21.5%“. «.44..
a r

 

.00..
.R

295315;}:

1
z

.D r
.3.
$5:

“.vamfll.

r.-
vai. x‘ nil-La...
. . I .
14".“! . I.Ir

$3.33.:wa

 

. ‘ figmuuqfiWifi:

 

 

>1” LIBRARY
2m, Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

THE ROLE OF THE HIF1 SIGNALING PATHWAY
IN TUMORIGENESIS

presented by

KangAe Lee

has been accepted towards fulfillment
of the requirements for the

Ph.D degree in Program in Cell and Molecular

 

fl BiologL
till 0 Nb
U Ma'orPr‘oTéssor's Signature
(“GEL (i j 300 5i

( Date

 

 

MSU is an affirmative-action, equal-opportunity employer

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DAIEDUE

DAIEDUE

 

MAY 0 21‘20‘99

 

éfivv

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:lClRC/DateDue.indd-p.1

THE ROLE OF THE HIF1 SIGNALING PATHWAY
IN TUMORIGENESIS

By

KangAe Lee

A DISSERTATION
Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of
DOCTOR OF PHILOSOPHY

Program in Cell and Molecular Biology

2007

THE RI

The E
the cellular
the micron:
glycolyxis. l
manner in
l’llFlCl lillitl
component
degradatiozi
tumor SUP:
PHD-medic
Wssibilin- .

ln Ll};
Cellular trar,

Wential. II.

 

 

ABSTRACT
THE ROLE OF THE HIF1 SIGNALING PATHWAY IN TUMORIGENESIS
By

KangAe Lee

The hypoxia-inducible factor 1 (HIF1) is the key transcription factor involved in
the cellular responses to hypoxia and it also plays an essential role in adapting a cell to
the microenvironment of a tumor through regulation of genes involved in angiogenesis,
glycolysis, and many other processes. HIF1 activity is regulated in an oxygen dependent
manner by a family of prolyl hydroxylases (PHDs), whose activity is required for
HIFIOL binding to the von Hippel-Lindau tumor suppressor protein (pVHL). pVHL is a
component of an E3 ubiquitin ligase complex responsible for instigating proteosomal
degradation of HIFla. This role of regulating HIF1 activity is one reason for pVHL’s
tumor suppressor activity. Interestingly, this process is completely dependant upon
PHD-mediated posttranslational hydroxylation of HIF10t and this requirement raises the
possibility of the involvement of PHD in tumor development.

In this study, we characterize the relationship between PHD and HIF1 activity and
cellular transformation using a lineage of the MSU] cell strains of varying tumorigenic
potential. We have shown that PHD2 represents the primary HIF prolyl hydroxylase in
regulating HIF1 signaling within this lineage of cells and that PHD2 levels decrease as
the cell exhibits more transformed characteristics. When PHD2 levels were altered with

RNAi in non-tumorigenic fibroblasts we found that moderate decreases in PHD2

activity ca;
PHD2 \tei
inhibition
characterix
fibroblasts
HlFl-at'lix'
stress. The
aCilI'll_V am
gain 21 gm'
malignant]:
miSlit been
Tilfi dug} Ill
balance lift
The
reamYe 0X

 

0': .
“0' “9 a?

 

activity can lead to malignant transformation, whereas cells with a more severe loss of
PHD2 were unable to form tumors. Consistent with these results, direct chemical
inhibition of PHD2 activity in transformed cells reverses a cell’s transformed
characteristics. Moreover, we found that overexpression of PHD2 in malignant
fibroblasts leads to loss of their tumor-forming ability. These changes correlated with
HIF1-activated glycolytic rates, vascularization, and the ability to grow under hypoxic
stress. These findings suggest a biphasic model for the relationship between PHD2
activity and malignant transformation: With a slight decrease in PHDZ activity the cells
gain a growth advantage, such as enhanced glycolysis and angiogenesis and become
malignantly transformed. As PHD2 activity is further decreased, the pro-death response
might become the dominant signal and the increased adaptation would be overwhelmed.
The dual nature of this response is presumably due to PHDZ’s ability to alter the cellular
balance between hypoxic-induced adaptation and pro-death responses.

The hypoxia signaling pathway has many input signals, including low oxygen,
reactive oxygen species, TCA metabolites, and various growth factors. It is currently not
known how these disparate signals influence HIF1 activity. To address this knowledge
gap, we attempted to characterize the PHD2-protein interaction network (PHD2-PIN)
using tandem affinity purification and liquid chromatography coupled tandem mass
spectrometric analysis (LC-MS/MS). Our result suggested the possibility that PHD2
might have no specific interaction with other proteins under normoxia and that PHD2

might act to coordinate the various input signals without the aid of accessory factors.

I \t'ou
John}
Studyi

seienti:

Susan E

for their

1 would

COlliil‘Om

ll “35 H
l
‘al‘Oraton-

3C0” LVn'

Imank \K.
K" w
dné‘HC"

Speclial [I '

 

 

ACKNOWLEDGEMENTS

I would first like to recognize with high regard the guidance and mentorship of Dr.
John J. LaPres. His encouragement and advice was vital for my educational and career.
Studying under his direction has provided me with the chance to learn the attitude of a

scientist which I will take with me through whole my career.

I greatly appreciate the advise and instruction of my guidance committee members, Dr.
Susan E. Conrad, Dr. R. William Henry, Dr. Laura R. McCabe, and Dr. Jetze J. Tepe

for their support over the years.

I would also like to thank Dr. J. Justin McCormick and Dr. Sandra O’Reilly for their

collaborative efforts.

It was very enjoyable to study with former and current members of Dr. LaPres
laboratory: Babara G. Wood, Ajith Vengellur, Dorothy Tappenden, Yogesh Saini, Dr.

Scott Lynn, Steve Proper, Jeremy Lynd, Amy Bays, and Julie Adkins.

I thank with all my heart my family: parents, Ki-Jun Lee and Young-Ja Pyo, sister,

Kang-Hee Lee and brother, Kang-Chual Lee for their love and support. Finally, my

special thanks to my fiancee Yong-Bum Kim.

iv

LIST OF I

LIST OF I

 

 

Cll-XPTH
lnlroductii
l.

burs

 

Re
CHAPTEI

{Wining p

lnt

 

TABLE OF CONTENTS

LIST OF FIGURES ............................................................................. vii

LIST OF TABLE ................................................................................. ix
CHAPTER 1:

Introduction .......................................................................................... 1

1. Hypoxia ................................................................................. 1

1.1. Normoxia versus hypoxia ................................................ l

1.2. Occurrence of hypoxia. . .L ................................................ 3

2. Hypoxia signaling; the hypoxia-inducible factor (HIF) system .................. 6

3. The integrative HIF1-mediated responses to hypoxia ........................... 23

4. Hypoxia (or HIF1) signaling and cancer .......................................... 26

4.1. HIF1 activity as a unique characteristic of tumor .................... 27

4.2. Positive role of HIF1 signaling in tumor progresses... .. . .. ...29

4.3. Hypoxia signaling in cancer therapy ................................... 32

Hypothesis and Specific Aims ......................................................... 36

References for the introduction ........................................................ 40
CHAPTER 2:

The biphasic role of the HIF prolyl-4-hydroxylase, PHD2, in modulating tumor-

forming potential ................................................................................. 53

Abstract ................................................................................... 54

Introduction .............................................................................. 55

Material and Methods ................................................................... 59

Results .................................................................................... 65

Discussion .............................................................................. 101

References .............................................................................. 105
CHAPTER 3:

The protein interaction network of HIF prolyl-4-hydroxylase, PHD2, in human

fibroblast .......................................................................................... 110

Abstract ................................................................................. 1 1 1

Introduction ............................................................................. l 12

Material and Methods ................................................................. 115

Results ................................................................................... 122

Discussion .............................................................................. 136

C11
Sun

IAPP
Iden'
unkl

ATTHEN
Hypoxi.

(J)

 

 

References .............................................................................. l 3 8

CHAPTER 4:
Summary ......................................................................................... 142
APPENDIX A:
Identification and characterization of genes susceptible to transcriptional cross-
talk between the hypoxia and dioxin signaling cascades ............................... 153
Abstract .................................................................................. 154
Introduction ............................................................................. 1 55
Material and Methods ................................................................. 158
Results ................................................................................... 168
Discussion .............................................................................. 185
References .............................................................................. 1 89
APPENDIX B:
Hypoxia, drug therapy and toxicity ......................................................... 194
Abstract ................................................................................. 195
1. Introduction .......................................................................... 196
2. Hypoxia signaling .................................................................. 197
2.1. Normoxia vs. hypoxia ................................................. 197
2.2. Occurrence of hypoxia ................................................ 198
2.3. The hypoxia-inducible factor (HIF) system ........................ 202
2.4. The integrative response to hypoxia ................................. 214
2.5. The balance between adaptation and cell death .................... 217
3. Hypoxia and pharmacological implications ..................................... 224
3.1. Hypoxia-related drug therapy ........................................ 225
3.2. Hypoxia and drug disposition ........................................ 232
3.3. Hypoxia and cancer chemotherapy .................................. 234
4. Hypoxia and toxicity ............................................................... 236
4.1. Link between HIF1 activation, metal-induced toxicity
and the prodeath response ............................................ 239
4.2. Indirect toxicity and unregulated hypoxic signaling ............... 240
5. Neuronal protection by stimulation of the hypoxia adaptive response. . .....243
6. Conclusions .......................................................................... 245
References .............................................................................. 250

vi

Figure 1-1

Figure 1-2
Figure 1-3
Figure 1-4

Figure 2-1
Figure 2.2l

Fi‘s’ure 2-3
Figure 2-4

“211m 2-
‘l
F igum 2.7
Figure 2-8
FigWe 2-9
Figure 3.]
Figllre 3.1

Figu’. e 3.1

 

Figure 1-1.

Figure 1-2.
Figure 1-3.
Figure 1-4.

Figure 2-1.

Figure 2-2.

Figure 2-3.
Figure 2-4.

Figure 2-5.

Figure 2-6.
Figure 2-7.
Figure 2-8.
Figure 2-9.
Figure 3-1.
Figure 3-2.

Figure 3-3.

LIST OF FIGURES

Structure of the HIFIOL protein and oxygen-dependent modification

of HIFla ......................................................................... 9
PHD-mediated hydroxylation of HIFIO. .................................... 12
The hypoxia signaling system ................................................ 15
Link between intermediary metabolism and HIFl signaling ............ 19

Expression levels of HIFIOL and HIF1 activity in MSUI lineage
of cells ........................................................................... 66

Expression levels of PHDs and the effects of modulating PHD

levels on HIF1-hypoxia signaling in MSUI lineage of cells ............ 69
Characterization of shPHD2 infected MSU 1.1 strains .................. 73
Tumor-forming ability of shPHD2-a strains .............................. 78

A graphic representation of the relationship between PHD2 levels

and tumor forming potential ................................................. 80
HIF-mediated cellular responses in shPHD2 strains ..................... 84
Inhibition of PHD2 activity in transformed cells ......................... 88
Over-expression of PHD2 in PH3MT cells ................................ 93
Characterization of tumor forming potential in PHD2 strains ........... 98
Tandem Affinity Purification ............................................... 118
Characterization of tPHD2-3MT and tGFP-3MT cells .................. 123
Verification of Tandem Affinity Purification (TAP) .................... 128

vii

Figur

Figun

Figure

Figure

Figure

Figure .

Flgure ‘1

Figure 11
FIgUre B
Figure B

Flgu re B

Figure 3-4.

Figure 4-1.

Figure A-l.

Figure A-2.

Figure A-3.

Figure A-4.

Figure A-S.

Figure B-l.
Figure B-2.
Figure B-3.

Figure B-4.

Figure B-5.

Figure B-6.

Protein profile of TAP-tagged purified PHD2 .......................... 130

Relationship between PHD activity and HIF1 target gene
Expression ..................................................................... 144

HRE and DRE position weight matrices and the sequences used
in their construction ......................................................... 163

Analysis of genomic data and comparison of 33 target genes ......... 169

qRTPCR verification of expression changes for selected cross-talk
gene ........................................................................... 174

Alignment and sequence analysis of the promoters and 5’ UTR
of core genes .................................................................. 177

Promoter analysis for over-represented motifs and correlation

with expression ............................................................... 180
PHD-mediated hydroxylation of HIF1 ................................... 204
The hypoxia signaling system ............................................. 207
Link between intermediary metabolism and HIF1 signaling .......... 212

Relationship between oxygen concentration and HIF1 target
gene expression ............................................................... 219

HIFIOL stability and transcriptional activity is controlled by
hydroxylation ................................................................. 222

Relationship between HIF1 activity and cell survival under stress. . .246

viii

Table 3‘

Table 3-

Table 3-

Table 3..
Table A-
Table A-;

Table 8-;

Table 3-1.

Table 3-2.

Table 3-3.

Table 3-4.

Table A-1.

Table A-2.

Table B-1.

LIST OF TABLES

Identification of proteins within individual band in Figure 3-4A
using mass-spectrometry .................................................... 132

Protein profile of TAP-tagged purified PHD2
(on-bead trypsinization) ...................................................... 133

Protein profile of TAP-tagged purified GFP
(on-bead trypsinization) ................................................... 134

Protein profile of TAP-tagged purified PHDZ (TEV cleavage). . . . . ...135

Primer information for qRTPCR ........................................... 166
Microarray fold change and information of 33 cross-talk genes ...... 172
Drug that target the hypoxia signaling cascade .......................... 227

ix

The
greater th

oxidase. ii

 

Ol enerex'

|
cope \\ ith

tissues of
because '1‘
maintain
Patholog-l
CElmer. “f
has mad;
research.
apprecia-
cit-em3 ft

Cellular]

5

1. mp”

1.l_ NUr

CHAPTER 1

INTRODUCTION

The reaction of Complex IV of the electron transport chain (ETC) utilizes
greater than 95 % of the oxygen we breathe. This reaction, catalyzed by cytochrome
oxidase, is required to maintain proper ATP production within the cell. This coupling
of energy production to the consumption of oxygen has made the ability to sense and
cope with low oxygen tension essential for survival. Decreases in oxygen reaching the
tissues of the body, a condition known as hypoxia, is detrimental to cells and tissues
because it decreases their ability to produce energy, and thereby disrupts their ability to
maintain proper function. Hypoxia and hypoxia-related signaling has been linked to the
pathology of all the major causes of death including cardiovascular disease, stroke and
cancer. The fact that hypoxia signaling is directly linked to a tumor’s ability to thrive
has made it and several hypoxia-activated genes the focus of intense drug discovery
research (12, 44, 104). To target hypoxia signaling successfully for therapeutics or
appreciate its role in xenobiotic toxicity, it is necessary to understand the cellular
events following loss of normal oxygen tension and characterize the role of hypoxia in

cellular transformation.

1. Hypoxia

1.1. Normoxia versus hypoxia

 

microscoi
between t
carrying c
is the gruc
0f the 0);
[31'6in d
alVEOll an

hemOglo’h

 

 

 

Oxygen is transported from the air that we breathe throughout the body via a
respiratory system. The primary function of this system is to deliver inspired oxygen to
peripheral tissues and remove carbon dioxide that cells produce as a byproduct of
normal cellular function. When a breath is taken, air passes from the airways into
microscopic air sacs called alveoli. The exchange of oxygen and carbon dioxide
between the alveoli and blood is called external respiration. Although the oxygen
carrying capacity of the blood is increased markedly by the presence of hemoglobin, it
is the gradient of oxygen partial pressure (p02) between the blood plasma and the locus
of the oxygen consumption in the mitochondrion that drives 02 into the tissues by
passive diffusion. The p02 in the inspired air is 150 mm Hg, whereas the p02 in the
alveoli and arterial blood is around 100 mm Hg. In the tissues, oxygen dissociates from
hemoglobin and diffuses through capillary endothelium into parenchymal cells, so that
the p02 of the blood draining tissues (i.e. venous p02) is far less than arterial p02,
averaging about 40 mm Hg. Thus, the p02 in the cells of the tissues is much lower than
that of the arterial blood.

The normal partial pressure of oxygen (p02) varies in cells within and among
different tissues. For example, the normal p02 of skeletal muscle has been reported to
be 20 - 30 mm Hg and that of brain is 45 — 65 mm Hg (28, 60). Even within a single
organ, cells are exposed to a wide range of oxygen tension in different locations. For

example, the p02 of the blood in liver varies by site, so that p02 in the hepatic artery,

portal \‘Cl.

 

mm Hg. i
“cellular
of methot
the locatxa I
gradients
mitochon.
mixes up.
cells Clo‘i
cells. Nex
touhkhl
lei'e] 0f &
[ISSUQS all

Exist,

1.2. 0Ccl1
ml
fetal dew

reqtlIEm‘

 

portal vein, sinusoids and central vein are estimated to be 95 —- 105 mm Hg, 50 —- 65
mm Hg, 35 — 45 mm Hg, respectively (17, 58, 90). Although estimates of values for
“cellular p02” have been made in numerous tissues, uncertainty exists due to limitation
of methodologies used for the measurements. Moreover, cellular p02 must vary with
the location of parenchymal cells along the length of the capillary (or sinusoid), and
gradients must also exist intracellulary due to the continuous consumption of Oz in the
mitochondria. Thus, cells in the portal triad, where arterial and portal venous blood
mixes upon entering the liver, are exposed to a more oxygen-rich environment than
cells close to the central vein, since the blood has lost 02 by diffusion into upstream
cells. Nevertheless, each of the cells along the sinusoid considers the degree of oxygen
to which they are usually exposed to be “normal”. How cells establish or perceive this
level of oxygen as “normoxic” is not entirely clear; however, when the p02 in the
tissues and individual cells drops below this normal level, a state of hypoxia is said to

exist.

1.2. Occurrence of hypoxia

Hypoxia arises in biological systems for a wide array of reasons, including
normal physiological variation and pathological conditions. During embryonic and
fetal development, hypoxia occurs as naturally as the process proceeds and the cellular

requirements throughout the developing organism display varying demands for oxygen.

£11

Studies iz‘.

 

embryo l‘
physiolfi
addition.
oxygcn‘l‘

doting 1h

 

 

 

Htpoxic .
during er
localized
to anaert:
highligh‘
and the :
C311 OCCI
increase
l8 also C

R.-
from in
“than.

IUmor

.- r,

Studies in vitro have demonstrated that the optimum oxygen tension in a mammalian
embryo is approximately 23 - 38 mm Hg, and this oxygen environment is an important
physiological factor for vasculogenesis, angiogenesis, and tubulogenesis (68, 126). In
addition, normal organogenesis in the developing fetus takes place in a relatively
oxygen-poor environment (relative to adult tissues), and increasing oxygen tension
during this time is deleterious for heart, kidney, and lung development (70, 115, 126).
Hypoxic condition can also occur in the adult under various conditions. For example,
during exercise muscle demand for energy can outpace the supply of oxygen, causing
localized hypoxia. This type of stress leads to a buildup of lactic acid as the body turns
to anaerobic metabolism to address the energy debt. These normal biological processes
highlight the balance that is established between oxygen levels and energy productions
and the importance of a programmed response to disruption in this balance. Imbalance
can occur either through reduced oxygen availability, as in the developing embryo, or
increased energy demand, as in the exercising muscle. More importantly, this balance
is also central to hypoxia’s role in pathological conditions.

Recent interest in hypoxia and its effects in biological systems has stemmed
from its role in a wide variety of pathological conditions, most notably cancer.
Hypoxia is an important feature of most solid tumors, arising from the process of
tumor formation and progression that is characterized by rapid cell proliferation and

drastic changes in the local oxygen supply. Rapid cellular expansion quickly outpaces

the abilit.

 

hypoxic t
hypoxic 1
the tumo,
respond.

The resul

 

abnormal
(llSOl'ngix
I0 irregu
deVClOpc
SlllCC m5
(‘er 10 l
HlPOXia

For mar

the ability of a tumor to create new blood vessels for oxygen delivery, leading to a
hypoxic or even anoxic microenvironment within certain portion of the tumor. The
hypoxic microenvironment occurs very early during tumor development, beginning as
the tumor reaches approximately 2 - 3 mm in a diameter. To this the tumor must
respond, primarily by increasing its glycolytic rate and angiogenic potential (44, 47).
The resultant newly formed vasculature differs from that in normal tissues, displaying
abnormal structure and function. For example, the new tumor vessels are highly
disorganized and usually leaky, with incomplete endothelial lining, qualities that lead
to irregular blood flow and diminished nutrient and oxygen supply (18, 65). Newly
developed vasculature, therefore, does not always address the oxygen debt of the tumor
since many portions remain further than 150 um from the nearest blood vessels, which
due to tissue 02 consumption, is the approximate diffusion limit of oxygen (18).
Hypoxia, therefore, remains a constant feature of tumors even after neovascularization.
For many years, tumor hypoxia has been considered a challenge for cancer therapy
because of its adverse impact on the effectiveness of radiation and chemotherapy.
Moreover, low oxygen availability has recently emerged as a major factor that
enhances malignant progression, because tumor cells that have adapted to hypoxia gain
various advantages in growth. In most cases, this is a sign of poor prognosis (6).
Disruption in oxygen homeostasis also occurs during other pathophysiological

conditions such as cardiovascular disease, stroke, and chronic pulmonary disease (100).

Atherose'.
the vascu
muscle l

viability

 

 

 

disrupts tl

_ l
nutrient s
time. it c
pulmons
C1111 be
hipoxia
and It (
UllimmC
seen in
“'OUndi
betwee,
is,

(lama

U

o

(If?
f

“-3):

2' ”Jpn

Atherosclerosis often leads to arterial stenosis which ultimately impairs perfusion of
the vascular bed and further disruption oxygen and nutrient flow to regions of the heart
muscle (98). When this ischemia persists, it can irreversibly reduce myocardial
viability (95, 100). Stroke is a well-known cause of cerebral hypoxia because it
disrupts the normal flow of blood in the brain and leads to localized loss of oxygen and
nutrient supply. In brain tissues, even if this disruption only lasts for a short period of
time, it can lead to neuronal cell death and disability (55, 100). Chronic obstructive
pulmonary disease (COPD) is the most common form of pulmonary dysfunction and
can be subcategorized as asthma, chronic bronchitis, and emphysema. Alveolar
hypoxia is common in COPD because of the difficulties in expelling air from the lungs,
and it can lead to pulmonary hypertension, pulmonary arteriolar remodeling, and
ultimately heart failure (123, 124). The hypoxia-induced injury and tissue remodeling
seen in these pathophysiological conditions and others, including diabetes and
wounding, can be considered a response to the inability to maintain the balance
between oxygen delivery and requirement. The injurious process takes place because
the tissue can not adapt to the change in oxygen tension. Toxicant-induced cellular
damage that involves the hypoxia signaling system can also be considered in a similar

way.

2. Hypoxia signaling; the hypoxia-inducible factor (HIF) system

Ae
especiall.
energy d
energy in
response
fact. QCllt.

T l
to hypoxi-
interactio
lHlFSl \I‘
ll] rcglllu
ellihropt
hYPOXIa
regulate.
hUndred

T‘

“Elmer

p—a

ha

COH ”(ll 5

8:1le1] ll

Aerobic metabolism provides a significant advantage to multicellular organisms,
especially in the area of energy production. When oxygen supply cannot meet the
energy demand of the cells, however, an adaptive response must compensate for the
energy imbalance to maintain tissue function. A primary mechanism for this adaptive
response is the transcriptional regulation of a battery of hypoxia-responsive genes. In
fact, genomic screens have shown that thousands of genes are influenced by exposure
to hypoxia (108, 116). The most well studied mechanism identified in this process is an
interaction of a family of transcription factors, called the hypoxia-inducible factors
(HIFs) with a cis-acting element, called a hypoxia-responsive element (HRE), located
in regulatory region of target genes. This mechanism was first demonstrated for the
erythropoietin gene, the expression of which was upregulated more than 100-fold by
hypoxia via HIF-induced transcription (2, 92). Since then more than 70 target genes
regulated directly by HIF have been identified, and expression of over several
hundreds genes are known to be directly or indirectly influenced by HIF (104).

The family of HIFs belongs to the Per-ARNT-SIM (PAS) superfamily of
transcription factors and is characterized by the presence of the PAS domain that
controls dimerization (48, 50). HIFs are heterodimeric proteins comprising 0t and B
subunits. The alpha class is composed of HIF 10L, HIF20t, and HIF30L. The beta subunit
includes the aryl hydrocarbon nuclear translocator (ARNT, also known as HIF 1B) and

ARNT2. HIFl is the most widely studied HIF heterodimer and is the combination of

HlFla a:

 

most ma.
whereas l'
concentr.
is quiek':_
hypoxia 1

elicit a re

 

HIF ptott
|
aCllVaiim-
QVEr IIIC
hydI'OXI
ll

dimem

HIF 1a and HIFIB (118). HIF1 is the principal regulator of the hypoxic response in
most mammalian cells (103). HIFIB is a constitutively expressed, nuclear protein,
whereas the expression and activity of HIFla is tightly regulated by oxygen
concentration. It rapidly accumulates upon exposure to hypoxia and on reoxygenation
is quickly degraded with a half-life of less than 5 min (123). Given how detrimental
hypoxia is to a cell, this rapid response time highlights the speed at which a cell must
elicit a reaction to loss of oxygen tension. Equally important is the short half-life of the
HIF protein following reoxygenation, and this time frame suggests that prolonged HIF
activation might not be beneficial to the cell. Our understanding of the mechanism a
cell uses to sense hypoxia and convey this signal to the or subunit has greatly increased
over the last several years. This is due to the recent identification of a family of
hydroxylases that modify the or subunit in an oxygen dependent manner.

HIFIOL contain basic helix-loop-helix (bHLH) and PAS domains that mediate
dimerization and DNA binding (Fig. l-l). HIF 10L also contains transcriptional
activation domain (TAD) that regulated its transcriptional activity and oxygen
dependent degradation (ODD) domain that control the half—life of or subunit in an 02-
dependent manner (Fig. 1-1). The ODD domain includes conserved proline residues,
Pro 402 and Pro 564, which are post-tranaslationally hydroxylated under normoxia.
Three hydroxylases, known as prolyl hydroxylase domain containing proteins ((PHDs),

also known as egg laying abnormal 9 homologues (EGLNS) and hypoxia prolyl

Figure 1-1. Structure of the HIFla protein and oxygen-dependent modification
of HlFla.

The HlFlor subunit (826 amino acid) is divided into several functional domains:
Dimerization and DNA binding domain (bHLH and PAS domain), oxygen
dependent degradation domain (ODD), and transactivation (TAD). The expression of
HIF10t is regulated by oxygen-dependent post-translational modification. Under
normoxia, conserved proline residues (P) 402 and 564 within the ODD domain of
HIFIOt are hydroxylated by the PHD enzyme. This process is required for the
binding of the von Hippel-Lindau tumor suppressor protein (pVHL), the recognition
component of an E3 ubiquitin—protein ligase. Ubiquitination of HIF 10L targets the
protein for degradation by the 26S proteosome. pVHL binding might also be
promoted by acetylation of lysine (K) residue 532 by ARDl acetyltransferase.
Oxygen also regulates the interaction of HIF 10L with transcriptional co-activators.
Oxygen dependent hydroxylation of asparagine (N) residue 803 within the TAD of
HIFIOL by FIH, blocks the binding of CBP and p300 to HIF10t and consequently
inhibits HIF1-mediated transcription. Under hypoxic conditions, hydroxylation is
inhibited and pVHL cannot bind to the HlFlot leading to HIFIOL accumulation and
interact with CBP and p300 allowing transcriptional activation of HIF1.

 

:o=u>=o<
3:35:35;

 

 

Iv cos—wuacmoo

522533:

hydroxyl.
residues \
hydroxyl.
tumor an
E3 ubiqu:
30). PHD
common
hydroxfl.

aflllllly C

 

 

 

 

l'fiquire It

 

 

hydroxylases (HPHs)), control HIFIOL stability by modifying conserved proline
residues within the ODD of HIFIOL under normoxia (Fig. 1-1) (7, 30). The proline
hydroxylations are required for the interaction of HIFIOL with the von-Hippel Lindau
tumor suppressor gene product (pVHL). pVHL serves as the recognition component of
E3 ubiquitin-ligase that leads to HIFIOL ubiquitination and proteosomal degradation (7,
30). PHDs contain a double stranded B-helix (jelly roll) core and iron-binding residues
common to members of the dioxygenase family, which includes the collagen prolyl 4-
hydroxylase (C-P4H) (85, 97). The catalytic properties of PHDs, including kinetics and
affinity constants for the co-substrates, have been clarified (45). These hydroxylases
require four factors for activity: oxygen, iron, or-ketoglutarate (orKG) and ascorbate
(Fig. 1-2). The enzyme process involves decarboxylation of the orKG to yield succinate
and concomitant hydroxylation of the targeted residue. The oxygen, iron, and
enzymetic reaction products play a critical role in regulation of the hydroxylase. The
proline residues hydroxylated in HIFIOL are all present in the context of a Pro-Xxx-
Xxx-Leu-Ala-Pro consensus sequence (54). Human type I and type II C-P4H cannot
hydroxylate this evolutionary conserved proline (46, 54). All the PHDs hydroxylate the
C-terminal hydroxylation site, Pro 564, and Km values of the three PHDs for the
peptide are similar. PHDl and PHD2 showed higher Km values for the peptide
containing the N-terminal hydroxylation site Pro 402 and this peptide was not

hydroxylated by PHD3 (45, 46). Differences also exist between PHDs in their

11

Figure 1-2. PHD-mediated hydroxylation of HIFIO.

Prolyl hydroxylases (PI-IDs) require oxygen, iron, ascorbate and (x-ketoglutarate for
activity. During the enzymatic process, or—ketoglutarate is decarboxylated, yielding
succinate, and the HIFIOL ODD is hydroxylated on conserved proline residues. Once
hydroxylated, HIFIOL is quickly degraded in a pVHL-ubiquitin-26S proteosome-
dependent manner.

12

 

Oxygen
Fe2+
ascorbate
or-ketoglutarate

\Jsluccinate

HL
ubiquitin

/\ ‘O/H\@$>,§7*

.

HIF1

 

13

 

intracell L;
whereas
comparti
specific I

responsib

 

different
inducibil.
physiolo
HlFlor j
accumui
are No
1‘31. Fir
Substrat
aVaHab]
t0 mm,
{Or [lies
has [QQ
Piggy“
Habit”:

[he 0Xx_

intracellular localization. Human PHDI is predominantly expressed to the nucleus,
whereas PHD2 is present in the cytoplasm (45,82). PHD3 distributes evenly in both
compartment in human. The expression of PHDs are induced by hypoxia in cell type
specific manner, suggesting a role for these enzymes in a negative feedback pathway
responsible for enhanced degradation of HIFIOL after reoxygenation (3, 16, 24). These
differences in substrate specificity, intracellular distribution and tissue specific
inducibility under hypoxia imply distinct function for the three PHDs, however the
physiological impact of this diversity is not clear. The PHD-mediated hydroxylation of
HIFla is inhibited under low oxygen conditions, leading to HIF1 0t stabilization,
accumulation, and consequent induction of HIF1 activity (7, 30, 69). Currently, there
are two theories as to how the PHD senses the decrease in cellular oxygen tension (Fig.
1-3). First, since these enzymes require oxygen for activity, hypoxia acts as a loss of
substrate, and this might be all that is required for PHD inhibition. The loss of
available oxygen would inhibit the enzyme’s ability to modify HIF 10c and thereby lead
to increase transcription factor stability. The disparity between the oxygen requirement
for these enzymes and the level of hypoxia needed to induce HIFla stability in cells
has led others to consider alternative mechanisms. In enzyme preparations, a
progressive decrease in PHD activity from just below 20% Oz induces an increasing
stability of HIFlor. In contrast, HIFla stability is not increased in most cell types until

the oxygen levels are decreased to less than 7% Oz (30, 114). In the stabilization of

14

 

 

Figure 1-3. The hypoxia signaling system.

The hypoxia signaling cascade begins in the cytosol where the PHDs are continuously
responding to local oxygen concentrations. Once the available oxygen concentration
reaches a certain critical point, the PHD becomes inhibited. This can involve direct
inhibition of the PHD due to loss of its molecular oxygen substrate or indirect
inhibition due to changes in reactive oxygen species (ROS) produced at complex III of
the ETC. This change in ROS is caused by the inhibition of the ETC due to
decreased oxygen, the terminal electron acceptor at complex IV, and subsequent
accumulation of ubisemiquinone at complex 111. Once the PHD is inhibited,
HIFla translocates to the nucleus where it interacts with ARNT and becomes
transcriptionally active. HIF1 target genes include adaptive and cell death genes. The
cellular increase in anaerobic metabolism leads to the overproduction of lactate and a

decrease in cellular and local tissue pH.

15

 

 

 

 

     

 

 

 

 

 

 

 

>

Om.

\ Emsouam

88820336

3.3.603 a... co 00 .E
14.0.... t . U c a.

55%... ligase Q Crk

 

 

.383
\m:o_o:z
052 J

9:? ...an I mm: I «ma
«Ms. v 33323 $0026
Emma .30 «
Eugene—25$ I
m_me:emo_mc< — w {IV n.._.<
/ . .. Iv m_m>_oo>_0 l «
g 3923 3328..

16

HIF1 a. s
generatet
mitochor
suggestir]

on this.

 

 

proposed

l
Chain in
NADH ;
I0 mOlCc
of ROS
site of .
diSmptx
‘0 CB’Cle
in ROS
indUCtio
is jmm‘

begins. [I

H.

 

HIFIOL, several laboratories have reported a role of reactive oxygen species (ROS)
generated in the mitochondria. These experiments have documented an inability of
mitochondria—deficient cells to induce HIFIOL stabilization during hypoxic stress,
suggesting a role for this organelle in the response (13). Recent reports have elaborated
on this, establishing complex III as the likely site of action (8, 13, 41, 77). The
proposed mechanism accounts for evidence implicating the mitochondrial respiratory
chain in regulation of HIF1 activity (Fig. 1-3). In respiration, electrons donated by
NADH and FADH; flux through complex I-IV of the ETC and are finally transferred
to molecular oxygen at complex IV. Hypoxia causes ETC inhibition and the generation
of ROS from a number of potential sites, such as complex I and the ubisemiquinone
site of complex 111. It has been proposed that this oxidative stress within the cell
disrupts the catalytic activity of the PHI), possibly through inhibiting the ability of iron
to cycle between oxidation states. In recent years, it has become evident that changes
in ROS levels within the cytosol result in PHD inhibition, ultimately leading to the
induction of HIF1 transcriptional activity (14, 38, 41). Regardless of which mechanism
is involved in the inhibition of the PHD, hypoxia leads to HIF 10L stabilization, and this
begins the cell’s attempt to adapt to the decrease in available oxygen.

Hypoxia is not the only signal that can lead to HIFIOL stabilization. It has long
been known that a number of chemicals, such as cobalt, deferoxamine and orKG analog,

dimethyloxaloglutarate (DMOG), can induce the stabilization of HIFIOL experimentally

17

(30. 78)=
with iron
and llll:
during ti:
the gene
demonsti
people l:
humans
[he IUIII',
Signs Ol
cenain r

mOdulat;

 

 

(30, 78). They act by directly inhibiting PHD activity by either removing or competing
with iron or OLKG. More recently, a link has been established between key metabolites
and HIFIOL stabilization. As discussed above, PHD requires orKG for activity and,
during the catalytic process, generates succinate. Recently, families with mutations in
the gene encoding succinate dehydrogenase (SDH, complex II of the ETC) have been
demonstrated to display characteristics of a pseudo hypoxic response (5, 22, 99). These
people have a gene expression profile similar to that induced by hypoxia in normal
humans and are prone to pheochromocytomas. In addition, people with mutations in
the fumarate hydroxylase (FH) gene, another citric acid cycle enzyme, also display
signs of aberrant hypoxia signaling (53). People with FH mutations are also prone to
certain types of cancer, presumably this involves direct inhibition of PHD through
modulation of the decarboxylation step within the enzyme. The mutation in SDH
would tend to increase intracellular concentrations of succinate, leading to PHD
dysfunction via product inhibition (Fig. 1-4). The FH mutation would be expected to
lead to increased fumarate levels within the cell, and fumarate can act as a competitive
inhibitor for the OLKG binding site in PHD (Fig. 1-4). Finally, the glycolytic end
product, pyruvate, regulates HIFIOL stability, and this activation might be responsible
for the Warburg effect (75).

In the early part of the last century, Otto Warburg described a tumor’s increased

dependence upon fermentation or anaerobic glycolysis (119). He suggested that the

18

Figure 1-4. Link between intermediary metabolism and HIF1 signaling.

Recent reports have established a direct link between key metabolic intermediates and
HIF1 signaling. These metabolites include a—ketoglutarate (or-KG), which is necessary
for PHD activity. Recently, it was demonstrated that patients with mutations in their
succinate dehydrogenase gene (Complex II of the ETC) display characteristics of a
pseudo-hypoxic state. Another link was established between fumarate disregulation
and HIF1 signaling, and it was suggested that fumarate competes for or-KG binding
within the PHD enzyme. Finally, pyruvate acts in a feed-forward mechanism to
stabilize HIFIOL in the absence of hypoxia. The recent characterization of pyruvate
dehydrogenase kinase (PDK) as a HIF1 target gene might amplify this linkage between
pyruvate and hypoxia signaling, since PDK over—regulation would serve to trap

pyruvate in the cytosol.

l9

 

 

 

 

             

 

.3030

 

«mg
it £33.23

Maren—4M0 «
m_m¢_ono._ 55m ; d« {V QP<
«

flee—35292
«

m_m>_oo>_0
3303 ”egg—Er.

 

 

OmOO—a—mv

 

 

 

 

 

gnawed.

 

 

 

Team

       

I .lit +1] .7: 1+:

9
>— h 2...,

 

 

          

 

>
- N I N .
a: ao< o z o ._ .

owwfiuoam .. 90:5". .
o .00: 3a «S.
< o. m ._ . I“

        

      

   

'

\ 32:33.

 

   

 

.. “833—59338
22225231: All 395003
23.33
..«ou.

 

5ng

 

20

increased

 

cellular t
pyruvate
possibilit

prolongei

 

dehydrogi
for inact
Qtosol.
Qtosoli.
Warburg
These c
the cell

‘0 be UT

increased glycolytic activity and metabolic adaptation might be the reason for the
cellular transformation. The demonstration that metabolic intermediates of glycolysis,
pyruvate and oxaloacetate, can activate HIFlathrough inhibition of PHD raises the
possibility that it can participate in feed-forward activation of HIFIOL leading to a
prolonged hypoxic signal (74). In addition, recent reports have shown that the pyruvate
dehydrogenase kinase (PDK) gene is a HIF1 target gene (61, 116). PDK is responsible
for inactivating the pyruvate dehydrogenase complex, thus trapping pyuvate in the
cytosol. This would further serve to prolong the hypoxic signal by maintaining the
cytosolic concentration of pyruvate. The resultant extended signal might explain the
Warburg effect and directly links HIF1 activation to increased glycolytic activity.
These combined results suggest that a better characterization of the metabolic state of
the cell, as well as oxygen and PHD levels, is required if the HIF signaling cascade is
to be understood and targeted successfully by therapeutic agents.

