith Vengellur and Dr. KangAe Lee.
I especially would like to thank all the members of the Harkema lab for their countless
contributions to my studies and pathological analyses, especially Lori Bramble, Ryan
Lewandowski, Dennis Shubitowski and Dr. Christina Brandenberger. I also would like to thank
Dr. James Wagner and Dr. Daven Jackson-Humbles for their assistance in functional airway
analyses and expertise. I wish to thank members of the Schutte lab including Youssef Kousa,
Arianna Smith and Tamer Mansour for assistance and training in immunofluorescence and
mouse techniques. I also want to thank Dr. Stephen Carey for his mentorship these last few
years. I am indebted to the MSU Histopathology lab, especially Amy Porter and Kathy Joseph
for all their help over the years with my histology, teaching me new skills and helping with all
my puzzles. Further, I am very thankful to ULAR, especially Don Herrington and Ken Ledford for
all their work in caring for our mice.
I wish to express my sincerest gratitude to my former mentors Dr. Deborah Keil and Dr.
Ray Wells for encouraging me to pursue and continue research, and to all the teachers over the
years who have inspired in me the wonder of science and the satisfaction of learning, especially
Dr. Ali Zand, Dr. Stacey Seeley, Dr. Pat Atkinson, Dr. Daryl Doyle and Dr. Judy Harwood.

vii

I sincerely wish to thank all of my fellow graduate students and close friends – I
appreciate your empathy, humor and support during the stressful times, and I value deeply the
friendships we have made. Of special note are Nick Tebeau, Cherie and Sean Kifer, Tyler and
Lisa Voss, Michelle Manente Angrish and Krishan Angrish, Noshir Amaria, Nathan and Christina
Schuldt, Tony Brandau and Shanti Virupannavar Brandau, Eric Carter and Julie Brownell Carter,
Dustin and Maria Carver, Chad Shultz, Eric Schauberger, Jospeh Prinsen, Igor Korolev, Hae
Young Hawong, Youssef Kousa, Dionisia Quiroga, Tyrell Simkins, Darin Quach, Paul Beach, Ania
Kopec, Agnes Forgacs, Peer Karmaus and Rupinder Sayal.
I wish to thank my family for all their love and support, especially my parents, Doug and
Linda, for always believing in me. I want to thank my brother Chris and his wife Deb, my
parents-in-law John and Jeannine Bauer, my brothers-in-law Ryan and Zach Bauer, John and
Keri Reigger and Wayne Proper for all their support as well.
Most importantly, I wish to thank my wife Alexis for her constant support, patience and
understanding these many years. She has always been there for me, and I could not have done
this without her. This is a special year for us because we welcomed our son Jamie Russell into
th

the world November 30 . His smiles and laughter are a bright spot in our life, and we are very
happy that he is here with us.
No meaningful work in life is done (or possible) alone. Thank you all!

viii

TABLE OF CONTENTS
LIST OF TABLES .....................................................................................................................xii
LIST OF FIGURES ..................................................................................................................xiii
KEY TO ABBREVIATIONS AND SYMBOLS .............................................................................. xvii
CHAPTER 1 - INTRODUCTION AND SPECIFIC AIMS .................................................................. 1
Oxygen – historical and biological significance ........................................................................... 2
Review of lung anatomy, physiology and development ............................................................. 4
Premature birth, supplemental O2, bronchopulmonary dysplasia and asthma ................... 14
Immune mechanisms in the lung .............................................................................................. 16
Innate immunity .................................................................................................................... 16
Adaptive immunity ................................................................................................................ 23
Role of T cells in lung defense........................................................................................... 23
T-helper type 1 (TH1) response......................................................................................... 24
T-helper type 2 (TH2) response......................................................................................... 26
Other T-cell responses ...................................................................................................... 27
Mechanisms of lung neutrophilia and eosinophilia .............................................................. 27
Epithelial-immune crosstalk in inflammatory responses ...................................................... 30
Alveolar epithelial type II (ATII) cells and immunity ......................................................... 31
The Hygiene Hypothesis ........................................................................................................ 31
Normoxia, hypoxia and oxygen sensing mechanisms............................................................... 33
Hypoxia Signaling ...................................................................................................................... 36
Hypoxia inducible factors ...................................................................................................... 36
HIFs in lung inflammation ................................................................................................. 43
Other hypoxia-responsive signaling pathways in lung inflammation ................................... 45
Table 2. Non-HIF hypoxia-responsive transcription factors (modified from [216]). ....... 46
Nuclear factor kappa B (NF-κB)......................................................................................... 47
Purinergic (adenosine) receptors ..................................................................................... 48
Cobalt ........................................................................................................................................ 52
Occurrence and uses ............................................................................................................. 52
Toxicity ................................................................................................................................... 53
Hard metal lung disease ........................................................................................................ 53
Role of HIFs in cobalt toxicity ................................................................................................ 55
Hypotheses and Specific Aims................................................................................................... 56
CHAPTER 2 – MATERIALS AND METHODS............................................................................. 57
Description of mice ................................................................................................................... 58
Doxycycline treatment and animal husbandry ......................................................................... 62
Cobalt exposure ........................................................................................................................ 65

ix

PSB1115 exposure ..................................................................................................................... 65
TCH013 exposure ...................................................................................................................... 66
LPS exposure ............................................................................................................................. 66
Hyperoxia exposure .................................................................................................................. 66
Necropsy, tissue harvesting and processing ............................................................................. 68
Histopathology and immunohistochemistry............................................................................. 68
Gene expression analysis by qRT-PCR ....................................................................................... 69
Cytokine bead array .................................................................................................................. 72
Flow cytometry of lung immune cells ....................................................................................... 72
Statistical analytical methods ................................................................................................... 73
CHAPTER 3 – MECHANISTIC INVESTIGATIONS OF THE EOSINOPHILIC PHENOTYPE OF MICE
DEFICIENT IN HIF1α IN LUNG EPITHLIUM TREATED WITH COBALT ........................................ 74
Results ....................................................................................................................................... 75
Timing of HIF1α Deletion and Eosinophilia ........................................................................... 75
Determination of TH2 Inflammation by Flow Cytometry ...................................................... 78
PSB1115 and Adora2b in Cobalt-Induced lung injury ........................................................... 83
NF-κB in Cobalt-induced lung injury ...................................................................................... 89
TCH013 .............................................................................................................................. 89
LPS ..................................................................................................................................... 93
Hyperoxia during early postnatal phase................................................................................ 97
Discussion ................................................................................................................................ 103
CHAPTER 4 – COMPARATIVE PHENOTYPES OF COBALT-INDUCED RESPIRATORY
INFLAMMATION IN MICE DEFICIENT FOR HIF1α, HIF2α or HIF1α/2α IN LUNG EPITHELIUM . 112
Results ..................................................................................................................................... 113
Δ/Δ

HIF2α
mice 60 μg CoCl2 time course.............................................................................. 113
HIF2α IHC ........................................................................................................................ 113
BAL Cellularity ................................................................................................................. 113
Histopathology and Immunohistochemistry .................................................................. 118
Cytokine analysis ............................................................................................................. 125
HIF1α/2α mice treated at 5 days ......................................................................................... 129
HIF1α and HIF2α IHC ....................................................................................................... 129
BAL Cellularity ................................................................................................................. 129
Histopathology and Immunohistochemistry .................................................................. 133
Cytokine analysis of whole lung lysate by bead array .................................................... 135
Gene expression analyses ............................................................................................... 135
Discussion ................................................................................................................................ 142
CHAPTER 5 – SUMMARY AND CONCLUSIONS ..................................................................... 148
Summary of Research ............................................................................................................. 149
Major findings and implications .............................................................................................. 149
Proposed Working Model ....................................................................................................... 150
Knowledge gaps and future studies ........................................................................................ 153
x

REFERENCES ...................................................................................................................... 157

xi

LIST OF TABLES

Table 1. Partial summary of cytokine/chemokine-mediated recruitment of cellular
effectors in the lung. ................................................................................................... 28
Table 2. Non-HIF hypoxia-responsive transcription factors (modified from [216]). ................... 46
Table 3. List of primers for mouse genotyping and qRT-PCR. ..................................................... 70

xii

LIST OF FIGURES

Figure 1. Comparative epithelial constituents across regions of mouse and human lungs.......... 8
Figure 2. Comparison of human and mouse lung development. ................................................ 13
Figure 3. Overview of pulmonary epithelial cell influence in immunity. .................................... 18
Figure 4. T-Helper (TH) cell Subtypes ........................................................................................... 25
Figure 5. Cartoon of HIFα and HIFβ subunit domains and functions .......................................... 37
Figure 6. Review of HIF signaling ................................................................................................. 39
Figure 7. Possible crosstalk of HIF1α with Adora2b and NF-κB in CTL and HIF1α

Δ/Δ

cells ......... 50

Figure 8. Inducible, lung epithelium-specific transgenic knockout mouse model ...................... 59
Figure 9. Eosinophilia in HIF1α

fl/fl

mice by genotype and DOX treatment ................................. 60

Figure 10. Mouse experiment treatment and dosing schemes .................................................. 64
Figure 11. BAL total cells and macrophages from various DOX regimens in HIF1α mice
treated with 5d cobalt ............................................................................................. 76
Figure 12. BAL eosinophils and neutrophils (PMNs) from various DOX regimens in HIF1α
mice treated with 5d cobalt. .................................................................................... 77
Figure 13. BAL lymphocytes from various DOX regimens in HIF1α mice treated with 5d
cobalt. ...................................................................................................................... 78
Figure 14. TH2 immune cell populations in HIF1α mice with 5d cobalt ...................................... 80
Figure 15. TH1 immune cell populations in HIF1α mice with 5d cobalt ...................................... 81
Figure 16. TC immune cell populations in HIF1α mice with 5d cobalt ........................................ 82
Δ/Δ

Figure 17. IHC for Adora2b in lungs of cobalt treated CTL and HIF1α
xiii

mice ........................... 84

Figure 18. BAL total cells and macrophages from HIF1α mice treated with cobalt for 5 days
+/- PSB1115 .............................................................................................................. 87
Figure 19. BAL eosinophils and PMNs from HIF1α mice treated with cobalt for 5 days +/PSB1115 ................................................................................................................... 88
Figure 20. BAL lymphocytes from HIF1α mice treated with cobalt for 5 days +/- PSB1115 ....... 89
Figure 21. Structural schematic of TCH013 ................................................................................. 90
Figure 22. BAL total cells and macrophages from HIF1α mice treated with cobalt for 5 days
+/- TCH013 ............................................................................................................... 91
Figure 23. BAL eosinophils and PMNs from HIF1α mice treated with cobalt for 5 days +/TCH013. .................................................................................................................... 92
Figure 24. BAL lymphocytes from HIF1α mice treated with cobalt for 5 days +/- TCH013 ........ 93
Figure 25. BAL total cells and macrophages from HIF1α mice in P4-30 and P32-42 DOX
groups treated +/- LPS ............................................................................................. 95
Figure 26. BAL eosinophils and PMNs from HIF1α mice in P4-30 and P32-42 DOX groups
treated +/- LPS ......................................................................................................... 96
Figure 27. BAL lymphocytes from HIF1α mice in P4-30 and P32-42 DOX groups treated +/LPS ............................................................................................................................ 97
Figure 28. H&E Staining for P0-14 mice exposed to 75% O2 ....................................................... 99
Figure 29. HIF1α IHC for mice exposed to 75% O2 from P0-14 ................................................. 100
Figure 30. BAL total cells and macrophages from CTL mice exposed to 75% O2 from P0-14
and 5d cobalt as adults .......................................................................................... 101
Figure 31. BAL eosinophils and PMNs from CTL mice exposed to 75% O2 from P0-14 and
5d cobalt as adults. ................................................................................................ 102
Figure 32. BAL lymphocytes from CTL mice exposed to 75% O2 from P0-14 and 5d cobalt
as adults ................................................................................................................. 103
Figure 33. HIF2α IHC for CTL and P4-30 DOX treated mice. ...................................................... 114

xiv

Figure 34. BAL total cells and macrophages from HIF2α mice treated with cobalt for 1, 5
and 14 days ............................................................................................................ 116
Figure 35. BAL eosinophils and PMNs from HIF2α mice treated with cobalt for 1, 5 and
14 days ................................................................................................................... 117
Figure 36. BAL lymphocytes from HIF2α mice treated with cobalt for 1, 5 and 14 days .......... 118
Figure 37. HIF2α strain H&E stain for mice with 1, 5 and 14 days cobalt ................................. 120
Figure 38. HIF2α strain major basic protein IHC for mice with 1, 5 and 14 days cobalt ........... 122
Figure 39. AB-PAS stain for HIF2α mice with 1, 5 and 14 days cobalt....................................... 124
Figure 40. IL-4 and IL-5 cytokine levels from bead array in HIF2α mice. .................................. 126
Figure 41. IL-6 and KC cytokine levels from bead array in HIF2α mice ..................................... 127
Figure 42. IFNγ and IL-10 cytokine levels from bead array in HIF2α mice ................................ 128
Figure 43. HIF1α and HIF2α IHC for HIF1α/2α mice .................................................................. 130
Figure 44. BAL total cells and macrophages from HIF1α, HIF2α and HIF1α/2α mice at 5
days cobalt exposure ............................................................................................. 131
Figure 45. BAL eosinophils and PMNs from HIF1α, HIF2α and HIF1α/2α mice at 5 days
cobalt exposure...................................................................................................... 132
Figure 46. BAL lymphocytes from HIF1α, HIF2α and HIF1α/2α mice at 5 days cobalt
exposure................................................................................................................. 133
Figure 47. H&E stain from HIF2α and HIF1α/2α mice at 5 days cobalt exposure ..................... 134
Figure 48. MBP IHC from HIF2α and HIF1α/2α mice at 5 days cobalt exposure ...................... 136
Figure 49. AB-PAS stain from HIF2α and HIF1α/2α mice at 5 days cobalt exposure ................ 137
Figure 50. IL-4 and IL-5 cytokines from WLL of HIF1α, HIF2α and HIF1α/2α mice after 5
days cobalt exposure ............................................................................................. 138
Figure 51. IL-6 and IL-17 cytokines from WLL of HIF1α, HIF2α and HIF1α/2α mice after 5
days cobalt exposure ............................................................................................. 139
Figure 52. IFNγ and TNFα cytokines from WLL of HIF1α, HIF2α and HIF1α/2α mice after 5
days cobalt exposure ............................................................................................. 140

xv

Figure 53. WLL gene expression profiles of HIF1α, HIF2α and HIF1α/2α mice with 5 days
cobalt exposure...................................................................................................... 141
Figure 54. Working model of WT and HIFα

Δ/Δ

lung epithelial cells exposed to cobalt ............ 151

xvi

KEY TO ABBREVIATIONS AND SYMBOLS

2-OG

2-oxoglutarate (α-keto glutarate)

A2BR

see Adora2b

AB-PAS

Alcian blue – periodic acid Schiff (stain)

ADP

adenosine diphosphate

Adora2b

adenosine receptor A2B (aka A2BR)

ALI

acute lung injury

AM

alveolar macrophage

APC

antigen presenting cell

ARNT

aryl hydrocarbon receptor nuclear translocator

ATI

alveolar epithelial type I

ATII

alveolar epithelial type II

ATP

adenosine triphosphate

BAL

bronchoalveolar lavage

bHLH

basic helix loop helix

BPD

bronchopulmonary dysplasia

cAMP

cyclic adenosine monophosphate

CC

cysteine-cysteine family of chemokines (β chemokines)

CD

cluster of differentiation (cell surface marker)

CD40L

cluster of differentiation 40 ligand

CO2

carbon dioxide

CoCl2

cobalt (II) chloride

Cre

cre recombinase

CTAD

carboxy terminal trans activation domain

CTL

control

CXC

cysteine-x-cysteine family of chemokines (α chemokines)

DC

dendritic cell

xvii

pDC

plasmacytoid dendritic cell

mDC

myeloid dendritic cell

cDC

conventional dendritic cell

DOX

doxycycline

Δ/Δ

deficiency via recombination

EDHB

ethyl 3,4-dihydroxybenzoic acid

Eos

eosinophils

Epas1

endothelial PAS-domain protein 1 (aka HIF2α)

Epo

erythropoietin

ETC

electron transport chain

FADH2

flavin adenine dinucleotide (reduced and hydroquinone form)

FIH

factor inhibiting HIF

GA

gestational age

GATA3

GATA binding protein 3

Gi

G alpha inhibitory protein

GOE

Great oxygenation event

GPCR

G-protein coupled receptor

Gq

G alpha q/11 protein

Gs

G alpha stimulatory protein

H&E

hematoxylin & eosin (stain)

HIF

hypoxia inducible factor

HMLD

hard metal lung disease

HRE

hypoxia response element

IFNγ

interferon gamma

IHC

immunohistochemistry (antibody labeling of tissues)

IκB

inhibitor of kappa B protein

IKK

IκB kinase (complex)

IL

interluekin

xviii

iNOS

inducible nitric oxide synthase

Io

ionomycin

i.p.

intra-peritoneal (injection)

KC

keratinocyte-derived chemokine (aka CXCL1 or GROα)

LPS

lipopolysaccharide

LTα

lymphotoxin alpha (aka tumor necrosis factor beta)

MBP

major basic protein

MCE

mucociliary escalator

MEF

mouse embryonic fibroblast

MHCII

major histocompatibility complex II

NADH

nicotinamide adenine dinucleotide (reduced form)

NF-κB

nuclear factor kappa B

NO

nitric oxide

NTAD

amino terminal trans activation domain

ODD

oxygen dependent degradation domain

OVA

ovalbumin (usually refers to model of sensitization/challenge of ovalbumin)

PAS

Per-Arnt-Sim

Pi

inorganic phosphate

PHD

prolyl hydroxylase domain-containing enzyme

PKC

protein kinase C

PMA

phorbolmyristate acetate

PMN

neutrophil, or polymorphonuclear leukocyte

PNEC

pulmonary neuroendocrine cell

PRR

pattern recognition receptor

pVHL

protein product of Von Hippel Lindau tumor suppressor gene

RBC

red blood cell (erythrocyte)

ROP

retinopathy of prematurity

ROS

reactive oxygen species

rtTA

reverse tetracycline transactivator

xix

Stat

signal transducer and activator of transcription

SP-C

surfactant protein C

T0

naïve T lymphoyte

TAD

trans activation domain

T-bet

T-box expressed in T cells (aka T-box 21)

TC

cytotoxic T lymphocyte

TCA

tricarboxylic acid

TCR

T cell receptor

TGFβ

transforming growth factor beta

TH

T helper lymphocyte

T H1

T helper type 1 lymphocyte

T H2

T helper type 2 lymphocyte

TH17

T helper type 2 lymphocyte

TREG

T regulatory cell

TLR

Toll-like receptor

TNFα

tumor necrosis factor alpha

TSLP

thymic stromal lymphopoietin

Vegf

vascular endothelial growth factor

WLL

whole lung lysate

Ym1

chitinase 3-like 3

Ym2

chitinase 3-like 4

xx

CHAPTER 1 - INTRODUCTION AND SPECIFIC AIMS

1

Oxygen – historical and biological significance
The discovery of oxygen resulted from the early study of air, combustion and respiration
th

th

in 17 and 18 centuries [1-5]. As modern science took root and the age of alchemy gave way
to chemistry and beyond, decades of further experimentation has revealed the fundamental
role oxygen has played in the evolution of almost all multicellular life. Oxygen is the third most
abundant element in the universe based on spectroscopic analysis behind hydrogen and helium
[6], and the most abundant element in the Earth’s crust by mass [4]. Molecular oxygen
(dioxygen; O2) accounts for 20.9% of the atmosphere, the presence of which can be almost
entirely attributed to photosynthetic organisms. Geologic evidence to date suggests that nearly
2.4 billion years ago there was a Great Oxygenation Event (GOE) which brought the oxygen
concentrations above 0.001% [7, 8]. The rate of O2 increase from the GOE to more current
levels (>20%) seems to be gradual, but is still an area of debate among geologists and
paleobiologists [9]. A more detailed timeline of atmospheric O2 concentrations from the last
550 million years shows great variation, and claims that oceanic and atmospheric O2 levels have
driven evolution of multicellular life are supported not just by the fossil record and
geochemistry, but also through evolutionary and systems biology [10-13].
Biologically, oxygen’s most fundamental role is to be the final electron acceptor in the
mitochondrial electron transport chain (ETC). The proton gradient produced by the ETC is used
for production of adenosine triphosphate (ATP), the energy currency of nearly all eukaryotic

2

and multicellular life. Reducing equivalents, in the form of NADH and FADH2 produced by the
tricarboxylic acid (TCA) cycle and β-oxidation of fatty acids are donated to the ETC. The ETC
+

uses the transfer of electrons to drive a proton (H ) gradient across the inner mitochondrial
membrane. This gradient is utilized by ATP synthase (Complex V), producing ATP from ADP and
Pi using the energy inherent in the process of the protons leaking back into the matrix [14].
NADH donates electrons to Complex I (NADH dehydrogenase), and FADH2 donates electrons to
Complex II (succinate dehydrogenase). Both Complex I and II donate electrons to Coenzyme Q
(ubiquinone), which in turn donates electrons to Complex III (cytochrome b/c1). Cytochrome c
accepts electrons from Complex III and donates them to Complex IV (cytochrome a/a3 or
cytochrome c oxidase). Oxygen is utilized as the final electron acceptor in the ETC, as Complex
IV donates four electrons to O2 to form two molecules of water. This entire process of utilizing
O2 to accept electrons for ATP production is referred to as oxidative phosphorylation or aerobic
respiration.

3

Review of lung anatomy, physiology and development
Due to oxygen’s central role in energy production, multicellular animals (metazoans)
have evolved systems to ensure adequate transport of O2 to the cells of the body. The energy
demand of cells and body size represent key drivers for the resulting respiratory systems and
their complexity [15]. Respiratory systems are concerned primarily with gas exchange. Simple
tubular structures that diffuse oxygen directly to tissues are seen with insect tracheal systems
[16], whereas larger animals such as birds and mammals require a highly specialized centralized
gas exchange organ (lung) to transfer of O2 into a nutrient-rich fluid (blood) which is then
pumped by the heart to the tissues [17].
Mammals utilize a diaphragm, a thin muscle that horizontally separates the thoracic and
abdominal cavities, which when contracted creates negative pressure in the pleural space
(between the inner wall of the thorax (parietal pleura) and the outer wall of the lung (visceral
pleura)). This negative pressure, in addition to accessory muscles and the movements of the rib
cage, forces the expansion of lung tissue, filling the airway and lungs. Inhaled air is humidified,
heated and mixed with residual air in the lung. To maximize the efficiency of gas exchange with
2

blood, air is brought past a large surface area (approximately 75 m in adult humans) and also
within close proximity to the red blood cells (within 1 μm). Blood contains erythrocytes (red
blood cells or RBCs) which house large amounts of hemoglobin, the proteins which carry
oxygen to the tissues and return carbon dioxide (CO2) from the tissues to the lungs to be

4

exhaled as waste. Relaxation of the diaphragm creates positive pressure in the pleural space
which condenses lung tissue and causes air to be expelled from the lungs [18-20].
The airway is divided into two main compartments: the conducting regions and
respiratory (gas exchange) regions. As implied, the conducting regions are concerned with
conducting and conditioning air from the external environment to the respiratory regions. For
humans, the conducting airway includes everything from the nose to the terminal bronchiole;
more specifically, the nasal cavity, nasopharynx, glottis/larynx, trachea, bronchi, bronchioles
and terminal bronchioles. The respiratory regions are located entirely within the deep lung,
which begins at the respiratory bronchiole, and then to the alveolar ducts and alveoli.
Beginning in the trachea, several divisions (up to approximately 23) of branching occur which
increase the surface area for gas exchange. Bronchi occupy the first 3 divisions, and divisions 416 represent the journey of bronchioles to terminal bronchioles. Respiratory bronchioles
occupy divisions 17-19, alveolar ducts 20-22, and finally the alveolus itself at 22-23. The
journey from terminal bronchiole to alveolus is on average 5 mm [21].
As mice are used as the model organism in my studies in this dissertation, it is worth
noting some important differences between murine and human airways to put the results of
my experiments into context. Mice are obligate nose breathers which humans are not.
Additionally, in humans inspired air has to bend 90° to move into line with the trachea, which
rodent airways do not require. Though the nasal airway will not be discussed here, the mouse
has highly developed and complex nasal turbinates and olfactory system that differs from
humans. Regarding the gross anatomy of the lung, one major difference in gross architecture is

5

that the human lung contains 3 right lobes and 2 left lobes (with a small cardiac lingula), to
make room for the heart. Mouse lungs, in contrast, have 4 right lobes and a single large left
lobe. Mice have approximately 13-17 airway divisions before reaching alveoli (compared to
approx. 17-21 in humans). Bronchi in the lungs of humans have several generations of
cartilaginous walls, while mice have none within the lung proper. Submucosal glands are also
not present in the airways distal to the larynx in mice, whereas humans contain submucosal
glands through the 2

nd

division of pulmonary bronchi. Another major difference in mice is that

they do not have respiratory bronchioles – terminal bronchioles give way directly to alveolar
ducts and “skip” the respiratory bronchiole structure seen in humans and other mammals.
Mice also have much thinner pleura (outer lining of lung) with less connective tissue than
humans, probably due to size [22].
In regards to cells lining the airway, the trachea and proximal bronchi are initially
composed of pseudostratified columnar epithelium, primarily of ciliated cells and goblet cells,
with a few serous and neuroendocrine cells, and many basal cells. Human lungs maintain this
epithelial cell architecture through most of the bronchial tree, whereas mice switch at the first
bronchial branch to a simple columnar epithelium where Club cells are the majority cell type
with ciliated cells. Upon reaching the terminal bronchiole, human lungs transition to a
respiratory bronchiole, where some outpockets of alveoli appear, with sporadic re-introduction
of the bronchial epithelium until reaching the alveolar duct. In mice, the terminal bronchiole
leads abruptly to the alveolar duct, where the only epithelial cells that are observed are
alveolar type I (ATI) cell and the occasional alveolar type II (ATII) cell. Alveolar macrophages are

6

also seen throughout the alveolar regions. Figure 1 shows a summary of the cellular
components of both mouse and human lung epithelium [23].

7

Figure 1

Figure 1. Comparative epithelial constituents across regions of mouse and human lungs. Epithelial cells of airways in mouse lungs
(left side, says “Mouse ~5,000 branches”) and human lungs (right side, says “Human ~54,000 branches”) are compared. Major
8

Figure 1 (cont’d)
differences include the presence of basal cells and goblet cells in human airways all the way to the respiratory bronchioles, whereas
in mice they end at roughly the main stem bronchi. Serous cells in mice are only present in the proximal tracheal airway and
submucosal glands. In mice, Club cells and ciliated epithelium dominate the bronchi and bronchioles, leading the Club cell to occupy
a higher percentage of total airway surface in mice than is seen in humans. Serous cells = dark green; ciliated cells = purple; goblet
cells = pink; basal cells = blue; neuroendocrine cells = salmon; Club cells = light green; ATII cells = light orange; ATI cells = brown [23].
For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this thesis
(or dissertation).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

9

Lungs represent one of the most complex organs both because of the vast structural complexity
required for effective gas exchange, and also for the specificity their cells display to defend the
lung from injury. There are approximately 40 unique cell types in the respiratory tract, their
distribution reflecting the function of the local airway in which they reside [24]. In the
bronchial tree we find many epithelial cells, including ciliated cells, goblet (mucous) cells,
serous cells, basal cells, neuroendocrine cells, dendritic cells and Club cells. The primary
function of the ciliated epithelial cells is to operate the mucociliary escalator (MCE), a hygiene
and protective apparatus where cilia beat in unison to flow a mucous layer (contributed by the
mucous cells and containing any unwanted particles) on top of a periciliary layer (contributed
by the serous cells) back toward the oropharynx. This process is described in more detail in the
following section. Goblet cells function to provide the protective mucous which forms the
mucous layer of the MCE and also defense from chemical injury. Serous cells have a nonmucous role, which includes the production and secretion of many protective and antimicrobial
proteins such as lysozyme, IgA, peroxidase and lactoferrin [25]. Basal cells act as stem cells for
the bronchial epithelium, showing self-renewal and ability to differentiate into ciliated and nonciliated cell types [26, 27]. Pulmonary neuroendocrine cells (PNECs) are innervated basally and
contain many granules filled with neuroactive peptides. Of all lung cell types, PNECs have the
most varied putative functions including oxygen sensing & regulating pulmonary blood flow,
bronchial tone, and provide a stem cell niche for ciliated and Club cells after injury [28, 29].
Toward the more distal end of the bronchial tree we find Club cells, which are cuboidal
non-ciliated, non-mucous secretory cells of the distal bronchioles that perform many functions.
They secrete protective molecules such as Club cell secretory protein (CCSP) (member of the

10

secretoglobin family), surfactant proteins A, B and D, antimicrobial peptides, proteases and
mucins to the extracellular fluid. Club cells also function to metabolically detoxify xenobiotic
substances and can act as a stem cell population for damaged epithelial cell types such as
ciliated and mucous goblet cells [30]. Recent controversy surrounding Max Clara, major
contributor of initial characterization of the cell type that bore his name, and his use of tissue
samples from executed prisoners in Nazi Germany and outspoken involvement in the Nazi Party
have resulted in the renaming of the Clara cell to the “Club” cell [31, 32].
In the alveolar compartment we find ATI and ATII cells, and alveolar macrophages
(AMs). ATI cells are the thin, flat, gas-exchanging epithelial cells, while ATII cells are cuboidal,
producing surfactants which not only hold the alveolus open by reducing surface tension, but
can also act as an antimicrobial [33]. ATII cells also regulate fluid levels in the alveolar
compartment, and act as stem cells for the alveolus, reproducing and differentiating into ATI
cells after these cells have been lost to injury [34]. AMs play a key function of housekeeping
and innate defense, phagocytosing any used surfactant or particulates and patrolling the airway
for harmful microbes. Additionally, AMs play a role in adaptive immunity with the ability to act
as antigen-presenting cells (APCs) and by modulating many inflammatory responses [35-37].
Development of the lung in mammals is absolutely essential for transitioning to air
breathing from the previous placental source of O2. Some mammals, however, do not finish (or
even start) the process of alveolarization, the final stage of lung development, until sometime
after birth. Lung development occurs in 5 major stages: embryonic, pseudoglandular,
canalicular, saccular and alveolar, which differ in timing across species [23]. For example, the

11

embryonic stage in humans begins around weeks 3-4 post-conception, whereas in mice this
stage doesn’t start until embryonic day 9.5 (E9.5), considered mid-gestation (Figure 2). Lungs
originate as buds from the ventral primitive foregut (future esophagus), regulated primarily by
thyroid transcription factor 1. Branching morphogenesis, directed by a plethora of
transcription factors, including fibroblast growth factor, transforming growth factor β,
epidermal growth factor, bone morphogenic protein, wingless-related integration site, retinoic
acid, hedgehog and Notch signaling, follows in order to give the basic shape of the bronchial
tree [23, 38, 39]. The canalicular stage is characterized by thinning mesenchyme along the
airway branches, finishing the bronchial tree shape and differentiation of ciliated, secretory and
neuroendocrine cells. The saccular stage is defined by expansion of the distal lung into
saccules, larger versions of alveoli, as ATI and ATII cells start to differentiate [40, 41]. Finally,
alveolarization occurs until about P30 in mice, and in humans concludes anywhere between 221 years, depending on the source. Modern techniques have indicated that alveolarization
continues much later than initially thought [42]. This is not surprising when one considers body
length and thoracic size does not plateau until age 17 for girls and 20 for boys [43].

12

Figure 2

Figure 2. Comparison of human and mouse lung development. The five stages of lung
development are overlaid with their gestational ages, dominant transcriptional networks and
appearance of major cell types. Note that birth of mice (usually E19) places them in the
saccular stage, whereas humans brought to term (37 weeks) are farther along into the alveolar
stage [23].
Also note that text underneath the developmental stages (Embryonic,
Pseudoglandular, etc.) is: “Patterning and differentiation regulated via FGF, EGF, TGF-β/BMP,
WNT, RA, HH, and Notch signaling pathways.”
Reproduced with permission of the copyright owner. Further reproduction prohibited
without permission.