Other modifications, in addition to PHD-mediated hydroxylation, have been
linked to hypoxia signaling. The transcriptional activity of HIFIOL is also reduced
under normoxia by oxygen-dependent hydroxylation of asparagine (Asn) 803 within
the C—terminal transactivation domain (CTAD) of this transcription factor (Fig. 1-1)
(76). Factor inhibiting HIF (FIH) mediates this modification of the conserved Asn
residue, and the hydroxyl-asparagine prevents the binding of co-activator p300/CBP to
HIFIOL (69). This effectively removes the ability of HIF 10c to direct p300-dependent

transcription; however, it remains possible for the transcription factor to modulate

21

nanscnp
oflhidt
HlFltlst
(Fig. 1-1
Dunng 3

Hlfltta

Th
tipresst
homeost
receptor
achraut
mTOR .
enhance
Oxlgcn-

 

transcription in a p300—independent manner. It has also been reported that acetylation
of HIF 10L at lysine 532 by arrest defective-1 (ARDl) acetyltransferase might regulate
HlFlot stability by enhancing the interaction between hydroxylated HIFIOL and pVHL
(Fig. 1-1) (35, 56). Phosphorylation is also involved in regulating HIF1 activity.
During hypoxia, p42/p44 mitogen activated protein kinase (MAPK) phosphorylates

HIFla and enhances the transcriptional activity of HIF1 (93, 96).

HIF1 activity is also regulated in an oxygen-independent manner at the HIFIOL
expression level by certain growth factors that ensure the maintenance of oxygen
homeostasis in normal growing tissues. For example, signaling via HER2 and IGF—1
receptor tyrosine kinases increase HIF1 expression. These responses result from the
activation of signaling pathways involving phosph-inositide 3-kinase (PI3K)/AKT/
mTOR that lead to the increases in the translation factor elF-4E, which in turn
enhances HIFlamRNA translation (96, 104). HIF1 is also regulated by additional
oxygen-independent molecular processes. Recently, several studies have demonstrated
that exposure of cells to certain nitric oxide (NO) donors or gaseous N0 molecules
modulated HIF1 activity. For example, treatment of cells with S-nitrosoglutathione
(GSNO), DETA-NO, or NOC18 induces HIF1 activity under non-hypoxic conditions
(59, 88). In contrast, sodium nitroprusside (SNP) inhibits hypoxia-induced HIF1
activation (42, 111). The molecular mechanisms regulating HIF1 signaling in response

to N0 molecules are under investigation but appear to involve effects on both HIFIOL

22

regulatek
relallVc

remainfi

 

3.1119 lI

.\l
infrastn‘:
blood L‘s“
circular:
Decreas
upregula
producti
HlFl tl‘.
hypoxia
responv.
Angiog.
oxygen

Vascult.

 

synthesis and stability. In summary, there are many ways in which HIF activity can be
regulated, some of which are oxygen-dependent and others of which are not. The
relative importance of these various conditions on normal and pathophysiology

remains to be determined.

3. The integrative HIF1-mediated responses to hypoxia

Multicellular organisms have developed sophisticated physiological;
infrastructure to maintain oxygen homeostasis. Hemoglobin, a major constituent of red
blood cells, is responsible for carrying oxygen to peripheral tissues, and the number of
circulating erythrocytes is a major determinant of tissue oxygenation (9, 102).
Decreased oxygen availability results in compensatory stimulation of erythropoiesis by
upregulating the production of erythropoietin (EPO) (l02). EPO enhances the
production of red blood cells, and its hypoxia-induced transcription is regulated by
HIF1 through an HRE site in the 3’ flanking region in the EPO gene (2, 92). The
hypoxia-induced expression of the EPO gene is thought to be part of a systemic
response that an organism initiates to cope with the decrease in available oxygen.
Angiogenesis is a tissue-oriented response to hypoxia and is critical to restoring
oxygen transport to ischemic areas by improving blood supply to the affected tissues.
Vascular endothelial growth factor (VEGF) is a central factor controlling physiological
and pathological angiogenesis, and, in most cases, it is induced in response to hypoxia

(113). The expression of VEGF is regulated by the binding of HIF1 to HREs located in

23

 

   

created
due to It
increase
one plac
Efficient
ATP pg!
firnctior
“Pregul
Sllppl-V
@85an
bfien v
aCllVir
han\p
“6t
PHD

PHD:

the 5’-flanking region of VEGF (107). In addition, VEGF receptors and the pro-
angiogenic cytokine interleukin-8 are upregulated in response to hypoxia (33, 83).

The cellular response to hypoxia is focused on addressing the energy deficit
created by the decrease in available oxygen. When aerobic metabolism is diminished
due to the lack of molecular oxygen, the cell utilizes the HIF1 signaling cascade to
increase the expression of the complete battery of glycolytic enzymes, since it is the
one place the cell can turn to address its energy debt. The glycolytic process is not as
efficient as the ETC and ATP synthase in producing ATP, generating only 2 moles of
ATP per mole of glucose, and this inefficiency can lead to impairment of physiological
function (101). The increased dependence upon anaerobic glycolysis is coupled to the
upregulation of various glucose transporters and other growth factors in an attempt to
supply the cell with the necessary catabolic reagents to carry out this adaptative
response (15). The importance of HIF1 in this switch to anaerobic metabolism has
been well demonstrated with in vivo and in vitro models. Cells with a loss of HIF
activity display reduced hypoxia-induced expression of 13 different glucose
transporters and glycolytic enzymes, compare with their wild type counterpart (15, 55,
116). Finally, this adaptive response also involves HIF1-mediated transcription of the
PHD themselves (3, 16, 31). This ability is PHD isoform- and cell type-specific.
PHD2 and PHD3 are upregulated in response to hypoxia in a wide variety of cells,
whereas PHD] is only marginally regulated by loss of oxygen tension in most cells.

There has been speculation that this HIF1-mediated upregulation of PHD functions as

24

35

SU

Fl

fr

a shut-off mechanism for the hypoxic response once normal oxygen tension is restored.
This might be one role for such feedback inhibition; however, there may be a more
subtle role for this regulatory loop, that is, that this feedback loop actually establishes
the set-point for “normoxia” within the various cell types of body. The cells in various
tissues and even within a single tissue are exposed to a wide range of oxygen tensions.
For example, liver cells close to portal vein and hepatic artery are exposed to a greater
oxygen tension than those near the central vein, and yet each cell type perceives the
oxygen concentration to which it is exposed as normal. It is possible that the
decreasing oxygen concentration along the sinusoid leads to a partial activation of
HIF1-mediated transcription of the appropriate PHD. The increased PHD levels along
the sinusoidal oxygen gradient could compensate for reduced delivery of oxygen,
similar to the way an increased receptor concentration can compensate for lack of
available ligand in signal transduction. In support of this notion, recent studies have
suggested that HIF1-dependent regulation of PHD levels might provide a self
regulatory loop which defines a tissue specific threshold for HIF1 activation as a
function of p02 rather than simply accelerating HIFIOL degradation following
reoxgenation (110). In this manner, the HIFlzPHD-linked transcription can establish
the required levels of HIF1 signaling necessary for the cell’s microenvironment and
normal cellular function.

The adaptive response, including altered PHD level and upregulation of EPO,

VEGF and anaerobic glycolysis, is not always capable of addressing the hypoxia-

25

 

 

induced
various

stressed
various

family r
and Nl.‘
cycle ar.
“suicide
currentl

[he llsxt
death ct

arid ott

4- “In

 

 

induced disruption of cellular homeostasis. In these cases, the cell begins modulating
various cell death pathway presumably as a mechanism to eliminate irrecoverable
stressed cells (18, 106). This pathway involves HIF1-dependent transcription of
various pro-apoptotic B-cell chronic lymphocytic leukemia (CLL)/ lymphoma (BCL)
family members, such as BCL2/adenovirus ElBl9 kDa interacting protein 3 (BNIP3)
and NIX (26, 89). The activation of HIF is able to contribute to p53-mediated cell
cycle arrest and cell death by stabilizing p53 through cross talk (1, 11). This increased
“suicide” response seems in opposition to the adaptive response described above. It is
currently thought that this pathway to cytotoxicity is an attempt by the cells to maintain
the tissue as a whole. Removal of cells under severe hypoxic stress by programmed
death could increase the chance of survival for neighboring cells by increasing nutrient

and oxygen supply and maintaining appropriate tissue architecture.

4. Hypoxia (or HIF1) signaling and cancer

Hypoxia is frequently exhibited in human cancers especially in solid tumors
when its mass reaches a certain size and oxygen delivery becomes limited. Interest in
the role of hypoxia in cancer biology has grown exponentially in the two decade since
its identification (105). Hypoxia was initially studied because of its effects on response
to radiotherapy. Radiation treatment requires free radicals from oxygen to destroy
target cells, and cells in hypoxic area were found to be resistant to radiation-induced

cell death. Tumor cells within the hypoxic areas were observed to survive and continue

26

all

pro]
beti
doc
hit
can
of '

m:

as
for

[Ill

pr:

proliferating, in contrast to those in perfusion-limited areas (120). The association
between tumor hypoxia and malignant transformation has been extensively
documented for a variety of tumors (4, 47, 109). For example, cellular adaptation to
hypoxia and altered glucose metabolism are well known fundamental characteristics of
cancer biology. It is also known that tumors can not grow beyond several mm3 because
of the restricted capacity of oxygen, glucoses, and other nutrients from blood vessel
without proper adaptation. Once a tumor reaches this size, its volume is subject to
minor changes as the rate of cell death equals that of cell division. Several factors can
assist carcinomas in conquering it limits in tumor growth, for example new vessel
formation. As a matter of fact, angiogenesis is considered as an essential process for
tumor growth and metastasis. However, as mentioned above, rapid cancer cell
proliferation easily outpaces the rate of angiogenesis and newly formed tumor vessels
are much less functional. New vessel formation, therefore, does not address the oxygen
debt; consequently hypoxia remains a constant feature of tumor, forcing the tumor to

adapt to these environmental conditions.

4.1. HIF1 activity as a unique characteristic of tumor
The ability of cells to adapt to hypoxia is important for cancer cell survival. As
mentioned above, cells possess mechanisms to response to low oxygen condition and
one of the major regulators in this response is a transcription factor HIF1. HIF1-

mediated signaling would appear to be critical to promote tumor cell survival and

27

even tumor growth. In fact, significant HIFlOL levels are one of the unique
characteristic of various human cancers including lung, prostate, breast, colon
carcinomas, which are the leading causes of US. cancer mortality (73, 127).
Moreover, overexpresison of HIFIOL occurs very early in carcinogenesis before
histological evidence of angiogenesis or invasion. For example, HIFIOL protein is
detectable in preneoplastic and premalignant lesion such as breast ductal carcinoma,
colonic adenoma, and prostate intraepithelial neoplasia (127). HlFla, indeed, has
been assessed for its use as a novel biomarker for precancerous lesions that warrant
clinical surveillance or therapeutic intervention.

Significant basal HIFICX and its activity are seen not only in poorly oxygenated
solid tumors but also in well oxygenated tumor area and metastatic nodules. Moreover,
many cancer cell lines cultured in normal growing condition (i.e. 21% oxygen)
display significant basal levels of HIFIO. protein as well as HIF1 mediated gene
expression. Tumor cells possess an increased basal glycolytic activity (Warburg
effect) to supply required energy under the lack of oxygen availability. In addition,
cancer cells use glycolysis for energy production preferentially even in the presence
of oxygen (75, 119). The aerobic glycolysis of cancer cells, in fact, contributes
directly to tumor progression and malignant transformation by promoting HIF1
activation. As mentioned, glycolysis may represent a feed forward mechanism to the
expression of genes turned on by HIF1, since glycolytic metabolites can promote

HIF1 activatiOn and many genes encoding glycolytic enzymes, glucose transporters

28

and glucose regulatory hormones are induced by HIF1 (73, 75, 23, 102). Glycolysis
may also contribute to the hypoxia-independent induction of HIF1 by several
endocrine agents and environmental toxins. Elevated hexokinase and
phosphofructokinase activities are hallmarks of cancer cells and phosphofructokinase
activity is stimulated by insulin, insulin-like growth factor, and epidermal growth
factor (10), all of which can induce HIF1 under normoxia through a mechanism
involving the phosphatidylinositol 3-kinase (PI3K) signaling pathway (10, 32).
Cancer cells, consequently, display significant HIF1 activity throughout their
progression, suggesting that hypoxia adapted cells with HIF1-induced glycolytic
activity as well as others signaling pathway may maintain HIF1 activity even after
they escape the hypoxic environment to take an advantage of HIF1-mediated

processes.

4.2. Positive role of HIF1 signaling in tumor progresses

HIFl induces expression of various growth factors that are known to promote
cell proliferation. This is normally involved in initiating cell migration and
regeneration after hypoxia damage. Growth factors such as transformation growth
factor-or and -B (TGF-or and -B), insulin-like growth factor 2 (IGF2), and platelet
derived growth factor (PDGF) are known HIF1 target genes and whose role in cellular
transformation have been extensively studied (27, 62, 67). The most well studied HIF1

activated growth factors regulate endothelial—cell proliferation and blood vessel

29

fonna
recept
expres
menti.
the hr‘.
vascui
Him
paths
and ti

CTOSsI.

55min

llle Hl

formation. HIF1 induces expression of a key angiogenic factor, VEGF, and one of its
receptors, VEGF receptor 1 (VEGFRllFLTl). HIF1 signaling also leads to reduced
expression of anti-angiogenic proteins such as Thrombospondin-l and -2 (25, 64). As
mentioned, tumors cannot grow beyond a certain size without angiogenesis because of
the limited diffusion of 02, glucose, and other nutrients. In many cancers, the degree of
vascularization is strongly correlated with malignancy and patient death (23, 43).
Hypoxia-induced angiogenesis is blocked by inhibitors of oncogene signaling
pathways, such as agents that inhibit RAS, epidermal growth factor receptor (EGFR),
and the receptor tyrosine kinase ERBB2 (HER2/NEU), which indicates that there is
crosstalk between oncogenic and hypoxia response pathways.

Hypoxic cancer cells use glycolysis as a primary energy source and cellular
transformation and clonal expansion of cancer cells depends on enhanced glucose
transport and glycolysis (23, 119). In addition, recent studies indicated that cells also
use this pathway as an energy source during metastasis (81). As mentioned, HIF1
regulates the expression of all the enzymes in the glycolytic pathway, as well as
expression of the glucose transporter GLUTl and GLUT3, which mediate glucose
uptake (15). The intermediates of glycolytic pathway provide the precursor for
synthesis glycine, serine, purines, pyrimidines and phospholipids, all of which are
required for tumor cell growth and maintenance (44). Tumor cells that are deficient in

the HIF1 signaling have lower ATP and glycine concentration in vivo (44).

30

Tu

m:

Ca

rep

cel

EXI

EX;

The glycolytic activities of cancer cells also affect the overall pH of tumor.
Tumors have been shown to adapt to pH changes and grow at lower pH than are found
in normal tissues, giving the tumor growth advantage. Glycolysis is thought to be the
main mechanism by which tumors lower their pH, through generation of lactic acid.
Carbonic anhydrase, which reversely converts carbon dioxide and water to carbonic
acid, might also be involved. The activities of carbonic anhydrase 9 and 12, were
reported to be regulated by HIF] and strongly induced by hypoxia in a range of tumor
cells (121).

HIF1 signaling also contributes invasion and metastasis in tumor cells. Previous
studies showed direct link between HIF1 and a cells’ invasiveness. Cells transfected
with HIF10t expression vector resulted in significant increase in invasiveness of cells
under both normoxic and hypoxic conditions (63). Consistence with their results, the
complementary experiment using siRNA targeting l-IIF 10L decreased the colon
carcinoma invasion (63). Cancer cells produce proteases including the urokinase-type
plasminogen-activator receptor (uPAR), cathepsin D, and matrix metalloprotease-2
(MMP2), which enhance the cells’ motility by digesting basement membrane/
extracellular matrix (ECM) (29, 72). The analysis of mouse ES cells showed that the
expressions of these genes were regulated by HIF1. HIFl also increases the expression
of autocrine motility factor (AMF), transforming growth factor-0t (TGF—(x), and
surface receptor such as c-MET tyrosine kinase, fibronectin 1, keratins, and vimentin,

which promote the fluid structure that required for pathophysiology of invasion (63).

31

pl

CZ

ei

\l.

Cl

In addition, hypoxia signaling is involved in many other aspects of tumor
progression, further decreasing a patient’s chance of survival. It can select cells within
the tumor that harbor mutations via cross-talk between HIF1 and other oncogenes
including H-Ras and v-Sac (57). For example, HIF1 overexpression is associated with
aberrant tumor suppressor accumulation in human cancer (80, 127). It also provides
cancer cells resistance to apoptosis by expression of adrenomedulin (ADM), EPO,
endothelin 1 (ET 1), nitric oxide synthase 2 (NOS 2). HIF1 signaling, in addition,
indirectly affects cellular immortalization and genomic instability by enhancing
telomerase activity, inducing DNA breaks at fragile site and disrupting repair of DNA
damage (19, 125). Finally, HIF1-mediated cell cycle arrest through cyclin kinase such
as WAFI (p21) and KIP (p27) can also act as a selective force by inducing cell

differentiation (36).

4.3. Hypoxia signaling in cancer therapy

It has long been known that the hypoxic nature of solid tumors limits the
efficacy of chemo- and radio-therapies. The decreased level of molecular oxygen
within tumor has repeatedly thwarted the therapeutic effect of ionizing radiation to
create oxygen free radicals, which is the basis of the pharmacological efficacy (86,
112). It has been estimated that hypoxic cells are approximately three times less
susceptible to radiation-induced damage (112). The limitations of chemotherapy are

influenced bothby the decrease in available oxygen at the level of the organism as well

32

as site-specific loss, such as in the central portion of tumors. A decrease in tissue
oxygen tension can influence therapies in several ways. First, changing expression of
the P4503 within hypoxic tissues can serve to increase or decrease a drug’s
effectiveness. In addition, low oxygen concentration can influence the catalytic activity
of the P4503 by removing required cosubstrates. Second, distribution of the drug can
be decreased within the affected tissue due to inadequate vascularization and perfusion
(91). For example, within a tumor, the main cause of the localized hypoxia is an
inability of the angiogenic process to keep up with the fast-growing tumor cells.
Finally, the changes in intermediate metabolism that occur as a result of hypoxia can
also influence drug metabolism and effectiveness. For example, as discussed above,
metabolic switch to glycolysis lead to tissue lactic acidosis, especially within a tumor
and changes in local pH can influence drug delivery and metabolism. Each of these
factors will influence a drug’s effectiveness in the hypoxic microenvironment and, in
case of a tumor, that might ultimately determine prognosis (4).

The hypoxic microenvironment within a tumor might also select for a more
aggressive phenotype. It has been shown that hypoxia increases the rate of mutation
within cell (109, 125). This increased mutation rate might promote an already partially
compromised cell to full transformation, or it might start a normal cell down the path
to full tumorigenic potential. It has also been suggested that hypoxia can select for
mutated forms of p53 (39). This selective pressure within the tumor could lead to

tumor cells that have removed themselves from one principal regulator of cellular

33

homeostasis. In addition, hypoxia can select for cells that are resistant to apoptotic
signals (122). Finally, the inability of therapy to kill all of the cells within a tumor due
partly to a combination of the systemic and localized effects of hypoxia could produce
cells refractory to later treatment. Indeed, each of these factors might play a role in the
ability of hypoxia to promote more aggressive tumors and ultimately increase the
chance of a poor prognosis.

Recently many investigations have focused on targeting hypoxia and the HIF1
signaling cascade for cancer therapeutics. The goal of these new therapies is to
manipulate the HIF signaling pathway to inhibit tumor growth or to exploit the
hypoxic microenvironment of a tumor to make the therapies more specific and
efficacious. There are 4 major areas of research in this field: (1) designing drugs that
directly inhibit HIF1 signaling, (37, 52), (2) influencing other signaling cascades that
indirectly alter HIF1 signaling (66), (3) exploiting the hypoxic microenvironment to
increase specificity and decreasing toxicity (20, 34), (4) altering regulation of HIF
target genes that are critical for tumor growth or for the function of their protein
products (20, 117).

Taking advantage of the hypoxic environment as a drug target has its limitations.
The efficacy of these therapeutics depends upon the level of hypoxia, the cells exposed,
the tissue’s normal p02, and levels of P450 and P450 reductase and the activation
range of the drug itself. To begin to cope with these limitations, these bioreactive

compounds are now being used in conjunction with gene therapy vector (21, 71).

34

 

 

These
enzyr
Detei
the h
These
signal
also c
lOTIIlJ
Iheref

and to

These therapies have also exploited HRE-driven gene constructs to insure that the
enzyme is only expressed in hypoxic tissue, where the drug should be activated (40).
Development of each of these compounds as drugs has utilized our understanding of
the hypoxic environment and the HIF1 signaling system to target hypoxic tissues.
These therapies, however, are still restricted by our limited knowledge of the HIF1
signaling to regulate a wide range of cellular responses essential to the adaptation but
also cell death signaling and how the resulting changes in cells can affect to tumor
formation and progression. Characterizing the role of HIF1 signaling in tumorigenesis,
therefore, is important for understanding of the basic cause of cellular transformation

and tumor growth as well as targeting this signaling for cancer therapy.

35

 

lypothi

ilfl sip
lCllVll}
modulo
actitiu
and (ill;
i0rming
Signihc
Signalir
has the
Cancer
Finally
afiivitt

hl'dIOx

 

Hypothesis and Specific Aims

Research projects were designed and developed to investigate the role of the
HIF1 signaling pathway in tumor development. The central hypothesis tested was: The
activity of PHI) is related to the process of cellular transformation via its ability to
modulate HIF1 signaling. This hypothesis includes the possibility of altering PHDs
activity can modulate the HIFll regulated balance between adaptation and cell death
and direct manipulation of the PHD activity within a cell will alter the cell’s tumor
forming potential. Previous studies strongly support this hypothesis: (1) Hypoxia and
significant HIF1 activity are common characteristic of fast growing tumors and HIF1
signaling has a positive impact in tumor formation and progression. (2) PHD activity
has the ability to control the HIF1 transcriptional responses. (3) There is a link between
cancer and perturbation of PHD activity and subsequent HIF1 mediated signaling. (4)
Finally, VHL functions as a tumor suppressor, primarily because it can regulate HIF1
activity, and the latter activity is dependant upon PHD-mediated posttranslational

hydroxylation of HIFIOL.

36

 

The MS'
were es
iibrobl;
lGl cc
cell sir
spontal
cells til

llOlI‘etQ

 

 

 

 

 

“Emit
transloi
CXPOSC

iOImet

Ccll sr

 

The MSU] lineage of cell strains was used to test this hypothesis. These cell strains
were established in Dr. McCormick’s laboratory at MSU. Initially, normal human
fibroblast cells were derived form the foreskin of a neonate and designated LGl (84).
LG] cells were transfected with v-myc and selected using neomycin to generate the
cell strain with an infinite life span, MSU-1.0 cells (84). A fast growing and
spontaneous variant strain emerged from the MSU-1.0 cells, resulting in the MSU-1.1
cells (51). The MSU1.1 cell strain acquired partial growth factor independence;
however, they are negative in anchorage independent growth and tumor formation in
athymic mice, indicating their non-tumorigenic nature. To generate malignantly
transformed cells, MSU-1.1 cells were transfected with the V12-H-RAS oncogene or
exposed to 'y-radiation (51, 87). Transformed foci derived from these cell strains
formed solid tumors in athymic mice and these tumor cells were cultured resulting in

cell strains designated PH3MT and Y2-3a, respectively.

37

Ea

Air

tur

lllli

tar;

rel;

MS

ICg

Air

Ofr

attr'

mlCI

Each of the chapters in this dissertation is focused on investigating the hypothesis by

addressing the following specific aims.

Aim 1: Characterization of HIF1 signaling in a lineage of cell strains of varying
tumorigenic potential.

The correlation between HIF1 activity and tumorigenic potential in the MSUl
lineage of cells was assessed by determining the expression levels of HIF la and HIF1
target genes. PHDs’ expression levels were also characterized to investigate the
relationship between PHDs levels and tumorigenicity as well as HIF1 activity in
MSUl cell strains. Finally, individual PHD isoforms were evaluated for their ability to

regulate HIF1 activity in these cells.

Aim 2: Characterize the effects of reduced levels of PHD2 on the tumorigenicity
of non-transformed cells.

New cell strains with reduced levels of PHD2 were generated using lentiviral
infection of shRNA targeting PHD2 in non-tumorigenic MSU-1.1 cells and continuous
clonal selection. Newly created shPHD2 cell strains were examined for their HIF1
activity and tested for their tumorigenic potential and tumor forming ability in athymic
mice. Finally, the expression of HIF1 target genes and their functional consequence
were determined in shPHD2 cell strains. A PHD inhibitor was also applied to test the

reproducibility of the results.

38

Aim 3:
of maligl
‘ |
NI
gene rra;

selection

tumorig

Aim 4:

T
Allinit}
Pupil
identih

immun.

 

 

Aim 3: Characterize the effect of overexpression of PHD2 on the tumorigenicity
of malignantly transformed cells.

New cell strains with that overexpressed PHD2 were created using retroviral
gene transfer of a PHD2 cDNA in malignantly transformed PH3MT cells and clonal
selection. Newly created PHD2 cell strains were characterized for their HIF1 activity,

tumorigenecity and the functional expression of HIF1 regulated genes.

Aim 4: Identify the protein components associated with PHD2.

The protein network associated with PHD2 was characterized using Tandem-
Affinity Purification (TAP)-tagging method. Cells expressing a TAP-tag version of
PHD2 were created and verified for their functional activity. Purified proteins were
identified using mass-spectrometry and interactions were verified using co-

immunoprecipitation and Western blot analysis.

39

REFE

t.)
9‘3 a:

.LTCU

10, C

 

REFERENCES

10.

11.

An, W. G, M. Kanekal, M. C. Simon, E. Maltepe, M. V. Blagosklonny, and L.
M. Neckers. 1998. Stabilization of wild-type p53 by hypoxia-inducible factor
lalpha. Nature 392:405-408.

Beck, 1., S. Ramirez, R. Weinmann, and J. Caro. 1991. Enhancer element at the
3'-flanking region controls transcriptional response to hypoxia in the human
erythropoietin gene. J. Biol. Chem. 6:15563-15566.

Berra, E., E. Benizri, A. Ginouves, V. Volmat, D. Roux, and J. Pouyssegur.
2003. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state
levels of HIF-lalpha in normoxia. EMBO J. 22:4082-4090.

Birner, P., M. Schindl, A. Obermair, C. Plank, G. Breitenecker, and G
Oberhuber. 2000. Overexpression of hypoxia-inducible factor lalpha is a marker
for an unfavorable prognosis in early-stage invasive cervical cancer. Cancer Res.
60:4693-4696.

Briere, J., J. Favier, P. Benit, V. Ghouzzi, A. Lorenzato, D. Rabier, M. Di-
Renzo, A. Gimenez-Roqueplo, and P. Rustin. 2005. Mitochondrial succinate is
instrumental for HIFIOL nuclear translocation in SDHA-mutant fibroblasts under
normoxic conditions. Hum. Mol. Genet. 14:3263-3269.

Brown, J. M., and A. J. Giaccia. 1998. The unique physiology of solid tumors:
opportunities (and problems) for cancer therapy. Cancer Res. 58: 1408-1416.

Bruick, R. K., and S. L. McKnight. 2001. A conserved family of prolyl-4-
hydroxylases that modify HIF. Science 294:1337-1340.

Brunelle, J. K., E. L. Bell, N. M. Quesada, K. Vercauteren, V. Tiranti, M.
Zeviani, R. C. Scarpulla, and N. S. Chandel. 2005. Oxygen sensing requires
mitochondrial ROS but not oxidative phosphorylation. Cell Metab. 1:409-414.

Bunn, H. F., and R. O. Poyton. 1996. Oxygen sensing and molecular adaptation
to hypoxia. Physiol. Rev. 7 :839-880.

Canesi, L., C. Ciacci, M. Betti, M. Malatesta, G Gazzanelli, and G. Gallo.
1999. Gen. Comp. Endocrinol. 116:241-248.

Carmeliet, P., Y. Dor, J. M. Herbert, D. Fukumura, K. Brusselmans, M.
Dewerchin, M. Neeman, Bono, F., R. Abramovitch, P. Maxwell, C. J. Koch, P.

40

17.

30.

(3

m
c.)
(I)
A

 

12.

13.

14.

15.

16.

17.

18.

19.

Ratcliffe, L. Moons, R. K. Jain, D. Collen, and E. Keshertk. 1998. Role of
HIF-10L in hypoxia-mediated apoptosis, cell proliferation and tumour
angiogenesis. Nature 394:485-490.

Carmeliet, P., and R. K. Jain. 2000. Angiogenesis in cancer and other disease.
Nature 407 :249-257.

Chandel, N. S., E. Maltepe, E. Goldwasser, C. E. Mathieu, M. C. Simon, and
P. T. Schumacker. 1998. Mitochondrial reactive oxygen species trigger hypoxia-
induced transcription. Proc. Natl. Acad. Sci. USA 95: 1 1715-11720.

Chandel, N. S., and P. T. Schumacker. 2000. Cellular oxygen sensing by
mitochondria: old question and new insight. J. Appl. Physiol. 88: 1880-1889.

Chen, C., N. Pore, A. Behrooz, F. Ismail-Beigi, and A. Maity. 2001. Regulation
of glutl mRNA by Hypoxia-inducible Factor-1. Interaction between H-ras and
hypoxia. J. Biol. Chem. 276:9519-9525.

Cioffi, C. L., X. Q. Liu, P. A. Kosinski, M. Garay, and B. R. Bowen. 2003.
Differential regulation of HIF prolyl4-hydroxylase genes by hypoxia in human
cardiovascular cells. Biochem. Biophys. Res. Commun. 303:947- 953.

Coelho, A., A. Ferraz, R. Neto, A. Souza, and E. Ferraz. 2000.
Subdiaphragmatic venous stasis and tissular hypoperfusion as source of metabolic
acidosis during passive portal-jugular and cacal-jugular bypasses in dogs. Acta.
Cir. Bras. 15.

Coleman, C. N., J. B. Mitchell, and K. Camphausen. 2001. Tumor hypoxia:
chicken, egg, or a piece of the farm? Journal of Clinical Oncology 20:610-615.

Coquelle, A., F. Toledo, S. Stern, A. Beith, and M. Debatisse. 1998. A new role
for hypoxia in tumor progression: induction of fragile site triggering genomic
rearrangements and formation of complex DS and HSRs. Mol. Cell. 2:259-265.

20. Cowen, R. L., K. J. Williams, E. C. Chinje, M. Jaffar, F. C. D. Sheppard, B. A.

21.

Telfer, N. S. Wind, and I. J. Stratford. 2004. Hypoxia targeted gene therapy to
increase the efficacy of tirapazarnine as an adjuvant to radiotherapy: reversing
tumor radioresistance and effecting cure. Cancer Res. 64: 1396-1402.

Dachs, G. U. e. a. 1997. Targeting gene expression to hypoxic tumor cells. Nat.
Med. 3:51-5—520.

41

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

Dahia, P. L. M., K. N. Ross, M. E. Wright, C. Y. Hayashida, S. Santagata, M.
Barontini, A. L. Kung, G Sanso, J. F. Powers, A. S. Tischler, R. Hodin, S.
Heitritter, F. Moore, R. Dluhy, J. A. E. Sosa, I. T. Ocal, D. E. Benn, D. J.
Marsh, B. G. Robinson, K. Schneider, J. Garber, S. M. Arum, M. Korbonits,
A. Grossman, P. Pigny, P. Toledo, A., V. Noseacute, C. Li, and C. D. Stiles.
2005. A HIF1-alpha; Regulatory Loop Links Hypoxia and Mitochondrial Signals
in Pheochromocytomas. PLoS Genet. 1:e8.

Dang, C. V., and G. L. Semenza. 1999. Oncogenic alteration of metablosm.
Trends Biochem. Sci. 24:68-72.

D'Angelo, G., E. Duplan, N. Boyer, P. Vigne, and C. Frelin. 2003. Hypoxia up
regulates prolyl hydroxylase. A feedback mechanism that limits HIF—l responses
during reoxygenation. J . Biol. Chem. 278:3 8183-38187.

de Fraipont, F., A. C. Nicholson, J. J. Feige, and E. G. Van Meir. 2001.
Thrombosponding and tumor angiogenesis. Trends Mol. Med. 7:401-407.

Denko, N. C., L. A. Fontana, K. M. Hudson, P. D. Sutphin, S. Raychaudhuri,
R. Altman, and A. J. Giaccia. 2003. Investigating hypoxia tumor physiology
through gene expression pattern. Oncogen 22:5907-5914.

Denko, N. e. a. 2000. Epigenetic regulation of gene expression in cervical cancer
cells by the tumor microenvironment. Clin. Cancer Res. 6:480-487.

Duong, T. Q., and S. G. Kim. 2000. Quantitative MR measurements of
interstitial oxygen tension in intact rat brain; hyperoxia and hypercapnia. Proc.
Int]. Soc. Mag. Reson. Med. 8:440.

Elbashir, S. M., J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, and T.
Tuschl. 2001. Duplexes of 2l-nucleotide RNAs mediate RNA interference in
cultured mammalian cells. Nature. 411:494-498.

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

Erez, N., M. Milyavsky, R. Eilam, I. Shats, N. Goldfinger, and V. Rotter. 2003.
Expression of prolyl-hydroxylase-l (PHDl/EGLNZ) suppresses hypoxia-

42

inc‘

32. Fel

33.

34.

35.

36.

39.

40.

41.

Re

grr

Fe

8“
Fe

de
Rt

 

32

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

inducible factor 1a and inhibits tumor growth. Cancer Res. 63:8777-8783.

. Feldser, D., F. Agani, N. V. Lyer, B. Pak, G. Ferreira, and G L. Semenza. 1999.

Reciprocal positive regulation of hypoxia-inducible factor lalpha and insulin-like
growth factor 2. Cancer Res. 59:3915-3918.

Ferrara, N. 1999. Molecular and biochemical properties of vascular endothelial
growth factor. J. Mol. Med. 77:527-543.

Ferrara, N., K. J. Hillan, H. P. Gerber, and W. Novotny. 2004. Discovery and
development of bevacizumab, an anti-VEGF antibody for treating cancer. Nature
Reviews. Drug Discovery 3:391-400.

Fisher, T. S., S. D. Etages, L. Hayes, K. Crimin, and B. Li. 2005. Analysis of
ARDl function in hypoxia response using retroviral RNA interference. J Biol.
Chem. 280: 17749-17757.

Gardner, L. B., Q. Li, M. S. Park, W. M. Flanagan, G. L. Semenza, and C. V.
Dang. 2001. Hypoxia inhibits GI/S transition through regulation of p27
expression. J. Biol. Chem. 276:7919-7926.

Geoerger, B., K. Kerr, C. B. Tang, K. M. Fung, B. Powell, L. N. Sutton, P. C.
Phillips, and A. J. J anss. 2001. Antitumor activity of the rapamycin analog CCI-
779 in human primitive neuroectodermal tumor/medulloblastoma models as
single agent and in combination chemotherapy. Cancer Res. 61:1527-1532.

Gong, Y., and F. H. Agani. 2005. Oligimycin inhibits HIFICX. expression in
hypoxic tumor cells. Am. J. Physiol. Cell Physiol. 288: 1023-1029.

Graeber, T. G., C. Osmanian, T. Jacks, D. E. Housman, C. J. Koch, S. W.
Lowe, and A. J. Giaccia. 1996. Hypoxia-mediated selection of cells with
diminished apoptotic potential in solid tumours. Nature 379:88-91.

Griffiths, L. e. a. 2000. The macrophage: a novel system to deliver gene therapy
to pathilogical hypoxia. Gene Ther. 7 :255-262.

Guzy, R. D., B. Hoyos, E. Robin, H. Chen, L. Liu, K. D. Mansfield, M. C.
Simon, U. Hammerling, and P. T. Schumacker. 2005. Mitochondrial complex
III is required for hypoxia-induced ROS production and oxygen sensing. Cell

Metab. 1:401-408.

Hagen, T., C. T. Taylor, F. Lam, and S. Moncada. 2003. Redistribution of

43

 

 

 

50. j

51.

53.

 

 

43.

45

46

47.

48.

49.

50.

51.

52.

53

intracellular oxygen in hypoxia by nitric oxide effect on HIFla. Science
302:1975-1978.

Hananhan, D. V., and J. Folkman. 1996. Pattern and emerging mechanisms of
angiogenic switch during tumorigenesis. Cell 86:353-364.

Harris, A. L. 2002. Hypoxia - a key regulatory factor in tumor growth. Nat. Rev.
Cancer 2:38-47.

. Hirsila, M., P. Koivunen, V. Gunzler, K. I. Kivirikko, and J. Myllyharju. 2003.

Characterization of the human prolyl 4-hydroxylase that modify the hypoxia-
inducible factor. J. Biol. Chem. 278:30772-30780.

. Hirsila, M., P. Koivunen, V. Gunzler, K. I. Kivirikko, and J. Myllyharju. 2003.

Characterization of the human prolyl 4—hydroxylases that modify the hypoxia-
inducible factor. Proc. Natl. Acad. Sci. USA 278:30772-30780.

Hockel, M., and P. Vaupel. 2001. Tumor hypoxia: Definitions and current
clinical, biologic, and molecular aspects. J. Natl. Cancer Inst. 98:266-276.

Hogenesch, J. B., W. C. Han, V. H. Jackiw, R. C. Brown, Y. Z. Gu, M. Pray-
Grant, G H. Perdew, and C. A. Bradfield. 1997. Characterization of a subset of
the basic-helix-loop-helix-PAS superfamily that interact with components of the
dioxin signaling pathway. J. Biol. Chem. 272: 8581-8593.

Huang, J., Q. Zhao, S. M. Mooney, and F. S. Lee. 2002. Sequence
determination in hypxia inducible factor 1a for hydroxylation by the prolyl
hydroxylases PI-IDl, PHD2, and PHD3. J. Biol. Chem. 277 :39792-39800.

Huang, Z. J., l. Edery, and M. Rosbash. 1993. PAS is a dimerization domain
common to drosophila period and several transcription factors. Nature 364:259-
262.

Hurlin, P. J., V. M. Maher, and J. J. McCormick. 1989. Malignant
transformation of human fibroblasts caused by expression of a transfected T24
HRAS oncogene. Proc. Natl. Acad. Sci. USA 86:287-291.

Isaacs, J. S., Y. J. Jung, E. G Mimnaugh, A. Martinez, F. Cuttitta, and L. M.
Neckers. 2002. Hsp90 regulates a von Hippel Lindau-independent hypoxia-
inducible factor-l alpha-degradative pathway. J. Biol. Chem. 277:29936-19944.

. Isaacs, J. S., Y. J. Jung, D. R. Mole, S. Lee, C. Torres-Cabala, Y. L. Chung, M.

44

58.

S9

 

60

61.

62.

54.

55

56.

57.

58.

59

60.

61.

62.

Merino, J. 'h‘epel, B. Zbar, and J. Tom. 2005. HIF overexpression correlates
with biallelic loss of fumarate hydratase in renal cancer: Novel role of fumarate in
regulation of HIF stability. Cancer Cell 8: 143-153.

Ivan, M., T. Haberberger, D. C. Gervasi, K. S. Michelson, V. Gunzler, K.
Kondo, H. Yang, 1. Sorokina, R. C. Conaway, and J. W. Conaway. 2002.
Biochemical purification and pharmacological inhibition of a mammalian prolyl
hydroxylase acting on hypoxia-inducible factor. Proc. Natl. Acad. Sci. USA
99:13459-13464.

. 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. 1998.
Cellular and developmental control of Oz homeostasis by hypoxia-inducible
factor 1 alpha. Genes Dev. 12:149-162.

Jeong, J. W., M. K. Bae, M. Y. Ahn, S. H. Kim, T. K. Sohn, M. H. Bae, M. A.
Yoo, E. J. Song, K. J. Lee, and K. W. Kim. 2002. Regulation and destabilization
of HIF-10L by ARD-mediated acetylation. Cell 111:709-720.

Jiang, B. H., F. Agani, A. Passaniti, and G L. Semenza. 1997. V-SRC induces
expression of hypoxia-inducible factor 1 (HIF1) and tanscription of genes
encoding vascular endothelial growth factor and enolase 1: involvement of HIF1
in tumor progression. Cancer Res. 57:5328-5335.

Jiang, J., T. Nakashima, K. Liu, F. Goda, T. Shima, and H. Swartz. 1996.
Measurement of P02 in liver using EPR oximetry. J. Appl. Physiol. 80:552-558.

. Kasuno, K., S. Takabuchi, K. Fukuda, S. Kizaka-Kondoh, J. Yodoi, T. Adachi,

G L. Semenza, and K. Hirota. 2004. Nitric oxide induces hypoxia-inducible
factor 1 activation that is dependent on MAP kinase and phosphatidylinositol 3-
kinase signaling. J. Biol. Chem. 279:2550-2558.

Kiaer, T., and K. D. Kristensen. 1988. Intracompartmental pressure, P02, PC02
and blood flow in the human skeletal muscle. Arch Orthop Trauma Surg 107: 1 14-
1 16.

Kim, J.-w., I. Tchernyshyov, G L. Semenza, and C. V. Dang. 2006. HIF-l-
mediated expression of pyruvate dehydrogenase kinase: A metabolic switch

required for cellular adaptation to hypoxia. Cell Metab. 3: 177-185.

Koong, A. C. e. a. 2000. Candiate genes for the hypoxic tumor phenotype. Cancer
Res. 60:883-887.

45

63.

64.

65

66.

67.

68.

69.

70.

71.

72

73.

Krishnamachary, B., S. Berg-Dixon, B. Kelly, F. Agani, D. Feldser, G
Ferreira, N. Iyer, J. LaRusch, B. Pak, P. Taghavi, and G L. Semenza. 2003.
Regulation of colon carcinoma cell invasion by Hypoxia-Inducible Factor 1.
Cancer Res. 63:1138-1143.

Kung, A. L., S. Wang, J. M. cho, W. G Kaelin, and D. M. Livingstone. 2000.
Suppression of tumor grOWIh through disruption of hypoxia-inducible
transcription. Nat. Med. 6:1335-1340.

. Kunz, M., S. Ibrahim, D. Koczan, H. J. Thiesen, H. J. Koehler, T. Akcer, K. H.

Plate, S. Ludwig, U. R. Rapp, and E. B. Broecker. 2001. Activation of c-jun
NH2-terminal kinase/stress-activated protein kinase (J NK/SAPK) is critical for
hypoxia-induced apoptosis of human malignant melanoma. Cell growth Differ.

12: 137- 145.

Laderoute, K., P. Wardman, and A. M. Rauth. 198. Molecular mechanisms for
the hypoxiadependent activation of 3-amino-1,2,4-benzotriazine-1, 4-dioxide (SR
4233). Biochem Pharrnacol 37: 1487-1495.

La], A. e. a. 2000. Transcriptional response to hypoxia in human tumor. J. Natl.
Cancer Inst. 93:1337-1343.

Land, S. C. 2003. Oxygen sensing pathway and the development of mammalian
gas exchange. Redox Rep. 8:325-340.

Lando, D. 2002. Asparagine hydroxylation of the HIF transactivation domain a
hypoxia switch. Science 295:856-861.

Lee, Y. M., C. H. Jeong, S. Y. Koo, M. J. Son, H. S. Song, 13. S.K., J. A.
Raleigh, H. Y. Chung, M. A. Yoo, and K. W. Kim. 2001. Determination of
hypoxic region by hypoxic marker in developing mouse embryo in vivo: a
possible signal for vessel development. Dev.Dyn. 220: 175-186.

Lemmon, M. e. a. 1997. Anaerobic bacteria as a gene delivery system that is
controlled by the tumor microenvironment. Gene Ther. 4:791-796.

. Liotta, L. A. and Kohn E. C. 2000. Invasion and metastasis. Bast R. C., Jr. Kufe D.

W. Pollock R. E. Weichselbaum R. R. Holland J. F. Frei B., III eds. Cancer
Medicine, 5‘h Ed. pp121—131, B. c. Decker Hamilton, Ontario, Canada 2000.

Lu, H., C. L. Dalgard, A. Mohyeldin, T. McFate, A. S. Tait, and A. Verma.
2005. Reversible Inactivation of HIF-1 Prolyl Hydroxylases Allows Cell

46

74.

75.

76.

77.

78.

79.

80

81.

82.

Metabolism to Control Basal HIF- 1 a. J. Biol. Chem. 280:41928-41939.

Lu, H., R. A. Forbes, and A. Verma. 2002. Hypoxia-inducible factor 1 activation
by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J. Biol.
Chem. 277:23111-23115.

Lu, H., R. A. Forbes, and A. verma. 2002. Hypoxia-inducible factor 1 activation
by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J. Biol.
Chem. 277:23111-23115.

Mahon, P. C., K. Hirota, and G L. Semenza. 2001. FIH-l: a novel protein that
interact with HIF-10L and VHL to mediate repression of HIF-1 transcriptional
activity. Gene Dev. 15:2675-2686.

Mansfield, K. D., R. D. Guzy, Y. Pan, R. M. Young, T. P. Cash, P. T.
Schumacker, and M. C. Simon. 2005. Mitochondrial dysfunction resulting from
loss of cytochrome c impairs cellular oxygen sensing and hypoxic HIF-alpha
activation. Cell Metab. 1:393-399.

Marti, H. H., H. H. Jung, J. Pfeilschifter, and C. Bauer. 1994. Hypoxia and
cobalt stimulate lactate dehydrogenase (LDH) activity in vascular smooth muscle
cells. Pflugers Arch. 429:216—222.

Maxwell, P. H., G U. Dach, J. M. Gleadle, L. G Nicholls, A. L. Harris, I. J.
Stratford, O. Hankinson, C. W. Pugh, and P. J. Ratcliffe. 1997. Hypoxia—
inducible factor-1 modulates gene expression in solid tumor and influence both
angiogenesis and tumor growth. Proc. Natl. Acad. Sci. USA 94:8104-8109.

. 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. 1999.
The tumor suppressor protein VHL target hypoxia-inducible factors for oxygen-
dependent proteolysis. Nature 399:271-275.

Mazurek, S., C. B. Boschek, and E. Eigenbrodt. 1997. The role of
phosphometabolites in cell proliferation, energy metabolism, and tumor therapy. J.
Bioenerg. Biomembr. 29:315-330.

Metzen, E., U. B-Pfannschmidt, P. Stengel, J. H. Marxsen, I. Stolze, M.
Klinger, W. Q. Huang, C. Wotzlaw, T. H-Burgel, W. Jelkmann, H. Acker, and
J. F andrey. 2002. Intracellular localization of human HIFIOL hydroxylase:
implication for oxygen sensing. J. Cell Sci. 116: 13 19-1326.

47

83

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

. Mizukami, Y., W.-S. Jo, E.-M. Duerr, M. Gala, J. Li, X. Zhang, M. A. Zimmer,

O. Iliopoulos, L. R. Zukerberg, Y. Kohgo, M. P. Lynch, B. R. Rueda, and D.
C. Chung. 2005. Induction of interleukin-8 preserves the angiogenic response in
HIFIOL deficiency colon cancer cells. Nat. Med. 9:992-997.

Morgan, T. L., d. J. Yang, D. G Fry, P. J. Hurlin, S. K. Kohler, V. M. Maher,
and J. J. McCormick. 1991. 5 Characteristics of an infinite life span diploid
human fibroblast cell strain and a near-siploid strain arising from a clone of cells
expressing a tranfected v-myc oncogene. Exp. Cell Res. 197: 125-136.

Myllyharju, J., and K. I. Kivirikko. 1997. Characterization of the iron- and 2-
oxoglutarate-binding sites of human prolyl hydroxylase. EMBO J. 16:1173- 1180.

Nordsmark, M., M. Overgaard, and J. Overgaard. 1996. Pretreatment
oxygenation predicts radistion response in advance squamous cell carcinomas of
the head and neck. Radiother Oncol. 41:31-39.

O'Reilly, 8., M. Walicka, S. K. Kohler, R. Dunstan, V. M. Maher, and J. J.
McCormick. 1998. Dose-dependent transformation of cells of human fibroblast
cell strain MSU 1.1 by cobalt-6O gamma radiation and characterization of the
transformed cells. Radiat Res. 150:577-584.

Palmer, L. A., B. Gaston, and R. A. Johns. 2000. Norrnoxic stabilization of
hypoxia-inducible factor 1 expression and activity: redox-dependent effect of
nitric oxides. Mol.Pharmacol. 58: l 197- 1203.

Piret, J. P., D. Motter, M. Raes, and C. Michiels. 2002. Is HIF-let a pro- or an
anti-apoptotic protein? Biochem. Pharmacol. 64:889-892.

Poole, D. C., P. D. Wagner, and D. F. Wilson. 1995. Diaphragm microvascular
plasma P02 measured in vivo. J. Appl. Physiol. 79:2050-2057.

Primeau, A. J., A. Rendon, D. Hedley, L. Lilge, and I. F. Tannock. 2005. The
Distribution of the Anticancer Drug Doxorubicin in Relation to Blood Vessels in
Solid Tumors. Clin. Cancer Res. 11:8782-8788.

Pugh, C. W., C. C. Tan, R. W. Jones, and P. J. Ratcliffe. 1991. Functional
analysis of an oxygen regulated transcriptional enhancer lying 3' to the mouse

erythropoietin gene. Proc. Natl. Acad. Sci. USA. 88:10553-10557.

Richard, D. E., E. Berra, E. Gothie, D. Roux, and J. Poussegur. 1999. p42/p44
mitogen-activated protein kinases phosphorylate hypoxia-induced factor la and

48

94. l

95.

96. S

97. S

 

In.

 

enhance the transcriptional activity of HIF-1. J. Biol. Chem. 274:32631-32637.

94. Ryan, H. E., J. Lo, and R. S. Johnson. 1998. The hypoxia inducible factor 1-
alpha is required for solid tumor formation and embryonic vasculation. The
EMBO J. 17:3005-3015.

95. Sabia, P. J ., E. R. Powers, M. Ragosta, I. J. Sarembock, L. R. Burwell, and S.
Kaul. 1992. An association between collateral blood flow and myocardial

viability in patients with recent myocardial infaction. N. ENGL, J. Med.
327:1825-1831.

96. Sang, N., D. P. Stiehl, J. Bohensky, V. Leshchinsky, and J. C. Srinivas. 2003.
MAPK signaling up-regulated the activity of hypoxia-inducible factors by its
effects on p300. J. Biol. Chem. 278:14013-14019.

97. Schofield, C. J., and Z. Zhang. 1999. Structure and mechanistic studies on 2-
oxoglutarate-dependent oxygenases and related enzymes. Curr. Opin. Struct. Biol.
9:722-731.

98. Schultz, A., L. Lavie, I. Hochberg, R. Beyar, T. Stone, K. Skorecki, P. Lavie, A.
Roguin, and A. P. Levy. 1999. Inter-individual heterogeneity in the hypoxia
regulation of VEGF: Significance for the development of the coronary artery
collateral circulation. Circulation 100:547-552.

99. Selak, M. A., S. M. Armour, E. D. MacKenzie, H. Boulahbel, D. G Watson, K.
D. Mansfield, Y. Pan, M. C. Simon, C. 13. Thompson, and E. Gottlieb. 2005.
Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-[alpha]
prolyl hydroxylase. Cancer Cell 7 :77-85.

100. Semenza, G L. 2006. HIF1 and human disease: one highly involved factor.
Genes & Dev. 14:1983-1991.

101. Semenza, G L. 2004. Hydroxylation of HIF-1: Oxygen sensing at the molecular
level. Physiology 19:176-182.

102. Semenza, G L. 2000. Hypoxia, clonal selection, and the role of HIF-l in tumor
progression. Crit. Rev. Biochem. Mol. Biol. 35:71-103.

103. Semenza, G L. 1998. Hypoxia-inducible factor 1: master regulator of Oz
homeostasis. Curr. Opin. Genet. Dev. 8:588-594.

104. Semenza, G L. 2003. Targeting HIF-l for cancer therapy. Nature 3:721-731.

49

105.

106.

107.

108.

109.

110.

Semenza, G L., and G L. Wang. 1992. A nuclear factor induced by hypoxia via
de novo protein synthesis binds to the human erythropoietin gene enhancer at a
site required for transcriptional activation. Mol. Cell. Biol. 12:5447-5454.

Shimizu, S., Y. Eguchi, W. Kamiike, Y. Itoh, J. Hasegawa, K. Yamabe, Y.
Otsuki, H. Matsud, M. Shinkai, M. A. Pirker, S. Montedonico, and P. and
Puri. 2005. The role of oxygen tension in the regulation of embryonic lung
development. Journal of Pediatric Surgery. 40:32-35.

Shweiki, D., A. Itin, D. Soffer, and E. Keshet. 1992. Vascular endothelial growth
factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature
359:843-845.

Sonna, L. A., M. L. Cullivan, H. K. Sheldon, R. E. Pratt, and C. M. Lilly.
2003. Effect of hypoxia on gene expression by human hepatocytes (HepGZ).
Physiol Genomics 12:195-207.

Stackpole, C. W., L. Groszek, and S. S. Kalbag. 1994. Benign-to-malignant
B16 melanoma progression induced in two stages in vitro by exposure to hypoxia.
J. Natl. Cancer Inst. 86:361-367.

Stiehl, D. P., R. Wirthner, J. Koditz, P. Spielmann, G Camenisch, and R. H.
Wenger. 2006. Increased prolyl 4-hydroxylase domain proteins compensate for
decrease oxygen levels. J. Biol. Chem. 281:23482-23491.

111.Takabuchi, S., K. Hirota, K. Nishi, S. Oda, T. Oda, T. Shingu, A. Takabayashi,

112.

113.

114.

T. Adachi, G L. Semenza, and K. Fukuda. 2004. The inhibitory effect if sodium
nitroprusside on HIF] activation is not dependent on nitric oxide-soluble guanylyl
pathway. Biochem. Biophys. Res. Co. 324:417-423.

Teicher, B. A. 1995. Physiologic mechanisms of therapeutic resistance. Blood
flow and hypoxia. Hematol. Oncol. Clin. North Am 9:475-506.

Tsuzuki, Y., D. Fukumura, B. Oosthuyse, C. Koike, P. Carmeliet, and R. K.
Jain. 2000. Vasculae endothelial growth factor (VEGF) modulatin by targeting
Hypoxia-Inducible Factor-10L --> Hypoxia Response Element --> VEGF cascade.
Differentially regulates vascular reponse and growth rate in tumors. Cancer Res.
60:6248-6252.

'lluckerman, J. R., Y. Zhao, K. S. Hewitson, Y.-M. Tian, C. W. Pugh, P. J.

Ratcliffe, and D. R. Mole. 2004. Determination and comparison of specific
activity of the HIF-prolyl hydroxylases. FEBS Lett 576: 145-150.

50

115.1

._.
n
’1'!

l33_‘

124,

 

 

115.

116.

Thy], M., J. Liu, J. Wang, M. Kuliszewski, D. Tibboel, and M. Post. 2005.
Role of oxygen and vascular development in epitherlial branching morphogenesis

of the developing mouse lung. Am. J. Physiol. Lung Cell Mol. Physiol. 288:167-
178.

Vengellur, A., B. G Woods, H. E. Ryan, R. S. Johnson, and J. J. LaPres. 2003.
Gene expression profiling of the hypoxia signaling pathway in hypoxia inducible
factor 1a null mouse embryonic fibroblasts. Gene Expression 11:181-197.

117.Verna McErlane, A. Y., H. O. McCarthy, C. M. Hughes, L. H. Patterson, D. G

118.

119.

120.

121.

122.

123.

124.

Hirst, T. Robson, and S. R. McKeown. 2005. A cytochrome P450 2B6
meditated gene therapy strategy to enhance the effects of radiation or
cyclophosphamide when combined with the bioreductive drug AQ4N. Journal
of Gene Med. 7:851-859.

Wang, G L., B. H. Jiang, E. A. Rue, and G L. Semenza. 1995. Hypoxia-
inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by
cellular 02 tension. Proc. Natl. Acad. Sci. USA 92:5510-5514.

Warburg, O. 1956. On the origin of cancer cells. Science 123:309-314.

Wouters, B. G, and J. M. Brown. 1997. Cell at intermediate oxygen levels can
be more important than the hypoxic fraction in determining tumor response to
fractionated radiotherapy. Radiat Res. 147:541-550.

Wykoff, C. C. 2000. Hypoxia-inducible expression of tumor associated carbonic
anhydrases. Cancer Res. 60:7075-7083.

Yao, K., J. S. Gietema, e. S. Hida, M. Selvakumaran, X. Fonrose, N. Haas, J.
Testa, and P. O'Dwyer. 2005. In vitro hypoxia-conditioned colon cancer cell
lines derived from HCT116 and HT29 exhibit altered apoptosis susceptibility and
a more angiogenic profile in vivo. Br. J. Cancer. 93: 1356-1363.

Yu, A. Y., M. G Frid, L. A. Shimoda, C. M. Wiener, K. Stenmark, and G L.
Semenza. 1998. Temporal, spatial and oxygen-regulated expression of hypoxia-
inducible factor 1 in the lung. Am. J. Physiol. 275:L818-L826.

Yu, A. Y., L. A. Shimoda, N. V. Iyer, D. L. Huso, X. Sun, R. McWilliams, T.
Beaty, J. S. K. Sham, C. M. Wiener, J. T. Sylvester, and G L. Semenza. 1999.
Impaired physiological responses to chronic hypoxia in mice partially deficient
for hypoxia-inducible factor Inc. J. Clin. Invest. 103:691-696.

51

125. Yuan, J ., L. Narayanan, S. Rockwell, and P. M. Glazer. 2000. Diminished
DNA repair and elevated mutagenesis in mammalian cells exposed to hypoxia and
low pH. Cancer Res. 60:4372-4376.

126. Yue, X., and R. J. Tomanek. 1999. Stimulation of coronary vasculogenesis/
angiogenesis by hypoxia in cultured embryonic heart. Dev. Dyn. 216:28-36.

127. Zhong, H., A. M. Marzo, E. Laughner, M. Lim, D. A. Hilton, D. Zagzag, P.
Buechler, W. B. Issacs, S. GL., and J. W. Simons. 1999. Overexpression of

Hyoxia-inducible factor loc in common human cancer and theri metastasis.
Cancer Res. 59:5830-5835.

52

CHAPTER 2

THE BIPHASIC ROLE OF THE HIF PROLYL-4-HYDROXYLASE, PHD2,
IN MODULATING TUMOR-FORMING POTENTIAL

This chapter represents a manuscript that was submitted to Mol. Cell. Biol. in May,
2007. Authors included: KangAe Lee, Sandra O’Reilly, Matti Kiupel, J. Justin
McCormick, and John J. LaPres.

53

ABS"!

stress.
factor
cellul
a line
is the
tumor
tumor
and
lnhibi
We it
tumo
V'EG
a b'

[Tans

 

 

ABSTRACT

Hypoxia is a common feature of solid tumors. The cellular response to hypoxic
stress is controlled by a family of prolyl hydroxylases (PHDs) and the transcription
factor HIF1. To investigate the relationship between PHD and HIF1 activity and
cellular transformation we characterized the expression levels of PHD isoforms across
a lineage of cell strains with varying transformed characteristics. We found that PHD2
is the primary functional isoform in these cells and its levels are inversely correlated to
tumor-forming potential. When PHD2 levels were altered with RNAi in non-
tumorigenic fibroblasts, we found that small decreases can lead to malignant
transformation, whereas severe decreases do not. Consistent with these results, direct
inhibition of PHD2 was also shown to influence tumor-forming potential. What is more,
we found that overexpression of PHD2 in malignant fibroblasts leads to loss of the
tumorigenic phenotype. These changes correlated with HIFIOL activity, glycolytic rates,
VEGF expression, and the ability to grow under hypoxic stress. These findings support
a biphasic model for the relationship between PHD2 activity and malignant

transformation.

54

INTRODUCTION

Many solid tumors frequently exhibit areas of hypoxia because they have a high
rate of cellular proliferation and form aberrant blood vessels (8, 32). Since tumor
hypoxia was first identified, it has been well documented that the decreased oxygen
tension has a strong impact on tumor progression in a variety of ways. Most
prominently, hypoxia induces glycolysis and angiogenesis which are important changes
for tumor growth and clonal expansion (9, 10, 18). In addition, hypoxia promotes the
stepwise progression along a benign to malignant pathway by selecting cells that have
acquired transformed characteristics and have lost tumor suppressor function (4, 19,
38). Finally, the transcription factors that regulate the cellular response to hypoxia are
important for tumor growth and progression (31). Characterizing the role of hypoxia
signaling in tumor development, therefore, is important for understanding the basic
causes of cellular transformation and tumor growth.

The cellular responses to hypoxia are primarily regulated by the transcription
factor, hypoxia inducible factor 1 (HIF1) (24, 36). HIF1 is a heterodimer of HIF 1a and
HIFIB (also referred to as aryl hydrocarbon receptor nuclear translocator (ARNT)) and
both belong to a superfamily of basic helix-loop-helix (bHLH) Per-ARNT-Sim (PAS)
proteins. HIF 1B is a constitutive nuclear protein and interacts with other transcription
factors, such as the aryl hydrocarbon receptor. HIFIOL is specific to the response of

hypoxia and is constantly synthesized and under normoxia, it is rapidly degraded by the

55

ubiquitin-proteosomal pathway (25). The degradation of HIFIOL is mediated by the
product of the von Hippel-Lindau (VHL) tumor suppressor gene, which specifically
interacts with the oxygen-dependent degradation domain (ODD) of HIFlor. The
oxygen-dependence of this process is regulated by a family of prolyl hydroxylase
domain-containing enzymes (PI-IDs, also designated EGL 9 homologues (EGLNs) and
HIF1 prolyl hydroxylases (HPHs)) (7, 14).

Three mammalian PHDs (PHD1-3) regulate HIF1 signaling and each has a
distinct tissue distribution, pattern of subcellular localization, and substrate specificity
(23, 28). For proper activity, PHD3 require oxygen, iron, or-ketoglutarate, and ascorbate.
The oxygen requirement suggests that PHDs are “sensors” for hypoxia (16, 25, 33). In
the presence of adequate oxygen, PHDs hydroxylate HIF 10L at conserved proline
residues within the ODD domain. Once hydroxylated, HIFla becomes a substrate for
VHL-mediated ubiquitination and degradation (7, 14). Under hypoxic conditions,
PHDs are inactive and HIFIa is stabilized and translocates to the nucleus, where by
dimerizing with HIF 1B, it forms the functional transcription factor HIF 1. HIF1-
mediated transcription regulates many processes involved in cellular homeostasis and
transformation, including anaerobic metabolism, Oz-carrying capacity, and
angiogenesis (9, 35). Alternatively, under severe hypoxic stress, HIF1 can induce a pro-
death response through transcriptional activation of pro-apoptotic factors such as BCL

family members and modulation of the p53 signaling pathway (1, 8, 29, 40). HIF 1,

56

therefore, regulates a balance between cellular adaptation through upregulation of
survival genes, such as glycolytic enzymes and cell death through modulation of
various pathways.

Several studies show that modulation of PHD activity has the capability to
control the HIF1 transcriptional response (3, 15, 21, 22). Recent studies indicated a link
between cancer and perturbation of PHD activity and subsequent HIF1-mediated
signaling (16, 33). What is more, it has been suggested that VHL is a tumor suppressor,
at least under some circumstances, because it can modulate HIF1 activity, and the latter
activity is dependent upon a functional PHD. In view of these reports, we hypothesize
that PHD activity is linked to a cell’s tumor-forming potential, and that direct
manipulation of the PHD activity within a cell will alter the cell’s tumor-forming
capacity.

To test this hypothesis, we made use of three human fibroblast cell strains from
the MSUl lineage of cells. These cells, derived one from the other, each one having
acquired a characteristic related to malignant transformation, until the last change gave
rise to a cell strain that by acquiring one more change, e. g., a RAS oncogene, becomes
capable of forming malignant tumors in athymic mice with a short latency (26). The
three cell strains used for the present study are: MSU-1.0, MSU-1.1, and PH3MT.
MSU-1.0 is the infinite life span precursor to MSU-1.1. The MSU-1.1 acquired partial

growth factor independence. The PH3MT cell strain was derived from a malignant

57

tumor formed in an athymic mouse that was injected with MSU-1.1 cells transfected
with an overexpressed hRAS oncogene and selected for focus formation.

Using these cells we have shown that PHD2 represents the primary functional
HIF prolyl hydroxylase within the MSUl lineage of cells and that PHD2 activity
decreases as the cell exhibits more transformed characteristics. Moderate decreases in
PHD2 activity resulting from the use of RNAi in the non-tumorigenic cell strain, MSU-
l.l, resulted in malignant transformation. Interestingly, MSU-1.1 cells with a more
severe loss of PHD2 activity were unable to form tumors. Consistent with these results,
chemical inhibition of PHD2 activity in transformed cells decreases the cell’s tumor-
fonning potential and the overexpression of PHD2 in malignant fibroblasts lead to
inhibition of tumor growth. These results suggest a biphasic relationship between a
cell’s tumor-forming potential and its PHD activity and highlights potential difficulties

when targeting this signaling cascade for therapeutics.

58

U

dt

dc

MATERIAL AND METHODS
Cell culture, DMOG treatment, and Western blot analysis

MSU] cell strains were maintained in orMEM (Mediathech Inc, Hemdon, VA),
and human embryonic kidney (HEK) 293 cells and phoenix-ampho cells were cultured
in DMEM (Mediathech Inc) supplemented with 10% fetal calf serum (HyClone, Logan,
UT), 100 unit/ml penicillin, 100 pig/ml streptomycin, 2 mM L-glutamine and 1 mM
sodium pyruvate (Invitrogen, Carlsbad, CA). Cells were grown in a 37 oC incubator
with 5% C02 (Precision, Winchester, VA). Dimethyloxallyl glycine (DMOG, Sigma, St.
Louis, MO) was dissolved in 1x PBS prior to use in these studies.

Preparation of total and nuclear proteins and Western blot analysis were
performed as described previously (40, 43). The following antibodies were used; rabbit
polyclonal anti -PHD1, -PHD2, and -PHD3, mouse monoclonal anti-HIFla (Novus,
Littleton, CO), rabbit polyclonal anti-B-Actin (a generous gift from Dr. John Wang,
MSU), and goat anti-rabbit and mouse (Sigma).

Quantitative Real Time-PCR Analysis

For gene expression analysis, cells were exposed to normoxia (20% 02) or
hypoxia ( 1% 02) for 16 hr. Total RNA was extracted from cells using TRIzol reagent
(Invitrogen) according to the manufacturer’s protocol. RNA concentration was
determined by UV spectrometry and analyzed for integrity using spectrometry and

denaturing gel electrophoresis. 1 ug total RNA was reverse-transcribed using

59

SuperScript First-Stranded Synthesis System (Invitrogen) according to the
manufacturer’s protocol and primed with oligo(dT)ig primers. Gene expression was
measured by quantitative real time-PCR (qRT—PCR), based on Sybr-Green
methodology (Applied Biosystems, Foster City, CA) as previously described (42).
Analysis was performed using the ABI PRISM 7000 Sequence Detection System
(Applied Biosystems). Primers used for qRT-PCR were (5’ to 3’): VEGF:
tcctcacaccattgaaacca and gatcctgccctgtctctctg; PHDl: caggatgggagtggagagtt and
agtggtagaggtggctgtgg; PHD2: gagctgtgcgggaagatg and gcacacgagcttgtgcttct; PHD3:
agcttcctcctgtccctcat and acgtggcgaacataacctgt; HIF 1a: acaagtcaccacagggacqag and
agggagaaaatcaagtcg; GAPDH: cagcctcaagatcatcagca and gtcttctgggtggcagtgat; LDH:
aggcccgtttgaagaagagt and tgcacaacctccacctagaa; BNIP3: gctg gaacacgtaccatcct and
atctgcccatcttcttgtgg. The level of each gene was normalized to the expression level of
the hypoxanthine phosphoribosyltransferase (HPRT) gene and each primer set was
determined to be specific by BLAST and dissociation curve analysis.
Design of shRNA and shRNA-lentiviral constructs

The sequences of siRNA targeting PHDl, PHD2 or PHD3 (three independent
sequences per hydroxylase) were designed using the Ambion web-based design tool
(http://www.ambion.com/techlib/misc/siRNAfinder.); PHDl-a: tcagaactgggacgttaag;
PHDl-b: gactatatcgtgccctgcatg; PHDl-c: cgcaggaaggccatggtggcg; PHD2-a:

taaagactgggatgccaag; PHD2-b: gacgaaagccatggttgcttg; PHD2-c: cttcagattcggtcggtaaag;

6O

PHD3-a: tcggccctcactgaagact; PHD3-b: gtctaaggcaatggtggcttg; PHD3-c:
caggttatgttcgccacgtgg. Control scrambled shRNA, tgcgtcttgttcatctcct, was also
designed using a web-based tool (https://www.genscript.com/ssl-bin). The sequence of
each shRNA was analyzed by BLAST to ensure specificity for each target. Small
hairpin RNA (shRNA) constructs were generated using a two-step PCR approach.
Briefly, the first round PCR generated an amplicon of the U6 promoter with the sense
strand of the shRNA cassette and the loop. The second round of PCR added the
antisense strand of shRNA cassette. The shRNA cassettes that had been generated were
then cloned into the pGEM-T—Easy vector (Promega, Madison, WI). Once shRNA
cassettes were verified as functional, they were sub-cloned into the lentiviral vector,
pVCwPBam vector (a generous gift from Dr. David Looney, UCSD). pVCwPBam is
a lentiviral that was designed for the expression of shRNA cassette and it contains a
puromycin resistant sequence for the creation of stable cells.
Luciferase Assay

Each cell strain was transiently transfected utilizing Lipofectamine2000 via
manufacturer’s instructions (Invitrogen). DNA used in the transfections was diluted in
OptiMEM media (Invitrogen) and consisted of an HRE-driven luciferase reporter (17)
and shRNA-pGEMT—easy plasmids described above. A B-galactosidase expression

vector was also employed to control for transfection efficiency. Separate transfections

were performed utilizing a similar protocol with expression plasmid for PHD], PHD2,

61

OI

Ea

ll

or PHD3 (a generous gift of Dr. Steven McKnight, Univ. of Texas, Southwestern (7)).
Each transfection was adjusted to contain the same total concentration of transfected
DNA using an empty expression vector. Following transfection, the cells were
incubated for 16 hr and then exposed to normoxia (20% 02) or hypoxia (1% 02) for 18
hr. Cells were lysed and analyzed for luciferase activity using Luciferase Assay System
(Promega) according to manufacture’s protocol.
Viral packaging, infection and selection

The three independent shPHDs- and scrambled- shRNA-pVCwPBam vectors
were co-transfected into HEK293 cells with the packaging plasmids, pC34N and
pVSV-G as described above (a generous gift from Dr. David Looney, UCSD).
Following a 24 hour incubation, the media was replaced with fresh growth medium and
the cells were incubated for an additional 24 hr. Viral particles were purified by
centrifugation (650g for 5mins), followed by filtering with 0.45 pm membrane filter
(Millipore, Billerica, MA). MSU-1.1 cells (1x105) were plated 24 hr prior to infection
on 60 mm tissue culture dishes. Purified virus containing polybrene (5 rig/ml, Sigma)
was added to the cells and incubated for 24 hrs. Media was replaced and the cells
were cultured for 48 hrs. Infected cells were then selected with puromycin (0.4 rig/ml,
USBiologicals, Swampscott, MA) and individual clones were isolated.

The cDNAs for PHD2 and GFP were inserted into the retroviral vector pZOME-

1N and were packaged using Phoenix-ampho cells (a generous gift from Garry Nolan,

62

Stanford Univ.) (30). Purified viral particles were used to infect PH3MT cells and
infected cells were selected and individual clones were isolated as described above.
Anchorage Independence and Tumor Formation Assays

Assessment of a cells ability to grow in an anchorage-independent manner was
performed as previously described (20). Cells were fed weekly for 3 weeks and fixed in
gluteraldehyde (2.5%).

Athymic BALB/c mice, 5 weeks of age, were implanted subcutaneously in the
rear flank with 1 cm3 absorbable gelatin sponges (Pharmacia/Upjohn Company,
Kalamazoo, M1) to serve as a matrix for cell injection. One week after implantation 107
cells suspended in 0.2 m1 MEM medium were injected into each sponge. The mice
were monitored weekly for tumor growth and the size of tumor was measured. When
the tumor reached 1 cm in diameter, the mice were euthanized and the tumors were
removed. Tumors were then fixed with neutral buffered formalin for histological
examination. All mice were euthanized 5 months after injection, even if tumors were
not observed.

GAPDH, LDH and MTT Assays

Cells were lysed with 100 pl of 1% Triton X-100 (Sigma) and cell debris was
removed by centrifugation (3500g for 5min). 2.5 uL of supematants were transferred
into 96 well plate and 200 pl of GAPDH reagent (100 mM C7HI6NOSNa (TAPS), pH

8.6, 1 mM NAD“, 1.5 mM DL-glyceraldehyde 3-phosphate (Sigma), and 20 mM

63

NaHzPOa, 6 mM Cysteine (Fisher Scientific, Fair Lawn, NJ )) or LDH reagent (50 mM
KzHPOa, 200 M NADH.Na2 (Sigma) and 6.5 mM Pyruvate (Invitrogen)) were added
to each well for GAPDH or LDH assays, respectively. NADH kinetics was measured
by absorbance at 340nm for 5 min at 37 °C with a 12 second reading interval. Values
were normalized to protein concentration of sample. MTT assays were performed as

previously described (40).

64

RESULTS
Increasing HIF1 protein levels and activity and decreasing PHD2 levels in the
MSU] cell lineage as cells become more transformed

High levels of HIFIOI protein and HIF1 signaling are hallmarks of solid tumors
and cells derived from such tumors (4, 32, 45). To determine if this relationship exists
in the cell strains from the MSU] lineage, we compared the level of HIFla protein in
three key cell strains from the lineage using Western blot analysis. We found that the
HIFIOL mRNA levels were consistent across the cell strains, both in the presence and
the absence of oxygen (Fig. 2-1A). HIFla protein was undetectable under normoxia.
Its levels, however, were substantially increased following exposure to hypoxia, and
there was a more pronounced increase in the cells that had the acquired characteristics
of malignant cells (Fig. 2-lB). In addition, the levels of VEGF mRNA, a classic HIF1
target gene, were significantly increased in a manner that correlated to HIFIOL levels
under hypoxia (Fig. 2-1C). Although HIFIOL protein was undetectable under normoxia,
the tumor-derived PH3MT cells displayed higher basal levels of VEGF mRNA, than
the two non-tumorigenic cell strains (Fig. 2-1C). These results show that the levels of
HlFlot and HIF1 activity are higher in a tumor-derived cell strain than in the non-
tumorigenic precursor cells.

Since PHDs are the primary regulators of HlFlot stability we analyzed the levels

of these hydroxylases in the MSU] cell lineage. Comparative analysis of PHD mRNA

65

 

Figure 2-1. Expression levels of HIFIOI and HIF1 activity in MSU] lineage of cells.

(A) HIFIOL mRNA levels in three MSU] lineage of cells were determined using qRT-
PCR. Cells were exposed to normoxia (20% 02, white bar,) or hypoxia (1% 02,
black bar) for 16 hr. (n=6). (B) HIFIOL protein levels were determined in MSU]
lineage of cells by Western blot analysis. Cells were exposed to normoxia (N, 20% 02)
or hypoxia (H, 1% 02) for 6 hr, and nuclear proteins were prepared and analyzed for
HIFIOL protein levels. To verify equal loading, the blot was stripped and reprobed with
a B-Actin antibody. (C) VEGF mRNA levels were determined in each of the MSU]
lineage of cells using qRT-PCR. Cells were exposed to normoxia (20% 02, white bar)
and hypoxia (1% 02, black bar) for 16 hr (n=10, * p<0.05 ** p<0.01).

66

>

VEGF mRNA

El Normoxia
. I Hypoxia

PPT‘
neon

HIF1a. mRNA levels

 

 

 

 

 

0 ;
MSU 1.0 MSU 1.1 PH3MT

Cell Lines

Cell Lines

MSU1.0 MSU1.1 PH3MT
N H N H N H

.L’T.’ Bi t'*”"‘“

<— B-Actin

 

 

El Normoxia
’ 'I Hypoxia

15 **

l
i **

10' H

l :1
at. .L _
MSU1.0 MSU1.1 PH3MT
Cell Lines

 

 

 

 

expression

       

 

...J

67

levels revealed that PHD2 is the predominant isoform in these cells, comprising greater
than 75% of the total PHD mRNA in the cells (Fig. 2-2A). Western blot analysis
showed that PHD2 protein levels were inversely correlated with the increasing
transformation of the MSU] cell strains (Fig. 2-2B). These results show that PHD2 is
the predominant isoform within the MSU] cell strains, and its levels are inversely
correlated with the transformed characteristics and HIF1 activity of these cells.
HIF1 activity can be regulated by PHD2 in MSU] cell strains

To test PHD2’s role in HIF signaling in these cell strains, a series of transient
transfections was performed with an HRE-driven luciferase reporter construct. First,
the specific silencing of PHD isofoms was confirmed by Western blot analysis using
MSU 1.] cells transfected with each independent shRNA construct (Fig. 2-2C). In the
presence of a PHD2-specific shRNA construct, there was an increase in luciferase
activity under normoxic conditions (20% Oz) in the MSU-1.1 cells that was not
observed for the PHD] or PHD3 shRNAs (Fig. 2-2D). There was no significant
change in the presence of hypoxia for any of the shRNA constructs, suggesting that the
HRE-mediated transcription was maximally stimulated. In a similar set of experiments,
PHD1-3 were overexpressed in PH3MT cells to determine if increases in endogenous
levels of the various PHDs could inhibit hypoxia-induced transcription (Fig. 2-2E). The
results mirrored those of the shRNA data, in that only PHD2 was capable of inhibiting

HRE driven luciferase activity under hypoxia. The luciferase activity in the PHD] and

68

Figure 2-2. Expression levels of PHDs and the effects of modulating PHI) levels on
HIFl- hypoxia signaling in MSU] lineage of cells.

(A) The mRNA levels of each PHD isoform were analyzed in the MSU cell lines using
qRT-PCR and expressed as a percent of total PHD message. (n=12) (B) The protein
levels of the PHI) isoforms were determined in the MSU cell lines by Western blot
analysis with PHDl-3 specific polyclonal antibodies or a B-Actin specific antibody.
(C) Specific silencing of PHD isoforms using shRNA. MSU-1.1 cells were transiently
transfected with no transfectant (MSU-1.1/Ctrl), scrambled shRNA cassette (Scram) or
three independent shRNA cassettes targeting each PHD isoform (PHD1-3). Total
protein was isolated and analyzed by Western blot with specific antibodies for each
PHI) or B-Actin. (D) MSU-1.1 cells were transiently transfected with no shRNA
cassette (Ctrl), a scrambled shRNA (Scram), or shRNA cassettes targeting a specific
PHD isoform (PHDl-3), together with an HRE-driven luciferase reporter construct and
a B-gal expression vector for normalization. After transfection, cells were exposed to
normoxia (20% 02, white bar) or hypoxia (1% 02, black bar) for 16 hr and analyzed
for luciferase activity. (n=10, ** p<0.01) (E) PH3MT cells were transiently
transfected with nothing (Ctrl), an empty expression vector (Vector) or an expression
vector for the PHDs (PHD1-3) as described for MSU—1.1 in B. (n=10, ** p<0.01).

69

Cell Lines PHD1 PHDZ PHD3

 

MSU 1.0 25.1 % 74.8 °/o 0.1 %
MSU 1.1 16.6 % 88.1 % 0.3 °/o
PH3MT 12 % 87.7 % 0.3 %

 

B ,9 ,5 g.