13

Premature birth, supplemental O2, bronchopulmonary dysplasia and asthma
Babies born prematurely have had higher rates of survival the last 30 years, thanks in no
small part to the availability of exogenous surfactants and improved supportive care. The
gestational age (GA) at which babies can expect to survive is between 23-26 weeks, which is
during the canalicular stage of lung development, before any of the terminal saccules
(precursors to alveoli) are formed. The exogenous surfactants supplement the endogenous
ones still developing in ATII cells from babies born before GA 26 weeks, and help maintain
airway patency during exhalation. Another supportive measure is supplemental O2, though the
baby’s O2-saturation of hemoglobin (O2sat) is kept below 95% to avoid retinopathy of
prematurity (ROP), a complication discovered in the 1950s [44]. The pathogenesis of ROP is
dysfunctional vascular development in the retina, thought to be from high O2 which negatively
impacts hypoxia signaling (discussed later) [45].
Bronchopulmonary dysplasia (BPD) is the primary pulmonary manifestation of survivors
of severe premature birth (GA <28 weeks) and low birth weight (<1500 grams), and is thus
associated with prematurity and supplemental O2. BPD is a disease characterized by
underdeveloped lungs, specifically fewer and larger (“simplified”) alveoli, which limits the
amount of gas exchange. Individuals with BPD have many sequelae later in life, most
prominently asthma and wheezing disorders [46]. Some wheeze can be attributed structurally
to loss of parenchymal tissue (alveolar simplification) which reduces tethering of bronchi (holds
these airways open during exhalation), leading to obstructed airflow. Whether there are other

14

molecular contributors to allergic airway disease in patients with immature lungs or BPD
remains to be demonstrated, but several large cohort studies have solidified the increased
incidence of asthma in premature infants. Infants born in the extremely premature range of 24
to 28 weeks have higher incidence of asthma both in childhood and in early adulthood (age 2535) [47-49].
While no causative relationships between BPD or asthma and hypoxia signaling have
been proven to date, hypoxia signaling is known to significantly affect lung development. It is
tempting to speculate that the halted development seen in BPD is a result of a normally hypoxic
environment of the developing lungs being interrupted by premature birth, exposing the lungs
to higher amounts of oxygen than would be expected for this developmental period [50].
Perhaps changes in oxygen exposure can affect hypoxia signaling, and thus alter lung
development in such a way that increases the likelihood of developing allergic disease and
altering lung immunity. Recent studies by our lab have supported this notion, and have shown
that loss of HIF1α in lung epithelial cells causes an exacerbated response using the ovalbumin
sensitization/challenge model of asthma in mice [51]. Further investigation of hypoxia signaling
and its relationship to premature birth, supplemental O2, BPD and asthma are needed to
determine whether these associations are pathophysiologically meaningful.

15

Immune mechanisms in the lung
2

With an epithelial surface of >70m , the airway represents the second largest surface on
the human body, requiring a proportionate defense effort. As a result, several mechanisms
have emerged that protect the airway from chemical, particulate and other environmental
stresses, such as microbial pathogens. Broadly, these mechanisms of immunity fall into two
categories: innate and adaptive. Innate mechanisms are less specific, protect from a wide
range of threats, act quickly (or continuously) and the machinery required to recognize threats
does not require further adaptation for function. These mechanisms are in sharp contrast to
adaptive immunity which, while being much more specific, requires more time to act.
Presentation and matching of pathogenic antigens as well as expansion of the effector cell
populations and their specific antibodies are required before adaptive immunity can begin.
These immune mechanisms are executed by the lung epithelium and immune cells, including
alveolar macrophages, dendritic cells, bone-marrow derived granulocytes and monocytes, as
well as B and T cells. An understanding of innate and adaptive immune mechanisms in the
airway, and their overlap, is essential to putting any toxicological investigation of the airway
into the proper context.
Innate immunity
Innate mechanisms begin with the basics of mechanical and chemical barriers. One of
the most effective examples of this in the airway is the MCE of the tracheobronchial tree, which
has three functions: clearance of particulates, chemical protection and biological protection.
The surface fluid of the MCE in the airway, aided by turbulent air flow, traps inhaled

16

particulates, which are then carried out of the lung by the action of millions of synchronously
beating cilia to the glottis where they can be coughed and swallowed. The surface fluid also
provides chemical protection by providing a physical layer to trap certain compounds and
provides antioxidant potential through peroxidases [52]. Other secretory products in the
airway lining fluid of the MCE, as well as elsewhere along the bronchial epithelium, provide
biological protection by targeting common microbial pathogens. Compounds such as IgA and
IgG, lysozyme, lactoferrin, fibronectin, cathelicidin, β-defensins and antiproteases, such as
secretory leukoproteinase inhibitor, are secreted by epithelial cells to protect the airway [53,
54]. Antimicrobial action of these compounds assists the action of macrophages and other
immune effector cells, fulfilling the biological protective function of the MCE [55, 56].
Epithelial cells provide far more protection than simply a physical and chemical barrier
of the MCE. Along with macrophages and dendritic cells, epithelial cells contain an array of
pattern recognition receptors (PRRs). PRRs recognize conserved molecular moieties of
microbes and alert immune effector cells of danger via specific downstream signaling [57]. One
classic family of PRRs are the toll-like receptors (TLRs) [58]. TLRs recognize many different
types of pathogen-associated molecules. For example, recognition of lipopolysaccharide (LPS),
a key constituent of the outer membrane of gram-negative bacterial cell walls, is mediated
through surface receptors CD14 and TLR4 [59]. Epithelial cells use these and other systems to
alert the immune system in several ways, usually through activation of the nuclear factor kappa
B (NF-κB pathway) (Figure 3) [60, 61]. Recent focus on the epithelial role in allergic airway
disease has revealed that these cells can control the nature of the ensuing immune response
directly [62, 63].

17

Figure 3

Figure 3. Overview of pulmonary epithelial cell influence in immunity. Epithelial cells function both as effectors of innate

18

Figure 3 (cont’d)

immunity, but also in modulating adaptive immune responses, recruitment of immune cells, and production of anti-inflammatory
mediators. Epithelial secretion of cytokines and chemokines influence a variety of different cell types. (mDC and pDC indicate
myeloid and plasmocytoid dendritic cells, respectively) [60]. Note that text underneath “Innate Immune Responses” (top left) is:
“viruses, bacteria, fungi”. Note that text underneath the “Host Defense Molecules” (left bottom) is: “Anti-microbial, lysozyme,
defensins, collectins, pentraxins” and “Anti-viral, IFN-β, IFN-γs” and “Pro-inflammatory molecules, TNF, IL-1, IL-6, GM-CSF”. Text
underneath “Anti-Inflammatory Responses”(right side) is: “anti-inflammatory cytokines, receptor antagonists, protease inhibitors,
arachidonic acid metabolites”. Note that text underneath “Adaptive Immune Responses” (middle) is: “GM-CSF (+ IL-4), monocyte,
mDC; IL-15, monocyte, pDC; TSLP, IFN-β, IFNγs, DC; IL-33, IFN-β, IFNγs, B7-homologs, T cell; BAFF, APRIL, IL-6, IL-10, B cell”. Note
that text underneath “Cell Recruitment”(middle right) is: “CCL20, DC; CXCL9, CXCL10, CXCL11, Th1 cell; CCL1, CCL17, CCL22, Th2
cell”.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

19

Macrophages are derived from hematopoietic tissues such as bone marrow and
circulate as monocytes in the peripheral blood until they are called upon by tissues to fill roles
as resident macrophages or to replace resident macrophages that have died. Macrophages
patrol all of the portals of entry to the body, and while they can participate in direct killing of
microbes, their primary function in the body is cleanup of cellular debris via phagocytosis and
to signal infections to the rest of the immune system. This signaling is usually through antigen
presentation and secretion of cytokines and chemokines. Armed with an array of PRRs,
macrophages can become activated upon exposure to microbes (usually through interferon
gamma (IFN-γ) and tumor necrosis factor alpha (TNFα)) and can amplify signals to the rest of
the immune system [64, 65]. Recently it has been appreciated that activation of macrophages
can occur in several different ways, based on the stimulus, to affect one type of response over
others. Initially these were referred to as either “classically activated, M1” macrophages or
“alternatively activated, M2” macrophages. M1 macrophages are induced through IFN-γ and
TNFα/LPS and stimulate the production of inducible nitric oxide synthase (iNOS) and secretion
of interleukin (IL)-1β, IL-12 and TNFα with microbicidal activity. In contrast, “alternatively
activated, M2” macrophages are activated by IL-4, which stimulates the production of arginase.
Further research has clarified that macrophages can exhibit different phenotypical markers
simultaneously and also retain a large amount of plasticity, which has led to the suggestion that
the ‘M1 vs. M2 axis’ be considered more like a continuum [66-71]. Regardless of the name of
the macrophage, it is clear that they play a vital role in almost every level of immunity.
Neutrophils (polymorphonuclear cells or PMNs) are also derived from hematopoietic
stem cells, and are the primary cell type responsible for phagocytosis and killing of microbes,

20

being far more efficient at phagocytosis and bacterial killing than macrophages [72]. PMNs play
a central role in innate host defense, and circulate the peripheral blood until called upon by
chemokines and other signals to migrate into affected tissues and become activated [73]. One
key feature of PMNs are their azurophilic (colored strongly with azure stain) granules which are
filled with several antimicrobial peptides and oxidases. Upon fusion, either with phagosomes
or the plasma membrane (a process referred to as “degranulation”), these granules increase
the concentration of antimicrobial peptides and produce large amounts of reactive oxygen
species (ROS) in what is called a “respiratory burst”, unleashing superoxide, hydroxyl radical,
hydrogen peroxide, hypochlorous acid and chloramines [54, 74, 75]. In humans, the lung is
reported to house roughly 40% of circulating PMNs at any time, allowing them close access to
affected airway tissues [76]. PMNs are activated by several signals, including Toll-like receptors.
Administration of LPS to the lung of animals can rapidly recruit PMNs so much that in 3-4 hours
post-administration, 60-80% of cells in bronchoalveolar lavage (BAL) are PMNs [77]. Defects in
several aspects of PMN function leave humans susceptible to many infections, further verifying
the central role of PMNs in host defense [78, 79].
Eosinophils, like neutrophils, are derived from granulocyte precursors in the bone
marrow and patrol in peripheral blood until called to specific tissues. Unlike PMNs, eosinophils
have classically been attributed with defense against parasitic infections (e.g. helminthes or
“worms”) and associated with allergic responses and asthma. Several recent reviews have
challenged this classical view of eosinophils as damaging effector cells in these limited contexts
and have outlined more broad roles for these cells as immune modulators and orchestrators of
tissue repair and regeneration [80-82]. Eosinophils’ namesake derives from their cationic

21

granules which hold Eosin stain. These granules contain many cytotoxic cationic proteins
including Major Basic Protein (MBP) which makes up roughly 50% of the protein contained in
these granules [83-85]. Other proteins in these granules include eosinophilic cationic protein,
eosinophil peroxidase, and eosinophil-derived neurotoxin, which, along with MBP have
cytotoxic properties. Granules may also contain cytokines and other immunomodulators such
as IL-2, -3, -4, -5, -6, -8, -10, -12, -13, -16, -18, TGFα/β, GM-CSF, TNFα and IFNγ, among others
[86]. Interestingly, degranulation of eosinophils has been shown to be stimulus-specific, as
opposed to an “en masse” release of all granule contents, as evidenced by Lacy et al. [87].
While the exact mechanistic role of these cells is still under intense investigation, it is clear that
they can elicit diverse effector functions.
Another key cell type within and interspersed among epithelial cells is the dendritic cell
(DC). DCs share many properties of macrophages in that they are derived from the myeloid
lineage. Unlike macrophages, however, they display different resident subpopulations prior to
immune activation (for review, see [88, 89]). In general, DCs are positioned in proximity to all
epithelial surfaces, and they are the primary APCs of the body. In the lung, the CD103+
conventional DC (cDC) use filopodia that migrate in between epithelial cells to reach into the
airway lumen and take occasional samples of surrounding material (i.e. “periscoping”) to
monitor the environment. DCs are armed with many PRRs with which to monitor possible
presence of microbes. If microbes are sensed, DCs can become activated, which enhances
secretion of cytokines and chemokines that promote DC migration to draining lymph nodes for
activation of naïve T cells [90]. Like the epithelium, DCs are considered by many to be a part of
both innate and adaptive immunity due to their “professional antigen presentation” function.

22

Antigen presentation to naïve T helper (TH) cells is a vital starting point in transmitting adaptive
immune responses. To initiate an adaptive immune response to TH cells, DCs must first migrate
to draining lymph nodes where they express major histocompatibility complex II (MHCII), which
presents processed antigens to the T-cell receptor (TCR) and CD4 co-receptor. Presence of costimulatory molecules, classically CD28 on T cells and B7-1/B7-2 (CD80/CD86) on APCs are
necessary to complete activation of TH cells [65].

Adaptive immunity
While the innate immune mechanisms of the airway are essential and widespread, they
do not recognize all threats, and they are poor at reacting to intracellular pathogens or
extracellular pathogens that have developed evasion strategies. Thus, a more specific system
of protection is needed to complement innate immunity. The adaptive immune system fills this
functional gap, and is mediated by APCs, T and B cells, with signaling to other cellular effectors
that include macrophages, neutrophils, eosinophils, basophils, and mast cells to orchestrate the
effective immune responses.
Role of T cells in lung defense.
For the lung, T cells are positioned in the draining/hilar lymph nodes, where activation
of T cells by activated dendritic cells occurs [91]. T cells are responsible for most cell-mediated
immunity (direct action of cells), and mature T cells are classified by surface receptors CD4+ T
helper (TH) cells or CD8+ cytotoxic T lymphocytes (TC). During an immune response, naïve T
cells (T0) are exposed to peptides on MHCII from APCs which must match the T0 TCR in addition

23

to co-stimulatory binding of B7.1/B7.2 (CD80/CD86, on APCs) with CD28 (on T0 cells). These
two signals cause activation and survival of T cells, usually by IL-2 (T cell growth factor).
Without co-stimulatory input, T cells undergo inactivation or clonal deletion (immune
“tolerance”). In contrast, co-activation signals cause clonal expansion and differentiation of T
cells (T cell “priming”), which causes expression of CD40 ligand (CD40L), a co-stimulator
necessary for binding to B cells [92].

The process of naïve T cell (T0) priming and activation into functional TH cells is heavily
dependent upon the cytokine milieu of the APC-T0 interaction. Several pathways of
differentiation can be followed (Figure 4) [93]. The resulting TH cells can generally be defined
by the cytokines they release, which further direct immune responses. The most commonly
studied TH cell populations in the lung to date are the TH1 and TH2 cells.

T-helper type 1 (TH1) response
TH1 responses are often seen with infections of intracellular pathogens such as viruses
or Listeria. TH1 cell priming requires the presence of IFNγ and IL-12. Production of these
cytokines is generally performed by APCs, caused by activation of a combination of TLR3, 4, 7-9
or 11 or any one of these in the presence of a type I interferon (IFNα or IFNβ) [94, 95]. IFNγ and
IL-12 drive transcription factors Tbet and Stat4 which further upregulate IFNγ (positive
feedback loop), IL-2 and lymphotoxin α (LTα). To combat intracellular pathogens, IFNγ
stimulates macrophages to increase their microbicidal functions and stimulates many cells to
24

express MHCI and MHCII to aid in antigen presentation. Importantly, the presence of IFNγ
inhibits the differentiation of both TH2 and TH17 cells. IL-2 promotes T cell and natural killer
(NK) cell activation and proliferation. Lymphotoxin (TNFβ) is similar to TNFα, and recruits and
activates PMNs [93, 96]. A properly functioning TH1 response is absolutely required for proper
clearing of intracellular pathogens.
Figure 4

Extracellular Bacteria
Fungi
Autoimmunity
IL-21
IL-17a
IL-17f
IL-22

TGFβ
IL-35
IL-10

Intracellular
Pathogens
Autoimmunity

RORγt
Stat3

T-bet
Stat4

TH17

TH 1

TREG

Activated

T0

TH 2

FoxP3
Stat5

Gata3
Stat5

Immune Tolerance
Downregulating Immune
Responses

Extracellular
Parasites
Allergy/Asthma

IFNγ
IL-2

IL-4
IL-5
IL-6
IL-10
IL-13

Figure 4. T-Helper (TH) cell Subtypes. Naïve T cells (T0, center) mature into effector TH cells
after a positive activation with an APC (not shown) based on presence of specific cytokines
(shown above large arrows). Transcription factors responsible for the downstream maturation
of each effector cell are labeled inside each cell. The small set of arrows indicates the secreted
cytokines which further direct the various kinds of immune responses, and are most commonly
used when attempting to identify TH subtype. Boxes describe general immune functions of
each subset. Modified from [93].

25

T-helper type 2 (TH2) response
TH2 responses are seen in extracellular parasitic infections (e.g. helminths), and are also
the mechanisms behind allergic diseases such as asthma. TH2 cell priming requires the
presence of IL-4 and IL-2. IL-4 activates STAT6, which in turn upregulates GATA-binding protein
3 (GATA3), the master regulator of the TH2 response. GATA3 acts as a master regulator by
binding to the gene loci of several TH2 cytokines to activate them while simultaneously
represses TH1 responses by repressing IL-12 receptor β2 and STAT4. The function of IL-2 in the
TH2 response is to activate STAT5, which in turn up-regulates IL-4, further driving the TH2
population to differentiate. In recent years it has been determined that IL-2 can be bypassed
by either IL-7 or thymic stromal lymphopoietin (TSLP) to activate STAT5. GATA3 and STAT5
cooperate to express TH2 cytokines, including IL-4, -5, -9, -13 and -25. IL-4 stimulates B cell
class-switching from IgM/IgG to IgE, which in turn activate basophils and mast cells to
degranulate upon antigen recognition to cause vascular permeability, smooth muscle
constriction and further recruitment of inflammatory cells. IL-5 acts to recruit and promote
survival of eosniophils, whereas IL-9 has a similar function for mast cells. IL-13 works in concert
with IL-4 on epithelial cells to induce mucous production and goblet cell metaplasia, and in
smooth muscle cells to promote airway hyper-responsiveness. IL-25 promotes TH2 responses
by initiation and activation of IL-4, IL-5 and IL-13, though several experiments in RAG knockout
(functional loss of B and T cells) mice have demonstrated that it can induce IL-5 and IL-13

26

expression from non-B, non-T cells, which have also been referred to as innate lymphoid cells
or nuocytes [97]. IL-33 promotes alternatively activated macrophages and secretion of TSLP
from epithelial cells in addition to production of other TH2 cytokines, but like IL-25, more
research is needed to elucidate its role more fully [98].
Other T-cell responses
While other T cell responses such as TREG and TH17 are important, induction of either
TH1 or TH2 responses greatly inhibits the formation of the other populations by the dual action
of their cytokines IFNγ and IL-4, respectively. In the absence of IFNγ or IL-4, high levels of TGFβ
activate the TREG response, which serves to dampen the overall action of activate T cells, which
induces tolerance and helps to resolve inflammation. Lower levels of TGFβ can activate TH17,
which has been shown to be crucial for defense against some extracellular bacteria and fungi
[93].
Mechanisms of lung neutrophilia and eosinophilia
Several factors are responsible for recruitment of different cell populations to the lung
(Table 1). The most common cell types seen in airway inflammation are macrophages and
PMNs, because of the general ability of these cells to kill pathogens and remove cellular debris.
However, several molecular factors are responsible for recruiting specific cell types to the lung.
The major groups of molecules that perform this function are the chemokines, several of which
have been characterized to belong to the CC or CXC-family of chemokines (L designates ligand,
R designates receptor). The CC and CXC receptors show ligand promiscuity, so there is often
27

Table 1. Partial summary of cytokine/chemokine-mediated recruitment of cellular effectors in
the lung.
Cell Type
Recruited/Activated
ALL – mixed
leukocyte
recruitment
Macrophages

Chemokine

Cellular Source(s)

References

MIP-1α (CCL3)

Almost all induced immune
and hematopoietic cells

[102]

MCP-1 (CCL2)
TNFα

Neutrophils
(PMNs)

Eosinophils,
Basophils
Eosinophils
Mast Cells
Eosinophils, T cells

DCs (pro-TH2)

T H1
T H2

IL-1
IL-8
MIP-2α
KC (CXCL1, GROα)
Eotaxin (CCL11),
Eotaxin2 (CCL24)
Eotaxin3 (CCL26)
IL-5
IL-9
RANTES (CCL5)
CCL20/β-defensin
TSLP
IL-25
IL-33
GM-CSF
IL-1β
Osteopontin
MMP9
IP10 (CXCL10)
TARC (CCL17)
PARC (CCL18)
MDC (CCL22)

ATII cells
Macrophages (endothelial
adhesion markers)
Macrophages (stim. PMN
production)
Epithelial Cells, Macrophages
Macrophages
Macs, Epithelial Cells, PMNs

[103]

T cells, Epithelial cells

[105]

TH2 cells

[98]

Macrophages

[106]
[107]
[98, 108]
[98, 109]
[98, 110]
[111]
[108]
[112]
[113]
[114]

Epithelial Cells

DCs, Epithelial cells, Macs
Epithelial Cells

28

[65]
[65]
[101]
[103]
[104]

[115]

redundancy of function. The CC family has several members that direct mixed leukocyte
recruitment, including CCL2 (MCP-1), 3 (MIP-1α), 5 (RANTES), 7 (MCP-3), 8 (MCP-2), 12, 13
(MCP-4) and 15 (MIP-1δ). The CXC family usually directs neutrophils and effector T cells,
though CXCL12 (SDF-1α/β) has mixed leukocyte recruitment [65].
PMNs are more prevalent in the pulmonary capillaries than systemic blood, and while
the reason is not fully understood, it is speculated to be from the nature of their cell shape in
the narrow capillaries [99]. During an inflammatory stimulus from TNFα, IL-1β or lymphotoxin
(TNFβ), activated neutrophils express a high-affinity integrin receptor and CD11/CD18, which
causes them to bind more tightly to the endothelial cells [100].
In turn, the endothelial cells express adhesion molecules such as P-, E-, and L-selectins,
vascular cell adhesion molecule 1 (VCAM-1), and intercellular adhesion molecule 1 (ICAM-1).
The CXC receptors CXCR1 and CXCR2 on the surface of PMNs bind to GROα (CXCL1), GROβ
(CXCL2), GROγ (CXCL3), IL-8 (CXCL8) to direct chemotaxis to the inflammation site [73, 101].
Eosinophilia is also seen in the lungs, usually in the context of allergic airway disease or
parasitic worm infection (also known as helminths, a classic example in humans being several
Schistosoma species). Recruitment of eosinophils is largely mediated by TH2 cells through IL-5,
though epithelial cells can secrete chemokines that target eosinophils directly (CCL11 (Eotaxin1), CCL24 (Eotaxin-2), CCL26 (Eotaxin-3), CCL5 (RANTES), CCL7 (MCP-3), CCL13 (MCP-4) and
CCL3 (MIP1α)) or indirectly by recruiting TH2 cells (CCL22 (MDC), CCL17 (TARC) and CCL18
(PARC)) [115, 116].

29

Epithelial-immune crosstalk in inflammatory responses
In the wake of all the recent advances made by immunologists in understanding
adaptive immune mechanisms, the epithelium is finally getting some deserved attention as an
active participant in the immune process. Epithelial and dendritic cells are now being viewed as
key functional units of the innate/adaptive immunity transition, especially in the lungs [117].
Most studies to date have focused on communication between epithelial cells and dendritic
cells.
Generally the direction of cytokines seems to be heavily one-way, that is from epithelial
cells to DCs. Some of the documented cytokines (summarized in Table 1) released by epithelial
cells to recruit and polarize DCs to induce a TH2 response are CCL20, β-defensin, TSLP, IL-25, IL33, GM-CSF, MMP9, IL-1β and osteopontin [118, 119]. The TH1/TH17 response is less favored
directly by epithelial cells, and seems to require both epithelial stimulation of the innate
immune response, as well as, direct contact of DCs with the pathogen in question [120]. If the
stimulation by local macrophages secreting IFNγ, (sometimes IL-1β or TNFα) combined with the
epithelial/DC innate response is enough to activate DCs, TH1 responses can be induced. If IL17A is secreted during this process, the response is skewed toward TH17. Other key players in
the epithelial induction of TH1 are the “inflammasome” proteins, Caspase-1 and NOD-like
receptors (NLRs); for a review see Mariathasan and Monack, 2007 [121].

30

Alveolar epithelial type II (ATII) cells and immunity
In addition to their role in the production of surfactant and maintenance of the alveolar
compartment, ATII cells have been recently shown to express MHCII and participate in the
creation of FoxP3+CD4+ T cells (TREGs) [122, 123]. The active role of epithelium in directing
immunity has been increasingly important because the lung is no longer assumed to be a sterile
environment. While lower in overall organism load, the lung microbiome is similar to upper
respiratory tract microbiome in bacterial constituents [124, 125]. Other studies by Chuquimia
et al. show the likely involvement of ATII cells in recognizing and amplifying response to
infection [103]. One important consequence of the inflammatory response in the lung is a
change in O2 availability and its role in T cell and macrophage activation. Investigations into the
contribution of ATII cells in immunity will be crucial in further developing this field, especially in
regards to how the immune dynamics change based on presence of different cell types and the
microenvironment (i.e. O2 availability) of the alveolar space.

The Hygiene Hypothesis
The rise of allergic diseases in industrialized nations over many decades has prompted
researchers to investigate the cause. In 1989, Strachan saw an inverse relationship between
the number of siblings and the incidence of hay fever and atopy. He speculated that exposure
to infections in childhood, either through decreased hygiene (caused by increases in family size
or other causes), was required for proper development of the immune system and that the
cleanliness of our modern lives was contributing to the rise of allergic disease [126]. This was
dubbed the “hygiene hypothesis”. Years of study have helped develop this idea into a more

31

nuanced understanding that environmental exposures in early postnatal life having significant
impact on later risk of allergic disease. While strides have been made in elucidating innate
immune mechanisms, more understanding of the microbiome and epithelial signaling in
regulating immune development is needed to clarify this hypothesis into a working molecular
model of allergic disease risk [127].

32

Normoxia, hypoxia and oxygen sensing mechanisms
Air is composed of approximately 21% oxygen (O2), 78% nitrogen (N2), 0.9% argon (Ar)
and 0.04% carbon dioxide (CO2) with the remainder as very small amounts of several other
gases. At sea level the pressure of air is 760 mmHg. When air is inspired into the respiratory
tract, it is saturated by water vapor at a pressure of 47 mmHg, and thus the partial pressure of
O2 (pO2) in inspired air is:

pO2 = 0.21 x (760 mmHg – 47 mmHg) = 150 mmHg

As the inspired air is conducted deeper into the airway, it mixes with existing gases, and
the alveolar pO2 is approximately 105 mmHg. After diffusion across ATI cells and pulmonary
endothelial capillaries, O2 is bound to hemoglobin of the RBCs and the pO2 of pulmonary
venule blood (and arterial blood in the systemic circulation) is 100 mmHg. As blood carries O2
to the peripheral tissues, the localized O2 concentration decreases dramatically, and along with
pH decreasing and allosteric regulation of hemoglobin drive the dissociation of O2 and the
association of CO2 to hemoglobin (known as the Haldane effect), and O2 diffuses through the
local cells [128]. As venous blood returns to the right heart and pulmonary arteries, the pO 2 is
40 mmHg [20].

33

The exposure of each cell to oxygen depends greatly on many factors, including the
distance of the cell to the nearest capillary, the specific tissue location of the cell (and
extracellular milieu) and the metabolic demands of the tissue (rate of O2 consumption) [129].
Levels of O2 have been measured at 5 mmHg in the retina [130], 20-30 mmHg in skeletal
muscle [131] and 45-65 mmHg in the brain [132]. Normal cells all develop and acclimate to a
unique “normal” pO2 (normoxia) through mechanisms that are not yet fully understood. When
the supply of O2 drops below this normal level, the cell is considered hypoxic [133, 134]. The
lung epithelium, despite having proximity to the organism’s O2 source, still can experience
hypoxia in various disease states [135].
Hypoxia is the main cause of cell death in cardiovascular diseases, such as myocardial
infarction and stroke, and is also seen at the sites of inflammation, cancer, and in varying
degrees from normal physiological variation (physical exercise, altitude variation). Maintaining
a steady supply of O2 to cells is required for homeostasis, and organisms have evolved
mechanisms to sense and physiologically regulate O2 [136]. These include neuronal sensing of
O2 in the glomus cells of the carotid body [137], the neuroepithelial cells in bronchial airways
[28], and smooth muscle cells in systemic and pulmonary vasculature [138]. Thought to
+

operate through neuronal recognition of O2 through K -channels sensitive to hydrogen
peroxide (H2O2) or ROS [139], these systems act in concert to optimize respiratory rate,

34

shunting of blood to well-ventilated regions, cardiac output and blood pressure to maintain O2
homeostasis [136, 140]. While these systemic feedback systems are essential for homeostasis,
they act in an acute setting and do not in and of themselves impart increased cellular tolerance
directly. At the level of the cell, the ability to adapt to hypoxia is derived primarily by the
hypoxia-inducible factors (HIFs).

35

Hypoxia Signaling
Hypoxia inducible factors
Hypoxia inducible factors (HIFs) are heterodimeric transcription factors composed of an
α and β subunit. HIF1α was first characterized by Wang and Semenza in 1993 as a protein
binding DNA upstream of the erythropoietin (Epo) gene in hypoxic Hep3b cells [141]. The
active transcription factor was found to be a dimer and the term HIF-1 beta (HIF1β) was given
to the new half of this protein. Subsequently, it was discovered HIF1β was identical to the aryl
hydrocarbon receptor nuclear translocator (ARNT), a member of the Per-Arnt-Sim (PAS)
superfamily of environmental sensors [142]. In 1997, hypoxia inducible factor-2 alpha (HIF2α)
was discovered and originally named endothelial PAS-domain protein 1 (Epas1) [143]. In 1998
the first description of HIF3α was reported [144]. Both HIF2α and HIF3α show sequence and
functional similarities to HIF1α.
In humans, there are three known HIFα subunits (HIF1α, HIF2α and HIF3α) and two
known HIFβ subunits (Arnt and Arnt2). The alpha subunits reside in the cytoplasm, while the
beta subunits remain the nucleus. All contain a basic helix-loop-helix (bHLH) domain which is
primarily responsible for DNA binding and dimerization (Figure 5). The characteristic PAS
domain consists of an A and B domain, which assist in dimerization selectivity and binding of
chaperones [145, 146]. The oxygen-dependent degradation domain (ODD) is a feature of the
HIFα subunits and contains the structures necessary for O2 sensitivity. Arnt and Arnt2 have a
single transactivation domain (TAD), whereas all HIFα subunits have an N-terminal activation

36

Figure 5

ARNT

Nuclear

ARNT2
HIF1a

Cytoplasmic

HIF2a
HIF3a
bHLH

ODD NTAD CTAD

PAS

Figure 5. Cartoon of HIFα and HIFβ subunit domains and functions. The hypoxia inducible
factors consist of alpha subunits that reside in cytoplasm and beta subunits that reside in the
nuclear compartment. Functional HIFs contain one alpha and one beta subunit. All contain an
N-terminal basic Helix-Loop-Helix (bHLH) domain (horizontal lines) for DNA binding and
dimerization as well as the PAS A and PAS B (PAS) domains for dimerization and binding other
cofactors. Only the alpha subunits contain the oxygen-dependent degradation domain (ODD,
black region) which contains conserved proline residues targeted for O 2-dependent
hydroxylation and confers oxygen sensitivity to HIFα. All subunits contain transactivation
domains (diagonal lines). The ODD overlaps with the N-terminal activation domain (NTAD) of
alpha subunits, but not the C-terminal activation domain (CTAD) of HIF1α and HIF2α. Arnt and
Arnt2 contain a single transactivation domain (TAD).

domain (NTAD, that overlaps with the ODD) and HIF1α and HIF2α have a second C-terminal
activation domain (CTAD) [144].

HIFs are primarily regulated posttranslationally by a family of O2-, 2-oxoglutarate- (22+

OG), ascorbate- and Fe -dependent enzymes called prolyl hydroxylase-domain containing
enzymes (PHDs). PHD1-3 have unique activities and tissue/cellular distribution [147, 148].
Under normoxic conditions, key proline residues (P402 and P564 in human HIF1α [149]) in the

37

ODD are hydroxylated [150, 151]. Upon hydroxylation, the E3 ubiquitin ligase von-Hippel
Lindau tumor suppressor protein (pVHL) targets HIFα for ubiquitination and subsequent
degradation via the 26S proteasome [152-154] (Figure 6). Thus, under normoxic conditions,
HIFα subunits are unstable and degraded. In contrast, under hypoxic conditions the PHD
enzymes become limited by lack of O2, and HIFα subunits stabilize, translocate to the nucleus
and meet their heterodimeric partner Arnt [155, 156]. There has been much debate in the field
about whether ROS are responsible for HIF activation or if it is due to loss of substrate. Recent
experiments have begun to favor loss of substrate, although ROS-mediated inhibition of PHDs
might play a role in certain cell types or conditions. The consensus opinion seems to be that
the mitochondrial involvement of HIF signaling has more to do with their role as O2 sinks than
as producers of ROS, a debate which has been recently summarized expertly by Dr. Thilo Hagen
[157].
As a functional dimer, HIFs seek out hypoxia response elements (HREs) in the genome,
containing the core sequence of A/GCGTG. With the help of other co-activators, the HIFs drive
expression of hypoxia-regulated genes. Well characterized HIF target genes include vascular
endothelial growth factor (Vegf) which initiates angiogenesis, erythropoietin (Epo) which
increases RBC production to increase O2 carrying capacity, and many glycolytic enzymes which
increase anaerobic metabolism to assist with cell adaptation to hypoxia [158, 159]. In addition
to these adaptive enzymes, HIFs also upregulate pro-death genes such as Bnip3, Nix and Noxa
which has led researchers to speculate that HIFs direct controlled cell death if adaptation is
impossible, though how this distinction is made in the cell is still being investigated [160-162].