—-- «PHD1
.‘w » ~- «PHD2

4—PHD3

- c- — +- B-Actin

 

 

 

 

C
\
vi: shRNA
.9
3' gs PHD1 PHD2 PHD3
‘9 e
é 'o a b c a b e a b c
-- .. o ' " -—---- fi-PHD‘I

i“---- .. ..---.—PH02
----....-,.. I «PHD3

--v---’-—-- 4— B-Actin

70

HRE- Lucifease Activity

HRE- Lucifease Activity

DNormoxia

IHypoxia

.t
O

 

O

#0!
CO

on
O

20‘

.x
O

 

O

 

 

Ctrl Scram PHD1 PHDZ PHD3
shRNA

 

EINormoxia

. IHypoxia

   

Ctrl Vector PHD1 PHDZ PHD3

 

Over-Expression

71

PHD3 transfected cells was similar to that found using the empty expression vector or
the mock transfected controls. These results are evidence that PHD2 is the major
functional isoform in regulating HIF1 activity in the MSU] cell strains and that direct
modulation of PHDZ activity can alter a cell’s ability to respond to hypoxic stress.
Decreases in PHD2 levels alter a cell’s tumor-forming potential

Our data show that the MSU] cell strains are a suitable system to directly test the
link between PHD levels, HIF1 activity, and tumorigenesis and prompted us to examine
whether the loss of PHD2 could bestow tumor forming ability upon a non-tumorigenic
cell strain. To answer this question, a series of stable cell strains was created that have
decreased levels of PHD2. Non-tumorigenic MSU-1.1 cells were infected with the
lentiviral vector, pVCwPBam, encoding three distinct shRNA targeting PHD2 or
scrambled shRNA. Initially, cell strains (shPHD2-a, -b, and -c) were assessed for
PHD2 levels using Western blot analysis. Among those, four clonal cell strains were
chosen from each shPHD2 strain that exhibited decreased PHD2 levels compared with
the parental MSU-1.1 cells and scrambled shRNA expressing controls (Fig. 2-3A). The
strains showed differences in PHD2 levels, with shPD2-a strains #7 and #21, shPHD2-
b strains #5 and #6, and shPHD2-c strains #40 and #41 having moderate reduction in
PHD2, which is similar to the levels of PHD2 in PH3MT, malignantly transformed
MSU-1 lineage of cells and shPHD2-a strains #2 and #5, shPHD2—b strain #1 and #10,

and shPHD2 strain #27 and #30 showing an almost complete loss of PHD2 expression.

72

Figure 2-3. Characterization of shPHD2 infected MSU 1.] strains.

(A) MSU-1.1 cells were infected with three independent lentiviral construct that
expresses distinct shRNA cassette targeting PHD2. The levels of PHD were
characterized in newly created shPHD2-a strains (#2, #5, #7, and #21), shPHD2-b
strains (#1, #10, #5, and #6), and shPHD2-c strains (#27, #30, #40, and #41) by
Western blot analysis using isoform-specific PHD antibodies or a B-Actin specific
antibody. The parental cell line (MSU-1.1) and a scrambled shRNA cell strain
(Scram) were included as controls. (B) HIFIOI protein levels were analyzed in
shPHD2-a strains exposed to normoxia (20% 02) or hypoxia (1% 02) for 6 hrs by
Western blot analysis. To verify equal loading, the blot was stripped and reprobed
with a B-Actin antibody. (C) shPHD2 strains (shPHD2-a #2, #5, #7 and #21,
shPHD2-b #1, #10, #5 and #6, and shPHD2-c #27, #30, #40 and #41) and
scrambled shRNA clone (Scram) were assessed for anchorage independent growth
by forming colonies in agarose. Parental MSU-1.1 cells and A210, hRAS-
transforrned cells, were included as a negative and positive control, respectively.
(n=10).

73

 

 

Encarniv -.I'" '
«Drill-.-

NDInIv .".. .. '

FDIn—Iv .""'

0e

3e 34 8a. Ru e
an...

 

cram—Imp?

"F" I...

I .8 ‘1‘ '1‘ a. ..I- It: ’1. If... T TI.

TI... N ...F
o

'I..4.J- _'. -.-

lIl .

..o

ee Eu; 2 an es 0% ed

a. .e A., an.
Pun—Eco 1.. J.

0% m% a; E

 

 

QINDImp—m

74

K
K

o
s“

l . “in. __ 34.—HIF1“

shPHDZ-a
#2 #5 #7 #21

 

f
o
9.)

Luau-I. --' Ihhh—r

 

I. d <— B-Actin

I r-trQQHEi‘-HW

n—Q—fi-i-_i d—fl-Actin

Hypoxia Normoxia

MSU 1.1 Scram A210
F—‘"‘ .11 .. ‘

, “d 4...-.. _- .' ’ O

shPHDZ-a
# 7 # 21

 

 

 

Ffl‘TQl-t‘l” - ' .I, 7' “q‘ in- . . . i.
l . ,_ l. , ' O .i

3 o . i . , .fl . - .
:Io . i ' . . i: .

 

 

shPHDZ-b
#1 #1o #5 _ #6

 

 

 

shPHDZ-c
# 27 # 30 g# 40

. 9A6? EFZET..7 . T q;

"r v.4»-v u u. roam”
r

 

75

The levels of PHD1 and PHD3 were unaffected in any of the shPHD2 infected cell
strains (Fig. 2-3A). To determine if the decreased PHD2 levels within the four strains
had a functional consequence, we characterized HIFIOL protein levels (Fig. 2-3B).
There was substantial HIFIOL protein in strains shPHD2-a #2 and #5 under normoxic
conditions, compared to that of the parental or scrambled shRNA cell strains. The level
of HIF10t protein was moderately up-regulated in strains shPHD2-a #7 and #21 under
normoxia. All of the shPHD2 strains displayed hypoxia-induced HIF] 0t stabilization.
To examine whether our shPHD2 cell strains had acquired transformed
characteristics, we first characterized their ability to form colonies in an anchorage-
independent manner (soft-agar assay) (Fig. 2-3C). The strains with the least PHD2 (i.e.
shPHD2-a #2 and #5, shPHD2-b #1 and #10, and shPHD2-c #27 and #30) were capable
of forming colonies only marginally better than the scrambled shRNA strain and the
parental MSU-1.1 cells; however, they did not perform as well as the positive control,
the RAS-transformed A210 cells . The strains with a moderate reduction in PHD2 (i.e.
shPHD2-a #7 and #21, shPHD2-b #5 and #6, and shPHD2-c #40 and #41) exhibited
strong anchorage-independent growth, forming colonies larger than the positive
controls (Fig. 2-3C). Additional shPHD2 strains with moderate or severe decreases in
PHD2 levels also displayed similar results (data not shown). These results indicate that
a small loss in PI-[D2 expression can aggressively promote a cell’s ability to grow in an

anchorage-independent manner, however, further loss of PHD2 does not significantly

76

change the cells’ anchorage independent growth phenotype.

To determine if the shPHD2 strains were capable of forming tumors, five
BALB/c athymic mice (5 weeks of age) were injected at two sites per mouse for each
shPHD2-a cell strains (shPHD2-a #2, #5, #7, and #21) or the scram cell strain, as a
control (Fig. 2—4). As expected, five months after injection, the scram cell strain did not
show any tumor growth, like the parental cell strain, MSU-1.1(27). In contrast, two
shPHD2-a strains, #21 and #7, yielded high grade fibrosarcomas at all ten injection
sites in weeks 3 and 5, respectively (Fig. 2-4). shPHDZ-a strains #2 and #5, which
exhibited the lowest levels of PHD2, were negative for tumor-forming ability even after
five months (Fig. 2-4). These results show that moderate decreases in PHD2 activity
lead to malignant transformation, while further loss of PHD2 activity produces cells
that do not form tumors.

These results suggest that a biphasic role exists for PHD2 in tumor formation
(Fig. 2-5). At normal levels of PHD2 (e.g. MSU-1.1), the hypoxic response is
regulated properly and no tumors are formed. With a slight decrease in PHD2 activity
(e.g. shPHD2-a strains #7 and #21 and PH3MT cells) the cells gain an advantage and
become malignantly transformed. As PHD2 activity is further decreased (e.g.
shPHD2-a cell strains #2 and #5), the pro-death response might become the dominant
signal and the increased adaptation would be overwhelmed. The dual nature of this

response is presumably due to PHD2’s ability to alter the cellular balance between

77

Figure 2-4. Tumor-forming ability of shPHD2-a strains.
shPHD2-a strains (#2, #5, #7 and #21) and scrambled shRNA clone (Scram) were

injected into athymic mice and tumor growth was monitored weekly for 5 months.
(n=10) (Images in this thesis/dissertation are presented in color)

78

 

 

Fufi Nora—cm IAVI

5 SE5 .0: .........................

mu Narnia LGI
Nu Nora—cm ID! .........................
£20m IOI

 

 

 

oN

at

cm

cm

2:.

our

(twin) awnjoA Joann;

79

Figure 2-5. A graphical representation of the relationship between PHD2 levels
and tumor forming potential

80

NO TUMORS TUMOR NO TUMORS

 

 

 

Excess pro-death

response Pre-adaptation Normal
' shPHDZ #2, #5 ] shPHDZ #7, #21 ] MSU 1.1 ]
LPH3MT-shPHoz PH3MT PHDZ strains

 

  
 

//

Tumor Forming Potential

 

/////A

PHDZ Levels

\

 

 

 

81

hypoxic-induced adaptation and pro-death responses.
Biphasic role of PHD in balancing HIF1-mediated adaptation and cell death

The proposed model presented in Figure 5 suggests that PH3MT and shPHD2-a
strains #7 and #21 (small decrease in PHD2) have a growth advantage, such as a higher
rate of glycolysis and angiogenesis via HIF1-mediated pre-adaptation. HIF1 also
regulates cell-death through transcriptional activation of pro-apoptotic factors (1, 41).
The model also suggests that shPHD2-a strains #2 and #5 (severe loss of PHD2) have
higher expression of pro-death genes, such as Bcl-2/adenovirus E13 19 kd-interacting
protein 3 (BNIP3), and a decreased viability under hypoxic stress.

To characterize the level of glycolysis and angiogenesis, the clones were
analyzed for the expression of a glyceraldehyde-3-phosphate dehydrogenase (GAPDH),
lactate dehydrogenase (LDH) and vascular endotheliaI growth factor (VEGF). To
determine the potential cell death signals within the cell strains, BNIP3, a known HIF1
regulated BCL2 family member, was also measured. PH3MT and shPHD2-a strains #7
and #21 showed a higher basal expression of GAPDH and LDH when compared to the
parental MSU-1.1 cells or scrambled control (Fig. 2-6A). In contrast, shPHD2-a strains
#2 and #5 had higher BNIP3 expression levels than the other clones tested (Fig. 2-6A).
Interestingly, VEGF basal expression was increased in the four PHD2 shRNA cell
strains compared to the parental strain (Fig. 2-6A). To determine if these expression

patterns had functional significance, GAPDH and LDH enzyme assays were performed

82

(Fig. 2-6B). These assays confirmed the mRNA data and showed that shPHD2-a strains
#7 and #21 have a higher rate of glycolytic activity. This activity was similar to that of
the malignantly-transformed PH3MT cell strain. Cell growth assays were also
performed on each of the cell strains, in the presence and absence of hypoxia. The
malignant strains, shPHD2-a #7 and #21 and PH3MT, had an increased growth
characteristics under hypoxic stress as compared to the parental MSU-1.1 and
scrambled control (Fig. 2-6C). shPHD2-a #2 and #5 exhibited increased doubling
time and an inability to grow under hypoxic stress (Fig. 2-6C). Finally, the functional
consequence of increased VEGF expression in shPHD2-a clone #21 was confirmed by
immunostaining of the tumors derived from these clones with the endothelial specific
antibodies for CD31 (PECAM) and Factor VIII. The tumors resulting from cells with
moderately decreased PHD2, PH3MT and shPHD2 #21, showed a range of
vascularization (Fig. 2-6D). These data support the biphasic model and are evidence
that small decreases in PHD2 activity can lead to malignant transformation, increased
glycolysis and vascularization, and severe loss of PHD2 activity can inhibit cell
viability and tumor-forming potentiaI through stimulation of cell-death pathways.
The inhibition of PHD activity reverses a cell’s transformed characteristics

To further support the biphasic model and provide evidence that the results
described in figures 3-6 were directly related to PHD activity, we inhibited PHDZ in

malignantly transformed PH3MT cells and shPHD2-a strains #7 and #21 using the

83

Figure 2-6. HIF1-mediated cellular responses in shPHD2 strains.

(A) mRNA levels of GAPDH, LDH, BNIP3, and VEGF were determined in MSU-
1.1, PH3MT, scrambled shRNA infected strain (Scram), and shPHD2-a strains (#2,
#5, #7 and #21) using qRT-PCR. (n=6, * p<0.05, ** p<0.01). (B) GAPDH and
LDH activity were determined in MSU-1.1, PH3MT, scrambled shRNA infected
strain (Scram), and shPHD2-a strains (#2, #5, #7 and #21). Kinetic activity was
normalized to protein concentration. (n=8, * p<0.05, ** p<0.01). (C) Each cell
strain was analyzed by MTT assay following exposure to normoxia (20% 02, white
bar) or hypoxia (1% 02, black bar) for 3 days. (n=4). (D) CD31 (PECAM) and
Factor VIII immunostaining was used to visualize vascularization in formalin fixed
tumor derived form shPHD2 strain #21, PH3MT, and a benign growth derived from
an early MSU lineage (Ctl).

84

**
** H
[j :1 [LU
#2 #5 #7 #21
shPHD2
**
[m
#2 #5 #7 #21
shPHD2
‘ **
Ltjj
#2 #5 #7 #21
shPHD2

 

as 8
$60 3 ,_H k 0‘06 A $00
“I“ as». a m U. «a... * é?
W Du *Q m ID NR0 Q NR0 Q
L u l l
2 owlb..0.\wav«. SZMMO 6* WNOOO 9%
1 1. 1 8 4
m_o>o_<sz m_o>o_<sz 1 bfl>30< In...
1 1 __. 1
“IH n 2 EMU n 2 .. n m
..LHW m u m m .. w m
P
m % m u % .m ,% .m
m n u n l n
8
H D saga. . age I see
D *THAQ Mikes“. ilk a.
P . \v. as
A , Q . at
G . o.

E
Q o;
to 4b

25 VEGF

5

0 I1
,3
Q
,4”

l

l
KN

o

$°

3 2 1
A 222 <sz «..2, B b_>=o< :920

85

El Normoxia
0.7 . I Hypoxia

E
C
O
m
e.
0
2
N
>
D
o .
o
N
o'" «$6
*9 .36
co 31

warezv‘rz‘r
w' '- ~

    
 
 

. (a
J."

.“ (J

 

..2}

     

86

# 5 # 7
shPHDZ

 

 

dimethyloxallyl glycine (DMOG). If the model is correct, further inhibition of PHDZ in
these cells should force them to phenotypically resemble shPHD2-a strains #2 and #5.
DMOG was capable of inhibiting PHDs in a dose dependent manner as evidenced by
the increasing stability of HIFla, with a maximal effect observed at 1 mM (Fig. 2-7A).
To exam if the inhibition of PHD activity can alter the tumor forming potential of
PH3MT cells and shPHD2-a strains #7 and #21, we analyzed their ability to form
colonies in an anchorage-independent manner in the presence of DMOG (Fig. 2-7B).
As predicted by the model, each of these cell types lost their ability to grow in soft agar
when PHD activity was diminished by DMOG. To insure this was not due to DMOG
toxicity, cells were plated on standard culture dishes and incubated in the absence or
presence of DMOG (1 mM) for 10 days and stained (Fig. 2-7C). There was a reduction
in the size of clones on the DMOG treated plates, however, removal of DMOG lead to
growth recovery within 5 days suggesting that direct inhibition of PHD activity
decreases cellular proliferation. These results suggest that DMOG is capable of
inhibiting the anchorage independent growth of the transformed cells and that this
inhibition may due to growth inhibition or cellular senescence. Presumably, this is
caused by high basal expression of pro-death gene, such as BNIP3, as described for
shPHD2-a strains #2 and #5 (Fig. 2-6A). Indeed, inhibition of PHD2 activity by
DMOG in tumorigenic shPHD2-a strains #7 and #21 and PH3MT cells led to a

significant (increase in BNIP3 expression (Fig. 2-7D). These results suggest that

87

Figure 2-7. Inhibition of PHD2 activity in transformed cells.

(A) HIFla protein levels were determined in PH3MT cells treated with various
concentrations of DMOG by Western blot analysis. HIFla levels in PH3MT cells
exposed to normoxia (N, 20% 02) or hypoxia (H, 1% 02) were also analyzed as a
control and B-Actin antibody was used as a loading control. (B) PH3MT,
shPHD2 strains #7 and #21, and MSU [.1 cells were assessed for anchorage-
independent growth using soft-agar assay, in the absence (ctl) and presence of
DMOG (1 mM in media). (n=lO) (C) Cell viability assessment following DMOG
treatment. 103 cells were plated in 100mm culture dish and cultured in the presence
or the absence of DMOG (1 mM) for 10 days and stained. A separate set was
exposed to DMOG followed by 5 days recovery in the normal growing media.
(n=10) (D) BNIP3 mRNA levels were determined in MSU-1.1, shPHD2 strains
#7 and #21, and PH3MT cells left untreated (ctl, white bar) or exposed to DMOG
(black bar). (n=6, * p<0.05)

88

DMOG
K ‘5 ‘3 ’\
H N o9 o9 o9o~° Q56? N w (mM)

.- 31.... «mm

M <- B-Actin

 

 

MSU1.1

 

PH3MT

shPHDZ #7

 

 

shPHDZ #21

 

 

89

#21

PH3MT

CTL

 

M
m
G
o
M
D

 

DCtl
. IDMOG (1mM)

l

*

2. 8. 4
1 o 0..

m_o>0_ <Z¢E nn=zm

 

90

malignantly transformed cells (e.g. shPHD2-a #7 and #21 and PH3MT) can lose their
tumor-forming potential in the presence of PHD inhibitors and a pro-death response is
involved in this phenotypic alteration.

Effects of alterations in PHD2 activity in the transformed cell strain, PH3MT

The model also predicts that altering the PHD2 activity in malignantly
transformed cells would alter the cell’s tumor-forming potential. To test this hypothesis,
we altered the PHDZ activity in PH3MT cells. First, PH3MT cells were infected with
the shPHD2 cassettes described above, and strains were selected. We were unable to
expand any of the 25 cell strains, suggesting these PHD2 shRNA expressing cells were
prone to premature cell death (data not shown). This observation is supported by the
previous results in which DMOG treatment lead to inhibition of anchorage independent
growth in malignantly transformed cells with a corresponding increase in pro-cell death
factor, BNIP3, expression. Each of these results is evidence that decreasing PHD2
activity within tumorigenic cells lead to a loss in tumor-forming potential.

The model proposed in Figure 2-5 also suggests that by increasing PHD2 levels
within these same cells will Iead to a similar decrease in tumor-forming potential. To
test this, PH3MT cells were infected with the retroviral vector, pZome—IN, encoding
the cDNA for PHD2. pZome-lN produces a tagged PHD2 protein and its expression
can be distinguished from endogenous PHD2 based on their molecular size. In addition,

a cDNA for GFP was also inserted into separate pZome-lN and was used to create a

91

control cell strains. PHD2 cell strains were screened for overexpression of PHDZ using
Western blot analysis and four of these (PHD2 strain #6, #8, #11, and #22) were
selected for further analysis (Fig. 2-8A). Tagged-PHD2 and GFP are visualized on the
B-actin Western blot due to the Protein A motif within the tag. We next examined
whether over-expression of PHD2 affects hypoxia-induced HIF10t stabilization (Fig. 2-
8B). As expected, HIFIOL was undetectable under normoxia and its levels were
increased by hypoxia in control, PH3MT and GFP strains, #3 and #4. However, the
PHD2 over-expressing cell strains, PHD2 #6, #8, #11 and #22, showed reduced
hypoxia-induced HIFIOL accumulation. These results are in agreement with previous
published reports showing that increased PHD expression can inhibit HIFla
accumulation and HIF1 activity under hypoxic stress (12, 13, 39). In addition over-
expression of PHD2 had functional consequences on HIF1-mediated upregulations of
GAPDH, LDH and VEGF. The hypoxia-induced expression of these genes was
diminished in all PHDZ strains, PHD2 #6, #8, #11 and #22 when compared to hypoxia
treated control cells, PH3MT and GFP strains #3 and #4 (Fig. 2-8C). GAPDH and LDH
enzyme assays confirmed the mRNA data and showed that PHD2 strains #6, #8, #11
and #21 lost their hypoxia-induced glycolytic activity (Fig. 2-8D).

To determine if PHD2 over-expression can alter a cell’s transformed phenotype,
each of the PHD2 cell strains was analyzed for its ability to form colonies in soft agar

(Fig. 2-9A). The tumorgenic parental PH3MT and control GFP strains, #3 and #4,

92

Figure 2-8. Over- expression of PHD2 in PH3MT cells.

(A) PH3MT cells were infected with a retroviral construct that expresses PHD2
cDNA. The expressions of PHD2-TAP were characterized in the parental PH3MT,
PHD2 strains (#6, #8, #11 and #22) and GFP cell strains (#3 and #4) by Western
blot analysis using a PHD2-specific antibody or B-Actin antbody. The PHD2 and
GFP proteins are visible due to the protein A tag. (B) HIFIOL protein levels in
PHD2 strains following exposure to normoxia (20% 02) or hypoxia (1% 02) were
analyzed by Western blotting with a HIFlCX monoclonal antibody or B-Actin
antibody. (C) mRNA levels of GAPDH, LDH, and VEGF were determined in
PH3MT, GFP strains, #3 and #4 and PHD2 strains #6, #8, #1 l and #22 using qRT-
PCR. Cells were exposed to normoxia (20% 02, white bar,) or hypoxia (1% 02,
black bar) for 16 hr. (n=8, * p<0.05, ** p<0.01). (D) GAPDH and LDH activity
were determined in MSU-1.1, PH3MT, GFP strain #3 and #4 and PHD2 strains #6,
#8, #11 and #22. Cells were exposed to normoxia (20% 02, white bar,) or
hypoxia (1% 02, black bar) for 16 hr and enzymatic activity was normalized to
protein concentration. (n=8, * p<0.05, ** p<0.01).

93

PHDZ -> ' 'l! I “u

Normoxia

Hypoxia

 

 

 

”$5 PHD2 GFP
41’ #6 #8 #11 #22 #3 #4

‘- PHDZ-Tag
+ GFP -Tag

 

--- - -— + GFP -Tag
”__-.--..___.-- -—-—-<-B-Actin

« PHDZ
”a GFP

db #3 #4 #6 #8 #11 #22

 

 

<- HIF1a

— ¢ — — --- + B-Acti"

p-d ~ «H'm

- 4- B-Actin

 

94

GAPDH C] Normoxia

 

 

 

“I * IHypoxia
.2!
O
>
2
<
z
m
E I
4 l .
$5 #3 #4 #6 #8#11#22
dz? GFP PHDz
LDH
l DNormoxia
4o. ** ** ** I Hypoxia

  

mRNA levels
N
O

 

 

 

 

 

 

 

 

10
o , I
$ #3 #4 #6 #8 #11#22
Q31" GFP PHDZ
20 _VEGF
fl, DNormoxia
IHypoxia
_g 15
0
3
< 10
E l
E 5 ‘
. l I I
$ #3 #4 #6 #8 #11 #22

4.5.5 GFP PHDZ

El Normoxia
I Hypoxia

60 "

 

 

 

**

 

**

111111

 

>=>zo< IDQ<G

#6 #8 #11 #22
PHDZ

#4

3

 

 

 

E] Normoxia
I Hypoxia

 

 

 

 

 

 

 

b_>_uu< In...

PHDZ

 

96

exhibited large colonies when grown in agar, while the PHD2 clonal cell strains, #6, #8,
#11 and #22, lost this ability to grow in an anchorage-independent manner. To
determine whether over-expression of PHD2 lead to decreases in tumor forming
potential, PHD2 strains # 6, #8, #11, and #22 were examined for their ability to form
tumors in athymic mice (Fig. 2—9B). Parental PH3MT and GFP strains, #3 and #4, were
used as a control and yielded tumors within 6 weeks (Fig. 2-9B). In contrast, all of
the PHD2 strains, #6, #8, #11, and #22, were negative for tumor formation after 5
months (Fig 2-9B). These results show that the increase of the PHD2 activity in
malignantly transformed cells can inhibit a cells’ transformed phenotype and support

our biphasic model relating PHD2 with tumor forming potential.

97

1.1.3

Figure 2-9. Characterization of tumor forming potential in PHD2 strains.

(A) PHD2 strains (#6, #8, #11 and #22) and GFP strains (#3 and #4) were assessed
for anchorage-independent growth by forming colonies in soft agar. Parental
PH3MT cells and A210 and MSU-1.1 cells were used as positive controls and
negative control, respectively. (n=10) (B) PHD2 strains (#6, #8, #11 and #22) and
GFP strains (#3 and #4) were injected into athymic mice and tumor growth was
monitored weekly for 5 months. (n=10)

98

 

 

9.0

 

_.

5‘ 0|:

.r 3m:

99

 

nut NOTE IXI
3% «CIA. I®l
at unin— III
0% Naza— |®I
3‘ 3.0
at nan—0
#22...“— IAvI

 

 

 

 

G
V‘

O
Q

G
O
1-

c3.

60..

(ennui) awnjoA Jownl

100

DISCUSSION

HIF1 regulated genes are involved in many cellular processes including cell
proliferation, angiogenesis, metabolism, migration and many others which are known
to be required for adaptive survival of tumor cells (9, 10, 19, 35). Alternatively, severe
hypoxia exposure can lead to cell death through HIF1-mediated up-regulation of pro-
cell death factors or p53-dependent processes (1, 6, 11, 37). Hypoxia signaling,
therefore, regulates a delicate balance between life and death through cellular
adaptation and a programmed death response. A cancer cell’s survival is dependent
upon its ability to maintain cell growth and/or decrease its programmed death response
once it is exposed to the hypoxic microenvironment of a tumor. The correlation
between PHD2 levels and tumor-forming potential suggests that these hydroxylases
might be involved in altering this balance (Fig. 2-l~2-5). Presumably, small decreases
in PHD activity would promote an adaptive response without increasing pro-death
signals (Fig. 2-6). In addition, the decreased hydroxylase activity and subsequent
increase in HIF1—mediated signaling can explain the observation that tumors and
corresponding cell strains have an increase in hypoxia signaling, even in the presence
of normal oxygen concentrations. The direct link between decreased PHD activity,
increased HIF1 signaling, and increased glycolytic activity might also explain the
Warburg effect (44). Previous reports have shown that HIF1 is necessary for the

Warburg effect, and a cellular decrease in PHD activity would explain a tumor’s

101

increased dependence on aerobic glycolysis (5, 16, 19, 34). It is possible that one step
in the transformation process is the sustained decrease in PHD activity, through genetic
or epigenetic mechanisms. This would serve to pre-adapt the cells (eg. increased
glycolytic rate) to the hypoxic environment found in many, and perhaps all, tumors and
gives them a growth advantage upon tumor development. It is also possible that this
loss of PHD activity and subsequent increased glycolytic activity causes the malignant
transformation, as Warburg had proposed (44).

Interestingly, cells with a severe loss of PHD2 showed no ability to form tumors
in athymic mice (Fig. 2-3 and 2-4). These cell strains, both MSU—1.1-derived and
PH3MT-derived, displayed growth abnormalities such as signs of premature cell death,
increased doubling time and an inability to grow under hypoxic stress (Fig. 2-6). It is
hypothesized that this is due to an uncontrolled pro-death response. It is hypothesized
that this response is driven by direct HIF1-mediated transcription of genes such as
BNIP3 and NIX. The almost complete loss of PHD2 in shPHD2 strains #2 and #5, and
subsequent HIFl activation, would also presumably contribute to p53-mediated cell
cycle arrest and apoptosis. Given the overwhelming pro-death response following
almost complete loss of PHD2 activity, no amount of adaptive cell signaling can
support continued expansion in the tumor microenvironment.

These two groups of cell strains, mild decrease and severe decrease in PHD2

levels, led to the proposed biphasic model presented in Figure 2-5. This type of model

102

would also predict that transformed cells are within the phase of the curve that supports
tumor formation and movement in either direction (more or less PHD activity) will
alter the cell’s tumor forming potential. The tumorigenic PH3MT cells and modulation
of PHDZ levels in these cell strains strongly support our model. First, the PH3MT cells
have a decreased level of PHDs when compared to the MSU-1.0 or MSU—1.1 (Fig. 2-
23). Second, PH3MT cells that express the PHD2 shRNA cassette (movement left
along the abscissa, Figure 2-5) stop growing after colony selection and cannot be
expanded. Third, tumorigenic cells (PH3MT, shPHD2-a #7 and #21) lost their
transformed characteristics when PHD activity was drastically decreased by a
hydroxylase inhibitor (Fig. 2-7). Fourth, over-expression of PHD2 (movement right
along the abscissa, Fig. 2-5) in the PH3MT cell’s inhibits the cells ability to grow in an
anchorage-independent manner and form tumor in athymic mice, suggesting they have
lost a transformed phenotype (Fig. 2-8). Finally, a recent report has shown that over-
expression of PHDl in colon cancer cells inhibits tumor growth (15). Taken together,
these results provide evidence for the biphasic model of PHD activity and tumor
forming potential and suggest the clinical importance of characterizing the factors that
establish the boundaries dictating the separate areas within this model.

The biphasic model will have profound consequences on chemotherapeutic
agents that target PHD3 and the hypoxia signaling cascade for cancer therapy. Cells that

have amassed the cellular mutations necessary to progress towards complete

103

transformation (e. g. MSU-1.1 cells) might become fully tumorigenic when exposed to a
drug that only partially blocks this cascade, while other cells at a different
transformation stage (e.g. PH3MT cells) will undergo cell death. Alternatively, new
therapies that completely shut down PHD activity will have detrimental consequences
on normal cellular homeostasis by promoting an uncontrolled pro-death response.
Ideally, any drug targeting this pathway should be specific to precancerous and hypoxic
cells thus increasing the chance that decreases in PHD activity within the cell will push
these cells toward cell death. The data presented here implies that cells will have a
biphasic response to such agents and if this is the case, it will create difficulties when

targeting these enzymes for cancer therapeutics.

104

REFERENCES

1. An, W. G., M. Kanekal, M. C. Simon, E. Maltepe, M. V. Blagosklonny, and L.
M. Neckers. 1998. Stabilization of wild-type p53 by hypoxia-inducible factor
lalpha. Nature 392:405-408.

2. Berra, E., E. Benizri, A. Ginouves, V. Volmat, D. Roux, and J. Pouyssegur.
2003. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state
levels of HIF-lalpha in normoxia. EMBO J. 22:4082-4090.

3. Berra, E., Benizri, E., Ginouves, A., Volmat, V., Roux, D., and Pouyssegur, J.
2003. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady—state
levels of HIF-la in normoxia. EMBO J. 22: 4082-4090.

4. Birner, P., M. Schindl, A. Obermair, C. Plank, G. Breitenecker, and G.
Oberhuber. 2000. Overexpression of hypoxia-inducible factor lalpha is a marker
for an unfavorable prognosis in early-stage invasive cervical cancer. Cancer Res.

60:4693-4696.

5. Briere, J. J ., J. Favier, P. Benit, V. E. Ghouzzi, A. Lorenzato, D. Rabier, M. F.
Di-Renzo, A. P. Gimenez-Roqueplo, and P. Rustin. 2005 . Mitochondrial
succinate is instrumental for HlFla nuclear translocation in SDHA-mutant
fibroblasts under normoxic conditions. Human Mol. Genet. 14:3263-3269.

6. Bruick, R. K. 2000. Expression of the gene encoding the proapoptotic BNIP3
protein is induced by hypoxia. Proc. Natl. Acad. Sci. USA 97 :9082-9087.

7. Bruick, R. K., and S. L. McKnight. 2001. A conserved family of prolyl-4-
hydroxylases that modify HIF. Science 294:1337-1340.

8. Carmeliet, P., Y. Dor, J. M. Herbert, D. Fukumura, K. Brusselmans, M.
Dewerchin, M. Neeman, Bono, F, R. Abramovitch, P. Maxwell, C. J. Koch, P.
Ratcliffe, L. Moons, R. K. Jain, D. Collen, and E. Keshertk. 1998. Role of
HIF-la in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis.
Nature 394:485-490.

9. Carmeliet, P., and R. K. Jain. 2000. Angiogenesis in cancer and other disease.
Nature 407 :249-257.

10.Chen, C.,-N. Pore, A. Behrooz, F. Ismail-Beigi, and A. Maity. 2001. Regulation of
glutl mRNA by Hypoxia-inducible Factor-1. Interaction between H-ras and

105

11

12

13

14

15

16.

17.

18.

19

20.

hypoxia. J. Biol. Chem. 276:9519-9525.

. Chen, D., M. Li, J. Luo, and W. Gu. 2003. Direct interaction between HIF1-0t and

Mdm2 modulate p53 function. J. Biol. Chem. 18:13595-13598.

. Choi, K., T. Lee, N. Lee, J. Kim, E. G. Yang, J. M. Yoon, J. H. Kim, T. G. Lee,

and H. Park. 2005. Inhibition of the Catalytic Activity of Hypoxia-Inducible
Factor-lalpha-Prolyl-Hydroxylase 2 by a MYND-Type Zinc Finger. Mol.
Pharmacol. 68: 1803-1 809.

. Cioffi, C. L., X. Q. Liu, P. A. Kosinski, M. Garay, and B. R. Bowen. 2003.

Differential regulation of HIF prolyl4-hydroxylase genes by hypoxia in human
cardiovascular cells. Biochem. Biophys. Res. Commun. 303:947- 953.

. Epstein, A. C., J. M. Gleadle, L. A. McNeil], K. S. Hewitson, J. O'Rourke, D. R.

Mole, M. Mukherji, E. Metzen, M. 1. Wilson, A. Dahnda, 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. Ratcloffe. 2001. C.elegans EGL-9 and
mammalian homologs define a family of dioxygenases that regulate HIF by prolyl
hydroxylation. Cell 107:43-54.

. Erez, N., M. Milyavsky, R. Eilam, I. Shats, N. Goldfinger, and V. Rotter. 2003.

Expression of prolyl-hydroxylase-l (PHD1/EGLN2) suppresses hypoxia-inducible
factor la and inhibits tumor growth. Cancer Res. 63:8777-8783.

Esteban, M. A., and P. H. Maxwell. 2005. HIF, a missing link between
metabolism and cancer. Nat. Med. 11:1047-1048.

Gu, Y. Z., S. M. Moran, J. B. Hogenesch, L. Wartman, and C. A. Bradfield.
1988. Molecular characterization and chromosomal localization of a third alpha-
class hypoxia inducible factor subunit, HIF3alpha. Gene Expression 7 :205-213.

Harris, A. L. 2002. Hypoxia - a key regulatory factor in tumor growth. Nat. Rev.
Cancer 2:38-47.

. Hockel, M., and P. Vaupel. 20011. Tumor hypoxia: Definitions and current clinical,

biologic, and molecular aspects. J. Natl. Cancer Inst. 98:266-276.

Hurlin, P. J., D. G. Fry, V. M. Maher, and J. J. McCormick. 1987.
Morphological transformation, focus formation, and anchorage independence
induced in diploid human fibroblasts by expression of a transfected H-ras

106

21.

22

23.

24.

25.

26.

27.

28

oncogene. Cancer Res. 47:5752-5757.

Ivan, M., T. Haberberger, D. C. Gervasi, K. S. Michelson, V. Gunzler, K.
Kondo, H. Yang, 1. Sorokina, R. C. Conaway, and J. W. Conaway. 2002.
Biochemical purification and pharmacological inhibition of a mammalian prolyl
hydroxylase acting on hypoxia-inducible factor. Proc. Natl. Acad. Sci. USA
99:13459-13464. ‘

. Jaakkola, P., D. R. Mole, Y.-M. Tian, M. 1. Wilson, J. Gielbert, S. J. Gaskell, A.

Von Kriegsheim, H. F. Hebestreit, M. Mukherji, and C. J. Schofield. 2001.
Targeting of HIF-0t to the von Hippel-Lindau ubiquitylation complex by 02-
regulated prolyl hydroxylation. Science 292:468-472.

Lieb, M. E., K. Menzies, M. C. Moschella, R. Ni, and M. B. Taubman. 2002.
EGLN genes have distinct patterns of mRNA expression and regulation. Biochem.
Cell. Biol. 80:421-426.

Maxwell, P. H., G. U. Dach, J. M. Gleadle, L. G. Nicholls, A. L. Harris, I. J.
Stratford, O. Hankinson, C. W. Pugh, and P. J. Ratcliffe. 1997. Hypoxia-
inducible factor-1 modulates gene expression in solid tumor and influence both
angiogenesis and tumor growth. Proc. Natl. Acad. Sci. USA 94:8104-8109.

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. 1999.
The tumor suppressor protein VHL target hypoxia-inducible factors for oxygen-
dependent proteolysis. Nature 399:271-275.

McCormick, J., S. O'Reilly, and V. Maher. 2000. Analysis of the process of
malignant transformation of human fibroblasts in culture by ionizing radiation and
other carcinogens. In Radiation Research: Volume 2 Proceedings of the Eleventh
International Congress of Radiation Research. Ireland, D., (ed) Vol. 2, pp 564-567.
Allen Press.

McCormick, J. J., and V. M. Maher. 1996. Analysis of the multistep nature of
human carcinogenesis utilizing human fibroblasts. Radiat. Oncol. Invest. 3:387‘
391.

. Metzen, E., U. Berchner-Pfannschmidt, P. Stengel, J. H. Marxsen, I. Stolze, M.

Klinger, W. Q. Huang, C. Wotzlaw, T. Hellwig-Burgel, W. J elkmann, H. Acker,
and J.- Fandrey. 2003. Intracellular localisation of human HIF-1 alpha
hydroxylases: implications for oxygen sensing. J. Cell Sci. 116:1319-26.

107

1.1.-LE

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

Piret, J. P., D. Motter, M. Raes, and C. Michiels. 2002. Is HIF-10L a pro- or an
anti—apoptotic protein? Biochem. Pharrnacol. 64:889-892.

Rigaut, G, A. Shevchenko, B. Rutz, M. Wilm, M. Mann, and B. Seraphin.
1999. A generic protein purification method for protein complex characterization

and proteome exploration. Nat.Biotechnol. 17: 1030- 1032.

Ryan, H. E., J. Lo, and R. S. Johnson. 1998. Hif-la is required for solid tumor
formation and embryonic vasculation. .EMBO J 17:3005-3015.

Ryan, H. E., M. Poloni, W. McNulty, D. Elson, M. Gassmann, J. M. Arbeit,
and R. S. Johnson. 2000. Hypoxia-inducible factor-10L is a positive factor in solid
tumor growth. Cancer Res. 60:4010-4015.

Selak, M. A., S. M. Armour, E. D. Mackenzie, H. Boulahbel, D. G. Watson, K.
D. Mansfield, Y. Pan, C. Simon, C. B. Thompson, and E. Gottlieb. 2005.

Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-0t prolyl
hydroxylase. Cancer Cell 7:77-85.

Semenza, G. L. 2000. Hypoxia, clonal selection, and the role of HIF-1 in tumor
progression. Crit. Rev. Biochem. Mol. Biol. 35:71-103.

Semenza, G. L. 2003. Targeting HIF—1 for cancer therapy. Nature 3:721-731.