38

Figure 6

A

B
Cytosol

HIFα

CTAD

ODD

P

Normoxia

P

P

αKG,
Fe2+

26S Proteasome
Degradation

CTAD

ODD

O2

PHD

Cytosol

HIFα

HIFα
stabilization

ascorbate

P

N

Hypoxia

ODD

VHL
Ubiquitination

P-OH

P-OH

Nucleus
ARNT

Nucleus
HIFα

HRE

X

ARNT

Transcription

Regulation
of hypoxiadriven genes

HREs

HREs

Figure 6. Review of HIF signaling. The alpha subunit resides in cytoplasm and beta subunit in the nuclear compartment. Under
39

Figure 6 (cont’d)
normoxic conditions (A), conserved proline residues in the ODD and an asparagine residue in the CTAD are targeted for O 2dependent hydroxylation by PHD. ODD hydroxylation targets HIFα for ubiquitination by pVHL and subsequent degradation in the
26S proteasome. Under hypoxic conditions (B) or in the presence of cobalt HIFα escapes degradation and translocates to the
nucleus where it meets HIFβ (Arnt) and locates HREs in DNA, regulating gene transcription.

40

There has been much interest in the different HIFα isoforms and their potentially unique
roles. That first indication that HIF1α and HIF2α are not playing identical roles was illustrated in
homozygous null mouse models of HIF1α and HIF2α. Both show embryonic lethality by
approximately E10.5 [163-165] but have different phenotypic malformations. Mice lacking
HIF1α, the most ubiquitously expressed HIF, display disorganized yolk sac/neural fold/cephalad
vascularization, lack of neural tube closure and an array of cardiovascular defects [166-168].
HIF-2α knockout mice displayed a wider variety of phenotypes, thought to be strain-dependent,
including bradycardia from reduced catecholamine synthesis, yolk sac vascular defects similar
to HIF1α knockouts, impaired lung maturation, and multiple organ malformations [169-172].
Further studies into gene target specificity between HIF1 and HIF2 have revealed that
the NTAD confers some of this specificity, whereas the CTAD targets common genes between
the two [173]. Studies to date seem to suggest that HIF1α targets the glycolytic pathway [174,
175], where HIF2α targets genes involved in cell cycle regulation, tumor growth factors such as
cMyc [176] and pluripotency genes such as Oct4 [177]. In whole genome arrays, HIF1α and
HIF2α targets showed some overlap (90 shared), though HIF2α-specific targets were far fewer
(40 genes compared to 304); additionally, when Z-scores for HIF1α capture were relaxed, HIF1α
targets almost completely overlapped with HIF2α (only 4 unique HIF2α genes) [178]. These
studies all suggest that while there are definitely overlapping gene targets, HIF2α seems to be
more selective in its unique gene targets, which is likely reflected in tissue distribution and
function of these two HIFs. The role of HIF3α is less studied, although it has been shown that
splice variants of HIF3α are capable of inhibiting hypoxia-induced gene expression [179-185].

41

Interestingly, HIF3α is upregulated under hypoxia in a HIF-dependent fashion, suggesting HIF3α
as a potential feedback inhibition loop for the HIF system [186, 187].
Tissue expression of HIF isoforms seems to be in agreement with the overall ubiquitous
nature of HIF1α and more specific profile of HIF2α. In tissues under hypoxic conditions, HIF1α
was inducible in nearly every tissue, and in some cases (muscle) was stable at normoxic
conditions [188]. HIF2α protein is far less detectable under normoxic conditions, but was
inducible in brain, heart, lung, kidney, liver, pancreas, and intestine of mice, and also in several
cell lines [189]. Interestingly, RNA analysis showed that lung contained the highest level of
HIF2α RNA [190].
Both HIF1α and HIF2α protein are expressed in the developing lung, although their
distinct spatial and temporal expression patterns suggest differential roles in lung maturation.
In first trimester human lungs, HIF1α is found prominently in branching epithelium, while HIF2α
is found both in epithelial cells as well as mesenchymal tissues [191, 192]. In rodents, HIF2α
expression increases dramatically during lung maturation to become more highly expressed in
adult lung than any other organ. In contrast, HIF1α does not show this pattern of induction,
though is more ubiquitously expressed [190, 192-195]. In our hands, the adult lungs of mice
show HIF1α being more detectable at normoxia in the bronchial epithelial cells and alveolar
macrophages than HIF2α . HIF1α is also detectable in the arterial smooth muscle of the
bronchial arteries adjacent to the axial airway (unpublished results).

42

HIFs in lung inflammation
Sites of inflammation tend to be hypoxic for many reasons. Edema, thrombosis and
tissue injury increase the distance of parenchymal tissues to functional capillaries and disrupt
the normal delivery of O2 and nutrients to tissues. Infiltration of immune and inflammatory
cells, with their high metabolic demand, also contributes to hypoxia [196]. HIFs have many
functional roles in sites of inflammation, primarily for metabolic adaptation of all cells, and
various cell-specific roles [197, 198]. For example, HIF1α is essential for myeloid cell function
[199]. Myeloid lineage-specific knockout models of HIF1α prevents ovalbumin-induced allergic
responses, likely from the decreased ability for these cells to become activated in the hypoxic
microenvironment [200]. In phagocytes, HIF1α plays an active role in bactericidal capacity
[201] and nitric oxide (NO) homeostasis [202]. Additionally, HIF1α can induce TLR2 and TLR6,
which form a necessary protective role in allergic and fungal infections in the lung via IL-23 and
IL-17 and TH17 cells [203, 204].

HIF1α regulates CD4+CD25+ TREGs, which play important roles in resolution of acute
lung injury (ALI) [205, 206]. HIF1α is also essential for T cell survival [207]. It should be noted,
however, that excessive thymocyte-specific HIF signaling induced T cell death [208]. Other
studies have observed that hypoxia has a general negative regulatory role when it comes to T
cells, and this is speculated to provide tissue-protection by preventing excessive T cell
activation [209, 210].

43

Recent studies have attempted to target HIFs pharmacologically in the lung. In a study
by Ahmad et al. (2012), mice in an ovalbumin model of asthma were treated with ethyl 3,4dihydroxybenzoic acid (EDHB), an iron chelator and inhibitor of PHDs [211]. A low-dose was
somewhat protective, but higher doses exacerbated allergic inflammation. These results led
the authors to suggest that HIFs are mediators of allergy-induced inflammation. Another more
detailed ovalbumin study by Huerta-Yepez et al. utilized an inducible system to knockout Arnt
and showed that ARNT deficiency led to moderate protection following OVA sensitization and
challenge. In addition, exposure of mice to EDHB (i.e. HIF activation) exacerbated the response
[212]. Finally these authors also showed lavage fluid from asthmatic patients to have increased
HIF1α and VEGF. Although the data seems promising for causation, the authors did not have a
convincing Arnt deletion, nor did they control for cell type, leaving the data in question because
most of the results could be explained from HIF-signaling in myeloid and/or macrophage
populations. If the mechanism cannot be isolated or better understood (e.g. why one dose is
protective while another very harmful), the clinical utility of HIF-targeting pharmacologically is
limited.
The studies for HIF-signaling specifically in the lung epithelium are usually limited to cell
lines and in vitro models, most of which confirm that HIFs aid in hypoxic adaptation with some
role for adhesion molecule expression (for review, see [213]). To date, ours is the only lab that
has published the use of a lung-epithelial specific deletion of HIF1α. Using this mouse model,
we were able to show that loss of epithelial HIF1α led to a switch from neutrophilic to
eosinophilic inflammation following aspiration exposure to cobalt [214, 215]. When applied to
the ovalbumin sensitization/challenge exposure, HIF1α-deficient mice had an exacerbated

44

eosinophilic inflammation as well as increased total lung resistance. Both of these studies
suggest that HIF1α in the postnatal ATII and Club cells is important in airway immunity and
immune regulation [51].
Other hypoxia-responsive signaling pathways in lung inflammation
Although HIFs play a central role in hypoxia adaptation, several other transcription
factors also have responsiveness to hypoxia (Table 2) (reviewed in [216-218]). Two important
hypoxia-responsive signaling pathways in lung inflammation are nuclear factor kappa B (NF-κB)
and purine signaling through adenosine receptors, specifically the A2B receptor (Adora2b).

45

Table 2. Non-HIF hypoxia-responsive transcription factors (modified from [216]).
Transcription
Factor

NFκB

AP-1
(Jun, Fos, ATF
dimers)
p53

cMyc
Notch
CREBP (cyclic
AMP response
element binding
protein)
C/EBPβ
(CCAAT/enhanc
er-binding
protein, NF-IL6)
SP-1, SP-3

Egr-1

Molecular Activation

Downstream Response

References

↑ COX-2, IL-6, TNFα, MIP-2

[223]

IL-8 upregulation

[216, 224228]

Cell cycle arrest, apoptosis

[229]

Cell dependent
+/- transformation potential

[217]

↓ Notch, ↑HIF expression

[230]

Depending on stimulus duration:
Pro-inflammatory/acute hypoxia
Anti-inflammatory/chronic hypoxia

[231-233]

↑ IL-6, regulation

[234]

↑ carbonic anhydrase IX, Epo
(HIF/TGFβ-dependent)

[235, 236]

Tissue factor expression,
thrombosis, vascular remodeling

[237, 238]

↓ hydroxylation of IkB by
PHDs and FIH;
phosphorylation of IkB by
Ras/Raf
others unclear
Evidence for Jnk
2+

↑ Ca
cooperation with HIF/NF-κB
MDM2 and HIF interactions
Phosphorylation of HIF and
ROS implicated
Competition with HIF for
binding sites (p21)
or partners (SP-1, Max)
Hydroxylation,
HIF-interaction
Phosphorylation

Phosphorylation
Unknown, localization and
binding altered by hypoxia
Cooperation with
HIF/Smad3
PKC, Ras/Raf and ERK
HIF-independent

46

Nuclear factor kappa B (NF-κB)
NF-κBs are a family of transcription factors that are considered the master regulators of
the innate and adaptive immune response. NF-κB signaling is one of the downstream targets of
anti-inflammatory asthma medications, specifically through activation of the glucocorticoid
receptor. The role in of NF-κB signaling in the pathogenesis of asthma and other inflammatory
diseases is well established [219]. In mammals, there are five members that function as homoand/or hetero-dimers: NF-κB1 (p105/p50), NF-κB2 (p100/p52), RelA (p65), RelB and c-Rel. The
NF-κB response is complex, but in brief, all NF-κB proteins contain a Rel-homology domain,
which binds the inhibitor of kappa B (IκB) protein that sequesters NF-κB dimers in the cytosol
until appropriately activated. Upon stimulation through many possible signals, the IκB kinase
(IKK) complex (also known as the “canonical NF-κB pathway”) is activated and phosphorylates
IκB, which is subsequently polyubiquitinated and degraded by the 26S proteasome. The freed
NF-κB dimers translocate to the nucleus, bind DNA, and activate gene transcription, mostly of
pro- inflammatory genes such as IL-6. They also are important regulators downstream of TNFα
and IL-1 signaling [220-222].
Hypoxic activation of NF-κB, while milder than for HIFs, has been reported for nearly 20
years, and its role in lung disease has been well documented [216, 239, 240]. Initial studies of
mechanism indicated that phosphorylation via Ras/Raf was necessary [241], though more
recently PHDs and FIH can also impact NF-κB signaling [242, 243]. Other crosstalk with the HIF
system is evident. For example, NF-κB can affect basal levels of HIF1α expression and HIF1α
can induce NF-κB activation indirectly through the upregulation of cytokines IL-1β and TNFα
[244-247].

47

Purinergic (adenosine) receptors
Another family of sensory proteins which are susceptible to hypoxic regulation and
important to lung immunity is the adenosine receptors, specifically Adora2b. Adenosine
receptors are G-protein coupled receptors (GPCRs), and Adora2b is paired to a Gs and Gq
subunit, so that it leads to both increased cAMP and activation of PKC. Compared to the A1
(Gi), A2A (Gs), and A3 (Gi) receptors, Adora2b is the least sensitive to adenosine, thought to be
key in its ability to detect hypoxic environments where higher concentrations of adenosine
metabolites would occur [248, 249].
Despite being the least sensitive receptor to adenosine, Adora2b has demonstrated key
functions in the lung. WT mice live twice as long as Adora2b knockout mice in ventilatorinduced lung injury [250]. Adora2b knockouts also demonstrated increased vascular leak to
hypoxia, while knockouts of A1AR, A2AR and A3AR had no effect on these outcomes.
Specificity to Adora2b was confirmed by inhibiting the Adora2b with PSB1115, resulting in a
profound increase in hypoxia-induced vascular leakage, while stimulation of Adora2b with
BAY60-6583 resulted in reversal of leakage [251]. Further studies showed that the A2B
receptor is expressed highly in mouse ATII cells [252], and under hypoxia or ventilation-induced
injury it is highly induced in the lung [250, 251]. HIF1α regulates adenosine kinase, CD73 and
Adora2b, all of which act to increase extracellular adenosine signaling during hypoxic conditions
and amplify the protective Adora2b response [253-255]. The HIF1α-Adora2b signaling pathway
represents promising therapeutic potential in lung inflammation, especially acute lung injury
(ALI) [256]. This signaling pathway also seems to apply to other organ systems, such as

48

protecting against intestinal epithelial inflammation. Mechanistically the hypoxic regulation of
Adora2b in naïve T-cells is thought to promote TREGs which dampens the resulting immune
response [257, 258]. Taking the known interactions between HIF1α, NF-κB and adenosine
signaling together, it is possible to see how loss of HIF1α may disrupt this signaling to affect the
resulting inflammatory profile in airway epithelial cells (Figure 7).

49

Figure 7

A

adenosine
AMP

A2B

Cytosol

CD73

Nucleus

Other
protective
mechanisms ↑cAMP-PKA
ERK-MAPK
Co2+

Co2+

NF-κB

Cell Damage &
danger signals

B

adenosine

Cytosol

Co2+

Normal
immune
signaling
profile,
hypoxic
conditioning

AMP
A2B

↓cAMP-PKA
ERK-MAPK

CD73

Nucleus
↑cell
damage,
altered
immune
profile

Loss of positive
feedback loop

NF-κB

Cell Damage &
danger signals
Figure 7. Possible crosstalk of HIF1α with Adora2b and NF-κB in CTL and HIF1α

50

Δ/Δ

cells. Representation of an ATII epithelial cell

Figure 7 (cont’d)

during an inflammatory insult such as cobalt. A cell from HIF-sufficient mice has a normal complement of HIF1α which activates
genes to assist in hypoxic adaptation, as well as a positive feedback loop with adenosine signaling to activate protective
mechanisms during acute lung injury, and it also participates in crosstalk with NF-κB to properly regulate expression of
Δ/Δ

inflammatory cytokines (A). In contrast, cells from HIF1α
mice lack beneficial hypoxic adaptation from direct HIF1α signaling,
are prevented from making a HIF1α-dependent positive feedback loop with adenosine receptor A2B and CD73, and lose crosstalk
with NF-κB such that cells experience increased cell damage and an altered response to inflammatory stimuli (B).

51

Cobalt
Occurrence and uses
Cobalt is a rare element in the Earth’s crust, comprising anywhere from 20-23 ppm.
There are many industrial uses for this ferromagnetic transition metal, most of which are in
alloys. Recently, there has been an increased usage of cobalt in portable electronics and in the
manufacture of rechargeable batteries [259, 260]. Prosthetic joints (e.g. Vittalium, CoCrMo)
and sintered carbides (aka “hard metal”) used for drilling and high-speed applications like jet
engines also utilize cobalt in their alloys which imbues superior wear resistance and strength at
high temperatures. In 2011, over 82,000 tons were produced (up 91% from 43,000 tons in
2003). Of all the cobalt used, approximately 44% goes into metals (19% of which are
“superalloys” used for cutting tools, turbines and space vehicles), 30% to batteries, 14% to
colors/paints, 9% to catalysts, 3% to biological applications [261]. Newer applications include
using cobalt as a catalyst support used in the production of carbon nanotubes [262].

Biologically, cobalt is required in vitamin B12 (cobalamin, created only by microbes), a
cofactor of many methylation reactions, notably the creation of methionine from homocysteine
and the conversion of methylmalonyl-CoA to succinyl-CoA in fatty acid metabolism. Clinically,
due to its role in purine synthesis and myelin production, B12 deficiencies present as anemia
and neurologic deficits such as sensory impairment and ataxia. Supplementation success
depends on whether an intact absorptive surface is present, mainly the parietal cells of the
stomach producing acid and intrinsic factor and proper distal ileum absorption, the lack of

52

which produces “pernicious anemia” [14]. In the 1950’s, cobalt was used to treat certain
anemias [263, 264]. Several cases of cobalt toxicity have occurred from this application
prompting the removal of its use clinically.
Toxicity
Humans are generally exposed to low levels of cobalt through water and food sources.
Toxicity only occurs as accidental ingestion, direct contact causing allergic dermatitis, iatrogenic
use in medicine or with occupational exposures that are usually by inhalation of cobaltcontaining dusts. Several exposure events in humans, in addition to its use for treatment of
anemias [265], have aided greatly in our understanding of human toxicity. One notable route
which caused multiple toxic exposures was the use of cobalt as a foam stabilizer in beer, where
several high-volume consumers developed complications referred to as “beer-drinker’s
cardiomyopathy” [266]. Inhalation of cobalt-dusts can cause a chronic disease called hard
metal lung disease (HMLD) (discussed below) or “cobalt lung”. Additional exposures from
groups of diamond polishers, again from inhalation of dusts, have been documented [267-269].
Hard metal lung disease
The most widely known cobalt-specific pathology is hard metal lung disease (HMLD) or
“cobalt lung”. HMLD is a giant-cell interstitial pneumonitis caused from inhalation of cobaltcontaining dusts, typically from hard metal grinding and drilling. It is characterized by chronic
inflammation from repeated exposures, but also has a unique pathognomonic feature (with
proper exposure history) of giant-cell pneumonitis. Giant cells are created by phagocytic cells,
usually macrophages, thought to result from impaired clearance of foreign (dust in HMLD)

53

particles leading to chronic unresolved inflammation. HMLD symptoms include wheezing and
cough, and on spirometry, patients are likely to show obstructive patterns but may also show a
mixed obstructive/restrictive pattern [270, 271]. HMLD should be differentiated from hard
metal asthma or cobalt asthma [272-274], which is likely to be a cobalt:albumin specific IgEmediated type I hypersensitivity that is relieved when patients are removed from exposure
conditions [275]. Some patients with cobalt asthma progress to HMLD, and it is thought this is
due to multifactorial reasons causing some individuals to be more susceptible, though the
mechanisms behind this susceptibility have not been identified to date. The mechanism of
HMLD is also still largely unknown.
There are several mechanisms known by which cobalt can cause toxicity. Molecularly, it
binds to thiol groups with high affinity, such as those in α-lipoic acid (used in the pyruvate
dehydrogenase and α-keto glutarate dehydrogenase complexes) and glutathione (essential to
the pool of antioxidants and regulation of oxidative stress) [276]. Cobalt can cause further
oxidative stress by creation of ROS through Fenton chemistry [277]. More recently cobalt has
received attention for its disruption of Ca

2+

sequestration in cells [269, 278], though it is

unknown if this is also related to metabolic disruption affecting ATPase-dependent ion
transport. Finally, the displacement of Fe

2+

2+

from the active sites of enzymes by Co
2+

been documented [279]. PHD enzymes which regulate HIF require Fe
disruption of this Fe

2+

by Co

2+

has also

in their active site, and

is possible based on ionic radius [280]. One group has shown

that cobalt can cause ascorbate depletion and that this is the limiting factor for PHD activity

54

[281]. Regardless of the mechanism, cobalt causes PHD inhibition, which prevents HIF
hydroxylation, leading to HIF stability and activation of HIF-signaling.
Role of HIFs in cobalt toxicity
Researchers and physicians have known for decades that cobalt is also a hypoxia mimic
[282]. Subsequent research has shown that cobalt’s mechanism for hypoxia mimicry involves
activation of HIF-signaling. Cobalt has the ability to inhibit PHD activity, stabilize HIFs and
induce transcription of HIF target genes [151, 280, 283, 284]. Considering these connections,
our lab is interested in the role that HIFs may play in cobalt toxicity.
Previous work in our laboratory showed that HIF1α

-/-

mouse embryonic fibroblasts

(MEFs) were partially resistant to cobalt toxicity, suggesting HIF1α contributes to the toxicity of
cobalt in vitro [285]. The next logical step was to evaluate the in vivo role of HIFs in cobalt
toxicity. Considering the most widespread exposures were respiratory in nature, this research
focused on the lung. A series of transgenic mice were created that are capable of doxycyclineinducible, lung epithelial-specific deletion of HIF1α, HIF2α, or both HIF1α and HIF2α. Given the
in vitro results, it was predicted that loss of HIF1α might provide protection from cobaltinduced lung inflammation. Surprisingly, loss of HIF1α increased the overall cellularity and
significantly altered the inflammatory profile from neutrophilic to eosinophilic and increased
markers of asthma, such as YM1/2 as well as other histopathologic changes consistent with
allergic airway disease [214]. These results suggested that HIF1α in lung epithelium has a more
nuanced immune modulatory role in response to cobalt-induced inflammation, and may have
implications for HIF signaling in allergic airway disease.

55

Hypotheses and Specific Aims
My hypothesis is that a proper balance of HIF1α and HIF2α in lung epithelium is necessary for
the establishment of normal immunity in the lung through their influence on inflammationregulating signaling pathways, including Adora2b and NF-κB. To test this hypothesis, I have
three specific aims:
1. Characterize the relationship between loss of HIF1α and the changes in lung immunity
following cobalt exposure by altering the timing of HIF1α deletion, performing flow
cytometry of lung immune cells, and testing the clinical relevance of this inflammatory
switch by determining the impact of neonatal hyperoxia exposure on cobalt-induced
inflammation.
2. Determine the role of Adora2b and NFκB in an oropharyngeal aspiration model of
cobalt-induced lung inflammation in HIF1α-deficient mice using pharmacological
inhibitors.
3. Determine the role of epithelial-derived HIF2α signaling on cobalt-induced inflammation
and phenotypically compare these results to HIF1α-deficient and HIF1α/2α-deficient
mice.
This dissertation is a summary of experiments performed to fulfill these specific aims.

56

CHAPTER 2 – MATERIALS AND METHODS

57

Description of mice
Three strains of mice were used in these studies, all of which utilize the SP-CrtTA

+/tg

/(TetO)7-CMV-Cre

tg/tg

system, a doxycycline-inducible, lung epithelial-specific Cre-

recombinase system were a gift Jeffrey Whitsett (Cincinnati Children’s Hospital Medical
Center). Mating between these mice and HIF1α
of Cambridge) and/or HIF2α

flox/flox

flox/flox

(a gift from Randall Johnson, University

(a gift from M. Celeste Simon, University of Pennsylvania)

resulted in three strains:
1. SP-C-rtTA
2. SP-C-rtTA
3. SP-C-rtTA

+/tg

tg/tg

flox/flox

+/tg

tg/tg

flox/flox

+/tg

tg/tg

flox/flox

/(TetO)7-CMV-Cre
/(TetO)7-CMV-Cre

/HIF1α

/(TetO)7-CMV-Cre

/HIF2α
/HIF1α

(HIF1α
(HIF2α
/HIF2α

fl/fl

) [51, 214, 215, 286],

fl/fl

), and

flox/flox

(HIF1/2α

fl/fl

).

Recombination in the floxed HIF1α and/or HIF2α gene, specifically in the respiratory
epithelium, occurs when mice are exposed to doxycycline (DOX) (Figure 8) [287]. Throughout
this paper these DOX-treated strains will be referred to as HIF1α

∆/∆

∆/∆

, HIF2α

∆/∆

, or HIF1/2α

.

Also, the term “-deficient” will also be used interchangeably with the “Δ/Δ” to describe the
recombined/DOX-treated mice. The bias toward eosinophilic inflammatory response to cobalt
in the HIF1α

∆/∆

mice does not occur in any other strain or treatment combination except for

mice containing all three transgenes that were treated with doxycycline, thus eliminating the
possibility that these observed inflammatory changes are due to doxycycline alone or Cre
toxicity (Figure 9) [215]. Genotyping of the mice was performed by PCR for the four loci as

58

Figure 8

hSPC-rtTA

Conditional HIFa locus

Human SPC promoter

rtTA

Exon I

Exon II
loxP

Exon III
loxP

Exon II

= rtTA
Lung-specific
Exon I
(TetO)7-Cre recombinase

Exon III
loxP

Dox
CRE

Exon I

Exon III
loxP

rtTA rtTA rtTA rtTA rtTA rtTA rtTA

CRE

Functional HIFα Deletion

Figure 8. Inducible, lung epithelium-specific transgenic knockout mouse model. The SP-CrtTA

tg/-

/(TetO)7-CMV-Cre

tg/tg

model system in combination with HIF1α

alleles generates mice functionally deficient in either HIF1α (HIF1α

∆/∆

fl/fl

and/or HIF2α

), HIF2α (HIF2α

∆/∆

fl/fl

) or

∆/∆

HIF1 and HIF2 both (HIF1/2α ). The first transgene of the system, hSPC-rtTA, features the
reverse tetracycline trans-activator (rtTA) protein driven by the human surfactant protein-C (SPC) promoter, making rtTA expression lung-specific to Club and alveolar epithelial type II (ATII)
cells. The second gene in the system, (TetO)7-CMV-Cre recombinase, allows for Crerecombinase expression only in the presence of rtTA and doxycycline (DOX) given in food and
water. Finally, Exon II of HIF1α or HIF2α, required for DNA binding and dimerization of the
HIF:ARNT complex, is flanked by loxP sites. These triple transgenic mice are capable of
functional deletion of the respective HIFα in the presence of doxycycline.

59

Figure 9
3.0
3.0
2.5
2.5
2.0
1.5
1.5
1.0
1.0
0.5
0.5

Total Cells

B

Monocytes

stH sTH sth StH STH
STH
stH
sTH
sth
StH
Neutrophils

25
25
20
20
15
15
10
10
55

Genotype
Genotype

E
30
25
25
20
20
15
15
10
10
55

Lymphocytes
Lymphocytes

stH sTH sth StH STH
stH
sTH
sth
StH
STH

stH sTH sth StH STH
stH
sTH
sth
StH
STH

Genotype
Genotype

Genotype
Genotype

fl/fl

Figure 9. Eosinophilia in HIF1α

Eosinophils
Eosinophils

stH
sTH
sth
StH
STH
stH sTH sth StH STH
Genotype
Genotype

Lymphocytes

30

Cells (x 10 )

Neutrophils

stH sTH sth StH STH
sth StH STH
stH
sTH

(x
Cells/mL 4 10^4)

Genotype
Genotype

Cells (x 104)

Cells/mL (x 10^5)

D

0.77
0.66
0.55
0.44
0.33
0.22
0.11

Monocytes

Cells (x 105)

1.8
1.8
1.5
1.5
1.2
1.2
0.9
0.9
0.6
0.6
0.3
0.3

C
Cells (x (x
Cells/mL104) 10^4)

Total Cells

Cells (x 105)

Cells/mL (x 10^5)

A

DOXY Cobalt (60 mg)

+
+

+
+

S – no SPC-rtTA transgene (WT)
s - SPC-rtTA/+
T – no tet-Cre transgene (WT)
t – tet-Cre/+ or tet-Cre/tet-Cre
H – no HIF1a floxed alleles (WT)
fl fl
h – HIF1a +fl/fl
HIF1α /+

mice by genotype and DOX treatment. Mice from every possible genotype in the SP-C- rtTA/

60

Figure 9 (cont’d)

fl/fl

(TetO)7-CMV-Cre/HIF1α
were treated with cobalt or saline and BAL was assessed for total cells (A) and differentials:
macrophages/monocytes (B), eosinophils (C), neutrophils/PMNs (D) and lymphocytes (E). white box = No Dox, Saline; blue box =
/+

No Dox, Cobalt; yellow box = DOX, Saline; red box = DOX, Cobalt. S = no SPC-rtTA transgene; s = SPC-rtTA ; T = no tet-Cre
transgene; t = tet-Cre

/+

or tet-Cre

+/+

; H = homozygous WT HIF1α locus; h = HIF1α

61

fl/fl

described previously [286]. Genotyping primers for HIF2α locus are as follows: 5’-CAG GCA GTA
TGC CTG GCT AAT TCC AGT T-3’ (Forward) and 5’-CTT CTT CCA TCA TCT GGG ATC TGG GAC T-3’
(Reverse) creating fragments of 410bp in the wild-type HIF2α locus and 444bp in the floxed
HIF2α locus.
All of the mice used in this study were males, and were maintained in a mixed C57BL/6
and FVB/N background. All the procedures regarding the handling, maintenance, exposure and
necropsy protocols of the mice used in this study were approved by the university laboratory
animal resources (ULAR) regulatory unit at Michigan State University.

Doxycycline treatment and animal husbandry
+/tg

The SP-C-rtTA

tg/tg

/(tetO)7-CMV-Cre

model system, HIF1α

fl/fl

mice exposed in utero

to DOX show lethality upon parturition [286]. Postnatal exposure leads to almost complete loss
of HIF1α from Club and ATII cells without lethality or signs of gross morphologic differences
[215]. These studies utilize three different DOX timing protocols, designated as P4-14 (early
recombination group), P4-30 (standard recombination group) or P32-42 (adult recombination
group). To achieve these different groups, pups were first exposed to DOX through mother’s
milk, then directly after weaning. Specifically, lactating dams were given DOX-containing feed
(625 mg doxycycline/kg; Harlan Teklad, Madison, WI) and drinking water (0.8 mg/ml; MP
Biochemicals, Solon, OH) ad libitum beginning on postnatal day 4 (P4) until either cessation at
P14 (for P4-14 group) or weaning (occurring around P21) for the P4-30 group). After weaning,
mice from the P4-30 groups were maintained on the same DOX-containing food and water
directly (ad libitum) for no less than P30. The P32-42 group was given DOX food and water only

62

from P32-42, and then allowed no less than 10 days to clear the doxycycline from their systems
before the start of cobalt dosing (Figure 10). The dose of doxycycline used was slightly lower
than the concentration that has been used to induce recombination without any observable
toxicity or impact on alveolarization [288]. Control animals in this study were mice containing
all the same transgenes (HIF1α

fl/fl

fl/fl

, HIF2α

or both in addition to SP-C-rtTA

tg/tg

Cre

) but were given normal (without DOX) food and water ad libitum.

63

+/tg

/(tetO)7-CMV-

Figure 10

Birth P4

P30

A

P42

DOX
Saline or 60 µg CoCl2
Birth P4

B

P14

P42

DOX
Saline or 60 µg CoCl2

P42

P32

Birth

C

P52

DOX
Saline or 60 µg CoCl2
P14

Birth

D

P42

75% O2
Saline or 60 µg CoCl2

Figure 10. Mouse experiment treatment and dosing schemes. The experimental timeline
schemes that were used in these studies. At least one of the schemes in A-D was used in every
study. P = postnatal day of mouse age. Sacrifice of mice was made 24 hours after the final dose
(red dashed arrows), corresponding to Day 2 (1 dose), Day 6 (5 doses) and Day 14 (10 doses
given as 5 once daily doses, 2 day break, 5 once daily doses, 2 day break) (A). The P4-14 DOX
protocol (B) and P32-42 DOX protocol (C) were for HIF1α mice only, and sacrifice was all after 5
doses. Finally, hyperoxia (75% O2) was given to mouse pups from P0-P14, and samples were
taken both at P14 and also after 5 doses (D).

64

Cobalt exposure
HIF1α

Δ/Δ

Δ/Δ

, HIF2α

Δ/Δ

, HIF1/2α

mice and their respective floxed controls were

randomly assigned to receive either cobalt or saline vehicle. Prior to each daily aspiration, mice
were anesthetized with isoflurane using the drop jar method. Oropharyngeal aspirations of 25
μL of either 10 mM cobalt choride or sterile saline vehicle were given once daily. This dose of
cobalt corresponds to a daily dose of 60 μg CoCl2 per mouse, doses known to reliably
reproduce substantial inflammation [214, 215]. Most commonly, five consecutive daily doses
of cobalt were given after DOX has cleared from the system of mice (no less than 10 days).
However, for HIF2α mice, a single cobalt dose group (called “1 day” group) or 2 weeks of dosing
(5 days dosed, 2 days off, 5 days dosed, 2 days off, called “14 day” group) were also performed
to better understand the time-course of cobalt exposure in this strain. All mice were sacrificed
24 hours after their final dose (Fig. 10).

PSB1115 exposure
PSB1115 is a specific inhibitor of the adenosine A2B receptor. It was utilized in a pilot
study of acute (5 day) cobalt-induced lung inflammation in HIF1α-deficient mice. PSB1115
(Tocris Bioscience) was dissolved in saline and sterilized via a 0.22 μm filter each day before
injection. Fifteen (15) minutes prior to each cobalt exposure, mice were injected i.p. at 30
mg/kg. Previous studies by Eckle et al. using this compound at 10 mg/kg in mice have shown
that Adora2b antagonism increases vascular leak and prevents the usual fluid clearance from
the lung [251]. The 30 mg/kg dose was chosen to minimize the number of injections to the
mice and also ensure that adequate PSB1115 reached the lung.