Semenza, G. L., and G. L. Wang. 1992. A nuclear factor induced by hypoxia via
de novo protein synthesis binds to the human erythropoietin gene enhancer at a
site required for transcriptional activation. Mol. Cell. Biol. 12:5447-5454.

Sowter, H. M., P. J. Ratcliffe, P. Watson, A. H. Greenberg, and A. L. Harris.
2001. HIF—l-dependent regulation of hypoxic induction of the cell deathfactor
BNIP3 and NIX in human tumors. Cancer Res. 61:6669-6673.

Stackpole, C. W., L. Groszek, and S. S. Kalbag. 1994. Benign-to-malignant Bl6
melanoma progression induced in two stages in vitro by exposure to hypoxia. J.
Natl. Cancer Inst. 86:361-367.

Stiehl, D. P., R. Wirthner, J. Koditz, P. Spielmann, G. Camenisch, and R. H.

Wenger. 2006. Increased prolyl 4-hydroxylase domain proteins compensate for
decreased oxygen levels. J. Biol. Chem. 281:23482-23491.

108

Ann.

40.

41.

42.

43.

45.

Vengellur, A., and J. J. LaPres. 2004. The role of hypoxia inducible factor
1{alpha} in cobalt chloride induced cell death in mouse embryonic fibroblasts.
Toxicol. Sci. 82:638-646..

Vengellur, A., and J. J. LaPres. 2004. The role of hypoxia inducible factor
1{alpha} in cobalt chloride induced cell death in mouse embryonic fibroblasts.
Toxicol. Sci. 82:638-646.

Vengellur, A., B. G. Woods, H. E. Ryan, R. S. Johnson, and J. J. LaPres. 2003.
Gene expression profiling of the hypoxia signaling pathway in hypoxia inducible
factor 10L null mouse embryonic fibroblasts. Gene Expression 11:181-197.

Wang, G. L., and G L. Semenza. 1995. Purification and characterization of
hypoxia-inducible factor 1. J. Biol. Chem. 270: 1230-1237.

Warburg, O. 1956. On the origin of cancer cells. Science 123:309-314.
Zhong, H., A. M. Marzo, E. Laughner, M. Lim, D. A. Hilton, D. Zagzag, P.
Buechler, W. B. Issacs, S. GL., and J. W. Simons. 1999. Overexpression of

Hyoxia-inducible factor lot in common human cancer and theri metastasis. Cancer
Res. 59:5830—5835.

109

CHAPTER 3

THE PROTEIN INTERACTION NETWORK OF HIF PROLYL
HYDROXYLASE, PHD2, IN HUMAN FIBROBLASTS

KangAe Lee and John J. LaPres

110

ABSTRACT

The oxygen-dependent degradation of the alpha subunits of the hypoxia-
inducible factors (HIFa) plays an essential role in regulating gene expression system in
response to hypoxia. This process requires the prolyl hydroxylase (PHDs)-mediated
hydroxylation of conserved prolines within the HIFOL, facilitating its interaction with
the von Hippel-Lindau protein, the recognition component of the E3 ubiquitin lagase.
PHDs belong to the iron- and 2-oxoglutarate-dependent oxygenase family that require
dioxygen for activity. The oxygen dependence of PHD and the role of PHD in
regulating HIF signaling suggest a role for PHD as an oxygen sensor. Characterizing
the precise mechanism and regulation of these enzymes, therefore, is essential to
understand physiological and pathological responses to hypoxia. In this study, we
characterized the PHD2-protein interaction network (PHD2-PIN) using tandem affinity
purification and liquid chromatography coupled tandem mass spectrometric analysis
(LC-MS/MS). Our results suggest that PHD2 might have no specific protein interaction

under normoxic conditions in this particular cell type.

111

INTRODUCTION

A cell’s ability to respond to changes in oxygen availability is a fundamental
property of most aerobic organisms and involves a precisely regulated gene expression
system. Hypoxia-inducible factors (HIFs) are the key transcription factors in regulating
a wide range of genes responding to decreased oxygen tension (22, 30). HIFs are
heterodimers of HIFO. and HIFB subunits, both of which are member of the basic helix-
loop-helix Per-ARNT-SIM (bHLH-PAS) protein family of transcription factors (30).
HIFIB is constitutively expressed as a nuclear protein and has previously been
recognized as the dimerization partner of the aryl hydroxarbon receptor, where it is
termed the aryl hydrocarbon receptor nuclear translocator (ARNT). HIFs is are specific
to the hypoxic response and constantly synthesized and under normoxia, immediately
removed via ubiquitination and proteasomal degradation (23, 30). This process
involves the interaction of the HIFO, with the von Hippel-Lindau (VHL) tumor
suppressor protein (pVHL) which acts as a recognition component of the E3 ubiquitin-
ligase complex (23). The interaction between HIFOL and pVHL depends on the
posttranslational hydroxylation of two conserved proline resides within the oxygen-
dependent degradation domain (ODD) of HIFOL (13, 14).

A family of prolyl hydroxylase domain containing proteins (PHDs, also known
as egg-laying deficient nine-like proteins (EGLNs) and HIF—prolyl-hydroxylases
(HPHs)) has recently been identified that can catalyze the hydroxylation of the

conserved proline residues within the ODD in an oxygen-dependent manner (9, 20, 23).

112

 

Three human PHD isoforms, PHD 1-3, have been identified with distinct
characteristics in substrate specificity, intracellular distribution and tissue specific
inducibility under hypoxia (5, 7). In mammals, PHD2 appears to be the predominant
isoform that regulates HIFCX protein stability (4, 15). PHDs belong to an iron and 0t-
ketoglutarate (2-oxoglutarate)-dependent dioxygenase family that requ'ne oxygen for
proper hydroxylation of prolines within the HIF1(X(5, 7, 12). Unlike other members of
this class of dioxygenases, H202 cannot be substituted for molecular oxygen in the
PHDs (9, 25). The oxygen-dependence and relatively high Km for 02 (in vitro) of these
enzymes suggests a role for PHD3 as cellular oxygen sensor for the hypoxia signaling
system (9, 10, 18, 26). In addition to oxygen, ferrous iron (Fe2+) and ascorbate are also
required for proper PHDs function. The substrate and product, a-ketoglutarate and
succinate, respectively, also affect PI-ID’s activity. Succinate functions as a competitive
inhibitor for PHD, which can be overcome by increasing a-ketoglutarate
concentrations (17, 25). Both OL-ketoglutarate and succinate are tricarboxylic acid
(TCA) cycle intermediates, suggesting some level of crosstalk of hypoxia signaling
with mitochondrial activity and metabolic state. Several recent studies suggested that
abnormal levels of TCA cycle intermediates (i.e. succinate and furmarate) and
glycolytic metabolites (i.e. pyruvate), can perturbed PHD activity (11, 19, 21, 29). It
has also been proposed that mitochondrial generated reactive oxygen species (ROS)
can alter PHD activity suggesting a possible mechanisms that links the respiratory

chain to HIF signaling (8).

113

Although several studies have implicated the role of other signals in PHD activity,
the mechanisms underlying the link between them have not been characterized and one
possibility is that additional proteins and/or pathways may be involved in this crosstalk.
In addition, PHD-mediated HIFOL degradation in response to oxygen availability is very
prompt and specific suggesting possible contribution of other proteins in this process.
Indeed recent study reported that 08-9, the protein product of a widely expressed gene,
interacts with PHD2 and HlFla and promotes HIFla-hydroxylation, pVHL binding,
and proteosomal degradation (1). What is more, other upstream signals may exist that
can modulate PHD acitivty. Recent reports demonstrate that the Siah protein, a
homologue of the Drosophila seven-in-absentia protein, post-translationally regulates
PHD] and PHD3 levels and the peptidyl-prolyl cis/trans isomerase, FKBP38, is
involved in regulating PHD2 protein stability (2, 27). Creating a comprehensive library
of PHD-interacting proteins, therefore, is important to understanding PHD-mediated
oxygen sensing mechanism(s) and hypoxia signaling. In this study, we characterize a
map for the PHD2-protein interaction network (PHD2-PIN) using tandem affinity
purification (TAP) and liquid chromatography coupled mass spectrometric analysis
(LC-MS/MS). Using this approach, we did not identified any specific proteins
associated with PHD2 within the human fibrosarcomas, PH3MT cells, suggesting that

PHD2 might act alone in modulating HIFIOL activity under normoxia in this cells.

114

MATERIAL AND NIETHODS

Cell culture

Human fibroblasts PH3MT cells (a generous gift from Dr. J. Justin McCormick,
MSU., (24)) were cultured in OL-MEM (Mediatech Inc., Hemdon, VA) and Phoenix
ampho cells (a generous gift form Garry Nolan, Stanford Univ., (28)) were maintained
in DMEM (Mediatech Inc.), supplemented with 10 % fetal calf serum (HyClone,
Logan, UT), 100 unit/m1 penicillin, 100 ug/ml streptomycin, 2 mM L-glutamine and 1
mM sodium pyruvate. All supplements were purchased from Invitrogen (Carlsbad,
CA). Cells were incubated at 37 °C and 5 % CO2 (Precision, Winchester, VA).
Plasmid construction and retroviral infection

The cDNA for PHD2 (a generous gift form Dr. Steven McKnight, Univ. of Texas
Southwestern (5)) was inserted into pZomelN, a retroviral vector including sequence
for TAP-tag and puromycin resistance. A corresponding GFP-pZomelC was also
designed to monitor viral packaging and infection rate of the virus. It is also used as the
negative control for the TAP-tag protocol. Both viral constructs were sequenced-
verified and used to transfect Phoenix-ampho cells (a generous gift from Garry Nolan,

2000 via manufacturer’s instructions

Stanford Univ.(28)) utilizing Lipofectamine
(Invitrogen). Following a 24 hour incubation, the media was replaced with fresh growth
media and the cells were incubated for an additional 24 hours. Viral particle were

collected and purified by centrifugation (650g for 5mins), and filtered using a 0.45 um

membrane (Millipore, Billerica, MA). PH3MT cells (2x105) were plated on 60 mm

115

“a

 

tissue culture dishes 24 hours prior to infection. Prepared virus containing polybrene
(Sug/ml) (Sigma) were added to the cells and incubate for 24 hours. Media were
replaced and infected cells were selected using puromycin (0.4 ug/ml, USBiologicals,
Swampscott, MA).
Western blot analysis and quantitative Real Time -PCR analysis

Total proteins were prepared in lysis buffer (50 mM HEPES, pH7.5, 150 mM
NaCl, 1 mM EDTA, 2.5 mM EGTA, 0.1% Tween-20, 10% Glycerin, 1 mM
dithiothritol, 10 mM B-glycerophosphate, 1 mM NaF, 0.1 mM PMSF, 1 mM Na3VO4,
and protease inhibitor tablet (Roche, Indianapolis, IN)). Nuclear proteins were prepared
from cells lysed in Buffer A (lOmM Tris, pH 7.5, 1.5 mM MgCl2, 10 mM KCl, 1 mM
EDTA, 2 mM dithiothritol, 0.2 mM PMSF, 2 mM Na3VO4, and protease inhibitor tablet
(Roche)) and nuclei were collected by centrifugation (2000g, for 5min). Nuclei pellets
were then lysed in Buffer C (0.42 M KC], 20 mM Tris, pH7.5, 1.5 mM MgCl2, 20%
(vol/vol) glycerol, 1 mM EDTA, 0.1 mM dithiothritol, 0.2 mM PMSF, 2 mM Na3VO4,
and protease inhibitor tablet (Roche)). Antibodies used for Western blot analysis; rabbit
polyclonal anti-PHD2 (Novus, Littleton, CO), mouse monoclonal anti-HIF1a(Novus),
rabbit polyclonal anti-B-Actin (Sigma, St. Louis, MO), goat anti-rabbit and -mouse
secondaries (Sigma).

Total RNA was extracted form cells using TRIzol reagent (Invitrogen) according
to the manufacturer’s protocol. RNA concentration and integrity were analyzed using

UV spectrometry and denaturing gel electrophoresis. 1 pg total RNA was primed

116

with oligo(dT)13 and used for first-strand synthesis with the SuperScript First-Stranded
Synthesis System (Invitrogen) as described previously (31). Gene expression was
determined using quantitative real time-PCR (qRT-PCR), based on Sybr-Green
methodology (Applied Biosystems, Foster City, CA) as previously described (31).
Analysis was performed using the ABI PRISM 7000 Sequence Detection System
(Applied Biosystems). Primers used for qRT-PCR (5’ to 3’): VEGF:
tcctcacaccattgaaacca and gatcctgccctgtctctctg; GAPDH: cagcctcaagatcatcagca and
gtcttctgggtggcagtgat; HPRT: gaccagtcaacaggggacat and cctgaccaaggaaagcaaag.
Specificity of each primer set was verified by BLAST and dissociation curve analysis.
Tandem Affinity Purification (TAP)

TAP was performed as outlined in figure 3-1. Briefly, cells were washed three
times in ice cold PBS and manually collected from the culture plates in the presence of
lysis buffer (10 % glycerol, 50 mM Hepes, pH=8.0, 100 mM KCl, 2 mM EDTA, 0.1 %
NP-40, 2 mM DTT, protease inhibitors). Lysates were incubated for 30 min on ice
followed by two freeze/thaw cycles and supernatant was purified by centrifugation
(10,000g, for 20 min). IgG sepharose was added to cell lysates and incubated for 4
hours (4 °C, agitation). IgG sepharaose was collected by gravity sedimentation and
washed 3 times with cold lysis buffer and 3 times with TEV buffer (10 mM Hepes, pH
8.0, 150 mM NaCl, 0.1 % NP-40, 0.5 mM EDTA, 1 mM DTT). The clean beads were
then resuspended in TEV buffer containing 1 unit/ul of TEV and incubated for 4 hours

at 4 °C. The IgG-separose was separated by centrifugation (600g, for 1min) and the

117

 

 

Figure 3-1. Tandem Affinity Purification.

Cartoon representation of TAP protocol. The TAP-tag PHD2 is purified from total cell
lysate using IgG-sepharose. The beads are extensively washed and complex is cleaved
from the IgG beads by TEV protease. The released protein complex is then bound to
calmodulin-sepharose. Following washing and elution, proteins in the complex are
identified by SDS-PAGE and/or mass spectrometry.

118

 

 

PHDZ GFP

 

 

 

 

SOS-PAGE

 

 

Cleave and bind to
Calmodulin Sepharose

 

Calmodulln

Wash and
Elute wl EGTA

—

 

..Imlmll “J .Illl-
500 1an 1000

Mass Spectrometry

119

supernatant was moved to a clean tube. The IgG-sepharose beads were washed 2 times
with calmodulin binding buffer (10 mM B-mercaptoethanol, 10 mM Hepes, pH 8.0,
150 mM NaCl, 1 mM MgOAc, 1 mM imidazol, 0.1 % NP-40, 2 mM DTT) and the
supernatant was collected by centrifugation (600g for 1min) and combined with
original sample. Calmodulin-sepharose was washed 3 times with calmodulin binding
buffer and added to collected supernatant. CaCl2 was added to a final concentration of 2
mM and incubated for 90 min (4 °C, agitation). The calmodulin-sepharose was washed
3 times with calmodulin binding buffer. Associated proteins were finally eluted from
the calmodulin beads either by elution buffer (50 mM NH3HCO3, pH=8.0, 25 mM
EGTA) or SDS protein sample buffer (0.5 M Tris, pH=6.8, 4.4 % (WN) SDS, 20 %
Glycerol, 2% B-mercapto-ethanol, 0.2 % bromophenol blue).
Ti’ypsinization and Mass spectrometry

The samples being prepared for mass spectrometry were resuspended in 50 mM
di-ammonium phosphate (dAP), supplemented with 10 % acetonitrile (final pH :80).
The solution was supplemented with Tris (2-Carboxyethyl) Phosphine Hydrochloride
(TCEP-HCl) and heated to 95 °C for 10 min. The proteins were then incubated in the
presence of 15 mM iodoacetamide (1 hour, dark). The iodoacetamide was quenched by
the addition of dithiothreitol (final cons.= 5 mM) and the protein mixture was digested
with the addition of freshly diluted trypsin (30 ng, 37 °C for 18 hrs). An equal volume
of formic acid (5 %) was added to the final peptide mixture and sonicated for 10 mins.

After the digestion has been completed, the peptides were desalted on a 2 cm x

120

100 um peptide trap packed with Micron Magic C18 AQ packing material. The bound
peptides were flushed onto a 10 c m x 75 mm New Objecteives Picofrit colum packed
with a Micro Magic C18 AQ packed material and eluted over 50 mins with a gradient
of 5 % B to 70 % B in 45 mins (Buffer A: 0.1 % formic acid, Buffer B: 95 %
Acetonitrile 0.1 % formic acid) into a Micromass Qtof Ultima API mass spectrometry
with a flow rate of 250 nl/min. The top three ions in each survey scan was subjected to
automatic low energy collision induced dissociation (CID) and the resulting
uninterpreted MS/MS spectra was searched against an appropriate database using the
XlTandem searching algorithm. Identifications were considered positive if 2 peptides
per protein are identified with a significant XlTandem Score (log(e).-2). If only one
peptide is identified, they were subjected to a Mascot cutoff of less than 5 %
probability that the match could be considered chance identification. The data were
filtered by comparison with the TAP-GFP data set. Those interactions that occur in both
data sets were marked as non-specific and removed from further consideration. The
screening processes also remove the most common background proteins, such as

tubulin, actins and keratins.

121

RESULTS

Tandem affinity purification (TAP) was applied as a methodology to characterize
the PHD2 protein interaction network. Stable cells expressing TAP-tagged version of
PHD2 were created through retroviral infection of PHDZ-T AP in human fibrosarcomas,
PH3MT cells, tPHD2-3MT cells. GFP-TAP expressing PH3MT cells, tGFP—3MT cells,
were also generated to monitor infection (Fig. 3-2A) and to act as a negative control for
the TAP procedure. The expression of PHD2- and GFP—TAP protein was verified
using Western blot analysis with a PHDZ-specific antibody (Fig. 3-2B). TAP-tag
proteins are detectable with all antibodies including pre-immune serum because of the
presence of the protein A portion at the end of tag. The tPHD2-3MT and tGFP-3MT
cells showed a strong band at the appropriate molecular weight of TAP-tagged PHD2
and GFP respectively. In addition, there was a band migrating at the correct molecular
weight for endogenous PHDZ across all three cell lines. The expressions of TAP-tagged
PHD2 and GFP in tPHD2-3MT cells and tGFP-3MT cells were confirmed by
additional Western blot analysis using pre-immune serum (Fig. 3-2B). The pre-immune
serum showed a band at the same molecular weight of TAP-tagged PHD2 and GFP as
shown in Western blotting with PHD2-specific antibody and it did not show any band
in non—infected PH3MT cells. These results show that the newly created tPHD2-3MT
and tGFP-3MT cells express significant amount of their respective TAP-tagged proteins.

PHD is a well known controller of HIFIOL stability and consequently HIF1-

mediated gene expression. To test whether ectopic expression of PHD2 has a functional

122

Ea

Figure 3-2. Characterization of tPHD2-3MT and tGFP-3MT cells

(A) PH3MT cells were infected with a retroviral construct that expresses the green
fluorescent protein (GFP). GFP expressions were observed via fluorescent microscopy.
(B) The expression of PHD2-TAP and GFP-TAP proteins were characterized in newly
created tPHD2-3MT and tGFP-3MT cells by Western blot analysis using PHD2-
specific antibodies or pre-immune. The parental PH3MT cell line was included as
control. (C) HIFIOL protein levels were analyzed in tPHD2-3MT and tGFP-3MT
cells exposed to normoxia (20% O2) or hypoxia (1% 02) for 5 hrs by Western blot
analysis. To verify equal loading, the blot was stripped and reprobed with a B-Actin
antibody. (D) mRNA levels of VEGF and GAPDH were determined in tPHD2-3MT
and tGFP-3MT cells using qRT-PCR. Cells were exposed to normoxia (20% O2, white
bar,) or hypoxia (1% 02, black bar) for 16 hr. (n=8, * p<0.05, ** p<0.01).

123

 

GFP

Cells

 

IB: PHD2

124

I3: pre-immune

<- PHD2-TAP
<- GFP-TAP

 

 

4- HIF1a
Normoxia '
-.¢ +- B-Actin
- <- HIF1a
Hypoxia ' "
.--- +B-Actin
EINormoxia
4o IHypoxia
m
E
u. o
0 _r
in
> g
m
E

 

GAPDH
mRNA Levels

 

125

consequence on HIF1 signaling, we first determined HIFIOL protein levels (Fig. 3-2C).
As expected, HIFICX. was not detectable under normoxia and induced by hypoxia in
control cells, PH3MT and tGFP-3MT, while hypoxia-induced HIFIOL accumulation was
inhibited in the tPHD2-3MT cells. In addition, over-expression of PHD2 had a
functional consequence on HIF1-mediated transcriptional of VEGF and GAPDH, two
classic hypoxia-regulated genes (Fig. 3-2D). The hypoxia-induced expression of these
genes was inhibited in tPHD2-3MT cells compared with that in control cells, PH3MT
and tGFP-3MT.

To characterize a map for PHD2-protein interaction network, we performed
tandem affinity purification (TAP) on each of the newly created cell strains. TAP-
tagging was initially developed for use in proteomic analysis of Saccharomyces
cerevisiae (28). The TAP-tag consists of a Staphyloccoccus aureus protein A and a
calmodulin binding peptide (CBP), separated by Tobacco Etch Virus (TEV) protease
site (Fig. 3-1). The TAP strategy has several advantages over conventional purification
schemes. It has permitted the rapid identification of protein interacting network with
less background contaminants. This approach can also be applied to link proteins of
unknown function to known pathways and gain detailed information regarding changes
in composition of protein complex that occur in different stimuli, e.g., various oxygen
concentrations. To identify the PHDZ-PIN, TAP analysis was performed with tPHD2-
3MT cells. As a control, tGFP-3MT cells were used to exclude non-specific proteins.

Each purification step was verified using Western blot analysis with a PHD2-specific

126

I J I...

antibody (Fig. 3-3). GFP-TAP protein can be seen by PHD2-specific antibody due to
the presence of the Protein A portion of the TAP-tag. After TEV protease cleavage,
tPHDZ-3MT cells showed a band (PHD2-CBP) migrating lower than PHD2-TAP and
tGFP-3MT cells did not show band due to the loss of Protein A part of the tag (Fig. 3-3).

To profile protein complex associated with PHD2, final elutes were separated
using Bis-Tris gradient (4-20%) gel electrophoresis and visualized by silver staining
(Fig 3-4A). The PHD2 and GFP band can be seen in tPHD2-3MT and tGFP-3MT cells,
respectively, and were confirmed by mass spectrometry. The tPHD2-3MT cells,
however, did not showed any significant difference on the profile of protein bands
compared with those in tGFP-3MT cells (Fig 3-4A). The individual bands were
analyzed by mass spectrometry and the results suggested that most were non-specific
contaminants (Table 3-1). In an attempt to limit the amount of non-specific
contamination, associated proteins were also eluted under mild condition with
optimized elution buffer and results showed no-specific protein band (Fig 3-4B). Since
gel electrophoresis and silver staining have certain limitations, we next attempted an
on-bead trypsinization protocol for PHD2 and GFP and digests were directly analyzed
by MS. Again, no specific, reproducible proteins were identified (Table 3-2 and 3-3).
Finally, we also analyzed the supernatant from the post-TEV protease treatment to
determine if we might be losing specific proteins with the Calmodulin sepharose. The

results from this fraction were similar to the previous attempts (Table 3-4).

127

 

 

 

 

Figure 3-3. Verification of Tandem Affinitu Purification (TAP).

Individual steps of TAP protocol were verified using Western blot analysis with PHDZ-
specific antibody.

128

mung. :3 $mp..mo& 283 on.

 

«2:3 as". 252. n_mo cam .23.. n5 mean:
“593:; t ‘ i '
“25.3 + l' ' . .. «9.:
m<....~o:n_ .v - ,

2 2 2 2 2 I 2 I I 2
an» 0% «web e042 . «ueoev a we 0% a we 0%
§ \ \ b a
a. c... a. a. «a. a... a. a... a. a.

129

Figure 3-4. Protein profile of TAP-tagged purified PHD2

(A) PHD2-associated proteins were isolated via Tandem Affinity Purification.
Calmodulin beads were eluted with 2xSDS protein sample buffer and proteins were
separated with Bis-Tris gradient gel (4-20%) and stained with silver. (B) PHD2-protein
complex were purified using TAP protocol and eluted with elution buffer.

130

131

N
D
I
n.
0

«— GFP

 

Table 3-1. Identification of proteins within individual band in Figure 3-4A using

mass-spectrometry.

 

Number in a silver
stained gel (Fig. 3-4A)

Protein Identified using Mass-Spectrometry

 

gi_34527698 unnamed protein product (Homo sapiens)
gi_34534305 unnamed protein product (Homo sapiens)
gi_34526220 unnamed protein product (Homo sapiens)
gi_50949554 hypothetical protein (Homo sapiens)
gi_34527453 unnamed protein product (Homo sapiens)
gi_40795897 homerin precursor (Homo sapiens)
gi_18031805 estrogen-induced tag 6 (Homo sapiens)

 

gi_28317 unnamed protein product (Homo sapiens)
gi_6470150 BIP protein (Homo sapiens)

 

gi_28317 unnamed protein product (Homo sapiens)
gi_908801 keratin type II

gi_21961227 keratin 6B (Homo sapiens)

gi_307086 keratin-10

gi_18999435 KRTS protein (Homo sapiens)

 

gi_55665459 tublin beta 5 (rattus norvegious)
gi_ll935049 keratin l (Homo sapiens)

gi_18999435 KRTS protein (Homo sapiens)
gi_10433717 unnamed protein product (Homo sapiens)
gi_34527453 unnamed protein product (Homo sapiens)

 

gi_55665459 egl nine homolog 1 (Homo sapiens) “PHD2”
gi_1346343 keratin type II cytoskeletal 1 (cytokeratin 1)
gi_435476 cytokeratin 9 (Homo sapiens)

gi_28317 unnamed protein product (Homo sapiens)

 

gi_l373325 Enhanced Green Fluorescent protein “GF P”
gi_14595132 Raichu404X (Homo sapiens)
gi_11935049 keratin l (Homo sapiens)

 

132

Table 3-2. Protein profile of TAP-tagged purified PHD2 (on-bead trypsinizing)

 

Number

Description

 

gi_825635

gi_66360504
gi_61680528
gi_16974825

gi_31092
gi_55665459
gi_56090271
gi_229585
gi_l6554039
gi_1297274
gi_l6554039
gi_1143492
gi_21669491
gi_56789800
gi_11935049
gi_67664801
gi_7648673
gi_34069
gi_46254043
gi_106586

calmodulin (Homo Sapiens)

Chain T, Crystal Structure of Anthrax Bdena actor (Ef) In
Complex with Calmodulin

Chain A, Trapped Intermediate of Calmodulin

Chain A, Solution Structure of Calcium-Calmodulin N-Terrninal
Domain

unnamed protein product (Homo Sapiens)

egl nine homolog 1 (Homo Sapiens) “PHD2”

ribosomal protein 520 (Rattus norvegious)

Ig A1 bur

unnamed protein product (Homo Sapiens)

beta-tubulin (Hopmo Sapiens)

unnamed protein product (Homo Sapiens)

BIP (Homo Sapiens)

immunoglobulin lambda light chain TIJ region (Homo Sapiens)
MOC27165 protein (Homo Sapiens)

Keratin 1 (Homo Sapiens)

hypothetical protein bcen2424DRAFT_3653

Voltage-gated potassium channel Ev4.2 (Homo Sapiens)
unnamed protein product (Homo Sapiens)

Immunoglobulin heavy chain (Homo Sapiens)

Ig kappa chain V-III (KAU cold agglutinin) - human

 

133

Table 3-3. Protein profile of TAP-tagged purified GFP (on-bead trypsinizing)

 

Number

Description

 

gi_825635
gi_16974825

gi_61680528
gi_1143492

gi_66360504

gi_62896605

gi_l 1935049
gi_1297274
gi_20809886
gi_l 808 87 19
gi_5609027 l
gi_229585
gi_181402
gi_42744612
gi_16554039

gi_l373325

gi_3 1874190
gi_386850
gi_3 1074633

gi_230161

gi_15193213

gi_90883
gi_6981490

calmodulin (Homo Sapiens)

Chain A, Solution Structure of Calcium-Calmodulin

N -Terminal Domain

Chain A, Trapped Intermediate of Calmodulin

BIP (Homo Sapiens)

Chain T, Crystal Structure of Anthrax Bdena actor (Ef)
In Complex with Calmodulin

eukaryotic translation elongation factor lot 1 variant
(Homo Sapiens)

keratin 1 (Homo Sapiens)

beta tubulin (Homo Sapiens)

tubulin, beta, 2 (Homo Sapiens)

tubulin beta polypeptide (Homo Sapiens)

ribosomal protein 820 (Rattus norvegious)

Ig A1 Bur

epidermal cytokeratin 2 (Hopmo Sapiens)

unkown (protein for MOC:70893) (Homo Sapiens)
unnamed protein product (Homo Sapiens)

Enhanced Green Fluorescent Protein

(Cloning vector pEGFP-N3) “GFP”
hypothetical protein (Homo Sapiens)

keratin K5

keratin 6L (Homo Sapiens)

Chain X, immunoglobulin heterologous light chain diner
(MCG-WEIR Hybrid)

anti-hepatitis B surface antigen immunoglobulin heavy chain
(Homo Sapiens)

keratin type II

ribosomal protein 829 (Rattus norvegicus)

 

134

Table 34. Protein profile of TAP-tagged purified PI-ID2 (TEV sup)

 

 

Number Description
gi_825635 calmodulin (Homo Sapiens)
gi_66360504 Chain T, Anthrax Bdena actor (Ef) In Complex with Calmodulin

gi_61680528
gi__ 1 6974825
gi_62896605
gi_55665459
gi_229601
gi_l 34634
gi_46254043
gi_345 26073
gi_l 143492
gi_5609027 1
gi_37492
gi_7648673
gi_61680025
gi_67664801
gi_37777889
gi_345 26391
gi-l 803 l 805
gi_6273998 l
gi_34537233
gi_34l91054
gi_509798
gi_37694545
gi_75 13045
gi_30749422
gi_639910
gi_l 1935049
gi_345 1240
gi_67939669
gi_67752988
gi_67763452
gi_l 8204970

Chain A, Trapped Intermediate of Calmodulin

Chain A, Solution Structure of Calcium-Calmodulin N-Terminal Domain
Eukaryotic translation elongation factor la variant (Homo Sapiens)

egl nine homolog l (Homo Sapiens) “PHD2”

1g 61 N N ie

Keratin, type II cytoskeletal l (cytokeratin 1) (K1) (CKl)
Immunoglobulin heavy chain (Homo Sapiens)

unnamed protein product (Homo Sapiens)

BIP (Homo Sapiens)

ribosomal protein 820 (Rattus norvegicus)

Alpha-tubulin (Homo Sapiens)

Voltage-gated potassium channel KV 4.2 (Homo Sapiens)

Chain I, HIV-l neutralizing human rab 4e 10 in complex with A l3-residue peptide
Hypothetical protein BCEN2424DRART_3653

Immunoglobulin heavy chain variable region (Homo Sapiens)

unnamed protein product (Homo Sapiens)

Estrogen-induced tag 6 (Homo Sapiens)

DKFIP547r 072 protein (Homo Sapiens)

Unnamed protein (Homo Sapiens)

Hypothetical proteinFlJ 23878 (Hopmo Sapiens)

IgG heavy chain, variable region. rheumatoid factor (Homo Sapiens)
Immunoglobulin heavy chain variable region (Homo Sapiens)
Hypothetical protein KIAA0627—human (fragment)

ChainA, rbe fibrillin-l cbegt12-l3 fair of ca2+ binding epidermal growth factor-like
Heat shock protein

Keratin 1 (Homo Sapiens)

2’-2’ oligoadenylate synthase (p59OAC) (Homo Sapiens)

Sulfate transporterantisigma-factor antagonist Srac:sulfate transporter
COG0583:Transcriptiona| regulator (surkholderia pseudonallei Pasteur)
COG37952uncharacterized protein conserved in bacteria

Tubulin tyrosine ligase-like family member 4 (Homo Sapiens)

 

135

DISCUSSION

The PHDs regulate HIF-mediated transcriptional response by promoting HlFla
degradation through the ubiquitin-proteasome pathway. PHDs belong to the iron and
OL-ketoglutarate dependent oxygenase family and require oxygen, iron, a-ketoglutarate,
and ascorbate for their proper activity. In mammals, three PHD isoforms, PHD1, 2,
and 3, have been identified and PHD2 is recognized as the main functional isoform in
regulating hypoxia signaling in humans. Recently, several studies have suggested a
role for other proteins and/or signaling pathways in influencing PHD3 activity (1, 2, 27).
In addition, it has been implicated that mitochondria generated reactive oxygen species
(ROS) and several metabolic intermediates contribute to PHD activity (8, 11, 19, 21,
29). However, the mechanisms underlying these links have not been characterized
and one possibility is that additional proteins and/or pathways may be involved in this
crosstalk. In this study, we characterize PHD2-protein interaction network on the
composition of protein complex using tandem affinity purification (TAP) and mass
spectrometry. Our result suggested that PHD2 might have no specific interaction with
other proteins under normoxia and that PHD2 might act to coordinate the various input
signals without the aid of accessory factors.

Previously, it has been reported that two proteins, 08-9 (3) and FKBP38 (2),
interact with PHD2 and play a role in regulating its activity. However, both studies
were performed using a yeast-two hybrid under certain condition (1, 2). PHD2-

mediated oxygen sensing mechanism is susceptible to change of oxygen levels and

136

should be controlled tightly not to cause any adverse effects mediated by hypoxia
signaling. It is possible that the interaction between PHD2 and other proteins might
change very rapidly within cells under normoxia to control the influence of the various
input signals on PHD activity. This type of rapid turnover would make it difficult to
identify the PHD2-PIN in vivo. It is also possible that PHD2 might interact with other
proteins and/or signaling pathways in a cell type specific manner. The normal oxygen
tension varies in cells within and among different tissues (6, 16) and PHD2-mediated
regulation of HIF1 signaling may need more complexity in certain cell types than
others. This possibility suggests the requirement of analyzing PHD2-PIN in other cell
lines. In addition, PHDs hydroxylate HIF 10L only under normoxia and are inhibited by
hypoxia leading to HIFIOL stabilization and activation. Although any specific protein
associated with PHD2 have not been identified in human fibrosarcomas, PH3MT cells,
under normoxia, it is possible that there may be changes in the compostion of the
PHD2-PIN following exposure to different stimuli, e.g., hypoxia. It is therefore
necessary to investigate the changes in the PHD2-PIN under various stresses, such as
hypoxia, oxidative stress and TCA cycle inhibition. Characterizing the PHD2-PIN in
various tissues under different conditions will provide insight on mammalian oxygen

sensing mechanism and will also help target this pathway for therapeutic agent.

137

REFERENCES

1. Back, J. H., P. C. Mahon, J. Oh, B. Kelly, B. Krishnamachary, M. Pearson, D.
A. Chan, A. J. Giaccia, and G L. Semenza. 2005. 08-9 interacts with hypoxia-
inducible factor 1 alpha and prolyl hydroxylase to promote oxygen-dependent
degradation of HIFIalpha. Mol. Cell. 17:503-512.

2. Barth, S., J. Nasper, P. A. Hasgall, R. Wirthner, K. J. Nytko, F. Edlich, D. M.
Katschinski, D. P. Stiehl, R. H. Wenger, and G Camenisch. 2007. The peptidy-
prolyl cis.trans isomerase FKBP38 determines HIF prolyl-4-hydroxylase PHDZ
protein stability. Mol. Cell. Biol. Mar (Epub ahead of print).

3. Beck, 1., S. Ramirez, R. Weinmann, and J. Caro. 1991. Enhancer element at the
3'-flanking region controls transcriptional response to hypoxia in the human
erythropoietin gene. J. Biol. Chem. 6:15563-15566.

4. Berra, E., E. Benizri, A. Ginouves, V. Volmat, D. Roux, and J. Pouyssegur.
2003. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state
levels of HIF-lalpha in normoxia. EMBO J 22:4082-4090.

5. Bruick, R. K., and S. L. McKnight. 2001. A conserved family of prolyl-4-
hydroxylases that modify HIF. Science 294: 1337- 1340.

6. Coelho, A., A. Ferraz, R. Neto, A. Souza, and E. Ferraz. 2000.
Subdiaphragmatic venous stasis and tissular hypoperfusion as source of metabolic
acidosis during passive portal-jugular and cacal-jugular bypasses in dogs. Acta Cir
Bras 15.

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

8. Guzy, R. D., B. Hoyos, E. Robin, H. Chen, L. Liu, K. D. Mansfield, M. C.
Simon, U. Hammerling, and P. T. Schumacker. 2005. Mitochondrial complex
III is required for hypoxia-induced ROS production and oxygen sensing. Cell
Metab 1:401-408.

9. Hirsila, M., P. Koivunen, V. Gunzler, K. I. Kivirikko, and J. Myllyharju. 2003.
Characterization of the human prolyl 4-hydroxylase that modify the hypoxia-

138

10.

ll.

12.

13.

14

15.

16.

17.

18.

inducible factor. J Biol Chem 278:30772-30780.

Hirsila, M., P. Koivunen, V. Gunzler, K. I. Kivirikko, and J. Myllyharju. 2003.
Characterization of the human prolyl 4-hydroxylases that modify the hypoxia-
inducible factor. Proc Natl Acad Sci USA 278:30772—30780.

Isaacs, J. S., Y. J. Jung, D. R. Mole, S. Lee, C. Torres-Cabala, Y. L. Chung, M.
Merino, J. Trepel, B. Zbar, and J. Toro. 2005. HIF overexpression correlates
with biallelic loss of fumarate hydratase in renal cancer: Novel role of fumarate in
regulation of HIF stability. Cancer Cell 8: 143-153.

Ivan, M., T. Haberberger, D. C. Gervasi, K. S. Michelson, V. Gunzler, K.
Kondo, H. Yang, I. Sorokina, R. C. Conaway, and J. W. Conaway. 2002.
Biochemical purification and pharmacological inhibition of a mammalian prolyl
hydroxylase acting on hypoxia—inducible factor. Proc Natl Acad Sci USA

99: 13459-13464.

Ivan, M., K. Kondo, H. Yang, W. Kim, J. Valiando, M. Ohh, A. Salic, J. M.
Asara, W. S. Lane, and W. G Kaelin. 2001. HIF1 alpha targeted for VHL-
mediated destruction by proline hydroxylation: Implications for O2 sensing.
Science 292:464-468.

. Jaakkola, P., D. R. Mole, Y.-M. Tian, M. 1. Wilson, J. Gielbert, S. J. Gaskell, A.

Von Kriegsheim, H. F. Hebestreit, M. Mukherji, and C. J. Schofield. 2001.
Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by 02-
regulated prolyl hydroxylation. Science 292:468-472.

Lee, K., S. O'Reilly, M. Kiupel, J. J. McCormick, and J. J. LaPres. 2007. The
biphasic role of HIF prolyl hydroxylase PHD2 in modulating tumor-forming
potential. Mol. Cell. Biol. In review.

Lee, K., R. A. Roth, and J. J. LaPres. 2007. Hypoxia, drug therapy and toxicity.
Pharm & Ther 113:229-246.

Lee, S., E. Nakamura, H. Yang, W. Wei, M. S. Linggi, M. P. Sajan, R. V.
Farese, R. S. Freeman, B. D. Carter, W. G Kaelin, and S. Schlisio. 2005.
Neuronal apoptosis linked to EGLN3 prolyl hydroxylase asn familial
pheochromocytoma genes: developmental culling and cancer. Cancer Res. 8:155-
167.

Lieb, M. E., K. Menzies, M. C. Moschella, R. Ni, and M. B. Taubman. 2002.

EGLN genes have distinct patterns of mRNA expression and regulation. Biochem
Cell Biol 80:421-426.

139

19.

20.

21.

22.

23.

24.

25.

26.

27

Lu, H., C. L. Dalgard, A. Mohyeldin, T. McFate, A. S. Tait, and A. Verma.
2005. Reversible Inactivation of HIF-1 Prolyl Hydroxylases Allows Cell

Metabolism to Control Basal HIF-10L. J Biol Chem 280:41928-41939.

Lu, H., R. A. Forbes, and A. Verma. 2002. Hypoxia-inducible factor 1 activation
by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J Biol
Chem 277:23111-23115.

Mackenzie, E. D., M. A. Selak, D. A. Tennant, L. J. Payne, S. Crosby, C. M.
Frederiksen, D. G Watson, and E. Gottlieb. 2007 . Cell-perrneating alpha-
Ketoglutarare derivatives alleviate pseudohypoxia in succinate dehydrogenase-
deficient cells. Mol Cell Biol 21:3287-3289.

Maxwell, P. H., G U. Dach, J. M. Gleadle, L. G Nicholls, A. L. Harris, I. J.
Stratford, O. Hankinson, C. W. Pugh, and P. J. Ratcliffe. 1997. Hypoxia-
inducible factor-1 modulates gene expression in solid tumor and influence both
angiogenesis and tumor growth. Proc Natl Acad Sci USA 94:8104-8109.

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. 1999.
The tumor suppressor protein VHL target hypoxia-inducible factors for oxygen-
dependent proteolysis. Nature 399:271-275.

McCormick, J., S. O'Reilly, and V. Maher. 2000. Analysis of the process of
malignant transformation of human fibroblasts in culture by ionizing radiation and
other carcinogens. In Radiation Research: Volume 2 Proceedings of the Eleventh
International Congress of Radiation Research. Ireland, D., (ed) Vol. 2, pp 564-
567. Allen Press.

McDonough, M. A., V. Li, R. Flashman, C. Chowdhury, B. M. Mohr, J.
Lienard, N. J. Zondlo, N. J. Oldham, I. J. Clifton, J. Lewis, L. A. McNeill, R.
J. Kurzeja, K. S. Hewitson, E. Yang, S. Jordan, R. S. Syed, and C. J. Schofield.
2006. Cellular oxygen sensing: crystal structure of hypoxia-inducible factor prolyl
hydroxylase (PHDZ). Proc. Natl Acad. Sci. USA 103:9814-9819.

Metzen, E., U. B-Pfannschmidt, P. Stengel, J. H. Marxsen, I. Stolze, M.
Klinger, W. Q. Huang, C. Wotzlaw, T. H-Burgel, W. J elkmann, H. Acker, and
J. Fandrey. 2002. Intracellular localization of human HIFlalpha hydroxylase:
implication for oxygen sensing. J. Cell Sci. 116:1319-1326

. Nakayama, K., I. J. F rew, M. Hangensen, M. Skals, H. Habelhah, A. Bhoumik,

140

28.

29.

30.

31.

T. Kadoya, H. E-Bromage, P. Tempst, P. B. Frappell, D. D. Bowtell, and Z.
Romai. 2004. Siah2 regulates stability of prolyl-hydroxylases, controls HIF1
abundance, and modulates physiological responses to hypoxia. Cell 117:941-952.

Rigaut, G, A. Shevchenko, B. Rutz, M. Wilm, M. Mann, and B. Seraphin.
1999. A generic protein purification method for protein complex characterization
and proteome exploration. Nat Biotechnol 17: 1030-1032.

Selak, M. A., S. M. Armour, E. D. MacKenzie, H. Boulahbel, D. G Watson, K.
D. Mansfield, Y. Pan, M. C. Simon, C. B. Thompson, and E. Gottlieb. 2005.
Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-[alpha]
prolyl hydroxylase. Cancer Cell 7:77-85.

Semenza, G L., and G L. Wang. 1992. A nuclear factor induced by hypoxia via
de novo protein synthesis binds to the human erythropoietin gene enhancer at a
site required for transcriptional activation. Mol Cell Biol 12:5447-5454.

Vengellur, A., B. G Woods, H. E. Ryan, R. S. Johnson, and J. J. LaPres. 2003.
Gene expression profiling of the hypoxia signaling pathway in hypoxia inducible
factor la null mouse embryonic fibroblasts. Gene Expression 11:181-197.

141

CHAPTER 4

SUNINIARY

The ability to adapt to a hypoxic environment is critical to a cancer cell’s survival
and ability to form tumors. Central to this adaptation is the HIF1-mediated signaling
system. HIF1 regulates the expression of a wide variety of genes essential to the
adaptive response, including cell proliferation, angiogenesis, anaerobic metabolism,
migration and many others (6, 16, 17, 19). During tumor development a cell is capable
of modulating HIF1 activity to cope with transient changes in oxygen availability. This
adaptability is also important for attempting to re-establish normoxia for normal cells
within an organ during hypoxic stress. The adaptive response, however, is not always
capable of addressing the hypoxic-induced cellular imbalance. In this case, the cells
begin modulating various cell death pathway presumably as a mechanism to eliminate
irrecoverably stressed cells and maintaining the tissue as a whole (8, 18). Removal of
cells under severe hypoxic stress by programmed cell death could increase the chance
of survival for neighboring cells by increasing nutrients and oxygen supply and
maintaining appropriate tissue architecture. This cell death pathway involves HIF1-

dependent transcription of pro-apoptotic B-cell chronic lymphocytic leukemia (CLL)

142

/lymphoma (BCL) family members, such as BCL2/ adenovirus E28 19 kDa interacting

protein 3 (BNIP3) and NIX (9, 15, 20). The activation of HIF] is also able to contribute

to the p53-mediated cell cycle arrest and cell death (1, 5). This increased “suicide”

response seems in opposition to the adaptive response described above. The HIF1

signaling cascade, therefore, can promote either cell survival through adaptation or cell

death. One major unanswered question concerning HIF1 signaling is how this balance

between adaptation and cell death is established and maintained. How does a cell know

when it should survive through adaptive measures or when it needs to die because it

cannot cope with the hypoxic environment? The intracellular regulatory system that

makes this decision probably involves numerous factors, including oxygen levels of the

cell, metabolic state and levels of key metabolites, growth factor, and the endogenous

activity of the PHDs.

In this study, we have demonstrated that PHD activity can serve as a master

regulator of a cell’s fate under hypoxic stress by the same mechanism that it uses to

establish normoxia (Fig. 4-1). For example, under normal oxygen concentration, PHDs

enzymes are at maintenance levels, that is, their levels are maintained at those required

143

 

 

 

Figure 4-1. Relationship between PHD activity and HIF1 target gene expression.

At normal oxygen tensions, the PHDs are capable of initiating the degradation of
HIFla ad expression of HIF1 target genes invovlved in glycolysis, angiogenesis, and
apoptosis are maintained at low levels (top panels). As the oxygen concentration
becomes moderately decreased, PHD activity is diminished for lack of available
substrate and some HIF1 accumulates and drives the expression of target genes.
These genes include angiogenic factors and glycolytic enzymes, levels of which will
be recalibrated to the new metabolic conditions and thus encoding a new “normoxia”
(middle panels). Finally, under severe hypoxia, PHD activity is incapable of
compensating for the loss of oxygen and HIF1 signaling becomes hyper-activated,

leading to expression of pro-apoptotic factor and cytotoxicity (bottom panels).

144

33039. 0.330% 3

9:23. teammconEoo on

«cacao combs 292.3...
:. 83.0% 093

womcoawo. 9.3m?“ >3 .8
teammcoQEoo m. 20.9.3
com>xo :. 33.6% =me

m.o>o. .32. «a toEBENE
2m meson 393 E... 05
c228 .5998 .950: Love:

me:

«0.. 3.3.30

‘Ill

 

 

I ‘
...OOOIIOI

  

3x25.
. 395.05
..0 9.90m

    

 

 

 

 

e
I ‘IIo-IQOIIOIII

 

 

5.8%: 2.:

II\

 

 

 

 

 

 

 

 

o.como.mc< I I
ozofioa< I
0_U>_°U>—MU I I I I I I

 

 

 

 

 

 

mic—Ea:

 

auMzua CIHd am :0 Minnov

145

for normal cellular function and oxygen-sensing capacity (Fig. 4-1, top panel). When
the cell is exposed to moderate hypoxia, the activity of PHDs is slightly decreased and
consequently induced HIF1 signaling provide adaptive responses such as increased
glycolysis and agiogenesis to tumor cells to cope with the detrimental environment (Fig.
4-1, middle panel). This adaptive response can also promote normal cells to gain
growth advantages and become malignantly transformed. It is possible that cell death
factors are upregulated in the short term; however, once the HIF1 activity is increased
in the cell and the energy balance is restored, these genes are turned off and cell death
will be avoided. It is likely that these genes are not expressed under moderate hypoxia
from which the cell can recover. Under severe hypoxic stress, HIF1-mediated adaptive
responses are unable to compensate for the loss in available oxygen (Fig. 4-1, bottom
panel). At these oxygen levels, PHD is inhibited and this leads to the HIFla-mediated
pro-death response becoming dominant and this unmitigated expression leads to cell
death. These observations suggested that tumor development is dependent upon its
ability to induce intracellular adaptive response to hypoxia without supporting pro-cell
death mechanisms and PI-IDs play an essential role in setting up the boundary between

cell survival and cell-death.

146

Targeting HIF1 signaling for cancer therapy, therefore, must be done with the
appropriate concern as to the impact these new drugs might have on the HIF1-regulated
balance between cell and tissue. survival and programmed cell death. The research
presented here suggests that the boundary between too much and not enough HIF1
activity is an important determinant of our ability to cope with pathological conditions.
Moreover, this boundary can shift a cell towards a transformed phenotype or drive a
transformed cell towards cell death. Exploiting HIFl signaling for therapeutics will
require an understanding of how this balance will be shifted in the various tissues and
cells and under various environmental conditions. For example, increases in HIF1
activity might promote cellular transformation in certain cells that have amassed the
cellular mutation necessary to progress toward complete transformation, i.e. MSU 1.1.,
but a similar increase in another cell might promote cell death, i.e. PH3MT.
Alternatively, new therapeutics that completely shut down PHD activity will have
detrimental consequence on normal cellular homeostasis by promoting an uncontrolled
pro-death response. ldeally, any drug targeting this pathway should be cell type specific
and capable of manipulating PHD activity within this target cell with limited chance of

driving other cells towards transformation. Our study implies that cells will have a

147

A

 

 

biphasic response to agents manipulating PHD/HIF1 signaling and if this is the case, it

will create difficulties when targeting this pathway for cancer therapeutics. A complete

understanding of normal oxygen levels, endogenous HIF1 signaling, metabolic flux,

and PHD activity in a specific tissue will be critical to our ability to manipulate the

system in a meaningful therapeutic sense.

In this study, we also characterized the PHD2 protein interaction network

(PHD2-PIN) to understand the possible influence of other proteins and/or signaling

pathway on PHD2 activity. The responses to hypoxia include various cellular

mechanisms, for example respiratory chain and TCA cycle in mitochondria and

anaerobic metabolism (10, 11, 13, 14). Recently it has been demonstrated that these

responses can also affect HIFl signaling through regulating PHDs activity. In addition,

a link between mitochondria generated reactive oxygen species (ROS) and PHD

activity has been established. Although several studies have implicated the role of other

signaling pathways in PHD activity, the mechanisms underlying the link between them

have not been characterized and one possibility is that additional proteins and/or

pathways may be involved in this crosstalk. Characterizing the precise mechanism and

148

Al

 

regulation of these enzymes, therefore, is essential to understand physiological and

pathological responses to hypoxia. Herein, we attempted to characterize a map for the

PHDZ-PIN using tandem affinity purification (TAP) and liquid chromatography

coupled mass spectrometric analysis (LC-MS/MS). Our result suggested that PHD2

might have no specific interaction with other proteins in human fibrosarcomas under

normoxia and that PHD2 might act to coordinate the various input signals without the

aid of accessory factors. Although any specific protein associated with PHDZ have not

been identified in PH3MT cells under normoxia, it is possible that there may be

changes in the PHD2-PIN following exposure to different stimuli in a cell type specific

manner. Characterizing the PHD2-PIN in various cell types and under different

conditions, such as hypoxia, oxidative stress and TCA cycle inhibition, will provide

insight on mammalian oxygen sensing mechanism for HIF1 signaling and will also

help target this pathway with therapeutic agents for certain diseases that have hypoxia

as a major component of their pathophysiological progression, such as cancer.

149

REFERENCES

1. An, W. G, M. Kanekal, M. C. Simon, E. Maltepe, M. V. Blagosklonny, and L.
M. Neckers. 1998. Stabilization of wild-type p53 by hypoxia-inducible factor
lalpha. Nature 392:405-408.

2. Back, J. H., P. C. Mahon, J. Oh, B. Kelly, B. Krishnamachary, M. Pearson, D.
A. Chan, A. J. Giaccia, and G L. Semenza. 2005. 08-9 interacts with hypoxia-
inducible factor 1 alpha and prolyl hydroxylase to promote oxygen-dependent
degradation of HIFIalpha. Mol. Cell. 17:503-512.

3. Barth, S., J. Nasper, P. A. Hasgall, R. Wirthner, K. J. Nytko, F. Edlich, D. M.
Katschinski, D. P. Stiehl, R. H. Wenger, and G Camenisch. 2007. The peptidy-
prolyl cis.trans isomerase FKBP38 determines HIF prolyl-4-hydroxylase PHD2
protein stability. Mol. Cell. Biol. Mar (Epub ahead of print).

4. Beck, 1., S. Ramirez, R. Weinmann, and J. Caro. 1991. Enhancer element at the
3'-flanking region controls transcriptional response to hypoxia in the human
erythropoietin gene. J. Biol. Chem. 6:15563-15566.

5. Carmeliet, P., Y. Dor, J. M. Herbert, D. Fukumura, K. Brusselmans, M.
Dewerchin, M. Neeman, Bono, F., R. Abramovitch, P. Maxwell, C. J. Koch, P.
Ratcliffe, L. Moons, R. K. Jain, D. Collen, and E. Keshertk. 1998. Role of HIF-
la in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis.
Nature 394:485-490.

6. Chen, C., N. Pore, A. Behrooz, F. Ismail-Beigi, and A. Maity. 2001. Regulation
of glutl mRNA by Hypoxia-inducible Factor-1. Interaction between H-ras and
hypoxia. J Biol Chem 276:9519-9525.

7. Coelho, A., A. Ferraz, R. Neto, A. Souza, and E. Ferraz. 2000. Subdiaphragmatic
venous stasis and tissular hypoperfusion as source of metabolic acidosis during

passive portal-jugular and cacal-jugular bypasses in dogs. Acta Cir Bras p15.

8. Coleman, C. N., J. B. Mitchell, and K. Camphausen. 2001. Tumor hypoxia:

150

chicken, egg, or a piece of the farm? Journal of Clinical Oncology 20:610-615.

9. Denko, N. C., L. A. Fontana, K. M. Hudson, P. D. Sutphin, S. Raychaudhuri, R.

10.

11.

12.

Altman, and A. J. Giaccia. 2003. Investigating hypoxia tumor physiology through
gene expression pattern. Oncogen 22:5907-5914.

Guzy, R. D., B. Hoyos, E. Robin, H. Chen, L. Liu, K. D. Mansfield, M. C.
Simon, U. Hammerling, and P. T. Schumacker. 2005. Mitochondrial complex III
is required for hypoxia-induced ROS production and oxygen sensing. Cell Metab
1:401-408.

Isaacs, J. S., Y. J. Jung, D. R. Mole, S. Lee, C. Torres-Cabala, Y. L. Chung, M.
Merino, J. Trepel, B. Zbar, and J. Tom. 2005. HIF overexpression correlates with
biallelic loss of fumarate hydratase in renal cancer: Novel role of fumarate in
regulation of HIF stability. Cancer Cell 8:143-153.

Lee, K., R. A. Roth, and J. J. LaPres. 2007. Hypoxia, drug therapy and toxicity.
Pharm & Ther 113:229-246.

13.Lu, H., C. L. Dalgard, A. Mohyeldin, T. McFate, A. S. Tait, and A. Verma. 2005.

14.

15.

16.

17.

Reversible Inactivation of HIF-1 Prolyl Hydroxylases Allows Cell Metabolism to
Control Basal HIF-1a. J Biol Chem 280:41928-41939.

Mackenzie, E. D., M. A. Selak, D. A. Tennant, L. J. Payne, S. Crosby, C. M.
Frederiksen, D. G Watson, and E. Gottlieb. 2007. Cell-permeating alpha-
Ketoglutarare derivatives alleviate pseudohypoxia in succinate dehydrogenase-

deficient cells. Mol Cell Biol 21:3287-3289.

Piret, J. P., D. Motter, M. Raes, and C. Michiels. 2002. Is HIF-10L a pro- or an
anti-apoptotic protein? Biochem Pharmacol. 64:889-892.

Semenza, G L. 2000. Hypoxia, clonal selection, and the role of HIF-1 in tumor
progression. Crit Rev Biochem Mol Biol 35:71-103.

Semenza, G L. 2003. Targeting HIF-1 for cancer therapy. Nature 3:721-731.

151

18. Shimizu, S., Y. Eguchi, W. Kamiike, Y. Itoh, J. Hasegawa, K. Yamabe, Y.
Otsuki, H. Matsud, M. Shinkai, M. A. Pirker, S. Montedonico, and P. and Puri.

2005. The role of oxygen tension in the regulation of embryonic lung development.
Journal of pediatric surgery. 40:32-35.

19. Shweiki, D., A. Itin, D. Soffer, and E. Keshet. 1992. Vascular endothelial growth

factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature
359:843-845.

20. Vengellur, A., B. G Woods, H. E. Ryan, R. S. Johnson, and J. J. LaPres. 2003.

Gene expression profiling of the hypoxia signaling pathway in hypoxia inducible
factor 1a null mouse embryonic fibroblasts. Gene Expression 11:181-197.

152

APPENDIX A

IDENTIFICATION AND CHARACTERIZATION OF GENES SUSCEPTIBLE
TO TRANSCRIPTIONAL CROSS-TALK BETWEEN THE HYPOXIA AND
DIOXIN SIGNALING CASCADES

This chapter represents a manuscript that was published in Chem. Res. Toxicol. 19:
1284-1293 (2006). Authors included: KangAe Lee, Lyle D. Burgoon, Laura Lamb,
Edward Dere, Timothy R. Zacharewski, John B. Hogenesch, and John J. LaPres.

153

ABSTRACT

The aryl hydrocarbon receptor (AHR) and hypoxia inducible factors (HIFs) are
transcription factors that control the adaptive response to toxicants such as dioxins and
decreases in available oxygen, respectively. The AHR and HIFs utilize the same
heterodimeric partner, the aryl hydrocarbon nuclear translocator (ARNT) for proper
function. This requirement raises the possibility that cross-talk exists between these
critical signaling systems. Single gene and reporter assays have yielded conflicting
results regarding the nature of the competition for ARNT. Therefore, to determine the
extent of cross-talk between the AHR and HIFs, a comprehensive analysis was
performed using global gene expression analysis. The results identified 767 and 430
transcripts that are sensitive to cobalt chloride and 2,3,7,8-tetrachlorodibenzo-F-dioxin
(TCDD) stimulation, respectively, with 308 and 176, respectively, exhibiting
sensitivity to crosstalk. The overlap between these two sets consists of 33 unique
transcripts, including the classic targetgenes CYP1A1, carbonic anhydrase IX, and
those involved in lipid metabolism and coagulation.Computational analysis of the
regulatory region of these genes identified complex relationships betweenHIFs, AHR,
and their respective response elements as well as other DNA motifs, including the SRF,
Sp-l, NF—kB, and AP-2 binding sites. These results suggest that HIF-AHR cross-talk is

limited to genes with regulatory regions that contain specific motifs and architectures.

154

INTRODUCTION

The PAS (named for founding members; PER, ARNTl, SIM) family of
transcription factors act as sensors for various environmental stimuli, including
hypoxia and specific classes of pollutants (1). As transcription factors, their principal
reaction involves the modulation of gene expression that ultimately promotes an
adaptive response to these stimuli. PAS transcription factors generally function as
heterodimers that can have both cytosolic and nuclear components. The cytoplasmic
component generally acts as a sensor for environmental stimuli and includes the aryl
hydrocarbon receptor (AHR) and the alpha subunit of the hypoxia inducible factors
(HIF1-3) (2, 3). Once activated, these factors translocate to the nucleus and interact
with the second class of the superfamily, the nuclear component, such as the aryl
hydrocarbon nuclear translocator (ARNT, also known as HIFIOL) (4, 5). ARNT/HIF10t
is the predominant binding partner for the AHR and HIF la, and therefore, ARNT
might act as a point of competition or cross-talk following the activation of the AHR
and HIFIOL.

The AHR is a ligand activated transcription factor that responds to a broad range
of planar aromatic hydrocarbons (6). Classic AHR ligands include environmental
pollutants such as 2,3,7,8-tetrachlorodibenzo-F-dioxin (TCDD), naturally occurring
compounds such as indole-3-carbazole, and endogenous ligands such as tryptophan
metabolites. In the absence of a ligand, the cytosolic AHR is bound to the
immunophilin-like protein, aryl hydrocarbon receptor-associated protein (ARA9) and a
dimmer of heat shock protein 90 (Hsp90) (7, 8). Upon binding a ligand, the AHR

translocates to the nucleus where it heterodimerizes with ARNT. The transcriptionally

155

active AHR-ARNT complex drives the expression of genes containing dioxin response
elements (DREs, core sequence = GCGTG) such as the canonical AHR-responsive
gene, cytochrome P450, family 1, subfamily A, polypeptide 1 (CyplAl) (9).

Hypoxia is defined as a decrease in available oxygen reaching the tissues of the
body. The cellular response to hypoxia is a fine balance between adaptation and cell
death and is primarily controlled by HIFla (10). HIF10t is a cytosolic protein whose
stability is regulated by a family of prolyl hydroxylases (11, 12). These hydroxylases
are oxygen dependent sensors for the hypoxia-signaling cascade either directly or
indirectly through changes in reactive oxygen species generated from complex III of
the electron transport chain (13-16). In the presence of oxygen, HIFIR is hydroxylated
and degraded via the Von Hippel Lindau tumor suppressor protein and the 26S
proteosome pathway (17). In the absence of sufficient oxygen, the hydroxylase is
inactive and the HIF 1(X. protein becomes stabilized and translocates into the nucleus
where it interacts with ARNT to drive the expression hypoxia response elements
(HREs, core sequence = (G/A)CGTG) containing genes (18). Classic hypoxia
inducible genes include vascular endothelial growth factor (VEGF), erythropoietin,
and most of the glycolytic enzymes (19, 20).

The ability of ARNT to act as the heterodimeric partner for both HIFlor and
AHR raises the possibility of cross-talk between these signaling cascades. In addition,
the similarities between the HRE and DRE core sequences suggests HIFlR-ARNT and
AHR-ARNT might compete for the same regulatory sequence (21). Consequently,
hypoxia-AHR cross-talk might have profound consequences on treatments aimed at

hypoxic targets such as tumors and might influence the ability of the AHR to regulate

156

the expression of various drug metabolizing enzymes. Previous in vitro and in vivo
studies have shown varying degrees of competition between AHR and HIF signaling
systems, presumably because of competition for ARNT or another critical cofactor
(22-26). We hypothesized that this cross-talk may only apply to target genes that
harbor select response element combinations that include but are not limited to DREs
and HREs. Global gene expression analysis was performed in the human hepatoma
HepB3 cells following exposure to TCDD, cobalt chloride (CoCl2), and their
cotreatment. A set of genes responsive to cobalt or TCDD, with a subset influenced by
cotreatment was identified. Computational analysis of the regulatory regions of a
subset of these genes, influenced by individual treatment, identified motifs associated
with TCDD or cobalt chloride modulated gene expression. The results suggest that
cross-talk between the AHR and HIF1 signaling systems in Hep3B cells is not
regulated by ARNT levels and is limited to a group of genes with specific promoter

architecture.

157

EXPERIMENTAL PROCEDURES

Cell Culture. Hep3B cells were maintained in orMEM (Invitrogen, Carlsbad,
CA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT), 20 mM L-
glutamine, 1 mM MEM nonessential amino acids, 100 mM Hepes (pH 7.4), 1000
units/Ml penicillin G and 1000 z’g/mL streptomycin sulfate (Invitrogen). Cells were
approximately 70% confluent at the time of treatment. Cells were maintained at 37 °C,
5% CO2, and 21% 02 prior to treatment. 2,3,7,8-tetrachlorodibenzo-F—dioxin (TCDD,
10 nM) and cobalt chloride (100 11M) treatments were performed for 20 h at 37 °C, 5%
CO2, and 21% 02. These doses for cobalt and TCDD were chosen for their ability to
activate HIFla and the AHR, respectively. Cobalt chloride has an established history
as a hypoxic mimic and the chosen dose has a demonstrated ability to stabilize HIFIOL
and activate all classic HIFloL target genes, including VEGF and the glycolytic
enzymes. The dose of TCDD was chosen for its ability to maximally activate the AHR
(several times its Kd). Each treatment group was supplemented with dimethyl
sulfoxide (DMSO) to a final concentration of 0.01%. Samples were compared to
vehicle control (0.01% DMSO).

RNA Extraction. Following treatment, duplicate cell samples were removed
from the tissue culture dish by trypsinization and washed in PBS (4 °C). RNA was
extracted by homogenization (Polytron, Kinematica, Lucerne, Switzerland) in TRIzol
reagent (GIBCO/BRL, Gaithersburg MD) (added to cell pellet) at maximum speed for
90-120 3. Following a 5 min room temperature incubation, a 1/5 volume of chloroform
was added, agitated, and subjected to centrifugation at 12 000g for 15 min. The

aqueous phase was removed, and the RNA was precipitated upon the addition of a half

158

vol of isopropanol. The RNA was furthered purified with the RNAeasy Total RNA
isolation kit (Qiagen, Valencia, CA) according to manufacturer’s specifications.
Finally, the purified total RNA was eluted in 10 uL of diethylpyrocarbonate (DEPC)
treated H20, and quantity and integrity were characterized using a DU640 UV
spectrophotometer (Beckman, Fullerton, CA) and a Bioanalyzer 2100 (Agilent, Palo
Alto, CA).

RNA Labeling. Five micrograms of total RNA from two separate biological
replicates were used to make first strand cDNA using the Superscript Choice system
(Gibco/BRL) and a T7 promotor/ oligo-dT primer (Gibco/BRL). Second strand cDNA
was also made with the Superscript Choice system (Invitrogen). The resulting cDNA
was subjected to phenol/chloroform purification, arrunonium acetate precipitation and
used as a template to make biotinylated amplified antisense cRNA using T7 RNA
polymerase (Enzo kit, Affymetrix, Santa Clara, CA). Twenty micrograms cRNA was
fragmented to a range of 20 to 100 bases in length using fragmentation buffer (200
mM Tris-acetate, pH 8.1, 500 mM KOAc, 150 mM MgOAc) and heating for 35 min at
94 °C. The quality of cRNA and the size distribution of fragmented cRNA were
examined by both agarose and polyacrylarnide gel electrophoresis.

Hybridization. cRNA (20 ug) was hybridized to a U95A version 1 gene chip
(Affymetrix) with 1 x 4-morpholineethanesulfonic acid (MES) hybridization buffer
using standard protocols outlined in the Gene Chip Expression Analysis technical
manual (Affymetrix). Hybridization was conducted in a GeneChip Hybridization Oven
for 16 h at 45 °C. Following hybridization, the arrays were washed on a Genechip

Fluidics Station 400 according to manufacturer’s instructions (Affymetrix). The arrays

159

were scanned using a Hewlett-Packard 2500A Gene Array Scanner, and the raw
images were visually inspected for defects, proper grid alignment, and converted into
CEL files using the MASS Software Suite (Affymetrix). Finally, the quality of cRNA
was assessed by examining 3'/ 5’ ratios for GAPDH oligonucleotides present on the
arrays.

Data Analysis. Background subtraction and single intensity measure for each
transcript was derived from multiple probe sets by means of the GCRMA algorithm
using the full model tag in R (http://www.r-project.org). The GCRMA algorithm was
chosen for its performance in reporting low and high level expression over other
methods as well as its dynamic range for single probe sets (27-29). Differentially
expressed genes that are statistically significant were determined by analysis of
variance (ANOVA). Fold change calculations were performed in Excel on data that
was median-scaled to a global intensity target value of 100. For each treatment versus
control condition, genes that changed were assigned on the basis of a P value of <0.05
and an absolute fold change value of >1.5.

Quantitative Real-Time PCR (qRTPCR) Analysis. Changes in gene
expression were observed by microarray analyses and verified by real-time PCR
performed on an Applied Biosystems Prism 7000 Sequence detection System (Foster
City, CA) as previously described (30). Analysis was performed on triplicate samples
that were treated in a way identical to those used in the GeneChip experiments. Briefly,
cDNA was synthesized from total RNA (1 pg per sample per treatment, n = 6) in a

reverse transcriptase reaction in 20 ILL of 1x First Strand Synthesis buffer (Invitrogen)

containing 1 ug of oligo (5’-T21VN—3’), 0.2 mM dNTPs, 10 mM DTT, and 200 IU of

160

I"

 

Superscript II reverse transcriptase (Invitrogen). The reaction mixture was incubated at
42 °C for 60 min and stopped by incubation at 75 °C for 15 min. Amplification of
cDNA (1/20) was performed using SYBR Green PCR buffer (1 x AmpliTaqTM Gold
PCR Buffer, 0.025 U/mL AmpliTaqTM Gold (Perkin-Elmer, Wellesley, MA), 0.2 mM
dNTPs, 1 ng/uL 6-carboxy-X-rhodamine, 1:40 000 diluted SYBR Green Dye and 3%
DMSO) and 0.1 uM primers. The thermal cycling parameters were 95 °C for 10 min,
followed by 40 cycles of 95 °C for 15 s and 60 °C for 60 5. Prior to analyzing the
samples, standard curves of purified, target-specific amplicons were created. Briefly,
gene specific oligonucleotides were used to PCR amplify the gene product from a
pooled sample of prepared cDNA, the concentration of the amplicons was determined
by UV spectophotometery, and a standard curve was created (102-108 copies). The
mRNA expression for each gene was determined in comparison to its respective
standard curve. This measurement was controlled for RNA quality, quantity, and RT
efficiency by normalizing it to the expression level of the B-Actin gene. Each primer
set produced a single product as determined by melt-curve analysis, and amplicons
were of the appropriate size, as analyzed by agarose gel electrophoresis. Statistical
significance was determined using normalized fold changes and ANOVA. Primers
were designed using the web-based application Primer3 (httpzl/www-

genome.wi.mit.edu/cgi-bin/ primer/primer3_www.cgi) biasing toward the 3’ end of

the transcript to maximize the likelihood of giving a gene specific product. The
settings used in Primer3 were 125 base-pair amplicon, 20mers, 60 °C melting
temperatures, and all other as defaults. Primer sequences were analyzed by BLAST.

Gene names, accession numbers, and forward and reverse primer sequences are listed

161

in Table A- 1.

Computational Scanning for DREs and HREs. Computational identification
and matrix similarity (MS) scores (31) of putative HREs and DREs were performed
using 19 base-pair position weight matrices (PWMs) (Fig. A-1A and -1B) as
previously developed by Sun et al. (32). The PWMs were constructed using the
sequences of reported functional response elements that were positive in either
electrophoretic mobility shift or transient transfection assays. In total, 14 DREs (Fig.
A-lC) and 38 HREs (Fig. A-lD) were used in the development of the PWS. Each
matrix consists of the 5 base-pair core HRE or DRE, (G/A)CGTG (Fig. A-lA) and
GCGTG (Fig. A-lB), respectively, and the adjacent variant 7 base-pair flanking
sequences of the bona fide functional response elements. The sequences of these
known response elements were subsequently scanned using the PWMs to determine
the threshold MS scores (0.818 for DREs and 0.813 for HREs) (32). The MS scores
for the 14 functional DREs and 38 HREs are listed in Figure A-lC and -1D,
respectively. Furthermore, the consensus index (Ci) vector was calculated for both the
HRE and DRE PWMs, where the Ci vectors represents the conservation of the
individual nucleotide positions in the matrices (31). A complete list of information
regarding the sequences used in creating the PWMs can be found in Supporting

Information, Table A-1 and Sun et al. 2004. The genomic sequences (-5000 base-pair

to the transcriptional start site (TSS)) and the 5' untranslated region (UTR)) of the 33

RefSeq genes were extracted from the University of California, Santa Cruz (UCSC)
Genome Browser (http://www.genome.ucsc.edu) (Build #35) and scanned for exact

matches to the DRE and HRE core sequences on both the positive and negative strands

162

Figure A-l. HRE and DRE position weight matrices and the sequences used in
their construction.

The position weight matrices were created as described in Experimental Procedures.
The HRE (A) and DRE (B) PWMs are displayed relative to the G of the second half of
the binding site. The genes and sequences used for the creation of the DRE (C) and
HRE (D) PWMs and their respective MS scores are listed. Sequence information
regarding HREs used in construction of PWM can be found in Supporting Information,
Table 1.

163

 

 

 

 

901

em

704

3504

3..

640-

so-

201

1
Position-9 -7-6-54 4412345678910
A 3181020202389000001010102231515
c18231028443115010000062232831283134
G 2851494418414180100010018543631264136
T 231831818536200100010132636231315
Ci 15 26 27 231924 24 76100100100100 33 27 19 26142119

Cl Vector

 

 

 

 

m
Position -9 -8 -7 -6 -5 -3 -2 -1 1 2 3 4 5 6 7 8 9 10
A 7 21 43 21 0 O 21 0 0 O 0 0 43 21 64 57 ‘14 29 7
C 57 43 14 14 50 7 7 0 100 0 0 0 57 29 7 7 29 43 29
G 21 36 36 50 29 21 7 100 0 100 0 100 0 50 7 21 57 29 21
T 14 0 7 14 21 21 64 0 0 0 100 0 0 0 21 14 0 0 43
Ci 31 34 26 23 36 53 36 100 100 100 100 100 58 36 38 31 41 33 23

164

1
r GstaZ

 

165

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mom“ a u .000 m .0038on 80:20 or . 538 2.7 5.0.4.93
822 2.08.0 08. .0 .8802058 . mm. omioo .27 EEO:—
mmozoo .000 mm 8 . no a 0000.2 00.0.0 om. @3538 .27
1Wm509omwommmmwmmm omfifimmfimmfiogmmm 5.. 1
000.5 3005.8 m 003 .mm m m 00 mm?
0929000903008 mmmomammmmmomfimmom vm r
em.
mm.
mm.
Eouo.moomoowoo.8 moommmwommmmmmmmmfim mm?

 

 

ozomflmmoooummmfima

 

 

0000me009083030 03.

166

 

 

 

 

 

 

 

 

 

 

 

o ..zxoo or 6.5.2:. 0mg... .ccvcoawu-c__o>o.

 

 

ONJOO

 

 

 

 

Emwowmmmmmmaoomom

 

 

 

5.5:. 3.. 8:55... 05.5 .3. 2.3.

using the PWMs. The MS score for each match was computed, and those with scores
greater than the threshold values are expected to have both a greater probability of
possessing a measurable binding affinity and presumable biological relevance.
Identification of Over-Represented Short Sequence Motifs in Gene Regulatory
Regions. Gene regulatory regions, defined as starting -5000 kb relative to the

transcription start site through the 5' UTR, were obtained from the UCSC Genome

Browser for all known genes assigned a mature RefSeq mRNA accession. These
sequences were stored in the Gene Regulatory Subsystem of our toxicogenomic
database, deach (httpzlldbzach.fst.msu.edu, (33)), to facilitate further analysis. All 5-
10 nucleotide short sequence motifs were identified using a sliding window method
(34). An empirical Bayes implementation of the Wilcoxon’s Rank Sum Test, similar to
those used for microarray analysis (35, 36), was used to identify 5-10 nucleotide
motifs that are over-represented in one population compared to that in another. This
method computes a posterior probability, which represents the likelihood of that result
occurring (i.e., a posterior probability of 0.90 means that there is a 90% probability
that the result is true). The Transfac database (37) was queried to annotate the motifs
and identify potential binding proteins. Absolute numbers of individual motifs that
were determined to be over-represented in the cobalt or TCDD treatment groups were
then analyzed by hierarchical clustering (unweighted pair-group method using
arithmetic averages using correlation distances; http://gepas.bioinfo.cnio.es/cgi-
bin/cluster (38)). These values were clustered with the GeneChip expression data

found in Fig. A-4.

167

RESULTS

Identification of genes potentially susceptible to cross-talk between the AHR
and HIF signaling cascades was performed on a high-density oligonucleotide array.
Hep3B cells were treated with DMSO (0.01%, vehicle control), cobalt chloride (100
mM), TCDD (10 nM), or cobalt chloride and TCDD (100 mM and 10 nM
respectively) for 20 h, and global gene expression patterns were analyzed. The
procedure to identify these genes is outlined in Fig. A-2A. Briefly, significant (p <
0.05) gene expression changes following a single treatment were identified and further
pared using a 1.5-fold cutoff leaving 767 and 430 probe sets for cobalt and TCDD
treatment, respectively. These genes were then analyzed for significant differences
between single treatment and co-treatment as determined by paired student’s t-test.
With these criteria, 308 and 176 probe sets were identified for cobalt- and TCDD-
treated gene groups, respectively. These two groups constitute the genes that can be
influenced by cross-talk from either single treatment. Of these, 34 probe sets,
corresponding to 33 genes were represented in both lists (Fig. A-2A). These 33 genes
represent a group of targets that are affected by both single treatments (i.e., AHR and
HIF responsive) and whose expression is modulated by cross-talk.

Because the analysis was performed with no bias toward direction of change, it
is possible that some of the changes would be in the same or in opposite directions.
Analysis showed that a majority of these genes displayed the same direction changes,
either both up or both down. Only 33% of the genes exhibited an opposite pattern of
expression (Fig. A-ZB). These results suggest a more complex interaction between the

AHR and HIF signaling cascades and identify genes where this interaction may lead to

168

Figure A-2. Analysis of genomic data and comparison of 33 target genes.

The genomic data was analyzed in a four step process (A). First, genes were screened
for significance (p < 0.05). Second, these genes were filtered for those that displayed
greater than 1.5-fold change. Third, the list was analyzed for those that were
significantly altered (p < 0.05) when single treatment was compared to co-treatment.
Finally, the cobalt and TCDD lists were compared for overlap. (B) Direction of
expression changes was analyzed for the 34 probe sets identified in the screen. A
complete list of cobalt and TCDD influenced genes can be found in Supportng
Information, Tables 5 and 6).

169

 

I Cobalt I TCDD

- Total Probes Sets
1 l

P < 0.05
1 1
Fold Change > 1.5
1 1

Different from
Co-Treatment 176
Overlap

34 Probes Sets
33 Different Genes

 

 

 

 

 

  
    
   
  
   
 

35%
Both up

: .i.I’=-.

‘ 38% ;‘ '.
1391119911191,

   

Co2+ down
TCDD up
15%
Co2+ up
TCDD down

170

antagonism (i.e., less than additive) or synergy (i.e., more than additive).