65

TCH013 exposure
TCH013 is a compound comprised of an imidazoline scaffold with four benzene rings
created at Michigan State University by Dr. Jetze Tepe and colleagues. This compound
decreased TNFα concentrations and histopathologic severity of a collagen-induced rheumatoid
arthritis mouse model [289]. This compound was tested for its ability to alter the response to
cobalt-induced lung injury in CTL and HIF1α-deficient mice. Fresh TCH013 was prepared
immediately prior to each daily dosing by combining 97.5 mg of TCH013 (white powder) and 2
mL of 30% propylene glycol (vortexing vigorously until dissolved), and then adding a total of 4.5
mL of 5% dextrose (warmed to 37°C) in 1 mL aliquots (vortexing with each addition so each
aliquot was well incorporated into homogeneous solution). Concentration of final solution was
15 mg/mL, and each mouse received 150 mg/kg (or vehicle control). Mice were injected intraperitoneal (i.p.) with TCH013 one hour prior to receiving their daily doses of cobalt.

LPS exposure
HIF1α

fl/fl

mice treated with regular food and water or DOX from P4-30 or P4-32 were

dosed once intranasally (under regulated isoflurane anesthesia) with 15 μL in each nostril of LPS
(1.67 μg/μL) for a total of 50 μg LPS (Sigma, St. Louis, MO) per mouse. Mice were sacrificed 24
hours later and standard tissues taken (described below).

Hyperoxia exposure
Mice were exposed to hyperoxia using a custom airtight regulated plexiglass glove box
from Coy Laboratories (Grass Lake, MI). Temperature was controlled via the main control unit

66

that can be manually controlled to within 0.1°F or C. Humidity was automatically set to prevent
condensation based on temperature by the main control unit and is impossible to change
manually. O2 and CO2 concentrations were regulated by sensors opening to the inside of the
glove box and a control unit for each above the glove box, and N2 was used as a background
gas. O2 could be set from 0-100% in 0.1% increments. CO2 could be set from 0-20% in 0.1%
increments. Additionally, the glove box was fitted with a filtration system to avoid
accumulation of ammonia and carbon dioxide by animals. The filtration system was activated
by CO2 concentrations that reach 0.1% or greater, and when activated a pump pushed air from
chamber through the filtration system and back into the chamber, until CO2 again reached 00.1%.
Mice from litters born on the same day were randomized into litters of 10 before the
exposure began. Hyperoxia exposed mice were placed in the chamber (75% O2, 0% CO2 and
25°C). Lactating dams were removed from the hyperoxia and replaced every 12 hours to avoid
causing their hyperoxic stress, and animals were kept on a 12 hour light/dark cycle.
Additionally, pups were checked every 12 hours for signs of hyperoxic stress or death, and
removed if death occurred. Conditions were maintained throughout the experiment by using
the purge box for anything entering or leaving the glove box, replacing animal food, gases and
filtration material as needed. Exposures began on P0 (first post-natal day) and continued until
P14, at which point mice were returned to normoxia. No doxycycline was given at any point

67

during these experiments. Mice were dosed with a 5d cobalt oropharyngeal aspiration protocol
(described above), and sacrificed for necropsy 24 hours after the last dose of cobalt.

Necropsy, tissue harvesting and processing
Mice to be sacrificed were anesthetized by an i.p. injection of 1000-1300 mg/kg Avertin
(2,2,2-tribromoethanol, Sigma-Aldrich, St. Louis, MO). Upon successful blockade of afferent
pain sensation, a midline laparotomy was performed and mice were exsanguinated by
transecting the renal artery. The lungs were exposed, and trachea cannulated. Heart and lungs
were removed en bloc and sterile saline was used in two successive 1 mL lavages to make
bronchoalveolar lavage (BAL). Total cells were counted using a hemacytometer and differential
cell counts were made from CytoSpin samples of BAL stained with DiffQuick reagent (Baxter,
FL). Right lung lobes were snap-frozen in liquid nitrogen for later RNA and protein analysis.
Left lung lobes were gravity-inflation fixed at 30 cm pressure with 10% phosphate buffered
formalin for histopathological and immunohistochemistry (IHC) analyses.

Histopathology and immunohistochemistry
Changes in histopathological properties were assessed by using formalin-fixed left lung
th

th

lobes cut at the 5 and 11 generation (G5/11, to differentiate shallow from deeper lung).
Dissected lungs were paraffin embedded, cut to 5 μm sections, mounted on glass slides, and
stained with hematoxylin and eosin (H&E) for general histopathology. Alcian blue (pH 2.5)
periodic acid Schiff (AB-PAS) stain was used to detect mucosubstances. Immunostaining was
performed for major basic protein (MBP; polyclonal rabbit anti-mouse MBP, 1:500, Mayo Clinic,
Scottsdale, AZ) as described previously [214]. Immunostaining for HIF1α (polyclonal rabbit anti-

68

HIF1α, 1:100, Novus Biologicals NB100-479, Littleton, CO) or HIF2α (polyclonal rabbit antiHIF2α, 1:100, Novus Biologicals NB100-122, Littleton, CO) was performed using the Vectastain
Elite ABC Kit (Rabbit IgG) as described previously [286].

Gene expression analysis by qRT-PCR
Snap frozen lung tissue was homogenized in TRIzol reagent (Life Technologies, Carlsbad, CA)
using a Retsch MM200 bead beater system (Retsch, Haan, Germany). Total RNA was
quantitated spectrophotometrically (Nano-Drop ND-1000 UV-Vis Spectrophotometer). Total
RNA (1μg) was reverse transcribed using Superscript III (Life Technologies, Carlsbad, CA), and
quantitative real-time PCR reactions were performed using gene-specific primers (listed in
Table 3) and Power SYBR Green PCR Master Mix on an ABI PRISM 7000 (Life Technologies,
Carlsbad, CA). Gene expression was measured using the standard curve method with
hypoxanthine guanine phosphoribosyl transferase (Hprt) as the housekeeping gene. Expression
was normalized to Hprt for all samples, and the fold change of the control saline treated group
was set to 1.

69

Table 3. List of primers for mouse genotyping and qRT-PCR.
Gene

For (5’-3’)

Rev (5’-3’)

Amplicon
Size (bp)

SP-CrtTA

GACACATATAAGACCCTGGTCA

AAAATCTTGCCAGCTTTCCCC

350

Cre

TGCCACGACCAAGTGACAGCAATG

AGAGACGGAAATCCATCGCTCG

374

HIF1α

GCAGTTAAGAGCACTAGTTG

GGAGCTATCTCTCTAGACC

274(fl)
240 (wt)

HIF2α

CAGGCAGTATGCCTGGCTAATTCCA
GTT

CTTCTTCCATCATCTGGGATCTGGG
ACT

444 (fl)
410 (wt)

Hprt

TTCCGGAGCGGTAGCACCTCC

ATCATCGCTAATCACGACGCTGGG

113

Bnip3

GGCGTCTGACAACTTCCACT

AACACCCAAGGACCATGCTA

123

Csf2

ATTCGAGCAGGGTCTACGGGGC

TTTCACAGTCCGTTTCCGGAGTTGG

110

Adora2b

CGGACTTCCACGGCTGCCTC

TTTATACCTGAGCGGGACGCGA

125

p65/relA

TGCCTACCCGAAACTCAACT

TTGAGATCTGCCCTGATGGT

126

Hmox1

GGCCTGAACTTTGAAACCAG

GTCGTGGTCAGTCAACATGG

122

CCACGTCGTAGCAAACCACC

ACAAGGTACAACCCATCGGC

IFNγ

CACGGCACAGTCATTGAAAGC

TTCATGTCACCATCCTTTTGCC

125

IL-1β

GCAGCTGGAGAGTGTGGATCCCA

GGTGCTGATGTACCAGTTGGGGA

133

Ccl17

CATGCCAGAGCTGCTCGAGCC

ACACGATGGCATCCCTGGAACA

133

Ccl20

CCAGGCAGAAGCAGCAAGCAAC

TCACAAGCTTCATCGGCCATCTGTC

111

Arg1

AAAGGACAGCCTCGAGGAGGGG

AGGTCCCCGTGGTCTCTCACG

110

iNOS

GCAACTACTGCTGGTGGTGA

AAGGCCAAACACAGCATACC

129

IL-2

AGATGAACTTGGACCTCTGCG

TGTGTTGTCAGAGCCCTTTAG

130

IL-4

AGGTCACAGGAGAAGGGACGCC

TGCGAAGCACCTTGGAAGCCC

114

IL-5

CCGCCAAAAAGAGAAGTGTGGCGA

GCCTCAGCCTTCCATTGCCCA

112

TNFα

70

132

Table 3 (cont’d)
IL-6

TCCTCTCTGCAAGAGACTTCCATCC

ACAGGTCTGTTGGGAGTGGTATCC

127

IL-10

CCAGTTTTACCTGGTAGAAGTGATG

TGTCTAGGTCCTGGAGTCCAGCAG
ACTC

324

IL-13

GCTGGCGGGTTCTGTGTAGCC

ATCGGGGAGGCTGGAGACCG

132

IL-17A

GGACTCTCCACCGCAATGAAGACC

TCATGTGGTGGTCCAGCTTTCCC

111

IL-25

TGTACCAGGCTGTTGCATTCTTGGC

CACTCCTCCGGGGGTTCTTGC

127

IL-33

CAGGCGACGGTGTGGATGGG

CGTCACCCCTTTGAAGCTCCACG

117

TSLP

GGACTGTGAGAGCAAGCCAGC

CTGGGCAGTGGTCATTGAGGGC

134

Ym1

AGCTTTTGAGGAAGAATCTGTGG

AAGAGACTGAGACAGTTCAGGG

121

Ym2

TGAGGAAGAATCCACTTTGAACC

GAGAGACTGAGACAGTTCAGGG

115

Eotaxin

GCGCTTCTATTCCTGCTGCTCACGG

GTGGCATCCTGGACCCACTTCTTC

239

Eotaxin2

CTTGCTGCACGTCCTTTATTTCCA

GGTCAGTACAGATCTTATGGCCCT

135

71

Cytokine bead array
Cytokines were measured from BAL or whole lung lysates (WLL). WLL is generated from
snap-frozen tissues in RIPA buffer using the bead beater system described above. RIPA buffer
contained protease inhibitors aprotinin, leupeptin and pepstatin (1 μg/mL each, Sigma-Aldrich,
St. Louis, MO), complete Mini Protease Inhibitor Cocktail (1 pellet / 10 mL RIPA Buffer, Roche)
and phenylmethylsulfonyl fluoride (PMSF, 1 mM), phosphatase inhibitors sodium
orthovanadate (Na3VO4, 1 mM) and sodium fluoride (NaF, 1 mM), and EDTA (1 mM).
Homogenates were centrifuged at 16,000 g for 15 min at 4°C and the supernatant was
quantified for protein content using the Bradford Assay (BioRad, Hercules, CA) [290]. 50 μL of
each sample supernatant was used for maximum detection of cytokines IL-2, IL-4, IL-5, IL-6, IL10, IL-13, IL-17A, IFNγ and TNFα using the Th1/Th2/Th17 Cytometric Bead Array Kit, IL-5/IL-13
Flex Sets and analyzed on a FACSCalibur flow cytometer according to manufacturer’s protocols
(BD Biosciences, San Jose, CA). For WLL, cytokine content was then normalized to total protein
in each sample, whereas BAL was analyzed without normalization.

Flow cytometry of lung immune cells
Right lung lobes from each treatment group were disrupted using a cell dissociation
sieve kit (CD-1, Sigma) to obtain a single cell suspension and stored in ice-cold RPMI. After
suspensions were obtained, cells were restimulated with phorbol myristate acetate (PMA; 40
nM) and ionomycin (Io; 0.5 µM) (both from Sigma) for 6h in 2% fetal bovine serum (FBS) RPMI
and 1× Brefeldin A (Biolegend, San Diego, CA). All staining was performed in 96-well round
bottom plates (BD Falcon, Franklin Lakes, NJ). Cells isolated from the lungs were washed with

72

FACS buffer (1× HBSS, 1% bovine serum albumin, 0.1% sodium azide, pH 7.6), and surface Fc
receptors were blocked with anti-mouse CD16/CD32 (BD Biosciences, Franklin Lakes, NJ) for 15
min at 4°C. Cells were labeled for 30 min at 4°C with CD4 (clone RM4-5) and CD8 (53–6.7)
(Biolegend). Subsequently cells were washed thrice with FACS buffer, fixed with Cytofix (BD
Biosciences) for 15 min, and resuspended in FACS buffer. After surface staining, cells were
washed twice with 1× Perm/Wash (BD) and incubated with 1× Perm/Wash for 30 min at room
temperature in 96-well round bottom plates. Fluorescently labeled antibodies were added at
0.25–0.5 µg/mL in 200 µL for 30 min. The following fluorescently labeled antibody clones were
used: IFNγ (XMG1.2, Biolegend) and GATA-3 (TWAJ, eBioscience). Cells were washed twice with
1× Perm/Wash and subsequently resuspended in FACS buffer. After intracellular staining, cells
were analyzed the same day. Compensation and voltage settings of fluorescent parameters
were performed using single color staining controls. Cells were identified as positive for
cytokine secretion after gating on the negative population in unstimulated samples (without
PMA/Io or Brefeldin A). Fluorescent staining was analyzed using a BD Accuri C6 flow cytometer
(Franklin Lakes, NJ). Fluorescence was quantified using CFlow software (BD Accuri, San Jose,
CA).

Statistical analytical methods
All quantitative analyses were performed using the OriginPro 8 suite of statistical
software (OriginLab, Northampton, MA). Unless otherwise stated, analyses were by One-way
ANOVA using Fisher’s LSD method (α=0.05). All error bars are presented as standard error of
the mean (SEM). Any data with P value < 0.05 were considered significant.

73

CHAPTER 3 – MECHANISTIC INVESTIGATIONS OF THE EOSINOPHILIC PHENOTYPE
OF MICE DEFICIENT IN HIF1α IN LUNG EPITHLIUM TREATED WITH COBALT

74

Results
Timing of HIF1α Deletion and Eosinophilia
HIF1α

fl/fl

mice were randomly assigned to receive either regular food and water or DOX

food and water in 3 time points: P4-14, P4-30, or P32-42. After no less than 10 days to clear
DOX, mice received 5 consecutive once daily doses of 25 μL saline or 10 μM CoCl2 via
oropharyngeal aspiration, and were sacrificed 24 hours after final dose (~P58). Total cells and
differentials were counted from BAL. Total cells from BAL of saline-treated mice were
negligible and did not change with DOX treatment. Total cells from BAL of cobalt-treated mice
were higher than in their saline counterparts, except for the P32-42 group, suggestive that this
latter group of mice was protected. The number of total cells was highest in the P4-14 group,
which was statistically different than all other cobalt treated groups. Numbers of total cells
from the control and P4-30 groups treated with cobalt were both statistically lower than the
P4-14 group, though they did have higher number of total cells than their saline controls (Figure
11A). When looking at the differentials of these total cells, the P32-42 group had fewer
numbers of all types of cells (macrophages, eosinophils, PMNs and lymphocytes) (Figure 11B,
12 and 13). The macrophages in the P4-30 group were higher than the P4-14 and P32-42 cobalt
treated groups (Figure 11B). Numbers of eosinophils increased in both the P4-14 and P4-30
groups compared to the control (Figure 12A). The eosinophilic response from the P4-14 group
was higher than we generally see in our P4-30 model, and are likely the major contributing
factor to the P4-14 group having the highest number of total cells. Numbers of neutrophils did

75

Figure 11

Total Cells

cells/mL (*10^5)

A
6.00
5.00
4.00
3.00
2.00
1.00
0.00

*
*

*

NO DOX

B

P4-14

P4-30

Macrophages
2.50

cells/mL (*10^5)

P32-42

2.00
1.50

*

*
*

1.00
0.50

0.00

NO DOX

P4-14

Saline

P4-30

P32-42

60 μg Cobalt

Figure 11. BAL total cells and macrophages from various DOX regimens in HIF1α mice treated
with 5d cobalt. Mice treated once daily for 5 days with 60 μg Co/mouse/day. A. Total Cells; B.
Macrophages; white bars = SAL, black bars = COB; * = significant compared to SAL control, black
brackets = significant across groups of cobalt exposed mice (α=0.05 for One-way ANOVA with
Fisher’s LSD).

76

Figure 12

Eosinophils

cells/mL (*10^5)

A
3.00
2.50
2.00
1.50
1.00
0.50
0.00

*
*
NO DOX

P4-14

P4-30

P32-42

P4-30

P32-42

PMNs

B
cells/mL (*10^5)

1.20
1.00

0.80
0.60
0.40
0.20
0.00

NO DOX

P4-14

Saline

60 μg Cobalt

Figure 12. BAL eosinophils and neutrophils (PMNs) from various DOX regimens in HIF1α mice
treated with 5d cobalt. Mice treated once daily for 5 days with 60 μg Co/mouse/day. A.
Eosinophils; B. PMNs; white bars = SAL, black bars = COB; * = significant compared to SAL
control, black brackets = significant across groups of cobalt exposed mice (α=0.05 for One-way
ANOVA with Fisher’s LSD).
not change across the control, P4-14 or P4-30 groups treated with cobalt, but P32-42 had a
statistically lower amount than all the others (Figure 12B). Interestingly, lymphocytes in the P4-

77

14 and P4-30 groups increased when treated with cobalt, and in the P4-14 group were higher
than all other cobalt-treated groups (Figure 13).
Figure 13

Lymphocytes
cells/mL (*10^5)

0.25

*

0.20
0.15

*

0.10
0.05

0.00

NO DOX

P4-14

Saline

P4-30

P32-42

60 μg Cobalt

Figure 13. BAL lymphocytes from various DOX regimens in HIF1α mice treated with 5d cobalt.
Lymphocytes from BAL of mice treated once daily for 5 days with 60 μg Co/mouse/day; white
bars = SAL, black bars = COB; * = significant compared to SAL control, black brackets =
significant across groups of cobalt exposed mice (α=0.05 for One-way ANOVA with Fisher’s LSD).

Determination of TH2 Inflammation by Flow Cytometry
Mice that followed a standard 5 day cobalt exposure protocol had BAL taken and single
cell suspensions were made from the right lung lobes. BAL cells in addition to lung suspensions
were stained with different fluorescent markers. Extracellular staining for CD4 (TH) and CD8
(TC) was performed, followed by intracellular staining for IFNγ (TH1 marker) and GATA3 (TH2
marker) and analyzed by flow cytometry. Mice lacking one of the transgenes of our model, the

78

SP-C-rtTA (rtTA), were included to determine if there was an effect of DOX treatment alone.
Strikingly, the TH2 cells (CD4+GATA3+) were highest among the cobalt treated mice that
received DOX and were also rtTA+, showing direct evidence of an exacerbated TH2 response
(Figure 14). There were no statistically significant differences in the TH1 population across the
groups, though the absolute mean for cobalt treated HIF1α-sufficient mice (with all 3
transgenes, but no DOX) was highest. The overall percentage of the T-helper population that
was TH1 was much lower than the other two populations (Figure 15). The cytotoxic T cell
population (TC) was highest in the DOX/Cobalt/rtTA+ group (Figure 16).

79

Figure 14

%TH2 Cells
(%Gata3+ of CD4+ Pop.)

% Gata3+ (of CD4+)

30

*,a

25
20
15

10
5
0

rtTA-

rtTA+

Saline

rtTA-

rtTA+

Cobalt

CTL

DOX

Figure 14. TH2 immune cell populations in HIF1α mice with 5d cobalt. TH2 cells are quantified
as a percentage of CD4+ cells that show GATA3+ staining. Data above represents a combination
of both cells from BAL and whole lung suspensions. * = significant from CTL within treatment
and genotype, a = significant from all other cobalt treated groups, brackets = significant across
genotype or treatment, One-way ANOVA with Fisher’s LSD, α = 0.05.

80

Figure 15

%TH1 Cells
(%IFNγ+ of CD4+ Pop.)
8

% IFNγ+ (of CD4+)

7
6
5
4
3
2
1
0

rtTA-

rtTA+

Saline

rtTA-

rtTA+

Cobalt

CTL

DOX

Figure 15. TH1 immune cell populations in HIF1α mice with 5d cobalt. TH1 cells are identified
as a percentage of CD4+ cells also staining positive for IFNγ. Data above represents a
combination of both cells from BAL and whole lung suspensions. * = significant from CTL within
treatment and genotype, a = significant from all other cobalt treated groups, brackets =
significant across genotype or treatment, One-way ANOVA with Fisher’s LSD, α = 0.05.

81

Figure 16

%TC Cells
(%IFNγ+ of CD8+ Pop.)

% IFNγ+ (of CD8+)

30

*,a

25
20
15

10
5
0

rtTA-

rtTA+

Saline

rtTA-

rtTA+

Cobalt

CTL

DOX

Figure 16. TC immune cell populations in HIF1α mice with 5d cobalt. Cytotoxic T lymphocytes
(TC) are quantified as a percentage of CD8+ cells also showing IFNγ staining. Data above
represents a combination of both cells from BAL and whole lung suspensions. * = significant
from CTL within treatment and genotype, a = significant from all other cobalt treated groups,
brackets = significant across genotype or treatment, One-way ANOVA with Fisher’s LSD, α =
0.05.

82

PSB1115 and Adora2b in Cobalt-Induced lung injury
Δ/Δ

Previous investigation revealed that the HIF1α

mice lost staining for Adora2b in the

lung epithelium, though it was still visible in the alveolar macrophages (Figure 17). Both the
bronchial airway (BA) (Figure 17A and C) and parenchymal (Figure 17B and D) regions show the
same pattern of stain loss.

83

Figure 17

A

B

BA

a

WT
Cobalt
(60 μg)

AD
AD

50 mm

C

50 mm

D

BA

Δ/Δ
Cobalt
(60 μg)

AD
AD

a
50 mm
Δ/Δ

50 mm

Figure 17. IHC for Adora2b in lungs of cobalt treated CTL and HIF1α
mice. All mice were treated with cobalt. Control mice
(containing all transgenes necessary for HIF1α deletion but kept on regular food and water) are on top (A and B), while DOX-treated
84

Figure 17 (cont’d)
Δ/Δ

mice (HIF1α ) are below (C and D). Panels A and C taken from bronchial airway, B and D from the parenchymal regions.
Arrowhead = bronchial epithelial cell, solid arrow = ATII cell, dashed arrow = alveolar macrophage; BA=bronchial airway, AD=alveolar
duct, a=alveolus. Black line indicates 50 μm.

85

Given the loss of Adora2b staining in HIF1α

Δ/Δ

mice, it was hypothesized that Adora2b

inhibition would exacerbate cobalt-induced lung inflammation in controls, with little to no
effect on the HIF1α

Δ/Δ

mice. To test this, mice were exposed to the Adora2b-specific inhibitor,

PSB1115 (30 mg/kg injected i.p.) prior to a once daily dosing of 60 μg cobalt for 5 days. Upon
sacrifice, BAL fluid was taken from the mice to assess total cells and differentials as a measure
of inflammation. PSB1115, which has been shown to exacerbate models of inflammation in the
lung, had little to no exacerbating effect on cobalt-induced lung inflammation (Figures 18-20).
Total cells did increase in HIF1α

Δ/Δ

mice treated with cobalt, though PSB1115 treatment did not

worsen this effect (Figure 18A). There were no observable changes in macrophage populations
(Figure 18B). Unexpectedly, the number of eosinophils decreased with PSB1115 treatment in
Δ/Δ

cobalt-treated HIF1α

mice (Figure 19A), and numbers of neutrophils seemed to increase
Δ/Δ

with PSB1115 treatment, but only in HIF1α

mice (Figure 19B). There were no observable

changes in lymphocyte populations (Figure 20).

86

Figure 18

A

Total Cells
cells/mL (*10^5)

5.0
4.0

3.0
2.0
1.0
0.0
Veh

PSB1115
CTL

B

Veh

PSB1115

HIF1α Δ/Δ

Macrophages
cells/mL (*10^5)

1.6
1.2
0.8

0.4
0.0

Veh

PSB1115
CTL

Saline

Veh

PSB1115

HIF1α Δ/Δ

60 μg Cobalt

Figure 18. BAL total cells and macrophages from HIF1α mice treated with cobalt for 5 days
+/- PSB1115. Bronchoalveolar lavage (BAL) was counted for total cells (A) and macrophages
(B). Some groups contained 2 mice due to deaths which rendered the data insufficient for
statistical analyses.

87

Figure 19

A
cells/mL (*10^5)

Eosinophils
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Veh

PSB1115
CTL

B

Veh

PSB1115

HIF1α Δ/Δ

PMNs
cells/mL (*10^5)

0.25
0.20
0.15
0.10
0.05
0.00
Veh

PSB1115
CTL

Saline

Veh

PSB1115

HIF1α Δ/Δ

60 μg Cobalt

Figure 19. BAL eosinophils and PMNs from HIF1α mice treated with cobalt for 5 days +/PSB1115. Bronchoalveolar lavage (BAL) was counted for eosinophils (A) and neutrophils
(PMNs) (B). Some groups contained 2 mice due to deaths which rendered the data insufficient
for statistical analyses.

88

Figure 20

Lymphs
cells/mL (*10^5)

0.08
0.06

0.04
0.02
0.00
Veh

PSB1115
CTL

Veh

PSB1115

HIF1α Δ/Δ

Saline

60 μg Cobalt

Figure 20. BAL lymphocytes from HIF1α mice treated with cobalt for 5 days +/- PSB1115.
Bronchoalveolar lavage (BAL) was counted for lymphocytes. Some groups contained 2 mice due
to deaths which rendered the data insufficient for statistical analyses.

NF-κB in Cobalt-induced lung injury
TCH013
TCH013 is a compound composed of an imidazoline scaffold with four benzene rings
created at Michigan State University by Dr. Jetze Tepe and colleagues (Figure 21). According to
preliminary testing of this compound, it has proteasome inhibitory capabilities that appear
specific for inhibiting activation of NF-κB. When TCH013 was given with IL-1β, it prevented the
induction of TNFα and IL-6 production in vitro [291]. In a mouse model of rheumatoid arthritis,
this compound decreased TNFα concentrations and histopathologic severity of arthritic disease
[289].

89

Figure 21

Figure 21. Structural schematic of TCH013. Modified from [289].

HIF1α mice were exposed to TCH013 (150 mg/kg, i.p.) one hour prior to each daily
dosing of 60 μg cobalt. Mice were sacrificed one day after the last cobalt dose and standard
samples were taken. Treatment with TCH013 did not protect the mice from cobalt-induced
lung injury as noted by no change to total cells (Figure 22A). Macrophages increased with
cobalt treatment and the HIF1α

Δ/Δ

mice treated with TCH013 had higher macrophage numbers

than the CTL mice treated with TCH013 (Figure 22B). A similar pattern is seen with eosinophils
(Figure 23A). Neutrophils were lower in HIF1α

Δ/Δ

mice treated with cobalt, and treatment with

TCH013 seemed to have no effect (Figure 23B). Lymphocytes showed only an increase in CTL
mice treated with vehicle (Figure 24). Even though TCH013 caused statistically significant
increases in macrophages and eosinophils, this change was not significant when taken to total
cells. Overall, these data suggest that TCH013 had little effect on overall inflammatory profile
in cobalt induced lung inflammation.

90

Figure 22

cells/mL (*10^5)

A

Total Cells
6.0
5.0
4.0
3.0
2.0
1.0
0.0

*

CTL

HIF1α Δ/Δ

*

CTL

Saline

cells/mL (*10^5)

B

*

*

HIF1α Δ/Δ

Cobalt

Macs
3.0
2.5
2.0
1.5
1.0
0.5
0.0

*

*

* *

CTL

HIF1α Δ/Δ
Saline

CTL

HIF1α Δ/Δ
Cobalt

Vehicle

TCH013 (150mg/kg)

Figure 22. BAL total cells and macrophages from HIF1α mice treated with cobalt for 5 days
+/- TCH013. Bronchoalveolar lavage (BAL) was counted for total cells (A) and macrophages (B).
* = significant from saline-treated control, brackets = significant across DOX or VEH/TCH groups,
One-way ANOVA with Fisher’s LSD, α = 0.05. N=4-7 mice/group.

91

Figure 23

A

Eos

cells/mL (*10^5)

2.0

*

1.5
1.0
0.5
0.0
CTL

HIF1α Δ/Δ

CTL

Saline

B

Cobalt

PMNs

*

2.5

cells/mL (*10^5)

HIF1α Δ/Δ

2.0

*

1.5
1.0
0.5

0.0
CTL

HIF1α Δ/Δ

Saline

CTL

HIF1α Δ/Δ

Cobalt

Vehicle

TCH013 (150mg/kg)

Figure 23. BAL eosinophils and PMNs from HIF1α mice treated with cobalt for 5 days +/TCH013. Bronchoalveolar lavage (BAL) was counted for eosinophils (Eos) (A) and neutrophils
(PMNs) (B). * = significant from saline-treated control, brackets = significant across DOX or
VEH/TCH groups, One-way ANOVA with Fisher’s LSD, α = 0.05. N=4-7 mice/group.

92

Figure 24

cells/mL (*10^5)

Lymphs
0.12
0.10
0.08
0.06
0.04
0.02
0.00

*

CTL

HIF1α Δ/Δ

CTL

HIF1α Δ/Δ

Saline

Cobalt

Vehicle

TCH013 (150mg/kg)

Figure 24. BAL lymphocytes from HIF1α mice treated with cobalt for 5 days +/- TCH013.
Bronchoalveolar lavage (BAL) was counted for lymphocytes. * = significant from saline-treated
control, brackets = significant across DOX or VEH/TCH groups, One-way ANOVA with Fisher’s
LSD, α = 0.05. N=4-7 mice/group.

LPS
Δ/Δ

Previous experiments have shown a TH2 bias in the HIF1α

mice treated with cobalt,

and suggested that some element of the inflammatory response in these mice was disrupted.
Lipopolysaccharide (LPS) is a constituent of gram negative bacterial cell walls, and it is used
frequently in animal experiments as a challenge to the innate immune response or as a general
inflammatory stimulus. To investigate whether this vital system of innate immunity was
Δ/Δ

functioning in the HIF1α

mice, animals were given 50 μg of LPS total (25 μg LPS in 15 μL

given twice, once in each nostril, under isoflurane anesthesia) and sacrificed 24 hours later.
93

BAL was assessed or cellularity, and total cells increased in all LPS-treated mice (Figure 25A).
Macrophages, eosinophils and lymphocytes showed no statistically significant changes across
groups (Figure 25B, 26A and 27, respectively). Neutrophils dominated the cellular response to
LPS for all groups (Figure 26B).

94

Figure 25

A
cells/ mL (*10^5)

Total Cells
15

*

*

10

*

5
0

CTL

P4-30

P32-42

Doxycycline Treatment Group

Macrophages
cells/ mL (*10^5)

B
0.8
0.6
0.4
0.2
0.0

CTL

P4-30

P32-42

Doxycycline Treatment Group

Saline

LPS (50 μg)

Figure 25. BAL total cells and macrophages from HIF1α mice in P4-30 and P32-42 DOX groups
treated +/- LPS. Mice were treated once with 50 μg total LPS (25 μg LPS in 15μL, given once in
each nostril) under anesthesia and sacrificed 24 hours later. Total cells (A) and macrophages
(B) were counted. * = significant from saline-treated control; One-way ANOVA with Fisher’s
LSD, α = 0.05. N=5-9 mice/group.

95

Figure 26

A
cells/ mL (*10^3)

Eosinophils
1.5

1.0
0.5
0.0
CTL

P4-30

P32-42

Doxycycline Treatment Group

B
cells/ mL (*10^5)

PMNs
15

*

*

10

*

5
0
CTL

P4-30

P32-42

Doxycycline Treatment Group

Saline

LPS (50 μg)

Figure 26. BAL eosinophils and PMNs from HIF1α mice in P4-30 and P32-42 DOX groups
treated +/- LPS. Mice were treated once with 50 μg total LPS (25 μg LPS in 15μL, given once in
each nostril) under anesthesia and sacrificed 24 hours later. BAL eosinophils (A) and
neutrophils (PMNs) (B) were counted. * = significant from saline-treated control; One-way
ANOVA with Fisher’s LSD, α = 0.05. N=5-9 mice/group.

96

Figure 27

cells/ mL (*10^5)

Lymphocytes
0.15
0.10
0.05

0.00
CTL

P4-30

P32-42

Doxycycline Treatment Group

Saline

LPS (50 μg)

Figure 27. BAL lymphocytes from HIF1α mice in P4-30 and P32-42 DOX groups treated +/LPS. Mice were treated once with 50 μg total LPS (25 μg LPS in 15μL, given once in each nostril)
under anesthesia and sacrificed 24 hours later. BAL lymphocytes were counted. * = significant
from saline-treated control; One-way ANOVA with Fisher’s LSD, α = 0.05. N=5-9 mice/group.

Hyperoxia during early postnatal phase
Several studies have illustrated that premature infants are more likely to be diagnosed
with asthma through childhood, and infants in the “extremely premature” group, typically GA
24-28 weeks, are more likely to have asthma that continues into adulthood [47-49]. The lungs
of premature infants are exposed to higher levels of O2 than they would normally experience in
utero; additionally, one supportive therapy present in nearly all premature infants with
underdeveloped lungs is supplemental O2. Given that HIF1α loss in lung epithelium has shown
a predisposition for a TH2 response, and the high levels of O2 that premature infants

97

experience has the potential to negatively regulate HIF-signaling in the lungs, we wanted to
know whether high O2 levels for mice in the early postnatal phase would mimic the HIF1α

Δ/Δ

phenotype.
Early postnatal hyperoxia is known to produce pathologic effects, including alveolar
simplification, and is sometimes used in models of bronchopulmonary dysplasia. Mice exposed
to hyperoxia (75% O2, for 14 days as described in materials and methods section) displayed
alveolar simplification (Figure 28B). In contrast, mice exposed to room air, or dosed with DOX
from P0-14 displayed normal lung architecture (Figure 28A and C). Additionally, the impact of
75% O2 on HIF1α levels was assessed by immunohistochemistry. Hyperoxic exposure had little
impact on HIF1α levels in the lung (Figure 29). Both normoxia and hyperoxia mice were carried
out to adulthood and dosed with a 5d cobalt protocol. Results show that mice responded
similarly between groups. BAL total cells were increased in both groups (Figure 30A) and most
cells induced by cobalt challenge were macrophages (Figure 30B). Eosinophils, neutrophils and
lymphocytes were not increased with 75% O2 treatment (Figure 31A & B and Figure 32),
suggesting that early postnatal exposure to hyperoxia does not phenotypically match that of
HIF1α loss due to recombination.

98

Figure 28

REG

DOX

75% O2

Normoxia

Figure 28. H&E Staining for P0-14 mice exposed to 75% O2. Mice were either exposed to
normoxia (A) or 75% O2 (B) from birth for 14 days and sacrificed. A group of normoxia mice
were also given DOX to compare any effects of DOX on development of normal lung
architecture (C). Size bars are 50 μm for large images and 200 μm for insets.

99

Figure 29

75% O2

Normoxia

Figure 29. HIF1α IHC for mice exposed to 75% O2 from P0-14. HIF1α (NB100-479 Rabbit pIgG,
1:100) with Vectastain ABC Rabbit Kit, NovaRed coloring agent, and light hematoxylin
counterstain was performed on mice exposed to 75% O2 from P0-14 and sacrificed under
hyperoxic conditions at P14. Normoxia (A, C) and 75% O2 (B, D) were compared. (A) and (B)
show axial airways and adjacent bronchial arteries, with heaviest staining for HIF1α in the
bronchial artery smooth muscle cells (solid arrows). (C) and (D) show pleural sections, with
staining in macrophages (dashed arrows) and alveolar type II cells (arrowheads). Black line
indicates 50 μm.

100

Figure 30

A
cells/mL (*105)

Total Cells
30
25
20
15
10
5
0

*

*

Normoxia

B

2
75% O2

Macrophages
cells/mL (*105)

15

*

*

10
5
0

Normoxia

Saline

2
75% O2

60 μg Cobalt

Figure 30. BAL total cells and macrophages from CTL mice exposed to 75% O2 from P0-14 and
5d cobalt as adults. Mice were allowed to age until P42 when first cobalt dose occurred. Total
cells (A) and macrophages (B) are shown. Most cells observed were in cobalt treated mice and
were macrophages. * = significant from saline-treated control. One-way ANOVA with Fisher’s
LSD, α = 0.05. N = 3-5 mice.

101

Figure 31

A

Eosinophils
cells/mL (*105)

4.0

3.0
2.0
1.0
0.0

Normoxia

B

2
75% O2

PMNs
cells/mL (*105)

20

15
10
5
0

Normoxia

Saline

2
75% O2

60 μg Cobalt

Figure 31. BAL eosinophils and PMNs from CTL mice exposed to 75% O2 from P0-14 and 5d
cobalt as adults. Mice were allowed to age until P42 when first cobalt dose occurred.
Eosinophils (A) and neutrophils (PMNs) (B) are shown. * = significant from saline-treated
control. One-way ANOVA with Fisher’s LSD, α = 0.05. N = 3-5 mice.

102

Figure 32

Lymphocytes
cells/mL (*105)

1.5
1.0
0.5
0.0

Normoxia

Saline

75% O2
2

60 μg Cobalt

Figure 32. BAL lymphocytes from CTL mice exposed to 75% O2 from P0-14 and 5d cobalt as
adults. Mice were allowed to age until P42 when first cobalt dose occurred. Lymphocytes are
shown. * = significant from saline-treated control. One-way ANOVA with Fisher’s LSD, α = 0.05.
N = 3-5 mice.

Discussion
The final stages of lung development occur after birth in both mice and humans. The
early postnatal time period has also been shown to be crucial for development of the immune
system, especially in the context of risk of humans developing allergic airway disease (i.e. the
hygiene hypothesis). Therefore, when initial experiments in our laboratory indicated that loss
of HIF1α in the lung (over a 26 day period of doxycycline exposure from P4-30) caused an
eosinophilic inflammation in adult mice, it was important to know whether or not lung
development may have played a role. We attempted to isolate this developmental window by

103

performing DOX exposures prior to (P4-14) or after (P32-42) the consensus age of P30 for the
completion of postnatal mouse lung development.
Results from the DOX timing experiments here revealed that DOX given to mice from
P4-14 was adequate in reproducing the eosinophilic phenotype in response to cobalt-induced
lung inflammation. In addition, the P32-42 DOX treatment did not cause eosinophilia in cobalt
treatment, suggesting that loss of HIF1α by itself is not sufficient to drive eosinophilic
inflammation, but that the timing of the exposures was important in establishing this
Δ/Δ

phenotype. Further studies using a model of ovalbumin sensitivity and challenge with HIF1α

mice revealed that HIF1α loss caused an exacerbation of eosinophilic inflammation [51]. This
suggests that mice lacking HIF1α in their ATII and Club cells in the early postnatal phase are
predisposed to eosinophilic inflammation. This is highly relevant for the idea of the hygiene
hypothesis of asthma, where exposures to microbes during early development drives later
allergic disease, and implicates HIF1α as a potential target gene for developing allergic
inflammation.
Perhaps the most interesting finding of these timing studies is that the P32-42 DOX
treated mice seemed not only unresponsive, but almost protected from cobalt-induced lung
injury, in agreement with the very first hypothesis we made regarding HIF1α contributing to
cobalt toxicity. This may indicate that HIF-signaling is contributory to the pathogenesis of
HMLD. While several aspects of our cobalt exposure model are not ideal for mimicking HMLD
(lack of tungsten carbide or other particulates, aspiration instead of inhalation and a subchronic
exposure period), the idea that loss of HIF1α in epithelial cells is protective in cobalt-induced

104

inflammation must be confirmed and pursued with further studies. The presence of two
distinct epithelial-HIF1α-deficiency phenotypes, either a pro-eosinophilic phenotype in the early
postnatal phase or an unresponsive/protective phenotype in the adult phase warrants further
study, but is highly suggestive that another (non-HIF) signal is responsible for these distinct
phenotypes. Several possibilities exist for this other signal, including a neighboring cell type
(immune or otherwise) or some developmentally controlled signaling within the epithelial cell
that interacts with HIF1α or its targets. Along these lines, our first experiment was to
characterize the most likely cell type involved in eosinophilia of the lung, the TH2 population.

Flow cytometry studies revealed that the TH2 cell population was higher only in
HIF1α

fl/fl

mice that contained all necessary transgenes for our inducible knockout system and

that received doxycycline. Not only does this confirm our previous assumptions about T H2
inflammation, it also illustrates the robustness of the SP-C/Cre model system. Whether the TH2
bias experienced in HIF1α

Δ/Δ

mice extends beyond the pulmonary system has yet to be

determined, but these data are highly suggestive that whatever is happening in the lung
epithelium, there are far more TH2 cells migrating to the lung in cobalt-treated HIF1α

Δ/Δ

mice.

Additionally, this data provides the first direct evidence that cobalt-induced lung injury
does not, in fact, activate TH1 cells, as has been proposed in our previous studies. In previously
published work with HIF1α mice, IFNγ, the primary TH1 cytokine, is minimally increased in both

105

control and HIF1α

Δ/Δ

mice treated with cobalt. Further, the only reported IL-12 measurements

were non-detectable [215]. Data in Figure 15 shows that CTL mice do not have a higher
number of TH1 cells, and further data regarding the cytotoxic T cell (TC) population (CD8+) in
Figure 16 suggests that any IFNγ detected is more likely to come from these cells (or other cells
such as innate lymphoid cells) than TH1 cells. Our previous studies of cobalt-induced lung
inflammation have never directly measured T helper bias, and the idea that lack of TH2
cytokines IL-4 and IL-5 (together with presence of some detectable TH1 cytokines) was
equivalent to activation of TH1 cells relied too heavily on the idea of a TH1 vs TH2 axis where
one or the other must be activated. This mentality failed to take into consideration the
important fact that one can have either TH1 or TH2 cell activation without the other, and the
current prevailing understanding of lymphocyte activation which includes other T helper
subtypes such as TH17 and TREG, or the innate lymphoid populations (which lack CD4 entirely
but still contribute to the “classic TH1/TH2 cytokine” pool). The studies of T helper bias
described here provide some clarity to our previous lung cytokine measurements, and imply
that cobalt treatment alone does not induce any specific T helper response, but that loss of
HIF1α in lung epithelium leads to activation of TH2 cells.

The current literature is highly suggestive that HIF1α has important levels of crosstalk in
several signaling pathways during inflammation, including NF-κB and Adora2b signaling [242-

106

247, 253-255]. To determine if either of these play a role in the change in inflammation
Δ/Δ

observed in the HIF1α

mice, pharmacological inhibitors were used. Previous IHC for

Adora2b suggests that loss of HIF1α also causes loss of Adora2b, which could impact the ability
Δ/Δ

for the HIF1α

lungs to respond with the appropriate protective downstream signaling. The

Adora2b inhibitor, PSB1115, did not worsen the response of control mice to cobalt, as was
hypothesized; however, PSB1115 treatment did decrease the eosinophilia seen in HIF1α

Δ/Δ

mice. These results are difficult to explain, given the role that Adora2b has been shown to play
in epithelial fluid clearance in the lung and upregulation of FoxP3+ TREG cells to limit
inflammation; however, several confounding variables exist when incorporating pharmacologic
agents (affecting all cells) in a cell-type-specific system. Additionally, further studies with an
Adora2b agonist would likely be helpful in clarifying the role of Adora2b in this context.
Several limitations of the study of Adora2b may explain the unexpected nature of the
results. Firstly, sample sizes were small in the experiment utilizing PSB1115 (N=1-3). At the
very least, repeating the experiment is required before any conclusions can be drawn.
Secondly, pharmacologic treatment has effects on cells other than ATII cells (where HIF1α was
recombined). These confounding effects on other cell types, especially the alveolar
macrophage, are potentially substantial, considering that the alveolar macrophage had the
highest staining for Adora2b of all the lung cell types identified (Figure 17A, B and D). In the
context of these limiting factors, few conclusions can be drawn on the link between HIF1α and

107

Adora2b from these studies, but further experimentation, especially from isolated ATII cells or
alveolar macrophages from the HIF1α mice would be helpful.
Given its role in regulating the expression of many cytokines, blockade of NF-κB was also
attempted in combination with cobalt exposure in the HIF1α

Δ/Δ

mice. TCH013, a recently

designed proteasome inhibitor, is known to limit the inflammatory response in certain contexts
and was used in coexposure studies with cobalt in the HIF1α mouse model. The results show
that TCH013 did not protect mice from cobalt-induced lung inflammation, as evidenced by
unchanged numbers of total cells. Numbers of macrophages and eosinophils increased in the
TCH013 treated HIF1α

Δ/Δ

mice compared to their HIF1α-sufficient controls; however, these

combined increases were not enough to increase the number of total cells. It is possible that
TCH013 affected alveolar macrophages and eosinophils to increase their numbers in mice
lacking epithelial HIF1α, but the biological significance of these changes is inconclusive.
Several limitations on the studies of TCH013 and NF-κB warrant scrutiny of results.
Firstly, the specific mechanism of TCH013 is unknown. Additionally, only one time point was
used for cobalt exposure, and while the 5 day response was common for many of our cobalt
studies, the pharmacodynamics of TCH013 are not characterized, so timing, dose, and route of
administration were not optimized before the study was performed. More experiments would
be required to characterize the response to cobalt with TCH013, including more time points, an
assessment of cytokine production and direct probing for NF-κB in different cell constituents.
Further experiments with TCH013 were performed with BALB/c mice and the ovalbumin
sensitivity/challenge model, where TCH013 administration was identical to this experiment.
108

TCH013 prevented inflammation in BAL cellularity. However, further histopathologic
investigation of the lung revealed little effect on eosinophilic inflammation (unpublished
results). Thus, use of the TCH013 compound has produced mixed results, likely from an
incomplete understanding of its mechanism. Additionally, these studies were limited by lack of
measurement of NF-κB activity. Also, pharmacologic investigations suffer from lack of cell type
specificity, since TCH013 undoubtedly has effects on all cell types, not just the ATII cells. If
HIF1α and NF-κB had meaningful crosstalk within the ATII cell, the effects of TCH013 on other
cell types may have confounded the results. Use of a fully characterized NF-κB inhibitor, along
with its use in isolated ATII cells, may produce more consistent results in future studies.
Another way of probing the NF-κB system and innate immunity is through the use of
LPS, which stimulates the Toll-like receptor pathway, specifically TLR4, which acts through NFκB to initiate the innate immune response. It has been reported that small amounts of LPS can
produce TH2 inflammation, but before we were able to address this question we first had to
know if the detection and response to LPS was still intact. Mice in which HIF1α was recombined
during early postnatal development (P4-30) and during early adulthood (P32-42) were
phenotypically identical to control mice in their response to LPS. All three groups displayed a
robust neutrophil-dominant infiltrate to the lungs. These results showed that despite HIF1α
deletion, the LPS response was robust and similar.
Taking all of the pharmacologic, DOX timing and T-cell population data into account, it
seems the timing of HIF1α deletion is the most important determinant we have been able to
change and observe. Acute signaling (during the inflammatory response) is very important, but

109

the mixed results of the studies targeting Adora2b and NF-κB leave many questions. The data
thus far seem to suggest that there is something occurring in lung epithelial cells that lose
HIF1α during the early postnatal phase of development to prime the TH2 immune response.
Isolating the ATII cells will allow probing to one of the primary cell types affected by loss of
HIF1α.
Focusing our attention on the early postnatal developmental period, we wondered if
high O2 levels during the early postnatal phase could decrease the amounts of HIF1α present,
and potentially mimic our lung-specific HIF1α deletion phenotype. Considering the importance
of supplemental O2 use in extremely premature infants and their increased likelihood of
developing asthma as children and adults make this idea intriguing as a potential molecular
mechanism by which this predisposition occurs. Looking at lungs from mice receiving high O2
revealed their alveoli to be much larger and overall more simplified with less septations,
consistent with other models of high O2 during this time period. However, analysis of HIF1α via
IHC was unconvincing for a decrease in HIF1α from high O2 levels, even when tissues were
taken and fixed in the hyperoxic environment. Further, dosing these mice in adulthood with
cobalt reveals that high O2 from P4-14 did not exacerbate eosinophilia, reducing the likelihood
that high O2 can mimic this phenotype.

110

There are several potential reasons why the hyperoxia treatments did not mimic the
HIF1α

Δ/Δ

phenotype. With high concentrations of O2 there are several important changes

occurring in the lung other than just potentially lower HIFα levels in epithelial cells, such as
production of ROS, activation of nuclear factor-erythroid 2 related factor (Nrf2) pathway and
proinflammatory cytokines such as IL-1β, IL-6 and TNFα [292-295]. Much like with
pharmacologic interventions, off-target effects of high concentrations of O2 could affect any
number of cell types involved in the pathogenesis of hyperoxia-induced lung injury, which could
have altered the lung in different ways than our model of loss of HIF1α in ATII and Club cells.
In summary, early postnatal HIF1α deletion seems to be very important in producing the
eosinophilic phenotype, which has been shown to include induction of T H2 responses in cobaltinduced lung injury. It also shows that TH1 responses are not a part of normal response to
injury from 10 mM CoCl2, which has implications for the pathogenesis of HMLD. Pharmacologic
inhibition of the Adora2b and NF-κB had some effect on cobalt induced lung injury, though the
role of HIF1α in these processes remains unclear. What is evident is that HIF1α in lung
epithelial cells seems to function more in a protective role in cobalt injury, which is contrary to
our initial suspicions about HIF1α contributing to cobalt toxicity. Finally, hyperoxia exposure
during the early postnatal time period does not mimic the loss of HIF1α in cobalt-induced lung
injury, though cell-type specificity is likely a factor. Further studies, especially of epithelial cell
crosstalk to other populations are required to elucidate more carefully the crosstalk of these
signaling pathways and other possible cell populations in cobalt-induced lung injury.

111

CHAPTER 4 – COMPARATIVE PHENOTYPES OF COBALT-INDUCED RESPIRATORY
INFLAMMATION IN MICE DEFICIENT FOR HIF1α, HIF2α or HIF1α/2α IN LUNG
EPITHELIUM

112

Results
Δ/Δ

HIF2α

mice 60 μg CoCl2 time course

Mice containing all three transgenes necessary for lung epithelial HIF2α-specific deletion
were exposed to DOX from P4 to P30. DOX was allowed to leave system for 10 days afterward
at which point mice were exposed to either sterile saline vehicle or 60 μg CoCl2 via
oropharyngeal aspiration once daily for 1, 5 or 14 (5 days on, 2 days off, 5 days on, 2 days off)
days, and sacrificed 24 hours after last dose. Normal necropsy tissues were taken.
HIF2α IHC
Immunohistochemistry for HIF2α was performed to show specific deletion in the lung
tissue (Figure 33). Dark staining indicates the presence of HIF2α protein in the Club cells of the
bronchial airway in control mice (arrow, Figure 33A) which is vastly decreased in the Club cells
of DOX treated mice (arrow, Figure 33C). Similar positive staining was observed in the
parenchymal portions of the airway, predominantly in the ATII cells in control mice (arrows,
Figure 33B) which is lost in DOX treated mice (arrow, Figure 33D). This staining confirms the
epithelial-specific nature of our DOX-inducible recombination system.
BAL Cellularity
BAL fluid was taken and total cells counted and differentials obtained from control and
HIF2α

Δ/Δ

mice at 1 day, 5 day and 14 day time points (Figures 34-36). The number of total cells

increases following 5 days of cobalt treatment with no difference between CTL and HIF2α

113

Δ/Δ

Figure 33

CTL

DOX

Figure 33. HIF2α IHC for CTL and P4-30 DOX treated mice. HIF2α (NB100-122 Rabbit pIgG,
1:100) with Vectastain ABC Rabbit Kit, DAB coloring agent, and light hematoxylin counterstain
was performed on control (A and B) and DOX treated (C and D) mice. A and C show the
bronchial airway, and Club cells are indicated with solid arrows. B and D show the parenchymal
compartment. Black line indicates 100 μm.

mice. At 14 days the total cells of CTL mice go down but remain high in HIF2α

Δ/Δ

mice (Figure

34A). Macrophages show a similar increase at 5 days in the cobalt treated groups that
decreases by 14 days (Figure 34B). Interestingly, at 14 days eosinophils went up significantly in
the HIF2α

Δ/Δ

mice only, similar to what has been observed with HIF1α

Δ/Δ

mice at 5 days (Figure

35A). Neutrophils did not make up a significant portion of the cellularity, but at 5 days the
cobalt treated CTL mice had higher levels than HIF2α

114

Δ/Δ

mice that dropped down at 14 days

(Figure 35B). Lymphocytes in cobalt-treated control mice were higher at 5 days than 1 and 14
days, and in HIF1α

Δ/Δ

mice were higher at 5 days than 14 days (Figure 36).

115

Figure 34

Total Cells

cells/mL (*10^5)

A
4.0

*

3.0

*

2.0

*
*

1.0
0.0

1 day

5 day

14 day

Time Exposure

B
cells/mL (*10^5)

Macrophages

3.0
2.5
2.0
1.5
1.0
0.5
0.0

**
**

1 day

5 day

14 day

Time Exposure
CTL – Saline
CTL – 60 μg Cobalt
∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 34. BAL total cells and macrophages from HIF2α mice treated with cobalt for 1, 5 and
14 days. Mice were treated for 1, 5, and 14 days with 60 μg CoCl2 or sterile saline vehicle. BAL
was taken and total cells (A) and macrophages (B) are presented. * = significant from salinetreated control, brackets = significant across CTL/Δ/Δ groups or same group on different time
point, One-way ANOVA with Fisher’s LSD, α = 0.05.

116

Figure 35

Eosinophils

A
cells/mL (*10^5)

1.0

*

0.8

0.6
0.4

*

0.2
0.0

1 day

5 day

14 day

Time Exposure

B

PMNs

cells/mL (*10^5)

0.4

*

0.3
0.2
0.1
0.0

1 day

5 day

14 day

Time Exposure
CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 35. BAL eosinophils and PMNs from HIF2α mice treated with cobalt for 1, 5 and 14
days. Mice were treated for 1, 5, and 14 days with 60 μg CoCl2 or sterile saline vehicle. BAL
was taken and eosinophils (A) and neutrophils (PMNs) (B) are presented. * = significant from
saline-treated control, brackets = significant across CTL/Δ/Δ groups or same group on different
time point, One-way ANOVA with Fisher’s LSD, α = 0.05.

117

Figure 36

Lymphocytes
0.1

cells/mL (*10^5)

*
*

0.0

1 day

5 day

14 day

Time Exposure
CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 36. BAL lymphocytes from HIF2α mice treated with cobalt for 1, 5 and 14 days. Mice
were treated for 1, 5, and 14 days with 60 μg CoCl2 or sterile saline vehicle. BAL was taken and
lymphocytes are presented. * = significant from saline-treated control, brackets = significant
across CTL/Δ/Δ groups or same group on different time point, One-way ANOVA with Fisher’s
LSD, α = 0.05.

Histopathology and Immunohistochemistry
Saline mice exhibited no visible pathology or notable difference between CTL and DOX
treated mice (Figure 37A and E). At 1 day cobalt exposure (Figure 37B and F), lungs from both
CTL and DOX treated mice showed evidence of perivascular infiltrates to the bronchial regions.
At 5 days (Figure 37C and G) the infiltrates had expanded to the parenchymal regions, and
thickening of bronchial lamina propria was evident in cobalt treated lungs. Additionally,
thickening of the bronchial epithelium was noted at 5 days in both CTL and HIF2α

118

Δ/Δ

mice,

which resolved in the CTL mice at 14 days, but worsened in the HIF2α

Δ/Δ

mice at 14 days. The

overall peak for pathology of the control lungs treated with cobalt appeared to be 5 days, with
a slight reduction in the amounts of affected lung at 14 days (Figure 37D), in agreement with
the BAL cellularity data. DOX treated mice receiving cobalt looked similar in severity at 5 and
14 day time points (Figure 37F and G), suggesting that resolution of infiltrates and bronchial
epithelial thickening was somewhat dependent on presence of HIF2α.

119

Figure 37

CTL

HIF2α
Δ/Δ
Saline

1d

5d

14 d

60 μg CoCl2
Figure 37. HIF2α strain H&E stain for mice with 1, 5 and 14 days cobalt. Control mice (A-D) and HIF2α

Δ/Δ

(E-H) are treated with

sterile saline vehicle (A and E) or 60 μg CoCl2 after 1 day (B and F), 5 days (C and G) and 14 days (D and H). All images were taken
from the G5 section. BA=bronchial airway, bv=blood vessel, ad=alveolar duct, a=alveolus, *=site of mixed inflammatory cell
infiltrates. Black line indicates 100 μm.

120

MBP, a key constituent in eosinophil secretory granules, was probed using IHC to assess
eosinophilia. As expected, saline treated mice did not show visible signs of eosinophil
infiltration (Figure 38A and E). This was similar for 1 day cobalt exposure (Figure 38B and F),
where some positive cells existed, but did not make up an appreciable amount of the
inflammatory cells. At day 5, many more eosinophils are seen in the tissues of both CTL and
HIF2α

Δ/Δ

mice (Figure 38C and G, respectively), though these resolve somewhat in control mice

for day 14 (Figure 38D). Interestingly, the number of eosinophils greatly increases at 14 days in
the HIF2α

Δ/Δ

Δ/Δ

of HIF1α

mice (Figure 38H), in agreement with the BAL data. These data mimic the results
Δ/Δ

mice, except that the HIF1α

Another aspect of the HIF1α

Δ/Δ

mice develop eosinophilia earlier, around day 5.

phenotype that is also seen in these HIF2α

Δ/Δ

mice is

eosinophilia of parenchymal tissues. In allergic airway disease, eosinophils are somewhat
restricted to the bronchi and bronchioles, so this finding may be significant.

121

Figure 38

CTL

HIF2α
Δ/Δ

Saline

1d

5d

14 d

60 μg CoCl2
Figure 38. HIF2α strain major basic protein IHC for mice with 1, 5 and 14 days cobalt. Immunostaining was performed for major
Δ/Δ

basic protein (MBP; polyclonal rabbit anti-mouse MBP, 1:500, Mayo Clinic, Scottsdale, AZ). Control mice (A-D) and HIF2α

(E-H)

are treated with sterile saline vehicle (A and E) or 60μg CoCl2 after 1 day (B and F), 5 days (C and G) and 14 days (D and H). AA=axial
airway, bv=blood vessel, a=alveolus, *=site of mixed inflammatory cell infiltrates, eosinophils in red stain. Black line indicates 100 μm.

122

Alcian blue periodic acid Schiff (AB-PAS) stain was performed to visualize acidic and
neutral polysaccharides and mucosubstances in goblet cells. No mucosubstances were
observed in the saline treated mice (Figure 39A and E), and very few were seen in 1 day
exposures to cobalt (Figure 39B and F). The 5 day exposures began to reveal increases in
mucous material, though no real differences between the groups were apparent (Figure 39C
and G). At 14 days, a much clearer difference in mucous material could be visualized (solid
arrows), with CTL mice showing sporadic mucous material (Figure 39D), while HIF2α

Δ/Δ

mice

had substantial increases in visualized mucous substances of goblet cells of the epithelium
(Figure 39H).

123

Figure 39

CTL

HIF2α
Δ/Δ
Saline

1d

5d

14 d

60 μg CoCl2
Figure 39. AB-PAS stain for HIF2α mice with 1, 5 and 14 days cobalt. Control mice (A-D) and HIF2α

Δ/Δ

(E-H) are treated with sterile

saline vehicle (A and E) or 60μg CoCl2 for 1 day (B and F), 5 days (C and G) and 14 days (D and H). AA=axial airway, black
arrows=mucous material visualized in goblet cells. Black line indicates 100 μm.

124

Cytokine analysis
Cytokines were analyzed from the BAL of the 1 day and 14 day time points using a
cytokine bead array system. TH2 cytokine IL-4 (Figure 40A) did not show substantial changes
across groups, whereas TH2 cytokine IL-5 (Figure 40B), pro-inflammatory cytokine IL-6 (Figure
41A) and leukocyte chemokine KC (Figure 41B) all showed increases at 1 day more in the
HIF2α

Δ/Δ

mice than controls. At 14 days, these cytokines returned to decreased levels. TH1
Δ/Δ

cytokine IFNγ (Figure 42A) was not measurable in cobalt-treated HIF2α
points, suggesting that HIF2α

Δ/Δ

mice at both time

mice treated with cobalt were less likely to activate TH1 cells.

Similarly, TREG cytokine IL-10 (Figure 42B) was not measurable in the cobalt treated HIF2α
Δ/Δ

mice, potentially signifying a lack of TREG promotion in HIF2α
resolution of any inflammatory immune response.

125

Δ/Δ

mice, which would prolong the

Figure 40

A

IL-4

pg/mL

0.4
0.3

*

0.2
0.1

0.0
1 day

14 day
Time Exposure

B

IL-5

pg/mL

4.0

*,a

3.0
2.0
1.0

0.0
1 day

14 day
Time Exposure

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 40. IL-4 and IL-5 cytokine levels from bead array in HIF2α mice. BAL was tested for
cytokine levels using a cytokine bead array (CBA) and run on a FACSCalibur flow cytometer (BD
Biosciences). Shown are IL-4 (A) and IL-5 (B). N=4-5 mice/group. One-way ANOVA was
performed and significance using Fisher’s LSD of α=0.05 is indicated by: * (from saline treated
control), a (from CTL or ∆/∆ counterpart within time point) or bracket (from same treatment in
different time point).
126

Figure 41

A
pg/mL

IL-6
12
10
8
6
4
2
0

*,a

1 day

14 day
Time Exposure

B

KC
25

*,a

pg/mL

20
15

10
5
0
1 day

14 day

Time Exposure

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 41. IL-6 and KC cytokine levels from bead array in HIF2α mice. BAL was tested for
cytokine levels using a cytokine bead array (CBA) and run on a FACSCalibur flow cytometer (BD
Biosciences). Shown are IL-6 (A) and KC (B). N=4-5 mice/group. One-way ANOVA was
performed and significance using Fisher’s LSD of α=0.05 is indicated by: * (from saline treated
control), a (from CTL or ∆/∆ counterpart within time point) or bracket (from same treatment in
different time point).
127

Figure 42

IFNγ

A

*,a

0.4

pg/mL

0.3
0.2
0.1
0.0
1 day

14 day

Time Exposure

IL-10

pg/mL

B
1.0
0.8
0.6
0.4
0.2
0.0

*

1 day

14 day
Time Exposure

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 42. IFNγ and IL-10 cytokine levels from bead array in HIF2α mice. BAL was tested for
cytokine levels using a cytokine bead array (CBA) and run on a FACSCalibur flow cytometer (BD
Biosciences). Shown are IFNγ (A) and IL-10 (B). N=4-5 mice/group. One-way ANOVA was
performed and significance using Fisher’s LSD of α=0.05 is indicated by: * (from saline treated
control), a (from CTL or ∆/∆ counterpart within time point) or bracket (from same treatment in
different time point).
128

HIF1α/2α mice treated at 5 days
Given the change in inflammation observed in the HIF1α and HIF2α deficient mice, it
was of interest to determine if the phenotype could be exacerbated upon loss of both HIFα
isoforms in the same mouse. Therefore, HIF1α/2α mice were treated with regular food and
water or DOX food and water for the P4-30 time period, and after 10 days of DOX clearing,
were treated with the usual 5 days of 60 μg CoCl2 per mouse per day oropharyngeal aspiration
dosing scheme discussed previously, with sacrifice and samples taken 24 hours after final dose.
HIF1α and HIF2α IHC
To illustrate the HIF1α and HIF2α knockout, IHC was performed for both HIF1α and
HIF2α using lungs from control and DOX mice. In both cases, a clear loss of staining in the lung
is seen in Club cells (dashed arrows) and the parenchymal region in general, though staining in
macrophages and other inflammatory cells remains (solid arrows)(Figure 43).
BAL Cellularity
BAL fluid collected from the 5 day time point was compared to previous data from the
HIF1α and HIF2α mice. Total cells increased with cobalt treatment across all strains when
compared to saline treated mice. HIF1α/2α CTL mice treated with cobalt had less total cells
than their HIF1α and HIF2α counterparts, which was not expected, and may be due to slight
variation in the mixed C57/FVB strains (black bars, Figure 44A). Macrophages make up most
cells in BAL regardless of treatment (Figure 44B). Eosinophils from HIF1α/2α
Δ/Δ

like the HIF1α

Δ/Δ

mice are much

mice, increasing with cobalt treatment compared to the cobalt-treated HIF-

129

sufficient controls (black bars, Figure 45A). Numbers of neutrophils and lymphocytes were
highest in the HIF1α cobalt-treated mice than the other 2 strains (Figure 45B and Figure 46).
The BAL cellularity suggests the HIF1α/2α

Δ/Δ

mice behave much like the HIF1α

Δ/Δ

mice only at

5 days.

Figure 43

HIF1α

HIF2α

CTL

DOX

Figure 43. HIF1α and HIF2α IHC for HIF1α/2α mice. Mice received regular food and water
(CTL) or DOX food and water (DOX) from P4-30. Sections were stained for HIF1α (NB100-479,
1:100, Novus Biologicals) and HIF2α (NB100-122, 1:100, Novus Biologicals). No counterstain
was used in order to highlight the loss of HIF-staining.

130

Figure 44

A

Total Cells

cells/mL (*10^5)

4.0
3.0

**

*

*

2.0

*,a
*

1.0
0.0
HIF1α

B

HIF1/2α

Macrophages
3.0

cells/mL (*10^5)

HIF2α

2.0

**

*

*,a

*

*

1.0
0.0
HIF1α

HIF2α

HIF1/2α

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 44. BAL total cells and macrophages from HIF1α, HIF2α and HIF1α/2α mice at 5 days
cobalt exposure. BAL total cells (A) and macrophages (B) are shown. One-way ANOVA was
performed and significance using Fisher’s LSD of α=0.05 is indicated by: * (from saline treated
control), a (from CTL or ∆/∆ counterpart within strain), brackets (from same treatment across
HIF strain). N=5-8 mice/group.