The list of 33 genes include prototypical dioxin and hypoxia responsive genes,
notably CyplAl and heme oxygenase-1. These genes exhibit a classic cross-talk
expression pattern. For example, CyplAl is induced approximately 100-fold in the
presence of TCDD, whereas it is repressed almost 6-fold after cobalt treatment.
Following co-treatment, gene expression is only induced 65-fold (Table A-2). A
similar pattern is seen for heme oxygenase-1, where it is induced by cobalt, repressed
by TCDD, and only partially induced by co-treatment. This type of competition is not
the only type of regulation observed. SOX-9 and CDKNIC displayed clear evidence of
additive regulation in which individual treatments modify expression in the same
direction, and co-treatment yields an additive expression value (Table A-2). In addition,
UGTIAI displays a level of synergy between treatments. These results suggest that
simple regulation or competition for ARNT or other cellular factors cannot fully
explain all of the expression changes seen in the genomic screen.

The expression patterns of 19 different genes, including 11 from Table 2, were
verified by qRTPCR. These genes were a mixture of classic and novel cobalt- or
TCDD-inducible genes identified in the individual treatment groups (Supporting
Information, Table A-3 and A-4) and cross-talk analysis (Table A-2). In most cases,
these results verified those of the Gene Chip data (Fig. A-3). Forty-nine of the 57 gene
expression changes (19 different genes under 3 different treatments) were verified by
qRTPCR (Fig. A-3B). Interestingly, in the gene chip data, the fibrinogen alpha subunit
(FGA) was up-regulated upon cotreatment, whereas the fibrinogen beta subunit (FGB)

was downregulated under similar conditions (Table A-2). qRTPCR results showed that

171

Table A-2. Microarray fold change and information for 33 cross-th genes

 

 

 

fold change
probe set Co TCDD TCDD+C0 REF seq descriptions abbrev.
1025 g_at -5. 9 99.6 64.8 NM_000499 cytochrome P450,
CYP1A1
family 1, subfamily A
33436_at - 2. 9 - 3.9 — 5. 7 NM_000346 SRY (sex determining SOX9
region Y)-box 9
1787_at - 2. 6 - 2. 3 - 5. 6 N M_000076 cyclin-dependent CDKN 1C
kinase inhibitor 1C
373l9_at -2.3 -1.8 -4.3 NM_000598 insulin-like growth
IGFBP3
factor binding protein 3
39545_at — 2. 2 - 2. 3 -3. 9 N M_000076 cyclin-dependent kinase
CDKNIC
inhibitor 1C
40385_at - 2. 2 —1.8 - 2.1 NM_004591 chemokine (C-C motif)
CCL20
ligand 20
35303_at -2.0 1.9 1.1 NM_005542 insulin induced gene 1 INSIGI
38519_at -1.9 -1.6 -3.0 NM_005123 nuclear receptor subfamily 1, NRIH4
group H, member 4
39352__at - l .9 -l 1.9 -4.8 NM_000735 glycoprotein hormones, CGA
alpha polypeptide
41424_at -1.9 -1.7 - 3.1 NM_000940 paraxonase 3 PON 3
33701_at - 1.8 -1.8 - 2. 6 NM_000277 phenylalanine hydroxylase PAH
37235g_a1 -1.8 -2.1 -3.6 NM_000893 kininogen 1 KNG l
38586_at - 1.7 1.8 -3.1 NM_001443 fatty acid binding
FABPI
protein 1
38178_at - 1.7 - 2.8 -3.9 NM_002l53 hydroxysteroid (l7-beta) HSD17BZ
dehydrogenase 2
36135_at - 1.6 1.8 1.1 NM_006824 EBNAI binding protein 2 EBNAIBP2
37188_at -1.6 -2.1 - 1.8 NM_004563 phosphoenolpyruvate
PCK2
carboxykinase 2
31792__at —1.5 —3.1 -2.2 NM_005139 annexin A3 ANXA3
33260_at 1.7 1 1.4 6.8 NM_005633 son of sevenless SOS]
homologue l
38789_at l .8 1 .9 2.3 NM_OO 1064 transketolase TKT

172

370 l 9_at
39070_at
38376_at

3|824_at
39425_at
38545_at
38825_at
38637_at
39008_at
33802_at

1232s_at

39072_at
34777_at
40309_at
32392s_at

1.8
1.9
2.0

2.4
2.5
2.5
2.9
3.6
4.5
6.0

10.0

10.4
33.1
39.4
80.7

"4.0
1.9
1.7

2.1
1.6
-2.1
-1.9
-3.3
1.5
-l.9

2.2

1.6

1.8

2.2
77.7

71.6
2.0
1.3

4.7
3.2

5.9
*1.0
2.2
3.9

3.6

7.1

l 1.0
20.3
295.4

NM_005 l4 1
NM_003088
NM_00001 8

NM_002395
NM_003330
NM_002 l 93
NM_000508
NM_0023 17
NM_000096
NM_002 133

NM_000596

NM_005962
NM_OOI l 24
NM_OO l 21 6
N M_007 120

fibrinogen beta chain
fascin homolog 1
acyl-Coenzyme A
dehydrogenase

malic enzyme 1
thioredoxin reductase l
inhibin. beta B
fibrinogen alpha chain
lysyl oxidase
ceruloplasmin

heme oxygenase
(decycling) l
insulin-like growth
factor binding protein 1
MAX interactor l
adrenomedullin
carbonic anhydrase IX

UDP

glucuronosyltransferase l

FGB
FSCNI

ACADVL

MEI
TXNRDI
INHBB
FGA
LOX
CP

HMOX l

IGFBPI

MXll
ADM
CA9
UGT 1 Al

 

173

Figure A-3. qRTPCR verification of expression changes for selected cross-talk
genes.

qRTPCR was performed on 19 genes using B-actin for normalization. (A) Gene
abbreviations, accession numbers used in primer design, and fold changes relative to
vehicle control are listed. (B) A direct comparison between qRTPCR and Gene Chip
results was performed and analyzed by linear regression.

174

SYBR (L092)

      

  
 

 

 

  
 

 
  
  
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

CDKN1C AF079221 -1.9 -8.5 -13.8
CCL20 NM_004591 -1.8 -3.8 -2.9
SMADS NM_001001419 -1.8 -2.0 -4.0
Cyp1B1 U03688 -1.6 6.8 1.4
KCNJ8 050312 -1.5 1.3 -2.3
LUM U21128 -1.5 2.7 -1.2
PCK2 NM_004563 -1.3 -1.6 -2.4
MYC NM_002467 -1.2 2.5 1.4
Cyp1 A1 NM_000499 -1.1 102.9 70.5
DUSP5 U15932 1.0 1.3 -1.5
CGA NM000735 1.0 -3.2 -4.1
FGA M64982 1.7 -2.1 -2.3
CP NM_000096 1.9 -7.6 -3.0
mm NM_005952 2.7 -1.2 3.8
Lox L16895 3.1 -2.9 -2.0
FGB J00129 3.3 -2.8 -2.1
UGT1A1 NM_007120 5.0 2.8 7.8
BNIP3L AF079221 6.4 -1.5 3.0
ADM NMOO1124 6.8 -1.1 3.8
GeneChip vs SYBR
8
O
6 7—7— . 4
4 _. . J _#
o 9 9 o
2 ',
O
0 °'
. I
O O O
-2 4 ’
4 . y = 0.5223x - 0.0876
R2 = 0.5514
-6
-5 0 5 1 0
GeneChip (L092)

175

these two subunits were expressed in a similar pattern under each treatment condition
(Fig. A-3A). Because different pools of RNA were used for the qRTPCR and
microarray experiments, cell line or culture differences cannot be ruled out as an
explanation for this difference. One possible explanation for the different expression
patterns exhibited by the 33 genes (Table A-2) that display cross-talk regulation is
promoter context of response elements. To initially characterize differences between
these promoters, genomic sequences 5000 base pairs upstream of the transcription start

codon and the 5’ UTR were analyzed for homology, HREs, DREs, and GC content.
The promoter and 5’ UTRs from the 33 genes were aligned using the ClustalW

algorithm. Interestingly, the sequences clustered on the basis of their expression
patterns with four clusters containing four or more sequences (A-D, Fig. A-4). Group
A was predominantly up-regulated in both treatments to an equal level, and several
were additive. Group B was primarily inhibited and showed a tendency for
competition, where the co-treatment group was somewhere between the two individual
expression levels. Group C was predominantly down-regulated in both treatments, and
Group D showed higher expression in the cobalt treatment compared to that of TCDD

(Fig. A-4). This pattern was not unsystematic or due to some inherent base pair
composition found within the promoters and/or 5’ UTR of the genes analyzed because
randomization (http://www.cellbiol.com/cgi-bin/randomizer/sequence randomizer.
html) of the promoter sequence yielded no consistent pattern with relation to
expression. In addition, random sequences exhibited no relation to the alignment seen

in Fig. A-4 (data not shown). These results suggest that motifs within the analyzed

176

Figure A-4. Alignment and sequence analysis of the promoters and 5’ UTR of
core genes.

5000 bp of upstream sequences and the 5’ UTRs were aligned by ClustalW and
compared to expression patterns from oligonucleotide chips. The promoters were also
analyzed for HREs and DREs using a PWM (Experimental Procedures). Finally, the
GC content was also calculated. Up-regulated genes are shown in white, and down—
regulated genes are shown in dark gray. Four distinct clusters are also noted (A-D).

177

p

895 805595 626282
02 om, 8N

_ — _

 

178

A

regions contained the regulatory information that influenced the cross-talk expression
patterns observed. Differences in the regulation patterns of the various genes could
also be explained by the number of hypoxia or dioxin response elements located within
the regulatory region. Comparison of these motif sequences to HRE and DRE PWMs
suggests that DREs were better represented in Group D and that this bias did not
correlate with an increase in GC content (Fig. A-4). The analysis did identify more
HREs than DREs, primarily due to the fact that the HRE PWM was only exclusive at
four positions, compared to five positions in the DRE (Fig. A—1 and A-4). Given the
redundancy within the core sequences of the DREs and HREs, it is also possible that
several sequences could register as both DREs and HREs. To determine the level of
overlap, each DRE and l-IRE was further scored using the opposing PWM. For
example, each DRE listed in Figure 4 was scored with the HRE PWM, and a similar
cutoff (the MS score had to be higher than the lowest HRE used to create PWM) was
applied. The results show that 40% of the DREs also scored positive as HREs, whereas
only 14% of the HREs scored positive as DREs (Supporting Information, Tables 8 and
9). However, there was no relationship between HRE-DRE overlaps and the type of
cross—talk observed. Finally, it is possible that base-pairing relationships, G/C versus
A/T, may bias the association of a promoter into one group or another. On the basis of
GC content, there was no correlation between base-pairing relationships and gene
expression patterns or groupings, suggesting that other information within the
promoter regulated expression (Fig. A-4).

Given that it is not the absolute number of HREs or DREs within the regulatory

region or their basic composition that dictates expression, promoter analysis for other

179

A

Figure A-S. Promoter analysis for over-represented motifs and correlation with
expression.

The promoters of 65 different genes from the cobalt- and TCDD-treatment groups
were analyzed for motifs that were over-represented as described in Experimental
Procedures. Information about the motif, its corresponding response element as
determined by TRANSFAC, and the statistical values are represented for the Cobalt
group (A) and the TCDD group (B) as defined in Experimental Procedures. The total
number of response elements listed in A and B (including HREs and DREs) were then
identified in the regulatory regions of the 33 genes from Figure 4 and correlated with
expression data via hierarchical clustering. A dendrogram correlating the GeneChip
expression data (C02+, TCDD and TCDD+C02+) with the various response elements
is displayed (C). A complete Table of motifs identified in the analysis and the table
used to create the dendrogram can be found in Supporting Information, Tables 2—4.

180

u will

 

adjusted

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Site word p-value
AP-1 gagac 9.58
AP-1 tgaga 9.56
AP-2 tggggc 9.52
c/EBP alpha gatttt 9.59
c-ETS-1 gctct 9.56
E2F-1 cttggc 9.50
EBF cttga 9.61
Egr—1 cqccc 9.61
Egr—1 cccac 9.48
IRS ttttg 9.56
MARE gctgag 9.58
MyoD gtggc 9.50
MyoD 90390 9.59
MyoD ctggc 9.59
NF-kB cctct 9.50
NF-kB ggctt 9.58
NF-kB ttcca 9.61
NRF-2 alpha tcttcc 9.50
RXR ttcat 9.50
891 tgggc 9.50
Sp1 ccccc 9.58
Sp1 909ch 9.61
831 99990 9.56
Sp1 gcact 9.48
Sp1 09099 9.52
Sp1 tccct 9.52
SRE tcctca 9.58
_ I adjusted
Site, word p-value
GRE taaac 0.91
ISRE ggaaa 0.89
ISRE aaact 0.88
c/EBP alpha aagag 0.87
AP-1 agttt 0.85
GRE ataaac 0.85
c/EBP alpha gttga 0.85
GATA-3 gatat 0.85
GATA—1 tctca 0.85

 

181

 

 

182

EGR-1
Sp-1
DRE
MyoD
NF-KB
AP-2
HRE
SRE
TCDD
TCDD+Co2+
Co2+
NRF-2a
E2F-1
C-ETS-1
EBF
AP-1
MARE
GATA-3
RXR
ISRE
GRE
clEBPa
IRS
clEBPoc
AP-1
GATA-1

4‘.

motifs that might correlate with the expression pattern in Fig. A-4 were also examined.

A computational word search was performed to identify over-represented
motifs/elements 5000 base pairs upstream of the start codon and in the 5’ UTR. The

33 genes from Figure 4 were not sufficient to perform a word analysis; therefore, the
analysis was conducted on a nonredundant set of 65 active genes from each of the
cobalt and TCDD treatment groups for gene regulatory sequences. The lists of
overrepresented motifs in cobalt and TCDD groups were combined, and the
frequencies of each motif occurrences were calculated for each gene. Twenty-seven
different words, corresponding to 15 different known transcription response elements,
including Sp1, serum response elements (SRE) and NF—kB sites, were over-
represented in the cobalt treated group. Conversely, nine different words,
corresponding to six different elements were over-represented in the TCDD treated
group. These included a c/EBP—a, GREs, and two types of GATA response elements
(Fig. A-5).

The regulatory regions of the 33 genes identified in Figure 4 were analyzed for
the over—represented response elements, and the correlation between expression
patterns from the microarray, and the frequency of occurrence of these 23 major
response elements (15 from the cobalt—treated group, 6 from the TCDD group and
HREs and DREs), using hierarchical clustering. All of the expression data clustered
together with the SRE (Fig. A-5C). A second cluster showed strong correlation with
the expression data, which included EGR-l, Sp-l, DRE, MyoD, NF—kB, AP2, and
HRE elements. These results suggest that the expression patterns in Figure 4 are well

correlated with the number of SREs in the respective promoters and to a lesser extent

183

to a group of seven other response elements, including the HREs and DREs.

184

A

DISCUSSION

Understanding the nature and extent of the cross-talk between the hypoxia and
dioxin signaling cascades is important for determining the role of each signaling
cascade in directly modulating the function of the other. The results presented here
suggest a gene-specific cross-talk. In some instances, the crosstalk was shown to be
additive (e.g., IGFBP3, UGTlAl, and CDKNlC), whereas in others, it was
competitive (e. g., P450, IHBB, and LOX). To establish a possible mechanism for the
selective nature of the cross—talk, a small portion of the putative regulatory region of a
subset of these genes was analyzed for HREs and DREs. There was no apparent
correlation between the number of these types of response elements and the nature of
the cross-talk observed. Computational analysis of the regulatory regions proximal to
the transcription starting site (TSS) identified over-represented motifs within these
regions, suggesting that AHR-HIF1 cross-talk cannot be explained completely by
competition for ARNT and most likely involves other cofactors, response elements,
and promoter context.

Differences in expression patterns for those genes affected by TCDD and cobalt
cross—talk suggest that a variety of mechanisms are involved. If the cross-talk was
solely dependent upon competition for ARNT, then evidence of interactions would be
widespread and would not include additive responses. In contrast, only 33 genes
exhibited an interaction, with approximately 30% displaying additive tendencies,
including MEI, SRY, CDKN 1C, and NRIH4. This pattern suggests that ARNT is not
limiting and that alternative mechanisms are involved. These mechanisms probably

involve competition and/or synergy between other co-activators/co-repressors or other

185

signaling systems that are being influenced, and the cross-talk is a result of secondary
signaling. For example, TCDD might inhibit the expression of a transcription factor
necessary for cobalt-induced expression of another TCDD target gene. Alternatively,
cobalt exposure might activate a signal that leads to post-translational modification of
the AHR, thus altering its activity. Comparable interactions have been described
between steroid receptors and other signaling cascades (ref 39 and references theirein).
Finally, these alternative pathways likely involve other cascades that influence the
ability of AHR and HIF1 to alter expression patterns. The predominance of SREs
present in the regulatory region of genes exhibiting cross-talk suggests that one of
these alternative inputs is the serum response cascade.

The presence of the SRE motif correlates with the expression patterns elicited by
various treatments, suggesting a regulatory role in gene expression. Surprisingly, the
number of SREs across the 33 genes in Table 2 is more predictive of cross-talk than
the number of HREs or DREs (Fig. A-4). The 33 cross-talk genes were identified after
meeting two criteria: their expression was influenced by each individual treatment, and
this expression was further altered upon co-treatment. The original hypothesis that
cross-talk was due to ARNT competition meant that this two step process would
identify the subset of genes most prone to cross-talk. Given the different types of
interactions (e. g., additive and competitive), simple competition for ARNT does not
satisfactorily explain cross-talk in Hep3B cells, and these requirements might have
biased the results to select for other transcription factor pathways such as the SRE.

The serum response factor (SRF) binds the SRE and regulates the expression of

a variety Of genes, including plasminogen activator inhibitor 1 (PAI-l), early growth

186

response factor-1 (EGR-l), and glucose transporter-1 (Glut-1) (4042). Interestingly,
each of these genes is a target for hypoxia-mediated transcription. The over-
representation of SREs in the cobalttreated group suggests that HIFl and SRF
signaling pathways may converge upon a select group of genes, and the correlation of
SREs with the set of cross-talk genes from Table A-2 might suggest that these genes
are important for the response to these two signaling pathways

Genes exhibiting cross-talk fall into a variety of ontological categories, with
several distinct processes and functions being over—represented, including blood
coagulation, cell proliferation, fatty acid oxidation, metal binding, and oxidoreductases
(Supporting Information, Table A-7), suggesting that HIF-AHR interactions may alter
these endogenous processes. For example, hypoxia influences the disposition of a
variety of drugs through multiple mechanisms, including altering the expression of
phase 1 enzymes and several oxidoreductases, and a patient who has been pre-exposed
to AHR ligands might be further compromised in his or her ability to dispose of the
drugs (43). The putative role of cross-talk might be most evident during development
of the liver and heart. Mice lacking functional AHR have aberrant liver development
and are prone to hypertension and cardiac hypertrophy (44). Liver development is
abnormal in AHR null mice as a result of decreased peripheral profusion (44), possibly
due to an imbalance in hypoxia-induced fibrin expression that ultimately influences
angiogenesis. Over-production of fibrin and its subsequent deposition into the local
circulation might explain the excess hematopoeitic cells in the AHR -/- mouse liver
and the subsequent increase in peripheral resistance (2). Similarily, cardiac

hypertrophy in AHR -/- mice appears to be correlated with endothelin-1, angiotensin II,

187

A

and HIFIOL (45-47). Therefore, AHR-HIFIOL interactions may adversely affect heart
and liver development following signaling cross-talk.

The ability of HIFltx and AHR to influence each other’s transcriptional activity
on a genome scale appears to involve multiple factors, and our results suggest that
HREs, DREs, and promoter context play a role in mediating this cross-talk. It is also
likely that promoter context and other transcription factors play a major role in
determining gene expression behavior following hypoxia and/or TCDD treatment. In
fact, the direction and magnitude of change following exposure to either hypoxia or
TCDD compared to the exposure to hypoxia and TCDD is probably more dependent
upon additional transcription factors than previously suspected. Therefore,
understanding the role of cross-talk between hypoxia and the AHR will require a more
thorough characterization of HIFs and the AHR interactions with other transcription

factors and their respective ability to influence basic transcriptional machinery.

188

A

REFERECES

1. Gu, Y-Z., Hogenesch, J ., and Bradfield, C. 2000. The PAS Superfamily: Sensors
of EnVironmental and DeVelopmental Signals. pp 519-561, Academic Press, New
York.

2. Schmidt, J. V., Carver, L. A., and Bradfield, C. A. 1993. Molecular
characterization of the murine Ahr gene. Organization, promoter analysis, and
chromosomal assignment. J. Biol. Chem. 268:22203-22209.

LA)

. Semenza, G. L., Agani, F., Booth, G., Forsythe, J., Iyer, N., Jiang, B.H., Leung,
S., Roe, R., Wiener, C., and Yu, A. 1997. Structural and functional analysis of
hypoxia-inducible factor 1. Kidney Int. 51:553-555.

A

. Reyes, H., Reisz-Porszasz, S., and Hankinson, O. 1992. Identification of the Ah
receptor nuclear translocator protein (Amt) as a component of the DNA binding
form of the Ah receptor. Science 256: 1 193-1 195.

IJI

. Wang, G. L., Jiang, B. H., Rue, E. A., and Semenza, G. L. 1995. Hypoxia-
inducible factor 1 is a basic-he]ix-loop—helix-PAS heterodimer regulated by
cellular 02 tension. Proc. Natl. Acad. Sci. U.S.A. 92:5510-5514.

0\

. Bradfield, C. A., and Poland, A. 1988. A competitive binding assay for 2,3,7,8-
tetrachlorodibenzo-p-dioxin and related ligands of the Ah receptor. Mol.
Pharmacol. 34:682-688.

7. Carver, L. A., and Bradfield, C. A. 1997. Ligand dependent interaction of the Ah
receptor with a novel immunophilin homolog in ViVo. .1. Biol. Chem. 272:11452-
11456.

00

. Ma, Q., and Whitlock, J. P. 1997. A novel cytoplasmic protein that interacts with
the Ah receptor, contains tetratricopeptide repeat motifs, and augments the

transcriptional response to 2,3,7,8-tetrachlorodibenzo-p-dioxin. J. Biol. Chem.
272:8878-8884.

9. Whitlock, J. P., Denison, M. S., Fisher, J. M., and Shen, E. S. 1989. Induction of
hepatic cytochrome P450 gene expression by 2,3,7,8-tetrachlorodibenzo-p-dioxin.
Mol. Biol. Med. 6:169-178.

10. Wang, G. L., and Semenza, G. L. 1993. Characterization of hypoxia-inducible
factor 1 and regulation of DNA binding activity by hypoxia. J. Biol. Chem.
268:21513-21518.

11. Bruick, R. K., and McKnight, S. L. 2001. A conserved family of prolyl—4-
hydr0xylases that modify HIF. Science 294:1337-1340.

189

A

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

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

Berra, E., Benizri, E., Ginouves, A., Volmat, V., Roux, D., and Pouyssegur, J.
2003. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state
levels of HIF-lalpha in normoxia. EMBO J. 22:4082-4090.

Brunelle, J. K., Bell, E. L., Quesada, N. M., Vercauteren, K., Tiranti, V.,
Zeviani, M., Scarpulla, R. C., and Chandel, N. S. 2005. Oxygen sensing

requires mitochondrial ROS but not oxidative phosphorylation. Cell Metab.
1:409-414.

Guzy, R. D., Hoyos, B., Robin, E., Chen, H., Liu, L., Mansfield, K. D., Simon,
M. C., Hammerling, U., and Schumacker, P. T. 2005. Mitochondrial complex

III is required for hypoxia-induced ROS production and cellular oxygen sensing.
Cell Metab. 1:401-408.

Mansfield, K. D., Guzy, R. D., Pan, Y., Young, R. M., Cash, T. P., Schumacker,
P. T., and Simon, M. C. 2005. Mitochondrial dysfunction resulting from loss of

cytochrome c impairs cellular oxygen sensing and hypoxic HIF-alpha activation.
Cell. Metab. 1:393-399.

Mole, D. R., Maxwell, P. H., Pugh, C. W., and Ratcliffe, P. J. 2001. Regulation
of HIF by the von Hippel-Lindau tumour suppressor: implications for cellular
oxygen sensing. IUBMB Life. 52:43-47.

Semenza, G. L., Jiang, B. H., Leung, S. W., Passantino, R., Concordet, J. P.,
Maire, R, and Giallongo, A. 1996. Hypoxia response elements in the aldolase A,
enolase l, and lactate dehydrogenase A gene promoters contain essential binding
sites for hypoxia-inducible factor 1. J. Biol. Chem. 271:32529-32537.

Minchenko, A., Salceda, S., Bauer, T., and Caro, J. 1994. Hypoxia regulatory
elements of the human vascular endothelial growth factor gene. Cell. Mol. Biol.
Res. 40:35-39.

Semenza, G. L., Roth, P. H., Fang, H. M., and Wang, G. L. 1994.
Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-

inducible factor 1. J. Biol. Chem. 269:23757-23763.

Hogenesch, J. B., Chan, W. C., Jackiw, V. H., Brown, R. C., Gu, Y. Z., Pray-

190

22

23

24

25

26

27

28

29

30

Grant, M., Perdew, G. H., and Bradfield, C. A. 1997. Characterization of a
subset of the basic-helix-loop-helix-PAS superfamily that interact with
components of the dioxin signaling pathway. J. Biol. Chem. 272:8581-8593.

. Chan, W. K., Yao, G., Gu, Y. Z., and Bradfield, C. A. 1999. Crosstalk between
the aryl hydrocarbon receptor and hypoxia inducible factor signaling pathways.
Demonstration of competition and compensation. J. Biol. Chem. 274:12115-
12123.

. Gassmann, M., Kvietikova, 1., Rolfs, A., and Wenger, R. H. 1997. Oxygen- and
dioxin-regulated gene expression in mouse hepatoma cells. Kidney Int. 51:567-
574.

. Gradin, K., McGuire, J., Wenger, R. H., Kvietikova, 1., Whitelaw, M. L.,
Toftgard, R., Tora, L., Gassmann, M., and Poellinger, L. 1996. Functional
interference between hypoxia and dioxin signal transduction pathways:

competition for recruitment of the Arnt transcription factor. Mol. Cell. Biol.
16:5221-5231.

. Nie, M., Blankenship, A. L., and Giesy, J. P. 2001. Interactions between aryl
hydrocarbon receptor (AhR) and hypoxia signaling pathways. EnViron. Toxicol.
Pharmacol. 10:17-27.

. Pollenz, R. S., Davarinos, N. A., and Shearer, T. P. 1999. Analysis of aryl
hydrocarbon receptor-mediated signaling during physiological hypoxia reveals
lack of competition for the aryl hydrocarbon nuclear translocator transcription
factor. Mol. Pharmacol. 56:1127-1 137.

. Gentleman, R. C., Carey, V. J ., Bates, D. M., Bolstad, B., Dettling, M., Dudoit,
S., Ellis, B., Gautier, L., Ge, Y., Gentry, J ., Hornik, K., Hothorn, T., Huber,
W., Iacus, S., Irizarry, R., Leisch, F., Li, C., Maechler, M., Rossini, A. J.,
Sawitzki, G., Smith, C., Smyth, G., Tierney, L., Yang, J. Y., and Zhang, J.
2004.Bioconductor: open oftware development for computational biology and
bioinformatics. Genome Biol. 5: R80.

. Ihaka, R., and Gentlemen, R. (1996) R: A language for data analysis and
graphics. J. Comp. Graph. Stats. 5:299-314.

. Wu, Z., Irizarry, R. A., Gentlemen, R., Murillo, F. M., and Spencer, F. 2004.
A model based background adjustment for oligonucleotide expression arrays.
http://www.bioinformatica.unito.it/downloads/ complexity/gcpaper.pdf: 1-29.

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

191

31

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

197.

Quandt, K., Frech, K., Karas, H., Wingender, E., and Werner, T. 1995.
Matlnd and Matlnspector: new fast and versatile tools for detection of consensus
matches in nucleotide sequence data. Nucleic Acids Res. 23:4878-4884.

Sun, Y. V., Boverhof, D. R., Burgoon, L. D., Fielden, M. R., and Zacharewski,
T. R. 2004. Comparative analysis of dioxin response elements in human, mouse
and rat genomic sequences. Nucleic Acids Res. 32:4512-4523.

Burgoon, L. D., Boutros, P. C., Dere, E., and Zacharewski, T. R. 2006. deach:
A MIAME-compliant toxicogenomic supportive relational database. Toxicol. Sci.
90:558-568.

Tajima, F. 1991. Determination of window size for analyzing DNA sequences. J.
Mol. Evol. 33:470-3.

Eckel, J. E., Gennings, C., Chinchilli, V. M., Burgoon, L. D., and Zacharewski,

T. R. 2004. Empirical bayes gene screening tool for time-course or dose-response
microarray data. J. Biopharm. Stat. 14:647-670.

Efron, B., and Tibshirani, R. 2002. Empirical bayes methods and false discovery
rates for microarrays. Genet. Epidemiol. 23:70-86.

Wingender, E. 2004. Transfac, Transpath and Cytomer as starting points for an
ontology of regulatory networks. In Silico Biol. 4:55-61.

Herrero, J., Al-Shahrour, F., Diaz-Uriarte, R., Mateos, A., Vaquerizas, J. M.,
Santoyo, J ., and Dopazo, J. 2003. GEPAS: a web-based resource for microarray
gene expression data analysis. Nucleic Acids Res. 31:3461-3467.

Wang, L., Zhang, X., Farrar, W. L., and Yang, X. 2004. Transcriptional
crosstalk between nuclear receptors and cytokine signal transductionpathways in
immunity. Cell. Mol. Irnmunol. 1:416-24.

Chang, H., Shyu, K. G., Lin, S., Tsai, S. C., Wang, B. W., Liu, Y. C., Sung, Y.
L., and Lee, C. C. 2003. The plasminogen activator inhibitor-1 gene is induced
by cell adhesion through the MEK/ERK pathway. J. Biomed. Sci. 10:738-745.

Ebert, B. L., Firth, J. D., and Ratcliffe, P. J. 1995. Hypoxia and mitochondrial
inhibitors regulate expression of glucose transporter-1 via distinct cis-acting

sequences. J. Biol. Chem. 270:29083-29089.

Yan, S-F., Lu, J., Zou, Y. S., Soh-Won, J., Cohen, D. M., Buttrick, P. M.,
Cooper, D. R., Steinberg, S. F., Mackman, N., Pinsky, D. J ., and Stern, D. M.

192

A

1999. Hypoxia-associated induction of early growth response-1 gene expression.
J. Biol. Chem. 274:15030-15040.

43. Fradette, C., and Du Souich, P. 2004. Effect of hypoxia on cytochrome P450
activity and expression. Curr. Drug Metab. 5:257-71.

44. Harstad, E. B., Guite, C. A., Thomae, T. L., and Bradfield, C. A. 2006. Liver
deformation in Ahr-null mice: evidence for aberrant hepatic perfusion in early
development. Mol. Pharmacol. 69: 1534-1541.

45. Lund, A. K., Goens, M. B., Kanagy, N. L., and Walker, M. K. 2003. Cardiac
hypertrophy in aryl hydrocarbon receptor null mice is correlated with elevated
angiotensin II, endothelin-l, and mean arterial blood pressure. Toxicol. Appl.
Pharmacol. 193:177-187.

46. Lund, A. K., Goens, M. B., Nunez, B. A., and Walker, M. 2006. Characterizing
the role of endothelin-1 in the progression of cardiac hypertrophy in aryl
hydrocarbon receptor (AhR) null mice. Toxicol. Appl. Pharmacol. 212:127-135.

47. T hackaberry, E. A., Gabaldon, D. M., Walker, M. K., and Smith, S. M. (2002)
Aryl hydrocarbon receptor null mice develop cardiac hypertrophy and increased
hypoxia-inducible factor-lalpha in the absence of cardiac hypoxia. CardioVasc.

Toxicol. 2:263-274.

193

APPENDIX B

HYPOXIA, DRUG THERAPY AND TOXICITY

This chapter represents a manuscript that was published in Pharm &Ther. 113: 229-246.
19: 1284-1293 (2006). Authors included: KangAe Lee, Robert A. Roth and John J.
LaPres. '

194

ABSTRACT

Hypoxia is defined as a decrease in available oxygen reaching the tissues of the
body. It is linked to the pathology of cancer, cardiovascular disease, and stroke, the
leading causes of death in the United States. Cells under hypoxic stress either induce
an adaptive response that includes increasing the rates of glycolysis and angiogenesis
or undergo cell death by promoting apoptosis or necrosis. The ability of cells to
maintain a balance between adaptation and cell death is regulated by a family of
transcription factors called the hypoxia inducible factors (HIFs). HIF1, the most
widely studied HIF, is essential for regulating the expression of a battery of hypoxia-
responsive genes involved in the adaptive and cell death responses. The ability of HIF1
to balance these two responses likely lies in the regulation of HIFIO. stability and
transcriptional activity by post-translational hydroxylation and its ability to respond to
other cellular factors including key metabolites and growth factors. Targeting HIF1
signaling for therapeutics, therefore, requires an understanding of how these various
signals converge upon HIF1 and regulate its role in maintaining the balance between
adaptation and cell death. In addition, one must understand how this balance can be
perturbed during toxicant-induced tissue damage. This review will summarize our
current understanding of hypoxia signaling as it applies to drug therapy and toxicity
and describe how these processes can influence the HIF-mediated balance between

adaptation and cell death.

195

INTRODUCTION

The reaction of Complex IV of the electron transport chain (ETC) utilizes
greater than 95% of the oxygen we breathe. This reaction, catalyzed by cytochrome
oxidase, is required for maintaining proper ATP production within the cell. This
coupling of energy production to the consumption of oxygen through cytochrome
oxidase has made the ability to sense and cope with low oxygen tension essential for
survival. Decreases in oxygen reaching the tissues of the body, a condition known as
hypoxia, is detrimental to the cells and tissues because it decreases their ability to
maintain energy production and thereby disrupts their ability to maintain proper
function. In fact, hypoxia and hypoxia-related signaling has been linked to the
pathology of all the major causes of death, including cardiovascular disease, stroke and
cancer. The fact that hypoxia signaling is directly linked to a tumor’s ability to thrive
has made it and several hypoxia-activated genes the focus of intense drug discovery
research. To target hypoxia signaling successfully for therapeutics or to appreciate
fully its role in xenobiotic toxicity, it is necessary to understand the cellular events
following loss of normal oxygen tensions and to characterize the proteins and
pathways involved.

This review summarizes current thinking with regard to causes of hypoxia under
normal and pathophysiological conditions and the programmed response to decreases

in oxygen availability. This includes a detailed discussion of the proteins involved and

196

the cellular consequences of hypoxia-activated signaling, and also outlines a balance
between hypoxia-induced cellular adaptation and death. Finally, this balance and its
role in therapeutics and toxicity are discussed.
2. Hypoxia Signaling
2.1. Normoxia vs. Hypoxia

Oxygen is transported from the air that we breathe throughout the body via a
respiratory system. The primary function of this system is to deliver inspired oxygen to
peripheral tissues and remove carbon dioxide that cells produce as a byproduct of
normal cellular function. When a breath is taken, air passes from the airways into
microscopic air sacs called alveoli. The exchange of oxygen and carbon dioxide
between the alveoli and blood is called external respiration. Although the oxygen
carrying capacity of the blood is increased markedly by the presence of hemoglobin, it
is the gradient of oxygen partial pressure (p02) between the blood plasma and the
locus of oxygen consumption in the mitochondrion that drives 02 into the tissues by
passive diffusion. The p02 in the inspired air is 150 mmHg, whereas the p02 in the
alveoli and arterial blood is around 100 mmHg. In the tissues, oxygen dissociates from
hemoglobin and diffuses through capillary endothelium into parenchymal cells, so that
the p02 of the blood draining tissues (ie, venous p02) is far less than arterial p02,
averaging about 40 mmHg. Thus, the pOz in the cells of tissues is much lower than that

of the arterial blood.

197

The normal partial pressure of oxygen varies in cells within and among different
tissues. For example, the normal p02 of skeletal muscle has been reported to be 20-30
mmHg and that of brain is 45-65 mmHg (33, 66). Even within a single organ, cells are
exposed to a wide range of oxygen tensions in different locations (64). For example,
the p02 of the blood in liver varies by site, so that pOz in hepatic artery, portal vein,
sinusoids and central vein are estimated to be 95-105 mmHg, 50-65 mmHg, 35-45
mmHg and 30-40 mmHg, respectively (21, 64, 94). Although estimates of values for
“cellular p02” have been made in numerous tissues, uncertainty exists ‘due to
limitations of methodologies used for the measurements. Moreover, cellular p02 must
vary with the location of parenchymal cells along the length of the capillary (or
sinusoid), and gradients must also exist intracellulary due to the continuous
consumption of Oz in the mitochondria. Thus, cells in the portal triad, where arterial
and portal venous blood mixes upon entering the liver, are exposed to a more oxygen-
rich environment than cells close to the central vein, since the blood has lost 02 by
diffusion into upstream cells. Nevertheless, each of the cells along the sinusoid
considers the degree of oxygen to which they are usually exposed to be “normal”. How
cells establish or perceive this level of oxygen as “normoxic” is not entirely clear;
however, when the p02 in the tissues and individual cells drops below this normal
level, a state of hypoxia is said to exist.

2.2. Occurrence of hypoxia

198

Hypoxia arises in biological systems for a wide array of reasons, including
normal physiological variation and pathological conditions. During embryonic and
fetal development, hypoxia occurs as the process progresses and the cellular
requirements throughout the developing organism display varying demands for oxygen.
Studies in vitro have demonstrated that the optimum oxygen tension in a mammalian
embryo is approximately 23-38 mmHg, and this oxygen environment is an important
physiological factor for vasculogenesis, angiogenesis, and tubulogenesis (71, 124, 157).
In addition, normal organogenesis in the developing fetus takes place in a relatively
oxygen-poor environment (relative to adult tissues), and increasing oxygen tension
during this time is deleterious for heart, kidney, and lung development (74, 140, 157).

Hypoxic conditions can also occur in the adult under various conditions. For
example, during exercise muscle demand for energy can outpace the supply of oxygen,
causing localized hypoxia. This type of stress leads to a buildup of lactic acid as the
body turns to anaerobic metabolism to address the energy debt. These normal
biological processes highlight the balance that is established between oxygen levels
and energy production and the importance of a programmed response to disruptions in
this balance. Imbalances can occur either through reduced oxygen availability, as in the
developing embryo, or increased energy demand, as in the exercising muscle. More
importantly, this balance is also central to hypoxia’s role in pathological conditions.

Recent interest in hypoxia and its effects in biological systems has stemmed

199

A

from its role in a wide variety of pathological conditions, most notably cancer.
Hypoxia is an important feature of most solid tumors, arising from the process of
tumor formation and progression that is characterized by rapid cell growth and drastic
changes in the local oxygen supply. Rapid cellular expansion quickly outpaces the
ability of a tumor to create new blood vessels for oxygen delivery, leading to a hypoxic
or even anoxic microenvironment within certain portions of the tumor. The hypoxic
microenvironment occurs very early during tumor development, beginning as the
tumor reaches approximately 2-3 millimeters in diameter. To this the tumor must
respond, primarily by increasing its glycolytic rate and angiogenic potential (53, 54,
119). The resultant newly formed vasculature differs from that in normal tissues,
displaying abnormal structure and function. For example, the new tumor vessels are
highly disorganized and usually leaky, with incomplete endothelial lining, qualities
that lead to irregular blood flow and diminished nutrient and oxygen supply (22, 69).
Newly developed vasculature does not always address the oxygen debt of the tumor
since many portions remain further than 150 pm from the nearest blood vessel, which,
due to tissue 02 consumption, is the approximate diffusion limit of oxygen (22).
Hypoxia, therefore, remains a constant feature of tumors even after neovascularization.
For many years, tumor hypoxia has been considered a challenge for cancer therapy
because of its adverse impact on the effectiveness of radiation and chemotherapy.