131

Figure 45

A

Eosinophils

cells/mL (*10^5)

1.0
0.8

*,a

*,a

0.6
0.4
0.2
0.0
HIF1α

B

HIF2α

HIF1/2α

PMNs

cells/mL (*105)

1.6
1.2
0.8

*
*,a

0.4

0.0
HIF1α

HIF2α

HIF1/2α

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 45. BAL eosinophils and PMNs from HIF1α, HIF2α and HIF1α/2α mice at 5 days cobalt
exposure. BAL eosinophils (A) and neutrophils (PMNs) (B) are shown. One-way ANOVA was
performed and significance using Fisher’s LSD of α=0.05 is indicated by: * (from saline treated
control), a (from CTL or ∆/∆ counterpart within strain), brackets (from same treatment across
HIF strain). N=5-8 mice/group.

132

Figure 46

Lymphocytes
cells/mL (*105)

0.3

*
0.2

0.1
0.0
HIF1α

HIF2α

HIF1/2α

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 46. BAL lymphocytes from HIF1α, HIF2α and HIF1α/2α mice at 5 days cobalt exposure.
BAL lymphocytes are shown. One-way ANOVA was performed and significance using Fisher’s
LSD of α=0.05 is indicated by: * (from saline treated control), a (from CTL or ∆/∆ counterpart
within strain), brackets (from same treatment across HIF strain). N=5-8 mice/group.

Histopathology and Immunohistochemistry
With hematoxylin and eosin staining, saline mice showed no visible pathology or
notable difference between CTL and DOX treated mice (Figure 47A and B). HIF2α and HIF1α/2α
mice treated with cobalt were similar in severity of inflammation, in agreement with the BAL
cellularity (Figure 47C-E). Lesions were sporadic and mostly seen in G5 (upper airway) sections.
Major basic protein IHC revealed that both strains of HIF2α and HIF1α/2α mice displayed mild

133

Figure 47

SAL
HIF1/2αfl/fl

COB
HIF2αfl/fl

COB
HIF1/2αfl/fl

CTL

DOX

Figure 47. H&E stain from HIF2α and HIF1α/2α mice at 5 days cobalt exposure. Lungs at G5
were stained with hematoxylin and eosin (H&E). Saline treated lungs (SAL, from HIF1α/2α
strain) were similar. Black line indicates 50 μm. * = mixed inflammatory cell infiltrate;
BA=bronchial airway; bv=blood vessel; ad=alveolar duct; a=alveolus

eosinophilia with cobalt exposure (Figure 48C and D), though HIF1α/2α

Δ/Δ

mice showed the

highest concentration of eosinophils in focal lesions both near bronchial blood vessels/lamina
134

propria and in the parenchymal regions (dashed arrows, Figure 48F). Alcian blue periodic acid
Schiff (AP-PAS) staining was performed (Figure 49), and while mucosubstances were increased
in cobalt-treated strains, all seemed similar to one another.
Cytokine analysis of whole lung lysate by bead array
Cytokines were analyzed from whole lung lysate (WLL) generated from right lung lobes
flash-frozen in liquid nitrogen at necropsy (Figures 50, 51 and 52). There was a general
agreement between the HIF1α and HIF1α/2α mice in that IL-4 and IL-5 were both increased in
these strains treated with cobalt (Figure 50A and B). Additionally, the presence of high levels of
IL-6 in cobalt-treated HIF2α mice and low levels of IL-6 in the HIF1α and HIF1α/2α strains
(Figure 51A) demonstrate these strains could be following disparate cytokine profiles which
could be contributing to eosinophilia seen at 5 days in the mice lacking HIF1α or HIF1α/2α but
not HIF2α alone. Along these lines, the presence of IL-17 in HIF1α/2α-deficient mice treated
with notable (Figure 51B). Levels of IFNγ (Figure 52A) had some significant changes but did not
follow any observable pattern across groups. TNFα levels (Figure 52B) did not show any
changes across groups.
Gene expression analyses
Total lung mRNA was extracted from frozen right lung lobes and cDNA was generated
from 1 μg mRNA. Quantitative real time PCR was performed for several genes known to have
roles in lung inflammation and/or HIF-signaling using SYRB green as a detector (Figure 53).

135

Figure 48

SAL
HIF1/2αfl/fl

COB
HIF2αfl/fl

COB
HIF1/2αfl/fl

CTL

DOX

Figure 48. MBP IHC from HIF2α and HIF1α/2α mice at 5 days cobalt exposure. Lungs were
immunohistochemically stained for MBP (red) and counterstained with light hematoxylin.
Eosinophils (Eos) were present in all infiltrates of cobalt-treated mice (stars, C-F). Higher
concentrations of Eos were within some infiltrates (dashed arrows) compared to more mixed
inflammatory infiltrates (solid arrows). Black line indicates 200 μm.

136

Figure 49

SAL
HIF1/2αfl/fl

COB
HIF2αfl/fl

COB
HIF1/2αfl/fl

CTL

DOX

Figure 49. AB-PAS stain from HIF2α and HIF1α/2α mice at 5 days cobalt exposure. Lungs were
stained with Alcian Blue – Periodic Acid Schiff to visualize production of mucosubstances, and
counterstained with light hematoxylin. Black line indicates 50 μm. Solid arrows indicate
location of mucosubstances in the axial airway.

137

Figure 50

A

IL-4

pg/mL

0.3

*,a,b

0.2

0.1
0.0
HIF1α

B

HIF2α

HIF1/2α

IL-5
8.0

**

pg/mL

6.0

*

4.0

*

2.0
0.0

HIF1α

HIF2α

HIF1/2α

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 50. IL-4 and IL-5 cytokines from WLL of HIF1α, HIF2α and HIF1α/2α mice after 5 days
cobalt exposure. Right lung lobe homogenate was analyzed via cytokine bead array for IL-4 (A)
and IL-5 (B). One-way ANOVA was performed and significance using Fisher’s LSD of α=0.05 is
indicated by: * (from saline treated control), a (from CTL or ∆/∆ counterpart within strain), b
(from same treatment across HIF strains). N=5-8 mice/group.

138

Figure 51

A

pg/mL

IL-6
*,a,b

160
140
120
100
80
60
40
20
0
HIF1α

B

HIF2α

IL-17

0.6
0.5

pg/mL

HIF1/2α

b

0.4

a,b

0.3

0.2
0.1
0.0
HIF1α

HIF2α

HIF1/2α

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 51. IL-6 and IL-17 cytokines from WLL of HIF1α, HIF2α and HIF1α/2α mice after 5 days
cobalt exposure. Right lung lobe homogenate was analyzed via cytokine bead array for IL-6 (A)
and IL-17 (B). One-way ANOVA was performed and significance using Fisher’s LSD of α=0.05 is
indicated by: * (from saline treated control), a (from CTL or ∆/∆ counterpart within strain), b
(from same treatment across HIF strains). N=5-8 mice/group.

139

Figure 52

A

IFNγ
6.0

pg/mL

5.0

a,b

a

4.0

3.0
2.0
1.0

0.0
HIF1α

B

HIF2α

HIF1/2α

TNFα

5.0

pg/mL

4.0

3.0
2.0
1.0

0.0
HIF1α

HIF2α

HIF1/2α

CTL – Saline

CTL – 60 μg Cobalt

∆/∆ – Saline

∆/∆ – 60 μg Cobalt

Figure 52. IFNγ and TNFα cytokines from WLL of HIF1α, HIF2α and HIF1α/2α mice after 5
days cobalt exposure. Right lung lobe homogenate was analyzed via cytokine bead array for
IFNγ (A) and TNFα (B). One-way ANOVA was performed and significance using Fisher’s LSD of
α=0.05 is indicated by: * (from saline treated control), a (from CTL or ∆/∆ counterpart within
strain), b (from same treatment across HIF strains). N=5-8 mice/group.

140

Figure 53

HIF2α HIF1/2α

Δ/Δ - Saline
CTL - 60 µg Cobalt
Δ/Δ - 60 µg Cobalt
Δ/Δ - Saline
CTL - 60 µg Cobalt
Δ/Δ - 60 µg Cobalt
Δ/Δ - Saline
CTL - 60 µg Cobalt
Δ/Δ - 60 µg Cobalt

HIF1α

IL-1β
Bnip3
Csf2
Adora2b
IL-25
Tslp
IFN-γ
IL-33
Hmox1
Ccl17
Arg1
IL-6
IL-4
IL-2
IL-13
p65/RelA
IL-17A
Ccl20
iNOS
IL-5
Figure 53. WLL gene expression profiles of HIF1α, HIF2α and HIF1α/2α mice with 5 days
cobalt exposure. Heat map and cluster analysis of the gene expression data are shown. Within
genotype, each group was normalized to the CTL-Saline group. Red indicates relative upregulation compared to CTL-Saline group, whereas green indicates relative down-regulation.

141

According to the phenotype of eosinophilia, we anticipated the HIF1α
similarly to HIF1α/2α

Δ/Δ

mice, but the HIF1α/2α

Δ/Δ

Δ/Δ

mice to behave

mice seemed to have their own unique

gene expression profile. IL-4, IL-5 and IL-13, classic TH2 cytokines, clustered together, though
not in a way that agrees with the phenotype.

Discussion
These studies are the first to investigate the role of lung epithelial HIF2α in cobaltinduced lung inflammation. Though HIF2α shares sequence homology with HIF1α, the
differential expression of HIF2α, notably the lung-specific presence of large amounts of HIF2α
mRNA transcript, suggests that they play unique roles [189, 190]. This idea is further supported
by unpublished results in our lab from in utero knockouts of HIF1α and HIF2α, and though these
roles are almost certainly a result of transcription factor networks engaged in lung
development, the differences are notable. The data presented here supports the idea that
HIF1α and HIF2α play similar, but not completely overlapping roles in cobalt-induced lung
inflammation.

Interestingly, eosinophilia was induced in cobalt treated HIF2α
which is a similar response to that seen at 5 days in HIF1α

Δ/Δ

Δ/Δ

mice at 14 days,

mice. Explanations for this

eosinophilia might lie in cytokine secretion such as IL-5. In studies of HIF1α by Saini et al., IL-5
increases very early (after 1 day of exposure to cobalt) in the HIF1α

142

Δ/Δ

mice compared to

controls, which is soon followed at the day 5 time point by large eosinophil populations in these
Δ/Δ

lungs [214]. In HIF2α

mice, IL-5 is also induced early at 1 day (Figure 40B), though the higher

eosinophil BAL counts do not occur until later at 14 days (Figure 35A) and IHC results (Figure
38H). Whether IL-5 represents a functional link between HIF1α

Δ/Δ

and HIF2α

Δ/Δ

phenotypes

remains to be determined by future studies, though it is tempting to believe that a common
target of HIF1α and HIF2α (that is not induced or repressed normally in recombined mice) is
Δ/Δ

responsible for the induction of IL-5 in both HIF1α
Δ/Δ

increased early in HIF2α

and HIF2α

Δ/Δ

mice. Other cytokines were

mice treated with cobalt, including IL-6 and KC, though these

cytokines do not necessarily have a direct role in eosinophilia. IL-6 induces TH2 maturation
through the production of IL-4 and suppress TH1 maturation [296]. In the HIF1α mice treated
with cobalt, however, IL-6 was only induced strongly in control mice treated with cobalt [214].
Histopathologically it appears that at 14 days of cobalt treatment, control mice have
mild recovery of overall injury compared to HIF2α

Δ/Δ

mice (Figure 37D and H, respectively).

This observation is supported by total cell counts (Figure 34A). Mucous substances also seem
to decrease for control mice from 5 days to 14 days (Figure 39C and D, respectively) compared
Δ/Δ

to HIF2α

mice which show increased mucous substances from 5 to 14 days (Figure 39G and

H, respectively). When comparing these results to our previous studies of HIF1α

Δ/Δ

mice, we

see very similar amounts of total cells at 14 days for both controls and HIF-deficient mice,
indicating that a slight recovery or normalization is achieved at the 14 day time point which is

143

Δ/Δ

prevented in both HIF1α

Δ/Δ

and HIF2α

mice. It will be interesting to see whether these

histopathologic observations of recovery in the controls become important in future
investigations of the mechanisms of these injury dynamics are elucidated.

Comparing the combination HIF1α/2α
Δ/Δ

from HIF1α

Δ/Δ

mice to HIF2α

Δ/Δ

mice (and published results

mice) we see an eosinophilia similar to that observed by HIF1α

apparent at the 5 day time point, whereas the HIF2α

Δ/Δ

Δ/Δ

mice

mice are not yet displaying eosinophilia

at 5 days (Figure 45A and 48D). The timing of this phenotype suggests that HIF1α-deficiency is
dominant in inducing an earlier eosinophilia compared to HIF2α-deficiency, which could have
implications for specificity of HIF1α and HIF2α target genes. Gene expression profiles to date
do not share this phenotypic similarity, however, and HIF1α/2α

Δ/Δ

mice show a unique

expression signature at 5 days (Figure 53). Upregulation of IL-17 expression (Figure 53)
correlates with increased levels of this cytokine in WLL (Figure 51B) in HIF1α/2α

Δ/Δ

mice,

suggesting TH17 cells may be contributing to this strain’s phenotype. Despite the differences
between HIF1α/2α mice and the others, the TH2 cytokines IL-4, IL-5 and IL-13 clustered
together, away from IFNγ. Further verification of these data, possibly through whole genome
arrays of the postnatal HIF1α, HIF2α and combo HIF1α/2α alveolar type II cells will potentially
yield insight into the time-dependent changes in gene expression and may identify targets or
target networks responsible for influencing the phenotype.

144

These studies of HIF1α and HIF2α improve our understanding of the roles that these
transcription factors play in cobalt-induced lung injury. Primarily, we have learned that lack of
these transcription factors can alter the inflammatory and immune response, predisposing
Δ/Δ

tissues to an eosinophilic inflammation (and TH2 in HIF1α

mice) correlating with IL-5

expression. At the very least this prompts further investigation into the effect that loss of
HIF1α and HIF2α in epithelial cells may have in other allergic and inflammatory diseases
involving the epithelium such as inflammatory bowel disease and eczema. The alteration of the
immune response based on presence or absence of HIFαs, as well as their dual roles in
targeting pro-survival and pro-death genes suggests that overexpression of HIFs may also be
detrimental, as targeting the HIFs pharmacologically has shown mixed results [211].
Additionally, different knockout systems have shown contradictory roles for HIFs in epithelial
cells (our data), myeloid cells [200], and more widespread knockouts [212]. We have learned
that HIF1α and HIF2α deletion in the lung epithelium using our inducible knockout system is not
protective to cobalt injury per se as previously thought to be the case from in vitro data [285,
297]. In retrospect it is not surprising that the HIFα-deficiency in a specific group of cells would
be complicated, yet the results clearly warrant further investigation and clarification.
There are several limitations to the current studies. First and foremost, our
investigations to date on all HIFα-deficient mice have focused on whole lung analyses (either
protein or RNA), which, while useful, may not reveal meaningful insights into the mechanism of
the eosinophilic phenotypes observed. More targeted analysis of ATII cells, such as through
primary isolation, both early in postnatal development and later during cobalt injury will be

145

helpful in determining the sequence of events leading to eosinophilia. Further clarification of
which cytokines are involved and their source is warranted now more than ever, as the
repertoire of cytokines and chemokines that the epithelium is capable of secreting that
increases the likelihood of eosinophilia has grown substantially in recent years. These include
but are not limited to: IL-25, IL-33, TSLP, Eotaxin1-3, GM-CSF, Ccl20, β-defensin, IL-1β,
osteopontin, TARC (Ccl17), PARC (Ccl18) and MDC (Ccl22), which are the target of active
investigation by our lab.
The second limitation to these studies, especially in regards to the combination
HIF1α/2α mice, was lack of similar time-course data than that for the HIF2α mice. Focusing on
Δ/Δ

5 days was considered the best starting point, but many more gaps exist. Since HIF2α

mice

follow a delayed eosinophilia, mechanistic questions may be better answered by comparing all
strains at their peak eosinophilic time point, such as 14 day exposures from HIF2α mice to 5 day
exposures of HIF1α and HIF1α/2α mice. Additionally, in either strain, a 2 day time point and/or
10 day time point may provide more detailed information about the leukocyte dynamics in
these lungs.
The impact of these studies on our understanding of HMLD is difficult primarily because
the animal model we have chosen to incorporate has several important differences to that of
HMLD. Firstly, our model system utilizes cobalt that is highly water soluble (and thus
distribution throughout the body and away from the lung occurs much faster than cobalt seen
in tungsten carbide) and does not incorporate particulates, which have been theorized to
contribute to ROS production in the prolonged presence of cobalt [270]. Secondly, mimicry of

146

HMLD probably requires more chronic exposures than 14 days, since exposed people have a
history of chronic occupational exposures followed by many years of being asymptomatic
before presenting with pulmonary symptoms. Finally, while the allergic sensitivity of cobalt has
been defined as a separate entity than HMLD, this allergic component to cobalt exposure is
more prevalent than HMLD, and it is difficult to determine with confidence whether (and to
what extent) these two processes are distinct.
Thus, with the several differences between HMLD and the chosen model, and with a
majority of the investigations focusing on the altered inflammatory response to cobalt with and
without HIF1α and/or HIF2α, the clinical impact of these findings to patients with HMLD is
difficult to predict. It is safe to say that while HMLD was the primary driver of developing this
model system, incorporating particulates into this model was impractical for several reasons,
and cobalt was used more as an inducer of an inflammatory response than for mimicry of
HMLD. Impact of this research model on our understanding of HMLD will be higher if
particulates and more chronic exposures are incorporated in future studies. The most
important aspect of the investigations here was the altered eosinophilic infiltration, which
highlights the importance of epithelial signaling in recruitment of inflammatory and immune
cells into the lung. These dynamics must be considered in any future study of HMLD and
cobalt-induced lung injury.

147

CHAPTER 5 – SUMMARY AND CONCLUSIONS

148

Summary of Research
The work I have presented here clarifies the role of lung-epithelial specific HIF1αdeficiency in cobalt-induced lung injury. First, I have investigated the timing of HIF1α deletion
necessary for eosinophilic predisposition, and also characterized the T cell populations in
cobalt-treated control and HIF1α

Δ/Δ

mice. I have further performed preliminary investigations

into the roles of adenosine receptor A2B and NF-κB signaling in the creation of the eosinophilic
phenotype of HIF1α

Δ/Δ

mice. Investigations into the time course of HIF2α-deficiency as well as

the combination HIF1α/2α

Δ/Δ

mice at 5 days in cobalt-induced lung injury illustrate the active

role that these proteins play in directing the immune response in the lung. Finally, these results
have strong implications for overlap of cobalt asthma and hard metal lung disease, namely that
alteration of HIFα signaling can alter the nature and timing of the immune response to cobalt.
In this way, epithelial HIFαs represent a potential key bridge and control point for pathogenesis
of both cobalt asthma and HMLD.

Major findings and implications
1. Deletion of HIF1α in the early postnatal development period (P4-14 and P4-30) is
required for the eosinophilic phenotype of cobalt-induced lung inflammation. This
finding is supported by other work in our lab utilizing an allergic airway disease model,
and may implicate HIF1α as an important target gene for the hygiene hypothesis.
2. Eosinophilic inflammation seen in HIF1α

Δ/Δ

mice is driven in part by an increase in TH2

cells. In contrast, control mice treated with cobalt do not display TH1 responses.

149

3. Influence of NF-κB and Adora2b signaling on the HIF1α

Δ/Δ

phenotype are inconclusive,

and require further clarification and characterization.
4. Loss of HIF2α in lung epithelium also produces an eosinophilic phenotype following a 14
Δ/Δ

day occupational exposure. This is different from the HIF1α
Δ/Δ

eosinophilia after 5 days. Both HIF1α

Δ/Δ

mice and HIF2α

mice, which display

mice showed increased IL-5

cytokine levels in BAL and thus IL-5 may be a key molecular effector of these responses.
5. Loss of both HIF1α and HIF2α in lung epithelium has a phenotype similar to HIF1α

Δ/Δ

mice, with eosinophilia after 5 days of cobalt exposure, indicating that HIF1α-deficiency
drives an earlier eosinophilic infiltration.

Proposed Working Model
Given the findings and implications mentioned previously, a working model is given in
Figure 54. HIF2α-deficiency in lung epithelium behaves in a similar fashion as HIF1α-deficiency
during cobalt induced lung inflammation, confirming that both isoforms play a role in altering
inflammatory responses. Upon inflammatory stimulation from cobalt, crosstalk with other
inflammatory signaling proteins or transcription factors directs transcription of normal
cytokines such as TNFα, IL-1, and IL-8 to produce a primarily neutrophilic and macrophagedriven response (Figure 54A). In contrast, lung epithelium lacking HIFαs does not participate in
crosstalk, which generates a unique set of cytokines such as IL-5, TSLP, IL-25 and IL-33, which
recruit eosinophils and probably involves innate lymphoid type 2 cells (ILC2s) (Figure 54B).

150

Figure 54

Cobalt

A

Wild Type

HIFαΔ/Δ

Cobalt

ATII Cell

ATII Cell
NFκB +
Innate
immune
signals

NFκB +
Innate
immune
signals

HIF1α
HIF2α

Alternative
Targets

Normal
Targets

Pro-TH2 Cytokines,
Eosinophil recruitment
Innate Lymphoid Cell 2

Normal cytokines and
PMN recruitment

IL-8

B

IL-1

TNF

IL-5 TSLP
IL-25 IL-33

PMNs

TH 2

Eos
ILC2

Macs
Figure 54. Working model of WT and HIFα

Δ/Δ

lung epithelial cells exposed to cobalt. In normal cells exposed to an inflammatory

151

Figure 54 (cont’d)
stimulus, HIF1α and HIF2α have crosstalk with inflammatory signaling molecules, their transcriptional partners, or the products of
the combined signaling to produce a neutrophilic and macrophage-dominant response (A). In contrast, lung epithelium lacking
HIF1α or HIF2α creates an unbalanced transcriptional profile which leads to secretion of IL-5 and other pro-TH2 cytokines which
may involve innate lymphoid type 2 cells (ILC2s) to promote a primarily eosinophilic response (B).

152

Knowledge gaps and future studies
The most difficult obstacle in attempting to determine any mechanism caused by a celltype specific deletion is that initial analyses typically include whole tissues, which could
potentially confound data. We are actively performing isolations of alveolar type II cells from
our mice to obtain a more direct assessment of HIF1α- and HIF2α-dependent gene expression
changes in these cells. These studies have the greatest potential to yield mechanistic
information about the start of the eosinophilic phenotype, assuming pure populations can be
isolated. They will also help determine the recombination efficiency of the SP-C/Cre model.
Another useful approach to characterizing the eosinophilic inflammation in these mice is
to measure the immune cell populations in the lung. Recent studies have indicated the
existence and functional relevance of several new classes of immune cells in lung inflammation
and allergic disease. One of the cell types, called nuocytes or innate lymphoid cells (ILCs) have
been shown to respond to epithelial inflammatory signals. In the case of ILC2s (the TH2 version
of this population), high levels of IL-4, IL-5 and IL-13 are secreted despite lacking recombined
surface receptors (TCR) and are important for early responses to helminth infection [97].
Considering that initial studies of dendritic cells did not show many changes in populations
(Greenwood, unpublished results), and that these ILC2 cells are capable of inducing eosinophilic
inflammation, they are likely important players to consider in our model.
In addition to further exploration into ILC2s/nuocytes, ATII cells have recently been
shown to express MHCII and participate in immune tolerance by inducing FoxP3+TREG cells

153

[122]. Immune tolerance (forming active TREG cells) depends on presentation of antigens to
APCs without additional inflammatory stimuli. If the pool of these APCs decreases
considerably, it can in theory decrease TREGs and thus increase the likelihood that tolerance is
broken (active T-cell responses are initiated from innocuous stimuli). It is possible that loss of
HIFα subunits in ATII cells may be directly impacting their survival and function from a
metabolic standpoint, which secondarily prevents their ability to induce tolerance in the lung in
the early postnatal time period. ATII cells require large amounts of energy to synthesize
surfactant and participate in fluid transport during pulmonary edema, so disruption of
metabolic adaptations to hypoxia by loss of HIFαs could certainly occur. Probing the lungs of
HIFα-deficient mice for overall populations of ATII cells and any notable morphologic
differences will be important for future studies.
There are several knowledge gaps in the crosstalk of NF-κB and Adora2b in our model,
and further characterization of these pathways, especially in isolated ATII cells, will clarify their
relevance to the eosinophilic phenotype of HIFα loss. Dedicated studies with the use of several
pharmacologic inhibitors and stimulators for either pathway, as well as isolating cell types (such
as through lavage vs. lung mash, specific isolation of ATII cells or the use of in vitro systems to
supplement in vivo studies) will likely be required to probe further into their relevance in our
model of cobalt-induced lung inflammation.
The other noteworthy aspect of our studies is the use of a broad spectrum antibiotic,
doxycycline, to induce recombination of HIFαs. While we have performed controls with DOX-

154

only animals to establish eosinophilia dependence in HIF1α

fl/fl

mice on all 3 transgenes and

DOX, clinical and experimental evidence is mounting regarding the impact of the microbiome
and the use of antibiotics in early life having substantial impact on severity of allergic diseases.
While the mechanism of antibiotic interference in allergic diseases is still unknown and under
investigation, use of antibiotics besides doxycycline (non-tetracycline antibiotics) as well as
differing housing environments (specific pathogen free conditions or cleaner) may help to put
the effects of HIF1α loss in early postnatal development into context with these other variables
receiving attention in clinical studies.

155

REFERENCES

156

REFERENCES
1.

Proctor, D. F., A history of breathing physiology. Lung biology in health and disease.
1995, New York: M. Dekker. xxii, 391 p.

2.

John Mayow (1641-1679). JAMA, 1966. 197(5): p. 364-5.

3.

Priestley, J., An Account of Further Discoveries in Air. By the Rev. Joseph Priestley, LL.D. F.
R. S. in Letters to Sir John Pringle, Bart. P. R. S. and the Rev. Dr. Price, F. R. S.
Philosophical Transactions, 1775. 65: p. 384-394.

4.

Cook, G. A. L., Carol M., Oxygen, in The encyclopedia of the chemical elements, Hampel,
C. A., Editor. 1968, Reinhold Book Corp.: New York,. p. 499-512.

5.

Severinghaus, J. W., Fire-Air and Dephlogistication - Revisionisms of oxygen's discovery,
in Hypoxia : through the lifecycle - Advances in experimental medicine and biology,
Roach, R. C., P. D. Wagner, and P. H. Hackett, Editors. 2003, Kluwer Academic/Plenum
Publishers: New York. p. 7-19.

6.

Anders, E. and M. Ebihara, Solar-system abundances of the elements. Geochimica et
Cosmochimica Acta, 1982. 46(11): p. 2363-2380.

7.

Lyons, T. W., Palaeoclimate: oxygen's rise reduced. Nature, 2007. 448(7157): p. 1005-6.

8.

Farquhar, J., A. L. Zerkle, and A. Bekker, Geological constraints on the origin of oxygenic
photosynthesis. Photosynth Res, 2011. 107(1): p. 11-36.

9.

Lyons, T. W. and C. T. Reinhard, Early Earth: Oxygen for heavy-metal fans. Nature, 2009.
461(7261): p. 179-81.

10.

Powell, F. L., Studying biological responses to global change in atmospheric oxygen.
Respir Physiol Neurobiol, 2010. 173 Suppl: p. S6-12.

11.

Berner, R. A., J. M. Vandenbrooks, and P. D. Ward, Evolution. Oxygen and evolution.
Science, 2007. 316(5824): p. 557-8.

12.

Thannickal, V. J., Oxygen in the evolution of complex life and the price we pay. Am J
Respir Cell Mol Biol, 2009. 40(5): p. 507-10.

13.

Ward, P. D., Out of thin air : dinosaurs, birds, and Earth's ancient atmosphere. 2006,
Washington, D.C.: J. Henry Press. xiii, 282 p.

14.

Champe, P. C., R. A. Harvey, and D. R. Ferrier, Biochemistry. 3rd ed. Lippincotts'
illustrated reviews. 2005, Philadelphia: Lippincott/Williams & Wilkins. x, 534 p.

157

15.

Maina, J. N., Structure, function and evolution of the gas exchangers: comparative
perspectives. J Anat, 2002. 201(4): p. 281-304.

16.

Forster, T. D. and H. A. Woods, Mechanisms of tracheal filling in insects. Biol Rev Camb
Philos Soc, 2013. 88(1): p. 1-14.

17.

Powell, F. L. and S. R. Hopkins, Comparative physiology of lung complexity: implications
for gas exchange. News Physiol Sci, 2004. 19: p. 55-60.

18.

West, J. B., Respiratory physiology : the essentials. 9th ed. 2012, Philadelphia: Wolters
Kluwer Health/Lippincott Williams & Wilkins. viii, 200 p.

19.

Staub, N. C., Basic respiratory physiology. 1991, New York: Churchill Livingstone. ix, 242
p.

20.

Rhoades, R. and G. A. Tanner, Medical physiology. 2nd ed. 2003, Philadelphia: Lippincott
Williams & Wilkins. x, 781 p.

21.

Weibel, E. R., The pathway for oxygen : structure and function in the mammalian
respiratory system. 1984, Cambridge, Mass.: Harvard University Press. xv, 425 p.

22.

Tyler, W. S. and M. D. Julian, Gross and subgross anatomy of lungs, pleura, connective
tissue, septa, distal airways, and structural units, in Treatise on pulmonary toxicology:
Comparative Biology of the Normal Lung, Parent, R. A., Editor. 1992, CRC Press: Boca
Raton. p. v. <1 >.

23.

Rackley, C. R. and B. R. Stripp, Building and maintaining the epithelium of the lung. J Clin
Invest, 2012. 122(8): p. 2724-30.

24.

Franks, T. J., T. V. Colby, W. D. Travis, R. M. Tuder, H. Y. Reynolds, A. R. Brody, W. V.
Cardoso, R. G. Crystal, C. J. Drake, J. Engelhardt, M. Frid, E. Herzog, R. Mason, S. H. Phan,
S. H. Randell, M. C. Rose, T. Stevens, J. Serge, M. E. Sunday, J. A. Voynow, B. M.
Weinstein, J. Whitsett, and M. C. Williams, Resident cellular components of the human
lung: current knowledge and goals for research on cell phenotyping and function. Proc
Am Thorac Soc, 2008. 5(7): p. 763-6.

25.

Basbaum, C. B., B. Jany, and W. E. Finkbeiner, The serous cell. Annu Rev Physiol, 1990.
52: p. 97-113.

26.

Rock, J. R., M. W. Onaitis, E. L. Rawlins, Y. Lu, C. P. Clark, Y. Xue, S. H. Randell, and B. L.
Hogan, Basal cells as stem cells of the mouse trachea and human airway epithelium.
Proc Natl Acad Sci U S A, 2009. 106(31): p. 12771-5.

27.

Hong, K. U., S. D. Reynolds, S. Watkins, E. Fuchs, and B. R. Stripp, Basal cells are a
multipotent progenitor capable of renewing the bronchial epithelium. Am J Pathol, 2004.
164(2): p. 577-88.
158

28.

Linnoila, R. I., Functional facets of the pulmonary neuroendocrine system. Lab Invest,
2006. 86(5): p. 425-44.

29.

Song, H., E. Yao, C. Lin, R. Gacayan, M. H. Chen, and P. T. Chuang, Functional
characterization of pulmonary neuroendocrine cells in lung development, injury, and
tumorigenesis. Proc Natl Acad Sci U S A, 2012. 109(43): p. 17531-6.

30.

Reynolds, S. D. and A. M. Malkinson, Clara cell: progenitor for the bronchiolar
epithelium. Int J Biochem Cell Biol, 2010. 42(1): p. 1-4.

31.

Irwin, R. S., N. Augustyn, C. T. French, J. Rice, V. Tedeschi, S. J. Welch, and T. Editorial
Leadership, Spread the word about the journal in 2013: from citation manipulation to
invalidation of patient-reported outcomes measures to renaming the Clara cell to new
journal features. Chest, 2013. 143(1): p. 1-4.

32.

Winkelmann, A. and T. Noack, The Clara cell: a "Third Reich eponym"? Eur Respir J, 2010.
36(4): p. 722-7.

33.

Jobe, A. H. and M. Ikegami, Biology of surfactant. Clin Perinatol, 2001. 28(3): p. 655-69,
vii-viii.

34.

Fehrenbach, H., Alveolar epithelial type II cell: defender of the alveolus revisited. Respir
Res, 2001. 2(1): p. 33-46.

35.

Lyons, C. R. and M. F. Lipscomb, Alveolar macrophages in pulmonary immune responses.
I. Role in the initiation of primary immune responses and in the selective recruitment of T
lymphocytes to the lung. J Immunol, 1983. 130(3): p. 1113-9.

36.

Fels, A. O. and Z. A. Cohn, The alveolar macrophage. J Appl Physiol, 1986. 60(2): p. 35369.

37.

Reynolds, H. Y., Lung inflammation and fibrosis: an alveolar macrophage-centered
perspective from the 1970s to 1980s. Am J Respir Crit Care Med, 2005. 171(2): p. 98-102.

38.

Cardoso, W. V., Lung morphogenesis, role of growth factors and transcription factors, in
The lung : development, aging, and environment, Harding, R., P. K.E., and C. G. Plopper,
Editors. 2003, Acadmeic Press: San Diego, CA. p. p.3-11.

39.

Maeda, Y., V. Dave, and J. A. Whitsett, Transcriptional control of lung morphogenesis.
Physiol Rev, 2007. 87(1): p. 219-44.

40.

Burri, P. H., Structural aspects of postnatal lung development - alveolar formation and
growth. Biol Neonate, 2006. 89(4): p. 313-22.

41.

Zeltner, T. B. and P. H. Burri, The postnatal development and growth of the human lung.
II. Morphology. Respir Physiol, 1987. 67(3): p. 269-82.

159

42.

Narayanan, M., J. Owers-Bradley, C. S. Beardsmore, M. Mada, I. Ball, R. Garipov, K. S.
Panesar, C. E. Kuehni, B. D. Spycher, S. E. Williams, and M. Silverman, Alveolarization
continues during childhood and adolescence: new evidence from helium-3 magnetic
resonance. Am J Respir Crit Care Med, 2012. 185(2): p. 186-91.

43.

Kuczmarski, R. J., C. L. Ogden, S. S. Guo, L. M. Grummer-Strawn, K. M. Flegal, Z. Mei, R.
Wei, L. R. Curtin, A. F. Roche, and C. L. Johnson, 2000 CDC Growth Charts for the United
States: methods and development. Vital Health Stat 11, 2002(246): p. 1-190.

44.

Network, S. S. G. o. t. E. K. S. N. N. R., W. A. Carlo, N. N. Finer, M. C. Walsh, W. Rich, M.
G. Gantz, A. R. Laptook, B. A. Yoder, R. G. Faix, A. Das, W. K. Poole, K. Schibler, N. S.
Newman, N. Ambalavanan, I. D. Frantz, 3rd, A. J. Piazza, P. J. Sanchez, B. H. Morris, N.
Laroia, D. L. Phelps, B. B. Poindexter, C. M. Cotten, K. P. Van Meurs, S. Duara, V.
Narendran, B. G. Sood, T. M. O'Shea, E. F. Bell, R. A. Ehrenkranz, K. L. Watterberg, and R.
D. Higgins, Target ranges of oxygen saturation in extremely preterm infants. N Engl J
Med, 2010. 362(21): p. 1959-69.

45.

Tschudy, M. M., K. M. Arcara, and Johns Hopkins Hospital. Children's Medical and
Surgical Center., The Harriet Lane handbook : a manual for pediatric house officers. 19th
ed. 2012, Philadelphia, PA: Mosby Elsevier. xvi, 1132 p., 10 p. of plates.

46.

Martin, R. J. and A. A. Fanaroff, The Preterm Lung and Airway: Past, Present, and Future.
Pediatr Neonatol, 2013.

47.

Crump, C., M. A. Winkleby, J. Sundquist, and K. Sundquist, Risk of asthma in young
adults who were born preterm: a Swedish national cohort study. Pediatrics, 2011.
127(4): p. e913-20.

48.

Baraldi, E. and M. Filippone, Chronic lung disease after premature birth. N Engl J Med,
2007. 357(19): p. 1946-55.

49.

Jaakkola, J. J., P. Ahmed, A. Ieromnimon, P. Goepfert, E. Laiou, R. Quansah, and M. S.
Jaakkola, Preterm delivery and asthma: a systematic review and meta-analysis. J Allergy
Clin Immunol, 2006. 118(4): p. 823-30.

50.

Asikainen, T. M. and C. W. White, Antioxidant defenses in the preterm lung: role for
hypoxia-inducible factors in BPD? Toxicol Appl Pharmacol, 2005. 203(2): p. 177-88.

51.

Greenwood, K. K., S. P. Proper, Y. Saini, L. A. Bramble, D. N. Jackson-Humbles, J. G.
Wagner, J. R. Harkema, and J. J. LaPres, Neonatal epithelial hypoxia inducible factor1alpha expression regulates the response of the lung to experimental asthma. Am J
Physiol Lung Cell Mol Physiol, 2012. 302(5): p. L455-62.

52.

Christensen, T. G., The distribution and function of peroxidases in the respiratory tract.
Surv Synth Pathol Res, 1984. 3(3): p. 201-18.

160

53.

Travis, S. M., B. A. Conway, J. Zabner, J. J. Smith, N. N. Anderson, P. K. Singh, E. P.
Greenberg, and M. J. Welsh, Activity of abundant antimicrobials of the human airway.
Am J Respir Cell Mol Biol, 1999. 20(5): p. 872-9.

54.

Zhang, P., W. R. Summer, G. J. Bagby, and S. Nelson, Innate immunity and pulmonary
host defense. Immunol Rev, 2000. 173: p. 39-51.

55.

Wanner, A., M. Salathe, and T. G. O'Riordan, Mucociliary clearance in the airways. Am J
Respir Crit Care Med, 1996. 154(6 Pt 1): p. 1868-902.

56.

Chilvers, M. A. and C. O'Callaghan, Local mucociliary defence mechanisms. Paediatr
Respir Rev, 2000. 1(1): p. 27-34.

57.

Mizgerd, J. P., Acute lower respiratory tract infection. N Engl J Med, 2008. 358(7): p. 71627.

58.

Akira, S., S. Uematsu, and O. Takeuchi, Pathogen recognition and innate immunity. Cell,
2006. 124(4): p. 783-801.

59.

Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. Van Huffel, X. Du, D. Birdwell, E. Alejos, M.
Silva, C. Galanos, M. Freudenberg, P. Ricciardi-Castagnoli, B. Layton, and B. Beutler,
Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene.
Science, 1998. 282(5396): p. 2085-8.

60.

Kato, A. and R. P. Schleimer, Beyond inflammation: airway epithelial cells are at the
interface of innate and adaptive immunity. Curr Opin Immunol, 2007. 19(6): p. 711-20.

61.

Diamond, G., D. Legarda, and L. K. Ryan, The innate immune response of the respiratory
epithelium. Immunol Rev, 2000. 173: p. 27-38.

62.

Hammad, H. and B. N. Lambrecht, Dendritic cells and airway epithelial cells at the
interface between innate and adaptive immune responses. Allergy, 2011. 66(5): p. 57987.

63.

Holgate, S. T., The sentinel role of the airway epithelium in asthma pathogenesis.
Immunol Rev, 2011. 242(1): p. 205-19.

64.

Gordon, S. B. and R. C. Read, Macrophage defences against respiratory tract infections.
Br Med Bull, 2002. 61: p. 45-61.

65.

Abbas, A. K., A. H. Lichtman, and S. Pillai, Cellular and molecular immunology. 6th ed.
2010, Philadelphia: Saunders/Elsevier. viii, 566 p.

66.

Mills, C. D., K. Kincaid, J. M. Alt, M. J. Heilman, and A. M. Hill, M-1/M-2 macrophages
and the Th1/Th2 paradigm. J Immunol, 2000. 164(12): p. 6166-73.

161

67.

Mosser, D. M. and J. P. Edwards, Exploring the full spectrum of macrophage activation.
Nat Rev Immunol, 2008. 8(12): p. 958-69.

68.

Dasgupta, P. and A. D. Keegan, Contribution of alternatively activated macrophages to
allergic lung inflammation: a tale of mice and men. J Innate Immun, 2012. 4(5-6): p. 47888.

69.

Byers, D. E. and M. J. Holtzman, Alternatively activated macrophages and airway
disease. Chest, 2011. 140(3): p. 768-74.

70.

Gordon, S. and P. R. Taylor, Monocyte and macrophage heterogeneity. Nat Rev
Immunol, 2005. 5(12): p. 953-64.

71.

Sica, A. and A. Mantovani, Macrophage plasticity and polarization: in vivo veritas. J Clin
Invest, 2012. 122(3): p. 787-95.

72.

Sibille, Y. and H. Y. Reynolds, Macrophages and polymorphonuclear neutrophils in lung
defense and injury. Am Rev Respir Dis, 1990. 141(2): p. 471-501.

73.

Kobayashi, S. D., J. M. Voyich, C. Burlak, and F. R. DeLeo, Neutrophils in the innate
immune response. Arch Immunol Ther Exp (Warsz), 2005. 53(6): p. 505-17.

74.

Lehrer, R. I., T. Ganz, M. E. Selsted, B. M. Babior, and J. T. Curnutte, Neutrophils and host
defense. Ann Intern Med, 1988. 109(2): p. 127-42.

75.

Marcinkiewicz, J., Neutrophil chloramines: missing links between innate and acquired
immunity. Immunol Today, 1997. 18(12): p. 577-80.

76.

Nelson, S., C. M. Mason, J. Kolls, and W. R. Summer, Pathophysiology of pneumonia. Clin
Chest Med, 1995. 16(1): p. 1-12.

77.

Nelson, S., G. J. Bagby, B. G. Bainton, L. A. Wilson, J. J. Thompson, and W. R. Summer,
Compartmentalization of intraalveolar and systemic lipopolysaccharide-induced tumor
necrosis factor and the pulmonary inflammatory response. J Infect Dis, 1989. 159(2): p.
189-94.

78.

Lekstrom-Himes, J. A. and J. I. Gallin, Immunodeficiency diseases caused by defects in
phagocytes. N Engl J Med, 2000. 343(23): p. 1703-14.

79.

Rosenzweig, S. D. and S. M. Holland, Phagocyte immunodeficiencies and their infections.
J Allergy Clin Immunol, 2004. 113(4): p. 620-6.

80.

Lee, J. J., E. A. Jacobsen, M. P. McGarry, R. P. Schleimer, and N. A. Lee, Eosinophils in
health and disease: the LIAR hypothesis. Clin Exp Allergy, 2010. 40(4): p. 563-75.

162

81.

Hogan, S. P., H. F. Rosenberg, R. Moqbel, S. Phipps, P. S. Foster, P. Lacy, A. B. Kay, and M.
E. Rothenberg, Eosinophils: biological properties and role in health and disease. Clin Exp
Allergy, 2008. 38(5): p. 709-50.

82.

Jacobsen, E. A., R. A. Helmers, J. J. Lee, and N. A. Lee, The expanding role(s) of
eosinophils in health and disease. Blood, 2012. 120(19): p. 3882-90.

83.

Abu-Ghazaleh, R. I., S. L. Dunnette, D. A. Loegering, J. L. Checkel, H. Kita, L. L. Thomas,
and G. J. Gleich, Eosinophil granule proteins in peripheral blood granulocytes. J Leukoc
Biol, 1992. 52(6): p. 611-8.

84.

Gleich, G. J., C. R. Adolphson, and K. M. Leiferman, The biology of the eosinophilic
leukocyte. Annu Rev Med, 1993. 44: p. 85-101.

85.

Malik, A. and J. K. Batra, Antimicrobial activity of human eosinophil granule proteins:
involvement in host defence against pathogens. Crit Rev Microbiol, 2012. 38(2): p. 16881.

86.

Rothenberg, M. E. and S. P. Hogan, The eosinophil. Annu Rev Immunol, 2006. 24: p. 14774.

87.

Lacy, P., S. Mahmudi-Azer, B. Bablitz, S. C. Hagen, J. R. Velazquez, S. F. Man, and R.
Moqbel, Rapid mobilization of intracellularly stored RANTES in response to interferongamma in human eosinophils. Blood, 1999. 94(1): p. 23-32.

88.

Guilliams, M., B. N. Lambrecht, and H. Hammad, Division of labor between lung dendritic
cells and macrophages in the defense against pulmonary infections. Mucosal Immunol,
2013.

89.

Merad, M., P. Sathe, J. Helft, J. Miller, and A. Mortha, The dendritic cell lineage:
ontogeny and function of dendritic cells and their subsets in the steady state and the
inflamed setting. Annu Rev Immunol, 2013. 31: p. 563-604.

90.

Hammad, H. and B. N. Lambrecht, Lung dendritic cell migration. Adv Immunol, 2007. 93:
p. 265-78.

91.

Murphy, K., P. Travers, M. Walport, and C. Janeway, Janeway's immunobiology. 8th ed.
2012, New York: Garland Science. xix, 868 p.

92.

Moore, B. B., T. A. Moore, and G. B. Toews, Role of T- and B-lymphocytes in pulmonary
host defences. Eur Respir J, 2001. 18(5): p. 846-56.

93.

Zhu, J. and W. E. Paul, CD4 T cells: fates, functions, and faults. Blood, 2008. 112(5): p.
1557-69.

163

94.

Trinchieri, G. and A. Sher, Cooperation of Toll-like receptor signals in innate immune
defence. Nat Rev Immunol, 2007. 7(3): p. 179-90.

95.

Hiscott, J., Convergence of the NF-kappaB and IRF pathways in the regulation of the
innate antiviral response. Cytokine Growth Factor Rev, 2007. 18(5-6): p. 483-90.

96.

Dela Cruz, C. S., M. J. Kang, W. K. Cho, and C. G. Lee, Transgenic modelling of cytokine
polarization in the lung. Immunology, 2011. 132(1): p. 9-17.

97.

Scanlon, S. T. and A. N. McKenzie, Type 2 innate lymphoid cells: new players in asthma
and allergy. Curr Opin Immunol, 2012. 24(6): p. 707-12.

98.

Paul, W. E. and J. Zhu, How are T(H)2-type immune responses initiated and amplified?
Nat Rev Immunol, 2010. 10(4): p. 225-35.

99.

Doerschuk, C. M., J. P. Mizgerd, H. Kubo, L. Qin, and T. Kumasaka, Adhesion molecules
and cellular biomechanical changes in acute lung injury: Giles F. Filley Lecture. Chest,
1999. 116(1 Suppl): p. 37S-43S.

100.

Mizgerd, J. P., Molecular mechanisms of neutrophil recruitment elicited by bacteria in
the lungs. Semin Immunol, 2002. 14(2): p. 123-32.

101.

Mukaida, N., Pathophysiological roles of interleukin-8/CXCL8 in pulmonary diseases. Am
J Physiol Lung Cell Mol Physiol, 2003. 284(4): p. L566-77.

102.

Menten, P., A. Wuyts, and J. Van Damme, Macrophage inflammatory protein-1.
Cytokine Growth Factor Rev, 2002. 13(6): p. 455-81.

103.

Chuquimia, O. D., D. H. Petursdottir, M. J. Rahman, K. Hartl, M. Singh, and C. Fernandez,
The role of alveolar epithelial cells in initiating and shaping pulmonary immune
responses: communication between innate and adaptive immune systems. PLoS One,
2012. 7(2): p. e32125.

104.

Becker, S., J. Quay, H. S. Koren, and J. S. Haskill, Constitutive and stimulated MCP-1, GRO
alpha, beta, and gamma expression in human airway epithelium and bronchoalveolar
macrophages. Am J Physiol, 1994. 266(3 Pt 1): p. L278-86.

105.

Van Coillie, E., J. Van Damme, and G. Opdenakker, The MCP/eotaxin subfamily of CC
chemokines. Cytokine Growth Factor Rev, 1999. 10(1): p. 61-86.

106.

Petrek, M., P. Pantelidis, A. M. Southcott, P. Lympany, P. Safranek, C. M. Black, V. Kolek,
E. Weigl, and R. M. du Bois, The source and role of RANTES in interstitial lung disease.
Eur Respir J, 1997. 10(6): p. 1207-16.

107.

Robays, L. J., T. Maes, S. Lebecque, S. A. Lira, W. A. Kuziel, G. G. Brusselle, G. F. Joos, and
K. V. Vermaelen, Chemokine receptor CCR2 but not CCR5 or CCR6 mediates the increase

164

in pulmonary dendritic cells during allergic airway inflammation. J Immunol, 2007.
178(8): p. 5305-11.
108.

Lee, H. C. and S. F. Ziegler, Inducible expression of the proallergic cytokine thymic
stromal lymphopoietin in airway epithelial cells is controlled by NFkappaB. Proc Natl
Acad Sci U S A, 2007. 104(3): p. 914-9.

109.

Angkasekwinai, P., H. Park, Y. H. Wang, Y. H. Wang, S. H. Chang, D. B. Corry, Y. J. Liu, Z.
Zhu, and C. Dong, Interleukin 25 promotes the initiation of proallergic type 2 responses. J
Exp Med, 2007. 204(7): p. 1509-17.

110.

Nakae, S., H. Morita, T. Ohno, K. Arae, K. Matsumoto, and H. Saito, Role of interleukin-33
in innate-type immune cells in allergy. Allergol Int, 2013. 62(1): p. 13-20.

111.

Stampfli, M. R., R. E. Wiley, G. S. Neigh, B. U. Gajewska, X. F. Lei, D. P. Snider, Z. Xing,
and M. Jordana, GM-CSF transgene expression in the airway allows aerosolized
ovalbumin to induce allergic sensitization in mice. J Clin Invest, 1998. 102(9): p. 1704-14.

112.

Xanthou, G., T. Alissafi, M. Semitekolou, D. C. Simoes, E. Economidou, M. Gaga, B. N.
Lambrecht, C. M. Lloyd, and V. Panoutsakopoulou, Osteopontin has a crucial role in
allergic airway disease through regulation of dendritic cell subsets. Nat Med, 2007.
13(5): p. 570-8.

113.

Vermaelen, K. Y., D. Cataldo, K. Tournoy, T. Maes, A. Dhulst, R. Louis, J. M. Foidart, A.
Noel, and R. Pauwels, Matrix metalloproteinase-9-mediated dendritic cell recruitment
into the airways is a critical step in a mouse model of asthma. J Immunol, 2003. 171(2):
p. 1016-22.

114.

Romagnani, S., Regulation of the T cell response. Clin Exp Allergy, 2006. 36(11): p. 135766.

115.

Hammad, H., H. H. Smits, C. Ratajczak, A. Nithiananthan, E. A. Wierenga, G. A. Stewart,
A. Jacquet, A. B. Tonnel, and J. Pestel, Monocyte-derived dendritic cells exposed to Der p
1 allergen enhance the recruitment of Th2 cells: major involvement of the chemokines
TARC/CCL17 and MDC/CCL22. Eur Cytokine Netw, 2003. 14(4): p. 219-28.

116.

Lloyd, C., Chemokines in allergic lung inflammation. Immunology, 2002. 105(2): p. 14454.

117.

Lloyd, C. M. and S. Saglani, Asthma and allergy: the emerging epithelium. Nat Med,
2010. 16(3): p. 273-4.

118.

Lambrecht, B. N. and H. Hammad, The airway epithelium in asthma. Nat Med, 2012.
18(5): p. 684-92.

165

119.

Ramakrishna, L., V. C. de Vries, and M. A. Curotto de Lafaille, Cross-roads in the lung:
immune cells and tissue interactions as determinants of allergic asthma. Immunol Res,
2012. 53(1-3): p. 213-28.

120.

Fritz, J. H., L. Le Bourhis, J. G. Magalhaes, and D. J. Philpott, Innate immune recognition
at the epithelial barrier drives adaptive immunity: APCs take the back seat. Trends
Immunol, 2008. 29(1): p. 41-9.

121.

Mariathasan, S. and D. M. Monack, Inflammasome adaptors and sensors: intracellular
regulators of infection and inflammation. Nat Rev Immunol, 2007. 7(1): p. 31-40.

122.

Gereke, M., S. Jung, J. Buer, and D. Bruder, Alveolar type II epithelial cells present
antigen to CD4(+) T cells and induce Foxp3(+) regulatory T cells. Am J Respir Crit Care
Med, 2009. 179(5): p. 344-55.

123.

Tosiek, M. J., A. D. Gruber, S. R. Bader, S. Mauel, H. G. Hoymann, S. Prettin, T. Tschernig,
J. Buer, M. Gereke, and D. Bruder, CD4+CD25+Foxp3+ regulatory T cells are dispensable
for controlling CD8+ T cell-mediated lung inflammation. J Immunol, 2011. 186(11): p.
6106-18.

124.

Beck, J. M., V. B. Young, and G. B. Huffnagle, The microbiome of the lung. Transl Res,
2012. 160(4): p. 258-66.

125.

Charlson, E. S., K. Bittinger, A. R. Haas, A. S. Fitzgerald, I. Frank, A. Yadav, F. D. Bushman,
and R. G. Collman, Topographical continuity of bacterial populations in the healthy
human respiratory tract. Am J Respir Crit Care Med, 2011. 184(8): p. 957-63.

126.

Strachan, D. P., Hay fever, hygiene, and household size. BMJ, 1989. 299(6710): p. 125960.

127.

Bufford, J. D. and J. E. Gern, The hygiene hypothesis revisited. Immunol Allergy Clin
North Am, 2005. 25(2): p. 247-62, v-vi.

128.

Jensen, F. B., Red blood cell pH, the Bohr effect, and other oxygenation-linked
phenomena in blood O2 and CO2 transport. Acta Physiol Scand, 2004. 182(3): p. 215-27.

129.

Leach, R. M. and D. F. Treacher, Oxygen transport-2. Tissue hypoxia. BMJ, 1998.
317(7169): p. 1370-3.

130.

Birol, G., S. Wang, E. Budzynski, N. D. Wangsa-Wirawan, and R. A. Linsenmeier, Oxygen
distribution and consumption in the macaque retina. Am J Physiol Heart Circ Physiol,
2007. 293(3): p. H1696-704.

131.

Kiaer, T. and K. D. Kristensen, Intracompartmental pressure, PO2, PCO2 and blood flow
in the human skeletal muscle. Arch Orthop Trauma Surg, 1988. 107(2): p. 114-6.

166

132.

Duong, T. Q., C. Iadecola, and S. G. Kim, Effect of hyperoxia, hypercapnia, and hypoxia on
cerebral interstitial oxygen tension and cerebral blood flow. Magn Reson Med, 2001.
45(1): p. 61-70.

133.

Lee, K. A., R. A. Roth, and J. J. LaPres, Hypoxia, drug therapy and toxicity. Pharmacol
Ther, 2007. 113(2): p. 229-46.

134.

Taylor, C. T., Mitochondria and cellular oxygen sensing in the HIF pathway. Biochem J,
2008. 409(1): p. 19-26.

135.

Jain, M. and J. I. Sznajder, Effects of hypoxia on the alveolar epithelium. Proc Am Thorac
Soc, 2005. 2(3): p. 202-5.

136.

Ward, J. P., Oxygen sensors in context. Biochim Biophys Acta, 2008. 1777(1): p. 1-14.

137.

Lopez-Barneo, J., P. Ortega-Saenz, R. Pardal, A. Pascual, and J. I. Piruat, Carotid body
oxygen sensing. Eur Respir J, 2008. 32(5): p. 1386-98.

138.

Sommer, N., A. Dietrich, R. T. Schermuly, H. A. Ghofrani, T. Gudermann, R. Schulz, W.
Seeger, F. Grimminger, and N. Weissmann, Regulation of hypoxic pulmonary
vasoconstriction: basic mechanisms. Eur Respir J, 2008. 32(6): p. 1639-51.

139.

Kemp, P. J., V. Telezhkin, W. J. Wilkinson, R. Mears, S. B. Hanmer, H. C. Gadeberg, C. T.
Muller, D. Riccardi, and S. P. Brazier, Enzyme-linked oxygen sensing by potassium
channels. Ann N Y Acad Sci, 2009. 1177: p. 112-8.

140.

Michiels, C., Physiological and pathological responses to hypoxia. Am J Pathol, 2004.
164(6): p. 1875-82.

141.

Wang, G. L. and G. L. Semenza, Characterization of hypoxia-inducible factor 1 and
regulation of DNA binding activity by hypoxia. J Biol Chem, 1993. 268(29): p. 21513-8.

142.

Wang, G. L., B. H. Jiang, E. A. Rue, and G. L. Semenza, Hypoxia-inducible factor 1 is a
basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad
Sci U S A, 1995. 92(12): p. 5510-4.

143.

Tian, H., S. L. McKnight, and D. W. Russell, Endothelial PAS domain protein 1 (EPAS1), a
transcription factor selectively expressed in endothelial cells. Genes Dev, 1997. 11(1): p.
72-82.

144.

Gu, Y. Z., S. M. Moran, J. B. Hogenesch, L. Wartman, and C. A. Bradfield, Molecular
characterization and chromosomal localization of a third alpha-class hypoxia inducible
factor subunit, HIF3alpha. Gene Expr, 1998. 7(3): p. 205-13.

167

145.

Scheuermann, T. H., J. Yang, L. Zhang, K. H. Gardner, and R. K. Bruick, Hypoxia-inducible
factors Per/ARNT/Sim domains: structure and function. Methods Enzymol, 2007. 435: p.
3-24.

146.

Kewley, R. J., M. L. Whitelaw, and A. Chapman-Smith, The mammalian basic helix-loophelix/PAS family of transcriptional regulators. Int J Biochem Cell Biol, 2004. 36(2): p. 189204.

147.

Lieb, M. E., K. Menzies, M. C. Moschella, R. Ni, and M. B. Taubman, Mammalian EGLN
genes have distinct patterns of mRNA expression and regulation. Biochem Cell Biol,
2002. 80(4): p. 421-6.

148.

Metzen, E., U. Berchner-Pfannschmidt, P. Stengel, J. H. Marxsen, I. Stolze, M. Klinger, W.
Q. Huang, C. Wotzlaw, T. Hellwig-Burgel, W. Jelkmann, H. Acker, and J. Fandrey,
Intracellular localisation of human HIF-1 alpha hydroxylases: implications for oxygen
sensing. J Cell Sci, 2003. 116(Pt 7): p. 1319-26.

149.

Ivan, M., K. Kondo, H. Yang, W. Kim, J. Valiando, M. Ohh, A. Salic, J. M. Asara, W. S. Lane,
and W. G. Kaelin, Jr., HIFalpha targeted for VHL-mediated destruction by proline
hydroxylation: implications for O2 sensing. Science, 2001. 292(5516): p. 464-8.

150.

Bruick, R. K. and S. L. McKnight, A conserved family of prolyl-4-hydroxylases that modify
HIF. Science, 2001. 294(5545): p. 1337-40.

151.

Epstein, A. C., J. M. Gleadle, L. A. McNeill, K. S. Hewitson, J. O'Rourke, D. R. Mole, M.
Mukherji, E. Metzen, M. I. Wilson, A. Dhanda, Y. M. Tian, N. Masson, D. L. Hamilton, P.
Jaakkola, R. Barstead, J. Hodgkin, P. H. Maxwell, C. W. Pugh, C. J. Schofield, and P. J.
Ratcliffe, C. elegans EGL-9 and mammalian homologs define a family of dioxygenases
that regulate HIF by prolyl hydroxylation. Cell, 2001. 107(1): p. 43-54.

152.

Maxwell, P. H., M. S. Wiesener, G. W. Chang, S. C. Clifford, E. C. Vaux, M. E. Cockman, C.
C. Wykoff, C. W. Pugh, E. R. Maher, and P. J. Ratcliffe, The tumour suppressor protein
VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature, 1999.
399(6733): p. 271-5.

153.

Ohh, M., C. W. Park, M. Ivan, M. A. Hoffman, T. Y. Kim, L. E. Huang, N. Pavletich, V. Chau,
and W. G. Kaelin, Ubiquitination of hypoxia-inducible factor requires direct binding to the
beta-domain of the von Hippel-Lindau protein. Nat Cell Biol, 2000. 2(7): p. 423-7.

154.

Jaakkola, P., D. R. Mole, Y. M. Tian, M. I. Wilson, J. Gielbert, S. J. Gaskell, A. von
Kriegsheim, H. F. Hebestreit, M. Mukherji, C. J. Schofield, P. H. Maxwell, C. W. Pugh, and
P. J. Ratcliffe, Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by
O2-regulated prolyl hydroxylation. Science, 2001. 292(5516): p. 468-72.

155.

Greer, S. N., J. L. Metcalf, Y. Wang, and M. Ohh, The updated biology of hypoxiainducible factor. EMBO J, 2012. 31(11): p. 2448-60.
168

156.

Kaluz, S., M. Kaluzova, and E. J. Stanbridge, Regulation of gene expression by hypoxia:
integration of the HIF-transduced hypoxic signal at the hypoxia-responsive element. Clin
Chim Acta, 2008. 395(1-2): p. 6-13.

157.

Hagen, T., Oxygen versus Reactive Oxygen in the Regulation of HIF-1alpha: The Balance
Tips. Biochem Res Int, 2012. 2012: p. 436981.

158.

Semenza, G. L., P. H. Roth, H. M. Fang, and G. L. Wang, Transcriptional regulation of
genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem, 1994.
269(38): p. 23757-63.

159.

Wenger, R. H., D. P. Stiehl, and G. Camenisch, Integration of oxygen signaling at the
consensus HRE. Sci STKE, 2005. 2005(306): p. re12.

160.

Sowter, H. M., P. J. Ratcliffe, P. Watson, A. H. Greenberg, and A. L. Harris, HIF-1dependent regulation of hypoxic induction of the cell death factors BNIP3 and NIX in
human tumors. Cancer Res, 2001. 61(18): p. 6669-73.

161.

Kim, J. Y., H. J. Ahn, J. H. Ryu, K. Suk, and J. H. Park, BH3-only protein Noxa is a mediator
of hypoxic cell death induced by hypoxia-inducible factor 1alpha. J Exp Med, 2004.
199(1): p. 113-24.

162.

Schmid, T., J. Zhou, and B. Brune, HIF-1 and p53: communication of transcription factors
under hypoxia. J Cell Mol Med, 2004. 8(4): p. 423-31.

163.

Shimoda, L. A. and G. L. Semenza, HIF and the lung: role of hypoxia-inducible factors in
pulmonary development and disease. Am J Respir Crit Care Med, 2012. 183(2): p. 152-6.

164.

Patel, S. A. and M. C. Simon, Biology of hypoxia-inducible factor-2alpha in development
and disease. Cell Death Differ, 2008. 15(4): p. 628-34.

165.

Dunwoodie, S. L., The role of hypoxia in development of the Mammalian embryo. Dev
Cell, 2009. 17(6): p. 755-73.

166.

Ryan, H. E., J. Lo, and R. S. Johnson, HIF-1 alpha is required for solid tumor formation and
embryonic vascularization. EMBO J, 1998. 17(11): p. 3005-15.

167.

Iyer, N. V., L. E. Kotch, F. Agani, S. W. Leung, E. Laughner, R. H. Wenger, M. Gassmann, J.
D. Gearhart, A. M. Lawler, A. Y. Yu, and G. L. Semenza, Cellular and developmental
control of O2 homeostasis by hypoxia-inducible factor 1 alpha. Genes Dev, 1998. 12(2):
p. 149-62.

168.

Compernolle, V., K. Brusselmans, D. Franco, A. Moorman, M. Dewerchin, D. Collen, and
P. Carmeliet, Cardia bifida, defective heart development and abnormal neural crest
migration in embryos lacking hypoxia-inducible factor-1alpha. Cardiovasc Res, 2003.
60(3): p. 569-79.
169

169.

Tian, H., R. E. Hammer, A. M. Matsumoto, D. W. Russell, and S. L. McKnight, The
hypoxia-responsive transcription factor EPAS1 is essential for catecholamine
homeostasis and protection against heart failure during embryonic development. Genes
Dev, 1998. 12(21): p. 3320-4.

170.

Peng, J., L. Zhang, L. Drysdale, and G. H. Fong, The transcription factor EPAS-1/hypoxiainducible factor 2alpha plays an important role in vascular remodeling. Proc Natl Acad
Sci U S A, 2000. 97(15): p. 8386-91.

171.

Compernolle, V., K. Brusselmans, T. Acker, P. Hoet, M. Tjwa, H. Beck, S. Plaisance, Y. Dor,
E. Keshet, F. Lupu, B. Nemery, M. Dewerchin, P. Van Veldhoven, K. Plate, L. Moons, D.
Collen, and P. Carmeliet, Loss of HIF-2alpha and inhibition of VEGF impair fetal lung
maturation, whereas treatment with VEGF prevents fatal respiratory distress in
premature mice. Nat Med, 2002. 8(7): p. 702-10.

172.

Scortegagna, M., K. Ding, Y. Oktay, A. Gaur, F. Thurmond, L. J. Yan, B. T. Marck, A. M.
Matsumoto, J. M. Shelton, J. A. Richardson, M. J. Bennett, and J. A. Garcia, Multiple
organ pathology, metabolic abnormalities and impaired homeostasis of reactive oxygen
species in Epas1-/- mice. Nat Genet, 2003. 35(4): p. 331-40.

173.

Hu, C. J., A. Sataur, L. Wang, H. Chen, and M. C. Simon, The N-terminal transactivation
domain confers target gene specificity of hypoxia-inducible factors HIF-1alpha and HIF2alpha. Mol Biol Cell, 2007. 18(11): p. 4528-42.

174.

Hu, C. J., L. Y. Wang, L. A. Chodosh, B. Keith, and M. C. Simon, Differential roles of
hypoxia-inducible factor 1alpha (HIF-1alpha) and HIF-2alpha in hypoxic gene regulation.
Mol Cell Biol, 2003. 23(24): p. 9361-74.

175.

Wang, V., D. A. Davis, M. Haque, L. E. Huang, and R. Yarchoan, Differential gene upregulation by hypoxia-inducible factor-1alpha and hypoxia-inducible factor-2alpha in
HEK293T cells. Cancer Res, 2005. 65(8): p. 3299-306.

176.

Gordan, J. D., J. A. Bertout, C. J. Hu, J. A. Diehl, and M. C. Simon, HIF-2alpha promotes
hypoxic cell proliferation by enhancing c-myc transcriptional activity. Cancer Cell, 2007.
11(4): p. 335-47.

177.

Covello, K. L., J. Kehler, H. Yu, J. D. Gordan, A. M. Arsham, C. J. Hu, P. A. Labosky, M. C.
Simon, and B. Keith, HIF-2alpha regulates Oct-4: effects of hypoxia on stem cell function,
embryonic development, and tumor growth. Genes Dev, 2006. 20(5): p. 557-70.

178.

Mole, D. R., C. Blancher, R. R. Copley, P. J. Pollard, J. M. Gleadle, J. Ragoussis, and P. J.
Ratcliffe, Genome-wide association of hypoxia-inducible factor (HIF)-1alpha and HIF2alpha DNA binding with expression profiling of hypoxia-inducible transcripts. J Biol
Chem, 2009. 284(25): p. 16767-75.

170

179.

Makino, Y., R. Cao, K. Svensson, G. Bertilsson, M. Asman, H. Tanaka, Y. Cao, A.
Berkenstam, and L. Poellinger, Inhibitory PAS domain protein is a negative regulator of
hypoxia-inducible gene expression. Nature, 2001. 414(6863): p. 550-4.

180.

Makino, Y., A. Kanopka, W. J. Wilson, H. Tanaka, and L. Poellinger, Inhibitory PAS domain
protein (IPAS) is a hypoxia-inducible splicing variant of the hypoxia-inducible factor3alpha locus. J Biol Chem, 2002. 277(36): p. 32405-8.

181.

Jang, M. S., J. E. Park, J. A. Lee, S. G. Park, P. K. Myung, D. H. Lee, B. C. Park, and S. Cho,
Binding and regulation of hypoxia-inducible factor-1 by the inhibitory PAS proteins.
Biochem Biophys Res Commun, 2005. 337(1): p. 209-15.

182.

Hara, S., J. Hamada, C. Kobayashi, Y. Kondo, and N. Imura, Expression and
characterization of hypoxia-inducible factor (HIF)-3alpha in human kidney: suppression
of HIF-mediated gene expression by HIF-3alpha. Biochem Biophys Res Commun, 2001.
287(4): p. 808-13.

183.

Li, Q. F., X. R. Wang, Y. W. Yang, and H. Lin, Hypoxia upregulates hypoxia inducible factor
(HIF)-3alpha expression in lung epithelial cells: characterization and comparison with
HIF-1alpha. Cell Res, 2006. 16(6): p. 548-58.

184.

Maynard, M. A., A. J. Evans, T. Hosomi, S. Hara, M. A. Jewett, and M. Ohh, Human HIF3alpha4 is a dominant-negative regulator of HIF-1 and is down-regulated in renal cell
carcinoma. FASEB J, 2005. 19(11): p. 1396-406.

185.

Maynard, M. A., A. J. Evans, W. Shi, W. Y. Kim, F. F. Liu, and M. Ohh, Dominant-negative
HIF-3 alpha 4 suppresses VHL-null renal cell carcinoma progression. Cell Cycle, 2007.
6(22): p. 2810-6.

186.

Augstein, A., D. M. Poitz, R. C. Braun-Dullaeus, R. H. Strasser, and A. Schmeisser, Cellspecific and hypoxia-dependent regulation of human HIF-3alpha: inhibition of the
expression of HIF target genes in vascular cells. Cell Mol Life Sci, 2011. 68(15): p. 262742.

187.

Tanaka, T., M. Wiesener, W. Bernhardt, K. U. Eckardt, and C. Warnecke, The human HIF
(hypoxia-inducible factor)-3alpha gene is a HIF-1 target gene and may modulate hypoxic
gene induction. Biochem J, 2009. 424(1): p. 143-51.

188.

Stroka, D. M., T. Burkhardt, I. Desbaillets, R. H. Wenger, D. A. Neil, C. Bauer, M.
Gassmann, and D. Candinas, HIF-1 is expressed in normoxic tissue and displays an organspecific regulation under systemic hypoxia. FASEB J, 2001. 15(13): p. 2445-53.

189.

Wiesener, M. S., H. Turley, W. E. Allen, C. Willam, K. U. Eckardt, K. L. Talks, S. M. Wood,
K. C. Gatter, A. L. Harris, C. W. Pugh, P. J. Ratcliffe, and P. H. Maxwell, Induction of
endothelial PAS domain protein-1 by hypoxia: characterization and comparison with
hypoxia-inducible factor-1alpha. Blood, 1998. 92(7): p. 2260-8.
171

190.

Wiesener, M. S., J. S. Jurgensen, C. Rosenberger, C. K. Scholze, J. H. Horstrup, C.
Warnecke, S. Mandriota, I. Bechmann, U. A. Frei, C. W. Pugh, P. J. Ratcliffe, S.
Bachmann, P. H. Maxwell, and K. U. Eckardt, Widespread hypoxia-inducible expression of
HIF-2alpha in distinct cell populations of different organs. FASEB J, 2003. 17(2): p. 271-3.

191.

Groenman, F., M. Rutter, I. Caniggia, D. Tibboel, and M. Post, Hypoxia-inducible factors
in the first trimester human lung. J Histochem Cytochem, 2007. 55(4): p. 355-63.

192.

Wagner, K. F., A. K. Hellberg, S. Balenger, R. Depping, O. J. Dodd, R. A. Johns, and D. Li,
Hypoxia-induced mitogenic factor has antiapoptotic action and is upregulated in the
developing lung: coexpression with hypoxia-inducible factor-2alpha. Am J Respir Cell Mol
Biol, 2004. 31(3): p. 276-82.

193.

Bridges, J. P., S. Lin, M. Ikegami, and J. M. Shannon, Conditional hypoxia inducible factor1alpha induction in embryonic pulmonary epithelium impairs maturation and augments
lymphangiogenesis. Dev Biol, 2011. 362(1): p. 24-41.

194.

Xu, Y., Y. Wang, V. Besnard, M. Ikegami, S. E. Wert, C. Heffner, S. A. Murray, L. R.
Donahue, and J. A. Whitsett, Transcriptional programs controlling perinatal lung
maturation. PLoS One, 2012. 7(8): p. e37046.

195.

Yu, A. Y., M. G. Frid, L. A. Shimoda, C. M. Wiener, K. Stenmark, and G. L. Semenza,
Temporal, spatial, and oxygen-regulated expression of hypoxia-inducible factor-1 in the
lung. Am J Physiol, 1998. 275(4 Pt 1): p. L818-26.

196.

Roos, D. and J. A. Loos, Changes in the carbohydrate metabolism of mitogenically
stimulated human peripheral lymphocytes. II. Relative importance of glycolysis and
oxidative phosphorylation on phytohaemagglutinin stimulation. Exp Cell Res, 1973.
77(1): p. 127-35.

197.

Eltzschig, H. K. and P. Carmeliet, Hypoxia and inflammation. N Engl J Med, 2011. 364(7):
p. 656-65.

198.

Nizet, V. and R. S. Johnson, Interdependence of hypoxic and innate immune responses.
Nat Rev Immunol, 2009. 9(9): p. 609-17.

199.

Cramer, T., Y. Yamanishi, B. E. Clausen, I. Forster, R. Pawlinski, N. Mackman, V. H. Haase,
R. Jaenisch, M. Corr, V. Nizet, G. S. Firestein, H. P. Gerber, N. Ferrara, and R. S. Johnson,
HIF-1alpha is essential for myeloid cell-mediated inflammation. Cell, 2003. 112(5): p.
645-57.

200.

Crotty Alexander, L. E., K. Akong, and V. Nizet, Analysis Of The Role Of Hypoxia Inducible
Factor-1α In Eosinophil Function And Allergic Inflammatory Airways Disease, in American
Thoracic Society. 2012: San Francisco, CA, Area E (Hall D, North Building, Lower Level),
Moscone Center.

172

201.

Peyssonnaux, C., V. Datta, T. Cramer, A. Doedens, E. A. Theodorakis, R. L. Gallo, N.
Hurtado-Ziola, V. Nizet, and R. S. Johnson, HIF-1alpha expression regulates the
bactericidal capacity of phagocytes. J Clin Invest, 2005. 115(7): p. 1806-15.

202.

Takeda, N., E. L. O'Dea, A. Doedens, J. W. Kim, A. Weidemann, C. Stockmann, M. Asagiri,
M. C. Simon, A. Hoffmann, and R. S. Johnson, Differential activation and antagonistic
function of HIF-{alpha} isoforms in macrophages are essential for NO homeostasis.
Genes Dev, 2010. 24(5): p. 491-501.

203.

Kuhlicke, J., J. S. Frick, J. C. Morote-Garcia, P. Rosenberger, and H. K. Eltzschig, Hypoxia
inducible factor (HIF)-1 coordinates induction of Toll-like receptors TLR2 and TLR6 during
hypoxia. PLoS One, 2007. 2(12): p. e1364.

204.

Moreira, A. P., K. A. Cavassani, U. B. Ismailoglu, R. Hullinger, M. P. Dunleavy, D. A.
Knight, S. L. Kunkel, S. Uematsu, S. Akira, and C. M. Hogaboam, The protective role of
TLR6 in a mouse model of asthma is mediated by IL-23 and IL-17A. J Clin Invest, 2011.
121(11): p. 4420-32.

205.

Ben-Shoshan, J., S. Maysel-Auslender, A. Mor, G. Keren, and J. George, Hypoxia controls
CD4+CD25+ regulatory T-cell homeostasis via hypoxia-inducible factor-1alpha. Eur J
Immunol, 2008. 38(9): p. 2412-8.

206.

D'Alessio, F. R., K. Tsushima, N. R. Aggarwal, E. E. West, M. H. Willett, M. F. Britos, M. R.
Pipeling, R. G. Brower, R. M. Tuder, J. F. McDyer, and L. S. King, CD4+CD25+Foxp3+ Tregs
resolve experimental lung injury in mice and are present in humans with acute lung
injury. J Clin Invest, 2009. 119(10): p. 2898-913.

207.

Makino, Y., H. Nakamura, E. Ikeda, K. Ohnuma, K. Yamauchi, Y. Yabe, L. Poellinger, Y.
Okada, C. Morimoto, and H. Tanaka, Hypoxia-inducible factor regulates survival of
antigen receptor-driven T cells. J Immunol, 2003. 171(12): p. 6534-40.

208.

Biju, M. P., A. K. Neumann, S. J. Bensinger, R. S. Johnson, L. A. Turka, and V. H. Haase,
Vhlh gene deletion induces Hif-1-mediated cell death in thymocytes. Mol Cell Biol, 2004.
24(20): p. 9038-47.

209.

Zuckerberg, A. L., L. I. Goldberg, and H. M. Lederman, Effects of hypoxia on interleukin-2
mRNA expression by T lymphocytes. Crit Care Med, 1994. 22(2): p. 197-203.

210.

Lederer, J. A., M. L. Rodrick, and J. A. Mannick, The effects of injury on the adaptive
immune response. Shock, 1999. 11(3): p. 153-9.

211.

Ahmad, T., M. Kumar, U. Mabalirajan, B. Pattnaik, S. Aggarwal, R. Singh, S. Singh, M.
Mukerji, B. Ghosh, and A. Agrawal, Hypoxia response in asthma: differential modulation
on inflammation and epithelial injury. Am J Respir Cell Mol Biol, 2012. 47(1): p. 1-10.

173

212.

Huerta-Yepez, S., G. J. Baay-Guzman, I. G. Bebenek, R. Hernandez-Pando, M. I. Vega, L.
Chi, M. Riedl, D. Diaz-Sanchez, E. Kleerup, D. P. Tashkin, F. J. Gonzalez, B. Bonavida, M.
Zeidler, and O. Hankinson, Hypoxia inducible factor promotes murine allergic airway
inflammation and is increased in asthma and rhinitis. Allergy, 2011. 66(7): p. 909-18.

213.

Clerici, C. and C. Planes, Gene regulation in the adaptive process to hypoxia in lung
epithelial cells. Am J Physiol Lung Cell Mol Physiol, 2009. 296(3): p. L267-74.

214.

Saini, Y., K. K. Greenwood, C. Merrill, K. Y. Kim, S. Patial, N. Parameswaran, J. R.
Harkema, and J. J. LaPres, Acute cobalt-induced lung injury and the role of hypoxiainducible factor 1alpha in modulating inflammation. Toxicol Sci, 2010. 116(2): p. 673-81.

215.

Saini, Y., K. Y. Kim, R. Lewandowski, L. A. Bramble, J. R. Harkema, and J. J. Lapres, Role of
hypoxia-inducible factor 1{alpha} in modulating cobalt-induced lung inflammation. Am J
Physiol Lung Cell Mol Physiol, 2010. 298(2): p. L139-47.

216.

Cummins, E. P. and C. T. Taylor, Hypoxia-responsive transcription factors. Pflugers Arch,
2005. 450(6): p. 363-71.

217.

Kenneth, N. S. and S. Rocha, Regulation of gene expression by hypoxia. Biochem J, 2008.
414(1): p. 19-29.

218.

Tsai, Y. P. and K. J. Wu, Epigenetic regulation of hypoxia-responsive gene expression:
Focusing on chromatin and DNA modifications. Int J Cancer, 2013.

219.

Barnes, P. J., Biochemical basis of asthma therapy. J Biol Chem, 2011. 286(38): p. 32899905.

220.

Perkins, N. D., Integrating cell-signalling pathways with NF-kappaB and IKK function. Nat
Rev Mol Cell Biol, 2007. 8(1): p. 49-62.

221.

Hayden, M. S. and S. Ghosh, Signaling to NF-kappaB. Genes Dev, 2004. 18(18): p. 2195224.

222.

Vallabhapurapu, S. and M. Karin, Regulation and function of NF-kappaB transcription
factors in the immune system. Annu Rev Immunol, 2009. 27: p. 693-733.

223.

Perkins, N. D. and T. D. Gilmore, Good cop, bad cop: the different faces of NF-kappaB.
Cell Death Differ, 2006. 13(5): p. 759-72.

224.

Laderoute, K. R., The interaction between HIF-1 and AP-1 transcription factors in
response to low oxygen. Semin Cell Dev Biol, 2005. 16(4-5): p. 502-13.

225.

Minet, E., G. Michel, D. Mottet, J. P. Piret, A. Barbieux, M. Raes, and C. Michiels, c-JUN
gene induction and AP-1 activity is regulated by a JNK-dependent pathway in hypoxic
HepG2 cells. Exp Cell Res, 2001. 265(1): p. 114-24.

174

226.

Fantozzi, I., S. Zhang, O. Platoshyn, C. V. Remillard, R. T. Cowling, and J. X. Yuan, Hypoxia
increases AP-1 binding activity by enhancing capacitative Ca2+ entry in human
pulmonary artery endothelial cells. Am J Physiol Lung Cell Mol Physiol, 2003. 285(6): p.
L1233-45.

227.

Premkumar, D. R., G. Adhikary, J. L. Overholt, M. S. Simonson, N. S. Cherniack, and N. R.
Prabhakar, Intracellular pathways linking hypoxia to activation of c-fos and AP-1. Adv
Exp Med Biol, 2000. 475: p. 101-9.

228.

Salnikow, K., T. Kluz, M. Costa, D. Piquemal, Z. N. Demidenko, K. Xie, and M. V.
Blagosklonny, The regulation of hypoxic genes by calcium involves c-Jun/AP-1, which
cooperates with hypoxia-inducible factor 1 in response to hypoxia. Mol Cell Biol, 2002.
22(6): p. 1734-41.

229.

Suzuki, H., A. Tomida, and T. Tsuruo, Dephosphorylated hypoxia-inducible factor 1alpha
as a mediator of p53-dependent apoptosis during hypoxia. Oncogene, 2001. 20(41): p.
5779-88.

230.

Zheng, X., S. Linke, J. M. Dias, X. Zheng, K. Gradin, T. P. Wallis, B. R. Hamilton, M.
Gustafsson, J. L. Ruas, S. Wilkins, R. L. Bilton, K. Brismar, M. L. Whitelaw, T. Pereira, J. J.
Gorman, J. Ericson, D. J. Peet, U. Lendahl, and L. Poellinger, Interaction with factor
inhibiting HIF-1 defines an additional mode of cross-coupling between the Notch and
hypoxia signaling pathways. Proc Natl Acad Sci U S A, 2008. 105(9): p. 3368-73.

231.

Beitner-Johnson, D. and D. E. Millhorn, Hypoxia induces phosphorylation of the cyclic
AMP response element-binding protein by a novel signaling mechanism. J Biol Chem,
1998. 273(31): p. 19834-9.

232.

Taylor, C. T., G. T. Furuta, K. Synnestvedt, and S. P. Colgan, Phosphorylation-dependent
targeting of cAMP response element binding protein to the ubiquitin/proteasome
pathway in hypoxia. Proc Natl Acad Sci U S A, 2000. 97(22): p. 12091-6.

233.

Comerford, K. M., M. O. Leonard, J. Karhausen, R. Carey, S. P. Colgan, and C. T. Taylor,
Small ubiquitin-related modifier-1 modification mediates resolution of CREB-dependent
responses to hypoxia. Proc Natl Acad Sci U S A, 2003. 100(3): p. 986-91.

234.

Takiguchi, M., The C/EBP family of transcription factors in the liver and other organs. Int
J Exp Pathol, 1998. 79(6): p. 369-91.

235.

Sanchez-Elsner, T., J. R. Ramirez, F. Sanz-Rodriguez, E. Varela, C. Bernabeu, and L. M.
Botella, A cross-talk between hypoxia and TGF-beta orchestrates erythropoietin gene
regulation through SP1 and Smads. J Mol Biol, 2004. 336(1): p. 9-24.

236.

Kaluz, S., M. Kaluzova, and E. J. Stanbridge, Expression of the hypoxia marker carbonic
anhydrase IX is critically dependent on SP1 activity. Identification of a novel type of
hypoxia-responsive enhancer. Cancer Res, 2003. 63(5): p. 917-22.
175

237.

Yan, S. F., J. Lu, Y. S. Zou, W. Kisiel, N. Mackman, M. Leitges, S. Steinberg, D. Pinsky, and
D. Stern, Protein kinase C-beta and oxygen deprivation. A novel Egr-1-dependent
pathway for fibrin deposition in hypoxemic vasculature. J Biol Chem, 2000. 275(16): p.
11921-8.

238.

Lo, L. W., J. J. Cheng, J. J. Chiu, B. S. Wung, Y. C. Liu, and D. L. Wang, Endothelial exposure
to hypoxia induces Egr-1 expression involving PKCalpha-mediated Ras/Raf-1/ERK1/2
pathway. J Cell Physiol, 2001. 188(3): p. 304-12.

239.

Koong, A. C., E. Y. Chen, and A. J. Giaccia, Hypoxia causes the activation of nuclear factor
kappa B through the phosphorylation of I kappa B alpha on tyrosine residues. Cancer
Res, 1994. 54(6): p. 1425-30.

240.

Batra, S., G. Balamayooran, and M. K. Sahoo, Nuclear factor-kappaB: a key regulator in
health and disease of lungs. Arch Immunol Ther Exp (Warsz), 2011. 59(5): p. 335-51.

241.

Koong, A. C., E. Y. Chen, N. F. Mivechi, N. C. Denko, P. Stambrook, and A. J. Giaccia,
Hypoxic activation of nuclear factor-kappa B is mediated by a Ras and Raf signaling
pathway and does not involve MAP kinase (ERK1 or ERK2). Cancer Res, 1994. 54(20): p.
5273-9.

242.

Cummins, E. P., E. Berra, K. M. Comerford, A. Ginouves, K. T. Fitzgerald, F. Seeballuck, C.
Godson, J. E. Nielsen, P. Moynagh, J. Pouyssegur, and C. T. Taylor, Prolyl hydroxylase-1
negatively regulates IkappaB kinase-beta, giving insight into hypoxia-induced NFkappaB
activity. Proc Natl Acad Sci U S A, 2006. 103(48): p. 18154-9.

243.

Cockman, M. E., D. E. Lancaster, I. P. Stolze, K. S. Hewitson, M. A. McDonough, M. L.
Coleman, C. H. Coles, X. Yu, R. T. Hay, S. C. Ley, C. W. Pugh, N. J. Oldham, N. Masson, C. J.
Schofield, and P. J. Ratcliffe, Posttranslational hydroxylation of ankyrin repeats in
IkappaB proteins by the hypoxia-inducible factor (HIF) asparaginyl hydroxylase, factor
inhibiting HIF (FIH). Proc Natl Acad Sci U S A, 2006. 103(40): p. 14767-72.

244.

Rius, J., M. Guma, C. Schachtrup, K. Akassoglou, A. S. Zinkernagel, V. Nizet, R. S. Johnson,
G. G. Haddad, and M. Karin, NF-kappaB links innate immunity to the hypoxic response
through transcriptional regulation of HIF-1alpha. Nature, 2008. 453(7196): p. 807-11.

245.

van Uden, P., N. S. Kenneth, and S. Rocha, Regulation of hypoxia-inducible factor-1alpha
by NF-kappaB. Biochem J, 2008. 412(3): p. 477-84.

246.

Zhang, W., J. M. Petrovic, D. Callaghan, A. Jones, H. Cui, C. Howlett, and D. Stanimirovic,
Evidence that hypoxia-inducible factor-1 (HIF-1) mediates transcriptional activation of
interleukin-1beta (IL-1beta) in astrocyte cultures. J Neuroimmunol, 2006. 174(1-2): p. 6373.

176

247.

Kim, H. Y., Y. H. Kim, B. H. Nam, H. J. Kong, H. H. Kim, Y. J. Kim, W. G. An, and J. Cheong,
HIF-1alpha expression in response to lipopolysaccaride mediates induction of hepatic
inflammatory cytokine TNFalpha. Exp Cell Res, 2007. 313(9): p. 1866-76.

248.

Eltzschig, H. K., Adenosine: an old drug newly discovered. Anesthesiology, 2009. 111(4):
p. 904-15.

249.

Eltzschig, H. K., J. C. Ibla, G. T. Furuta, M. O. Leonard, K. A. Jacobson, K. Enjyoji, S. C.
Robson, and S. P. Colgan, Coordinated adenine nucleotide phosphohydrolysis and
nucleoside signaling in posthypoxic endothelium: role of ectonucleotidases and
adenosine A2B receptors. J Exp Med, 2003. 198(5): p. 783-96.

250.

Eckle, T., A. Grenz, S. Laucher, and H. K. Eltzschig, A2B adenosine receptor signaling
attenuates acute lung injury by enhancing alveolar fluid clearance in mice. J Clin Invest,
2008. 118(10): p. 3301-15.

251.

Eckle, T., M. Faigle, A. Grenz, S. Laucher, L. F. Thompson, and H. K. Eltzschig, A2B
adenosine receptor dampens hypoxia-induced vascular leak. Blood, 2008. 111(4): p.
2024-35.

252.

Cagnina, R. E., S. I. Ramos, M. A. Marshall, G. Wang, C. R. Frazier, and J. Linden,
Adenosine A2B receptors are highly expressed on murine type II alveolar epithelial cells.
Am J Physiol Lung Cell Mol Physiol, 2009. 297(3): p. L467-74.

253.

Morote-Garcia, J. C., P. Rosenberger, J. Kuhlicke, and H. K. Eltzschig, HIF-1-dependent
repression of adenosine kinase attenuates hypoxia-induced vascular leak. Blood, 2008.
111(12): p. 5571-80.

254.

Synnestvedt, K., G. T. Furuta, K. M. Comerford, N. Louis, J. Karhausen, H. K. Eltzschig, K.
R. Hansen, L. F. Thompson, and S. P. Colgan, Ecto-5'-nucleotidase (CD73) regulation by
hypoxia-inducible factor-1 mediates permeability changes in intestinal epithelia. J Clin
Invest, 2002. 110(7): p. 993-1002.

255.

Kong, T., K. A. Westerman, M. Faigle, H. K. Eltzschig, and S. P. Colgan, HIF-dependent
induction of adenosine A2B receptor in hypoxia. FASEB J, 2006. 20(13): p. 2242-50.

256.

Koeppen, M., T. Eckle, and H. K. Eltzschig, Interplay of hypoxia and A2B adenosine
receptors in tissue protection. Adv Pharmacol, 2011. 61: p. 145-86.

257.

Sitkovsky, M. V., T regulatory cells: hypoxia-adenosinergic suppression and re-direction
of the immune response. Trends Immunol, 2009. 30(3): p. 102-8.

258.

Colgan, S. P. and H. K. Eltzschig, Adenosine and hypoxia-inducible factor signaling in
intestinal injury and recovery. Annu Rev Physiol, 2012. 74: p. 153-75.

177

259.

Whittemore, C. R., Cobalt, in The Encyclopedia of the Chemical Elements, Hampel, C. A.,
Editor. 1968, Reinhold Book Corporation: New York. p. 154-163.

260.

Survey, U. G., Rare Earth Elements—Critical Resources for High Technology, US
Geological Survey (USGS).

261.

Cobalt. 2012, Cobalt Development Institute.

262.

Delpeux, S., K. Szostak, E. Frackowiak, S. Bonnamy, and F. Beguin, High yield of pure
multiwalled carbon nanotubes from the catalytic decomposition of acetylene on in-situ
formed cobalt nanoparticles. J Nanosci Nanotechnol, 2002. 2(5): p. 481-4.

263.

Coles, B. L., The use of cobalt in some common anaemias of childhood. Arch Dis Child,
1955. 30(150): p. 121-6.

264.

Coles, B. L. and U. James, Use of cobalt and iron in the treatment and prevention of
anemia of prematurity. J Lancet, 1955. 75(3): p. 79-82; passim.

265.

Schirrmacher, U. O., Case of cobalt poisoning. Br Med J, 1967. 1(5539): p. 544-5.

266.

Alexander, C. S., Cobalt-beer cardiomyopathy. A clinical and pathologic study of twentyeight cases. Am J Med, 1972. 53(4): p. 395-417.

267.

Gheysens, B., J. Auwerx, A. Van den Eeckhout, and M. Demedts, Cobalt-induced
bronchial asthma in diamond polishers. Chest, 1985. 88(5): p. 740-4.

268.

Nordberg, G., Handbook on the toxicology of metals. 3rd ed. 2007, Amsterdam ; Boston:
Academic Press. xlvii, 975 p.

269.

Simonsen, L. O., H. Harbak, and P. Bennekou, Cobalt metabolism and toxicology-A brief
update. Sci Total Environ. 432: p. 210-5.

270.

Lison, D., Human toxicity of cobalt-containing dust and experimental studies on the
mechanism of interstitial lung disease (hard metal disease). Crit Rev Toxicol, 1996. 26(6):
p. 585-616.

271.

Cugell, D. W., The hard metal diseases. Clin Chest Med, 1992. 13(2): p. 269-79.

272.

Coates, E. O., Jr. and J. H. Watson, Diffuse interstitial lung disease in tungsten carbide
workers. Ann Intern Med, 1971. 75(5): p. 709-16.

273.

Bech, A. O., Hard Metal Disease and Tool Room Grinding. J. Soc. Occup. Med., 1974.
24(1): p. 11-16.

274.

Scherrer, M. and J. M. Maillard, [Hard metal pneumopathies]. Schweiz Med
Wochenschr, 1982. 112(6): p. 198-207.

178

275.

Shirakawa, T., Y. Kusaka, N. Fujimura, S. Goto, and K. Morimoto, The existence of specific
antibodies to cobalt in hard metal asthma. Clin Allergy, 1988. 18(5): p. 451-60.

276.

Rona, G., Experimental aspects of cobalt cardiomyopathy. Br Heart J, 1971. 33: p.
Suppl:171-4.

277.

Kadiiska, M. B., K. R. Maples, and R. P. Mason, A comparison of cobalt(II) and iron(II)
hydroxyl and superoxide free radical formation. Arch Biochem Biophys, 1989. 275(1): p.
98-111.

278.

Simonsen, L. O., H. Harbak, and P. Bennekou, Passive transport pathways for Ca(2+) and
Co(2+) in human red blood cells. (57)Co(2+) as a tracer for Ca(2+) influx. Blood Cells Mol
Dis. 47(4): p. 214-25.

279.

Batie, C. J., E. LaHaie, and D. P. Ballou, Purification and characterization of phthalate
oxygenase and phthalate oxygenase reductase from Pseudomonas cepacia. J Biol Chem,
1987. 262(4): p. 1510-8.

280.

Maxwell, P. and K. Salnikow, HIF-1: an oxygen and metal responsive transcription factor.
Cancer Biol Ther, 2004. 3(1): p. 29-35.

281.

Salnikow, K., S. P. Donald, R. K. Bruick, A. Zhitkovich, J. M. Phang, and K. S. Kasprzak,
Depletion of intracellular ascorbate by the carcinogenic metals nickel and cobalt results
in the induction of hypoxic stress. J Biol Chem, 2004. 279(39): p. 40337-44.

282.

Fisher, J. W., A quest for erythropoietin over nine decades. Annu Rev Pharmacol Toxicol,
1998. 38: p. 1-20.

283.

Vengellur, A., B. G. Woods, H. E. Ryan, R. S. Johnson, and J. J. LaPres, Gene expression
profiling of the hypoxia signaling pathway in hypoxia-inducible factor 1alpha null mouse
embryonic fibroblasts. Gene Expr, 2003. 11(3-4): p. 181-97.

284.

Gleadle, J. M., B. L. Ebert, J. D. Firth, and P. J. Ratcliffe, Regulation of angiogenic growth
factor expression by hypoxia, transition metals, and chelating agents. Am J Physiol,
1995. 268(6 Pt 1): p. C1362-8.

285.

Vengellur, A. and J. J. LaPres, The role of hypoxia inducible factor 1alpha in cobalt
chloride induced cell death in mouse embryonic fibroblasts. Toxicol Sci, 2004. 82(2): p.
638-46.

286.

Saini, Y., J. R. Harkema, and J. J. LaPres, HIF1alpha is essential for normal intrauterine
differentiation of alveolar epithelium and surfactant production in the newborn lung of
mice. J Biol Chem, 2008. 283(48): p. 33650-7.

179

287.

Perl, A. K., S. E. Wert, A. Nagy, C. G. Lobe, and J. A. Whitsett, Early restriction of
peripheral and proximal cell lineages during formation of the lung. Proc Natl Acad Sci U S
A, 2002. 99(16): p. 10482-7.

288.

Whitsett, J. A. and A. K. Perl, Conditional control of gene expression in the respiratory
epithelium: A cautionary note. Am J Respir Cell Mol Biol, 2006. 34(5): p. 519-20.

289.

Lansdell, T. A., S. O'Reilly, T. Woolliscroft, L. M. Azevedo, D. K. Kahlon, S. Hovde, J. J.
McCormick, R. W. Henry, J. A. Cornicelli, and J. J. Tepe, Attenuation of collagen-induced
arthritis by orally available imidazoline-based NF-kappaB inhibitors. Bioorg Med Chem
Lett, 2012. 22(14): p. 4816-9.

290.

Bradford, M. M., A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal Biochem, 1976.
72: p. 248-54.

291.

Kahlon, D. K., T. A. Lansdell, J. S. Fisk, C. D. Hupp, T. L. Friebe, S. Hovde, A. D. Jones, R. D.
Dyer, R. W. Henry, and J. J. Tepe, Nuclear factor-kappaB mediated inhibition of cytokine
production by imidazoline scaffolds. J Med Chem, 2009. 52(5): p. 1302-9.

292.

Kallet, R. H. and M. A. Matthay, Hyperoxic acute lung injury. Respir Care, 2013. 58(1): p.
123-41.

293.

Cho, H. Y. and S. R. Kleeberger, Nrf2 protects against airway disorders. Toxicol Appl
Pharmacol, 2010. 244(1): p. 43-56.

294.

Ganter, M. T., J. Roux, B. Miyazawa, M. Howard, J. A. Frank, G. Su, D. Sheppard, S. M.
Violette, P. H. Weinreb, G. S. Horan, M. A. Matthay, and J. F. Pittet, Interleukin-1beta
causes acute lung injury via alphavbeta5 and alphavbeta6 integrin-dependent
mechanisms. Circ Res, 2008. 102(7): p. 804-12.

295.

Ogawa, Y., S. Tasaka, W. Yamada, F. Saito, N. Hasegawa, T. Miyasho, and A. Ishizaka,
Role of Toll-like receptor 4 in hyperoxia-induced lung inflammation in mice. Inflamm Res,
2007. 56(8): p. 334-8.

296.

Diehl, S. and M. Rincon, The two faces of IL-6 on Th1/Th2 differentiation. Mol Immunol,
2002. 39(9): p. 531-6.

297.

Karovic, O., I. Tonazzini, N. Rebola, E. Edstrom, C. Lovdahl, B. B. Fredholm, and E. Dare,
Toxic effects of cobalt in primary cultures of mouse astrocytes. Similarities with hypoxia
and role of HIF-1alpha. Biochem Pharmacol, 2007. 73(5): p. 694-708.

180