Moreover, low oxygen availability has recently emerged as a major factor that

200

enhances malignant progression, because tumor cells that have adapted to hypoxia gain
various advantages in growth. In most cases, this is a sign of poor prognosis (9).
Disruption in oxygen homeostasis also occurs during other pathophysiological
conditions, such as cardiovascular disease, stroke, and chronic pulmonary disease
(113). Atherosclerosis often leads to arterial stenosis which ultimately impairs
perfusion of the vascular bed and further disrupts oxygen and nutrient flow to regions
of the heart muscle (110). When this ischemia persists, it can irreversibly reduce
myocardial viability (105, 113). Stroke is a well known cause of cerebral hypoxia
because it disrupts the normal flow of blood in the brain and leads to localized loss of
oxygen and nutrient supply. In brain tissues, even if this disruption only lasts for a
short period of time, it can lead to neuronal cell death and disability (60, 113). Chronic
obstructive pulmonary disease (COPD) is the most common form of pulmonary
dysfunction and can be subcategorized as asthma, chronic bronchitis, and emphysema.
Alveolar hypoxia is common in COPD because of the difficulties in expelling air from
the lungs, and it can lead to pulmonary hypertension, pulmonary arteriolar remodeling,
and ultimately right heart failure (154, 155). The hypoxia-induced injury and tissue
remodeling seen in these pathophysiological conditions and others, including diabetes
and wounding, can be considered a response to the inability to maintain the balance
between oxygen delivery and energy production. The injurious process takes place

because the tissue cannot adapt to the change in oxygen tension. As will be discussed

201

later in this review, toxicant-induced cellular damage that involves the hypoxia
signaling system can also be considered in a similar way.
2.3. The hypoxia-inducible factor (HIF) system

Aerobic metabolism provides a significant advantage to multicellular organisms,
especially in the area of energy production. When oxygen supply cannot meet the
energy demand of the cells, however, an adaptive response must compensate for the
energy imbalance to maintain tissue function. A primary mechanism for this adaptive
response is the transcriptional regulation of a battery of hypoxia—responsive genes. In
fact, genomic screens have shown that thousands of genes are influenced by exposure
to hypoxia (127, 143, 144). The most well studied mechanism identified in this process
is an interaction of a family of transcription factors, called hypoxia-inducible factors
(HIFs) with a cis-acting element, called the hypoxia-responsive element (HRE),
located in regulatory regions of target genes. This mechanism was first demonstrated
for the erythropoietin gene, the expression of which was upregulated more than 100-
foldl by hypoxia via HIF-induced transcription (6, 96). Since then, more than 70 target
genes regulated directly by HIFs have been identified, and expression of over several
hundred genes are known to be directly or indirectly influenced by HIF.

The family of HIFs belong to the Per-ARNT-SIM (PAS) superfamily of
transcription factors is characterized by the presence of the PAS domain that controls

dimerization (55, 56, 73). HIFs are heterodimeric proteins comprising an 0t and

202

B subunit. The alpha class is composed of HIF 1a, HIF20t, and HIF30L (for the purpose
of this review, thesewill be collectively referred to as HIFOL). The beta subunit includes
the aryl hydrocarbon nuclear translocator (ARNT, also known as HIFIB) and ARNT 2.
HIF1 is the most widely studied HIF heterodimer and is the combination of HIF 101 and
ARNT (146). HIF1 is the principal regulator of the hypoxic response in most
mammalian cells (116). ARNT and ARNT2 are constitutive nuclear proteins, whereas
the expression and activity of the Qt subunit is tightly regulated by oxygen
concentration. It rapidly accumulates upon exposure to hypoxia and on reoxygenation
is quickly degraded with a half-life of less than 5 minutes (153). Given how
detrimental hypoxia is to a cell, this rapid response time highlights the speed at which
a cell must elicit a reaction to loss of oxygen tension. Equally important is the short
half life of the HIF protein following reoxygenation, and this time frame suggests that
prolonged HIF activation might not be beneficial to the cell. Our understanding of the
mechanism a cell uses to sense hypoxia and convey this signal to the alpha subunit has
greatly increased over the last several years. This is due to the recent identification of a
family of hydroxylases that modify the CL subunit in an oxygen-dependent manner.
Three hydroxylases, known as prolyl hydroxylase domain containing proteins
(PHDs, also known as egg laying abnormal (EGL) 9 homologues [EGLNs] and
hypoxia prolyl hydroxylases [HPHs]), control HIF10t stability by modifying conserved

prolines within the oxygen-dependent degradation domain (ODD) of HIFOL under

203

A

Figure B-l. PHD-mediated hydroxylation of HIF1

Prolyl hydroxylase (PHD) requires oxygen, iron, ascorbate and (x-ketogluturate for
activity. During the enzymatic process, OL-ketoglutarate is decarboxylated, yielding
succinate, and the HIFla ODD is hydroxylated on proline residues. Once
hydroxylated, HIF 1a is quickly degraded in a VHL—ubiquitin-26S proteosome-
dependent manner.

204

 

 

Oxygen
Fe2+
ascorbate
a-ketoglutarate

\Jsuccinate

VHL
ubiquitin

/\
\ \°“ 5

'3 Q
HIF1

205

normoxia (Fig. B-l) (10, 34). The proline hydroxylations are required for the
interaction of HIFoc with the von-Hippel Lindau tumor suppressor gene product
(pVHL). pVHL serves as the recognition component of E3 ubiquitin-ligase that leads
to HIFOL ubiquitination and proteosomal degradation (10, 34). The transcriptional
activity of HIFlot is also reduced under normoxia by oxygen-dependent hydroxylation
of asparagine 803 within the C-terrninal transactivation domain (CTAD) of this
transcription factor (82). Factor inhibiting HIF (FIH) mediates this modification of the
conserved Asn residue, and the hydroxyl-asparagine prevents the binding of co-
activator p300/CBP to HIFla (72). This effectively removes the ability of HIFIOL to
direct p300-dependent transcription; however, it remains possible for the transcription
factor to modulate transcription in a p300-independent manner, though this possibility
has not been explored. These hydroxylases require four things for activity, oxygen,
iron, a-ketoglutarate (OLKG) and ascorbate. The enzymic process involves
decarboxylation of the aKG to yield succinate and concomitant hydroxylation of the
targeted residue (Fig. B-l). The oxygen, iron and enzymic reaction products play a
critical role in regulation of the hydroxylase.

The hydroxylation of HIFIOL, both PHD and FIH mediated, are inhibited under
low oxygen conditions, leading to HIF 10L accumulation and consequent induction of
HIF1 activity (10, 34, 63, 72). Currently, there are two theories as to how the PHDs

sense the decrease in cellular oxygen tension (Fig. B-2). First, since these enzymes

206

Figure B-Z. The hypoxia signaling system

The hypoxia signaling cascade begins in the cytosol where the PHDs are continuously
responding to local oxygen concentrations. Once the available oxygen concentration
reaches some critical point, the PHD becomes inhibited. This can involve direct
inhibition of the PHD due to loss of its molecular oxygen substrate or indirect
inhibition due to changes in reactive oxygen species (ROS) produced at complex III of
the electron transport chain (ETC). This change in ROS is caused by the inhibition of
the ETC due to decreased oxygen, the terminal electron acceptor at complex IV, and
subsequent accumulation of ubisemiquinone at complex 111. Once the PHD is inhibited,
HIF 101 translocates to the nucleus where it interacts with ARNT and becomes
transcriptionally active. HIF1 target genes include adaptive and cell death genes. The
cellular increase in anaerobic metabolism leads to the overproduction of lactate and a
decrease in cellular and local tissue pH.

207

 

 

   

 

@2032 j
984.

 

nmn

 

 

  
  

 

 

 

39.333338
23303
ac".
<_xon_>: ll_ 52.6

1‘

5560 =60

» £06322

m_me_ono..£bm
«_mocomo_m:<

 

 

 

, l

.8650

       
    

$33.0

«
. w.
_ “/1 “Ea

 

 

$9.320
ldlcozs 6.2

 

1.2.1. .... .
...fluwn x..
...4......

\ 82.325

..2 ..o< o“... No

«

393:5 £2323

+1 AIL...

 

4:
L .
...2 __. , ..

 

 

 

 

        
       

{
Ohm

->’

catacosoogE

CC

 

 

 

 

 

208

require oxygen for activity, hypoxia acts as a loss of substrate, and this might be all
that is required for PHD inhibition. The loss of available oxygen would inhibit the
enzyme’s ability to modify HIFIOL and thereby lead to increased transcription factor
stability. The disparity between the oxygen requirement for these enzymes and the
level of hypoxia needed to induce HIF 10L stability in cells has led others to consider
alternative mechanisms. In enzyme preparations, a progressive decrease in PHD
activity from just below 20% Oz induces an increasing stability of HIFIa. In contrast,
HIF10t stability is not increased in most cell types until the oxygen levels are
decreased to less than 7% 02 (34, 139). In the stabilization of HIFla, several
laboratories have reported a role for reactive oxygen species (ROS) generated in the
mitochondria. These experiments have documented an inability of mitochondria-
deficient cells to induce HIF 1a stabilization during hypoxic stress, suggesting a role
for this organelle in the process (15). Recent reports have elaborated on this,
establishing complex III as the likely site of action(ll, 16, 51, 83). The proposed
mechanism accounts for evidence implicating the mitochondrial respiratory chain in
regulation of HIF activity (Fig. B-2). In respiration, electrons donated by NADH and
FADHz flux through complexes I-IV of the electron transport chain (ETC) and are
finally transferred to molecular oxygen at complex IV. Hypoxia causes ETC inhibition
and the generation of ROS from a number of potential sites, such as complex I and the

ubisemiquinone site of complex III. It has been proposed that this oxidative stress

209

within the cell disrupts the catalytic activity of the PHDs, possibly through inhibiting
the ability of iron to cycle between oxidation states. In recent years, it has become
evident that changes in ROS levels within the cytosol results in PHD inhibition,
ultimately leading to the induction of HIF transcriptional activity (17, 48, 52).
Regardless of which mechanism is involved in the inhibition of the PHD, hypoxia
leads to HIF stabilization, and this begins the cell’s attempt to adapt to the decrease in
available oxygen.

Hypoxia is not the only signal that can lead to HIF10t stabilization. It has long
been known that a number of chemicals, such as cobalt, deferoxamine and the (xKG
analog, dimethyloxaloglutarate (DMOG), can induce the stabilization of
HIF1 . experimentally (34, 84). They act by directly inhibiting PHD activity by either
removing or competing with iron or (XKG. More recently, a direct link has been
established between key metabolites and HIFIOL stabilization. As discussed above,
PHDs require (xKG for activity and, during the catalytic process, generate succinate.
Recently, people with mutations in their succinate dehydrogenase enzyme (SDH,
complex 11 of the ETC) have been demonstrated to display characteristics of a pseudo
hypoxic response (8, 27, 111, 112). These people have a gene expression profile
similar to that induced by hypoxia in normal humans, and affected people are prone to
pheochromocytomas. In addition, people with mutation in fumarate hydratase (FH), a

citric acid cycle enzyme, also display signs of aberrant hypoxia signaling (59). People

210

with FH mutations are also prone to certain types of cancer, and presumably this
involves direct inhibition of the PHD through modulation of the decarboxylation step
within the enzyme. The mutation in SDH would tend to increase intracellular
concentrations of succinate, leading to PHD dysfunction via product inhibition (Fig. B-
3). The FH mutation would be expected to lead to increased fumarate levels within the
cell, and fumarate can act as a competitive inhibitor for the ocKG binding site in PHD.
Finally, the glycolytic end product, pyruvate, regulates HIFIOL stability, and this
activation might be responsible for the Warburg effect (77).

In the early part of the last century, Otto Warburg described a tumor’s increased
dependence upon fermentation or anaerobic glycolysis (148). He suggested that the
increased glycolytic activity and metabolic adaptation might be the reason for the
cellular transformation. The demonstration that pyruvate can activate HIFla raises the
possibility that it can participate in feed-forward activation of HIFIOL, leading to a
prolonged hypoxic signal. Recent reports have shown that the pyruvate dehydrogenase
kinase (PDK) gene is a HIF1 target gene (67, 143). PDK is responsible for inactivating
the pyruvate dehydrogenase complex, thus trapping pyruvate in the cytosol. This
would further serve to prolong the hypoxic signal by maintaining the cytosolic
concentration of pyruvate. The resultant extended signal might explain the Warburg
effect and directly links HIFIOL activation to increased glycolytic activity. These

combined results suggest that a better characterization of the metabolic state of the cell,

211

Figure B-3. Link between intermediary metabolism and HIF1 signaling

Recent reports have established a direct link between key metabolic intermediates and
HIF1 signaling. These metabolites include a-ketoglutarate (Qt-KG), which is necessary
for PHD activity. Recently, it was demonstrated that patients with mutations in
succinate dehydrogenase (Complex II of the ETC) display characterics of a pseudo-
hypoxic state. Another link was established between fumarate disregulation and HIFl
signaling, and it was suggested that fumarate competes for (x-KG binding within the
PHD enzyme. Finally, pyruvate acts in a feed-forward mechanism to stabilize HIFIOL
in the absence of hypoxia. The recent characterization of pyruvate dehydrogenase
kinase (PDK) as a HIF1 target gene might amplify this linkage between pyruvate and
hypoxia signaling, since PDK over-regulation would serve to trap pyruvate in the
cytosol.

212

 

 

 

 

 

@2032

 

 

 

 

  

 

 

 

 

 

 

\lovuafiogm

 

 

 

33:6
8..
w £3322 3003.0
.383 :8 «
m_mo_oao._:tcw w {v
£35329? “ n_._.<
3.620020 «

:O_umumm—u< 230m.— ”053:?“

  

 

 

    

T238

‘ O: I .9: I 6.:

»

> .2 I B
__

No

 

     

 

     

..2. .22 of

      

 

 

33.33.323.83 .

 

‘ sflp-BOq-W 88g: n— a
' (001003” Ouflfiflz . .
v
32828“? (No
8.5.08. l l 3.58

 

. 4... . A... w. .. u. 3 m.» x. ”.3 ..r...

. .1 . .u r f l. . ..u w .7. r, .
v. .v» \ J\.\ . . U .rl. 0L“ rt. 3 al..» -J

.u A.“ l A... 3 lb 1. l

l i u .-w . . 6 vi 1. 9|.
3 a . 1.» 111 Z lav 4b in

213

as well as oxygen and PHD levels, is required if the HIF signaling cascade is to be
understood and targeted successfully by therapeutic agents.

Other modifications, in addition to hydroxylation, have been linked to hypoxia
signaling. Acetylation of HIFla at lysine 532 by arrest defective-l (ARDI)
acetyltransferase has been suggested to regulate HlFlor stability by enhancing the
interaction between hydroxylated HIFOL and pVHL; however, these interactions might
be cell type specific (63,40). Phosphorylation is also involved in regulating HIF
activity. During hypoxia, p42/p44 mitogen activated protein kinase (MAPK)
phosphorylates HIFIOL and enhances the transcriptional activity of HIF1 (97, 109).
HIF activity is also regulated in an oxygen-independent manner at the HIFla
expression level by certain growth factors that ensure the maintenance of oxygen
homeostasis in normal growing tissues. These responses result from the activation of
signaling pathways involving phosphoinositide 3-kinase (PIBK)/AKT/mTOR that lead
to the increases in the translation factor elF-4E, which in turn enhances HIF 10L mRNA
translation (109, 117). In summary, there are many ways in which HIF activity can be
regulated, some of which are oxygen-dependent and others of which are not. The
relative importance of these various conditions on normal and pathophysiology
remains to be determined.

2.4. The integrative response to hypoxia

Multicellular organisms have developed sophisticated physiological

214

infrastructures to maintain oxygen homeostasis. Hemoglobin, a major constituent of
red blood cells, is responsible for carrying oxygen to peripheral tissues, and the
number of circulating erythrocytes is a major determinant of tissue oxygenation (12,
114). Decreased oxygen availability results in compensatory stimulation of
erythropoiesis by upregulating the production of erythropoietin (EPO) (114). EPO
enhances the production of red blood cells, and its hypoxia-induced transcription is
regulated by HIF through an HRE site in the 3’ flanking region in the EPO gene (6, 96).
The hypoxia-induced expression of the EPO gene is thought to be part of a systemic
response that an organism initiates to cope with the decrease in available oxygen.
Angiogenesis is a tissue-oriented response to hypoxia and is critical to restoring
oxygen transport to ischemic areas by improving blood supply to the affected tissues.
Vascular endothelial growth factor (VEGF) is a central factor controlling physiological
and pathological angiogenesis, and, in most cases, it is induced in response to hypoxia
(138). Similar to EPO, the expression of VEGF is regulated by binding of HIF to
HREs located in the 5’-flanking region of VEGF (125). In addition, VEGF receptors
and the proangiogenic cytokine interleukin-8 are upregulated in response to hypoxia
(37, 88).

The cellular response to hypoxia is focused on addressing the energy deficit
created by the decrease in available oxygen. When aerobic metabolism is diminished

due to the lack of molecular oxygen, the cell utilizes the HIF signaling cascade to

215

increase the expression of the complete battery of glycolytic enzymes, since it is the
one place the cell can turn to address its energy debt. The glycolytic process is not as
efficient as the ETC and ATP synthase in producing ATP, generating only 2 moles of
ATP per mole of glucose, and this inefficiency can lead to impairment of physiological
function (115). The increased dependence upon anaerobic glycolysis is coupled to the
upregulation of various glucose transporters and other growth factors in an attempt to
supply the cell with the necessary catabolic reagents to carry out this adaptive response
(18). The importance of HIF in this switch to anaerobic metabolism has been well
demonstrated with in vivo and in vitro models. Cells with a loss of HIF activity display
reduced hypoxia-induced expressions of 13 different glucose transporters and
glycolytic enzymes, compared with their wild type counterparts (18, 60, 144).

Finally, this adaptive response also involves HIF1-mediated transcription of the
PHDs themselves (7, 20, 35, 144). This ability is PHD isoform- and cell type-specific.
PHDZ and PHD3 are upregulated in response to hypoxia in a wide variety of cells,
whereas PHD1 is only marginally regulated by loss of oxygen tension in most cells.
There has been speculation that this HIF1-mediated upregulation of PHD functions as
a shut-off mechanism for the hypoxic response once normal oxygen tension is restored.
This might be one role for such feedback inhibition, however, there may be a more
subtle role for this regulatory loop, i.e. that this feedback loop actually establishes the

set-point for “normoxia” within the various cell types of the body.

216

The cells in various tissues and even within a single tissue are exposed to a wide
range of oxygen tensions. For example, in the liver cells close to the portal vein and
hepatic artery are exposed to a greater oxygen tension than those near the central vein,
and yet each cell type perceives the oxygen concentration to which it is exposed as
normal. It is possible that the decreasing oxygen concentration along the sinusoid leads
to a partial activation of HlFl-mediated transcription of the appropriate PHD. The
increased PHD levels along the sinusoidal oxygen gradient could compensate for
reduced delivery of oxygen, similar to the way an increased receptor concentration can
compensate for lack of available ligand in signal transduction. In this manner, the
HlFlzPHD-linked transcription can establish the required level of HIF1 signaling
necessary for the cell’s microenvironment and normal cellular function.

2.5. The balance between adaptation and cell death

The adaptive response, including altered PHD levels and upregulation of EPO,
VEGF and anaerobic glycolysis, is not always capable of addressing the hypoxia-
induced metabolic imbalance. In these cases, the cell begins modulating various cell
death pathways presumably as a mechanism to eliminate irrecoverably stressed cells
(22, 123). This pathway involves HIF1-dependent transcription of various pro-
apoptotic B-cell chronic lymphocytic leukemia (CLL)/lymphoma (BCL) family
members, such as BCL2/adenovirus ElB 19kDa interacting protein 3 (BNIP3) and

NIX (32, 93, 144). The activation of HIF is also able to contribute to the p53-mediated

217

cell cycle arrest and cell death by stabilizing p53 caused by cross-talk with alpha HIFs
(1, 14). This increased “suicide” response seems in opposition to the adaptive response
described above. It is currently thought that this pathway to cytotoxicity is an attempt
by the cells to maintain the tissue as a whole. Removal of cells under severe hypoxic
stress by programmed death could increase the chance of survival for neighboring cells
by increasing nutrient and oxygen supply and maintaining appropriate tissue
architecture.

One major, unanswered question concerning hypoxia is how this balance
between adaptation and cell death is established and maintained. How does a cell know
when it should survive through adaptive measures or when it needs to die because it
cannot cope with the hypoxic environment? The intracellular regulatory system that
makes this decision probably involves numerous factors, including oxygen levels of
the cell, metabolic state, levels of key metabolites and growth factors, and the
endogenous activity of the PHDs. The latter can serve as a master regulator of a cell’s
fate under hypoxic response by the same mechanism that it uses to establish normoxia
(Fig. B-4). For example, under normal oxygen concentrations, PHD and glycolytic
enzymes are at maintenance levels, i.e., their levels are maintained at those required
for normal cellular function and oxygen-sensing capacity.

When the cell is exposed to moderate hypoxia, the expression of PHDs and

glycolytic enzymes is increased to cope with the detrimental environment (Fig. B-4).

218

Figure B-4. Relationship between oxygen concentration and HIF1 target gene
expression

At normal oxygen tensions, the PHDs are capable of initiating the degradation of
HIFIOL and expression of HIF1 target genes, including glycolytic enymes, PHDs and
apoptotic genes, are maintained at low levels (top panels). As the oxygen
concentration becomes moderately decreased, PHD activity is diminished for lack of
available substrate and some HIF1 accumulates and drives the expression of target
genes. These genes include PHDs and glycolytic enzymes, levels of which will be
recalibrated to the new metabolic conditions and thus encoding a new “normoxia”
(middle panels). Finally, under severe hypoxia, PHDs activity is incapable of
compensating for the loss of oxygen and HIF1 signaling becomes hyper-activated,
leading to expression of pro-apoptotic factors and cytotoxicity (bottom panels).

219

 

.0280... :E 3:0 Qt.
00:800..
0.333030 :30... :3 0 3
@5000. 03005an0 09
30::00 :0m>xo 0.n0=0>0
:. 0000.00: 090..

.338 :E 52505010.
00:2.03
035300 0...: “0:0 0.0>0.
DIn. :_ 00000.05 >2 .5“—
u300:0n.:._00 0. 5:05.
:0m>xo :_ 0000.000 =0Em

.9300 :E 3:0 9E.
0.0>0_ .0000 .0
305350.: 0.0 00:03 u_0
00000.0 00:5 05 :o_0:3
combs .350: 00::

No: 5.2.80

 

 

 

038%....
03:29:.
:0 0L0>0m

 

 

 

 

0.0802.
2.5.

 

 

 

 

0.33:2: I
0.320030 ......
9...". I I.

sdnorg sues aarql J0 uogssardxg

 

 

 

   

        

0.0.9502

 

auMzua GHd an: 10 KiiAnov

220

It is possible that cell death factors are upregulated in the short term; however, once
the PHD activity is increased in the cell and the energy balance is restored, these genes
are turned off and cell death will be avoided. It is likely that these genes are not
expressed under moderate hypoxia from which the cell can recover. If so, this would
suggest that HE‘l-target genes require complex transcriptional regulatory mechanisms.
It can be hypothesized that these transcriptional mechanisms could involve the
different oxygen requirements of the PHD and FIH enzymes, as PHD requires a
greater oxygen concentration for function than FIH (Fig. B-5) (130). This raises the
possibility that, at certain lowered oxygen tensions, HIFIOL could be stabilized because
the PHD is inhibited but unable to interact with p300, while FIH is still capable of
hydroxylating the critical asparagine (Fig. B-5). This would establish a stable HIF10t
transcription factor that regulates p300-independent transcription, thereby serving as
the selective switch. Indeed, using p300 mutant mice, Kasper et al. demonstrated that
HIFla is capable of modulating transcription in a p300-independent manner (65). In
further support of this model, Dayan et al. have recently established that FIH regulates
distinct genes sets and characterized a number of genes that are susceptible to oxygen
dose-dependent transcription (30).

Under severe hypoxic stress, increased HIF1-mediated transcription of PHDs is
unable to compensate for the loss in available oxygen (Fig. B-5). At these oxygen

levels, PHDs and FIH are inhibited and this leads to HIFIa-mediated regulation of its

221

..I. .I.. .07."..M'ni

 

Figure B-5. HIFla stability and transcriptional activity is controlled by
hydroxylation

HIFIO. contains conserved prolines (Pro) and asparagines (Asp) that are important for
its regulation. Prolyl hydroxylation, via PHD, controls the ability of VHL to direct
HIFIOL degradation (top). PHDs require more oxygen than FIH for activity. Therefore,
it is possible that HIF10t could be found in a stable form (with asparagine
hydroxylation but without proline hydroxylation) and unable to bind to p300 (center).
Finally, once the hypoxia becomes severe, both FIH and PHD may be inhibited, and
HIFla would become fully activated (bottom).

222

 

 

OH +
W it... Degradation
Normoxia
PHD and FIH ® 0"
active
Moderate Hypoxia
PHD inactive Limited
® FIH active ® Transcriptional
: activity
1W- ® ® 0" (p300-independent)

0
“21; 1mm
[0 PHD and 1k ® 1' P300 Full
Inacflve —' Transcriptional
v A- ® Activity

until

223

complete library of target genes, including those expressing cell death factors (Fig. B-
4). This unmitigated expression leads to cell death. Obviously, this concept is
simplified since the HIF-mediated adaptive responses are probably more predominant
in certain tissues and cell types under various environmental conditions. For example,
macrophages seem to survive and function in hypoxic tissue under conditions in which
death of parenchymal cells occurs, and this ability to survive depends on adaptive
responses mediated by HIFs (13, 25). Conversely, hypoxia-stimulated cell death may
be more important beyond the boundary of oxygen deprivation. Understanding the
factors involved in how HIF signaling establishes this balance is critical to our ability
to manipulate the system in a meaningful therapeutic sense. Drugs that target this
pathway must consider the target tissue and the factors involved within this tissue in
establishing this balance. Finally, this balance will also be critical to toxicity of any
compound that disrupts blood flow and to the pathology of any disease, such as cancer,
that unnaturally creates a hypoxic environment.
3. Hypoxia and pharmacological implications

The role of hypoxia and HIF1 in cancer, stroke and cardiovascular disease has
made it an attractive target for drug design. As discussed above, however, the HIF1-
regulated balance between adaptation and cell death contains many inputs, such as
PHD activity, oxygen concentration and metabolite concentrations. Targeting this

pathway for viable drug candidates must take these inputs and their effects on this

224

balance into consideration. For example, given that HIF1 is upregulated in various
cancers, inhibiting HIFl signaling directly might eliminate the ability of tumors to
promote an adaptive response. Conversely, patients at risk of stroke-induced
hypoxia/reoxygenation injury might benefit from HIF1 activation to promote adaptive
or neuroprotective effects. Moreover, the input signals that regulate HIF1 activity are
likely to vary among tissues and even among cells within the same tissue. Finally,
though HIF1 is the focus of most of the research related to this field and the focus of
this review, it can be assumed that HIF20t and HIF30L are also targeted by these same
pharmaceutical paradigms. A successful drug candidate must be able to cope with
these differences and still give the expected pharmacologic outcome within the target
tissue.
3.1. Hypoxia-related drug therapy

Most investigations have focused on targeting hypoxia and the HIF signaling
cascade for cancer therapeutics. The goal of these new therapies is to manipulate the
HIF signaling pathway to inhibit tumor growth or to exploit the hypoxic
microenvironment of a tumor to make the therapies more specific and efficacious.
There are four major areas of research in this field: 1) designing drugs that directly
inhibit HIF1 signaling, 2) influencing other signaling cascades that indirectly alter HIF
signaling, 3) exploiting the hypoxic microenvironment to increase specificity and

decrease toxicity, 4) altering regulation of HIF target genes that are critical for tumor

225

growth or for the function of their protein products (Table B-1).

Designing drugs that directly inhibit HIF1 signaling is not as easy as targeting
enzymatic processes. There are five principal ways that HIF1 can be selectively
targeted by therapeutics. The first direct mechanism involves specifically inhibiting
HIF10t transcription or translation (75, 133). PX-478, a drug developed by ProlX
Pharmaceuticals is billed as the first drug that selectively targets HIFIOL and has shown
efficacy in antitumor assays (81, 150). The second direct mechanism is to inhibit
HIFla’s ability to translocate to the nucleus. Currently, there are no drugs that have
exploited this avenue, and this is most likely because the process is not well
understood. Third is to block HlFla’s ability to interact with ARNT. This strategy has
been used experimentally in the design of HIF-dominant negative constructs. These
constructs lack a transactivation domain but are still capable of binding ARNT. This
sequesters the ARNT pool in a transcriptionally deficient complex, thereby inhibiting
normal hypoxia-induced gene transcription (41). Eliminating HIFI’s ability to bind
DNA would eliminate its ability to regulate its target genes transcriptionally (98).
Echinomycin, a quinoxaline family antibiotic, inhibits HIFIOL DNA binding; however,
it has not been determined if this is the only effect of echinomycin that makes it an
attractive cancer chemotherapy test compound, since it also induces apoptosis in
certain cell types (68). Finally, inhibiting HIFS ability modulate transcription by

blocking its ability to interact with the basic transcriptional machinery would also

226

Table B-1. Drugs that target the hypoxia signaling cascade

 

 

Class of drug Name Action
PX-478 decreases HIF expression
Direct inhibitor Echinomycin inhibits DNA binding
103D5R decreases HIF expression
Dominant negative inhibits ARNT binding
siRNA decreases HIF expression
Chetomin inhibits HIF 101/p300 interaction

 

Indirect inhibition

BAY-43-9006 (Sorafenib)
Berberine

CCI-779 (Temsirolimus)
Dibenzoylmethane and others
dimethyloxaloyl glycine
(DMOG)

FG-4095

Flavopiridol

Geldanamycin, Apigenin,
Radicicol and others

081 774 (Tarceva(erlotinib))
PD98059

PS-341 (Bortezomib)

SN38 (CPT-ll (irinotecan))
STIS71 (Gleevac (Imatinib))
trastuzumab (Herceptin)
Trichostatin A

Vincristine

YC-l

ZD-1839 (Iressa®

or gefitinib)

Raf Kinase inhibitor increased
HIF1 or protein degradation
mTOR inhibitor

iron chelators

non-selective prolyl

hydroxylase inhibitor

selective PHD inhibitor
cyclin-dependent kinase inhibitor

Hsp90 inhibitors

EGFR inhibitor

MEK inhibitor
proteosome Inhibitors
topo I inhibitor

c-KIT inhibitor

HER2 inhibition
histone deacetylase
microtubule disruption

soluble guanylate cyclase activator
EGFR inhibitor

 

Hypoxia-

regulated genes

2-Methoxyestradiol
AGO 13736
Bevacizumab (Avastin)

227

anti-angiogenic
VEGFR inhibtor
VEGF inhibitor

CBP-7055 VEGF inhibitor
CP-547,632 VEGFR inhibtor
hypericin anti-angiogenic
PTK787 (Vatalanib) VEGFR, c-Kit, PDGFR inhibitor
SCH66336 , _ ,

, , , anti-angiogenic
(farnesyltransferase 1nh1b1tor)
SU11248 (Sutent) VEGFR, c-Kit, PDGFR inhibitor
ZD6474 (Zactima) VEGFR, EGFR inhibitor

 

2-Cyclopropylindoloquinones hypoxia selective cytotoxin

Hypoxia— AQ4N hypoxia selective cytotoxin
activated KIN-847(2-nitroimidazole hypoxia selective cytotoxin
agents 1 -acetylhydroxymate)
KS 1 19 hypoxia selective cytotoxin
NLCQ-l hypoxia selective cytotoxin
Quinoxaline 1,4 dioxides hypoxia selective cytotoxin
Tirapazamine hypoxia selective cytotoxin
TX-402 hypoxia selective cytotoxin

 

228

”—7. . “Jun-all"

serve to eliminate hypoxia-mediated activity. For example, Chetomin, a natural product
from the fungus Chaetomium, disrupts the interaction of HIF 1a with p300 (Kung et al.,
2004). Therapies that directly influence HIF signaling comprise a relatively small
category for hypoxia-centered drug design.

The second and largest category of drugs that target hypoxia signaling influence
HIF1 signaling indirectly. Most of these compounds were not originally intended to -
influence HIF1 signaling but have been found to modulate HIF1 signaling at some
level. Recent work in this area has surrounded mammalian Target of Rapamycin
(mTOR) and its role in HIF10t translation. Inhibition of mTOR by CCl-779 decreases
HIF10t and other protein levels and is linked to decreased growth in certain cancer
models (46, 135). This indirect-acting class also includes compounds that influence
HIF1 by inhibiting Hsp90 function. Geldanamycin, apigenin and others can increase
HIFltx signaling at low doses but decrease HIF1-mediated transcription at larger
concentrations (36, 58, 79, 87, 91). The results suggest that Hsp90 might play a crucial
role in several different facets of HIF1 signaling, and the success of targeting Hsp90
for cancer therapy might depend upon how much drug can be delivered to target (57).
The final group of drugs that fall within this indirect category are the prolyl
hydroxylase inhibitors. Like Hsp90 inhibitors, these drugs increase HIF1 signaling and
include direct prolyl hydroxylase inhibitors, such as dimethyloxaloylglycine, and iron

chelators, such as dibenzoylmethane (3, 80). These drugs increase HIF1 signaling by

229

removing the enzymic activity necessary to direct HIF 101 degradation. The
hydroxylase inhibitors work by competing for the a-ketoglutarate binding site within
the PHD enzyme, whereas iron chelators remove the metal necessary for catalytic
activity. These types of inhibitors are neuroprotective in models of
hypoxia/reoxygenation injuries (31, 158).

Though this class of indirect inhibitors is currently the largest, it is not the one
making the most impact for patients. Drugs designed to inhibit HIF1 target genes have
gained FDA approval and are currently in various phases of clinical trials, alone or in
combination therapy, for the treatment of a wide range of cancer types. These drugs
focus on the tumor’s dependence on angiogenesis for oxygen and nutrients. Several
critical growth factors necessary for this vascularization response are hypoxia target
genes. Most notable is VEGF. The most widely recognized of these drugs is
bevacizumab (Avastin), a recombinant antibody designed to inhibit VEGF function
(39). Bevacizumab is currently the only approved drug in this class and has shown
promise in the treatment of colon cancer but only limited efficacy in breast cancer
when used as a single agent. Efficacy seems to be consistent across this class of
compounds, and they all may prove efficacious when used in conjunction with other
therapeutics. The majority of these anti-angiogenic drugs fall under the class of
receptor tyrosine kinase (RTK) inhibitors (38, 100). These RTK inhibitors have various

degrees of selectivity toward the receptors necessary for neovascularization, such as

230

VEGF receptor and the PDGF receptor (103).

One extension of this class of drugs that target HIF1-controlled genes is the use
of HRE-driven suicide genes as gene therapy agents (24). These agents utilize
knowledge of genomic structure of the regulatory regions of hypoxia-inducible genes
to engineer gene therapy vectors to deliver cytotoxic genes, such as herpes simplex
virus thymidine kinase (121, 147). These vectors contain multiple copies of HREs
derived from the VEGF or EPO gene promoters and in some cases additional
regulatory sequences. This type of approach is not limited to suicide constructs for the
treatment of cancer; it has also been explored to deliver angiogenic factors to the
ischemic heart in a hypoxia-inducible fashion (24, 131).

The final class of drugs actually exploits the hypoxia signal itself. These
compounds are hypoxia-activated cytotoxins and come in three main varieties,
including oxides, quinones and nitroheterocyclics (19, 92, 118). Each of these drugs
are nontoxic at physiological oxygen tension because cells are capable of reoxidizing
the reduced intermediate with available molecular oxygen. Under hypoxic conditions,
however, this futile cycle is inhibited due to the lack of available oxygen, and the
reduced intermediate reaches concentrations that are able to cause cellular damage by
interacting with various macromolecules, especially DNA. These drugs require
activation by cytochromes P450 and P450 reductases. The most widely studied of

these is tirapazamine, an aromatic N-oxide that is currently in clinical trials in

231

combination therapy for treatment of various solid tumors (70).

Taking advantage of the hypoxic environment as a drug target has its limitations.
The efficacy of these therapeutics depends upon the level of hypoxia, the cells exposed,
the tissue’s normal p02, and levels of P450 and P450 reductases and the activation
range of the drug itself. To begin to cope with these limitations, these bioreactive
compounds are now being used in conjunction with gene therapy vectors (24, 99, 145).
This type of gene-directed enzyme prodrug therapy (GDEPI‘) couples the bioreactive
drug, such as tirapazamine, with gene delivery vectors that express different P450-
related enzymes, such as NADPH cytochrome P450 reductase (24). These therapies
have also exploited HRE-driven gene constructs, as described above, to insure that the
enzyme is only expressed in hypoxic tissues, where the drug should be activated.

Development of each of these compounds as drugs has utilized our
understanding of the hypoxic environment and the HIF1 signaling system to target
hypoxic tissues. These therapies, however, are still restricted by our limited knowledge
of the ability of HIF1 to regulate metabolic flux and how the resulting changes in
metabolism ultimately link back to HIF1 activity. In addition, these compounds might
also be limited in their efficacy by several factors that make hypoxic tissues refractory
to conventional therapies, such as poor circulation, decreased cell cycle progression,
and changes in drug detoxifying enzyme profiles.

3.2. Hypoxia and drug disposition

232

Hypoxia influences drug effectiveness on both systemic and local levels. For
example, hypoxic patients have different clearance rates for a variety of drugs
compared to normoxic patients (42). A decrease in blood oxygenation influences the
distribution and effectiveness of drugs in part by changing the expression pattern of
many phase 1 drug metabolizing enzymes. These effects vary for each particular
enzyme, and the pattern of enzyme expression ultimately influences how each
particular drug will be affected. A wide range of studies have been performed on
human patients suffering from diseases that influence blood oxygenation, including
chronic obstructive pulmonary diseases. These studies have documented hypoxia-
related changes in drug clearance for theophylline, tolbutamide, and ethanol, to name a
few (26, 45, 128, 129). All of this research points to potential toxicity or inadequate
pharmacological effectiveness if hypoxia-associated diseases are not taken into
consideration in the determination of dosing regimen.

Hypoxia can also influence drug disposition by altering systemic blood flow
distribution. Since drugs are delivered to the liver in the blood, changes in hepatic
blood flow will influence the rate at which toxicants or drug substrates reach the liver
to be metabolized. Indeed, the hepatic metabolism of many drugs is limited by liver
blood flow rather than the amount of enzyme(s) in the liver that are responsible for
their metabolism (90). Breathing atmospheres of lowered p02 (“hypoxic hypoxia”)

causes redistribution of blood flow away from the splanchnic area, resulting in a

233

decrease in liver perfusion. Thus, in addition to reducing metabolism by limiting 02 as
a cofactor for drug oxidation, hypoxia can diminish drug clearance through its
propensity to reduce drug delivery to the liver by decreasing hepatic blood flow.
Altered hepatic blood flow has its greatest effect on drugs with high extraction ratios.
For example, the hepatic clearance of hexobarbital, a drug eliminated by cytochrome
P450 oxidation, was markedly reduced by exposure of rats to 8% 02, resulting in
prolongation of the anesthetic effect of the drug (101). This was associated with a
marked reduction in liver blood flow (102). Interestingly, an equal decrease in blood
oxygenation caused by breathing a carbon monoxide-containing atmosphere did not
result in reduced hepatic blood flow and caused a less pronounced decrease in
hexobarbital clearance. Thus, hypoxia can influence drug disposition by altering
delivery of the drug to the liver for metabolism, and the means by which hypoxia is
produced can influence its effectiveness in decreasing drug oxidations in vi v0.
3.3. Hypoxia and Cancer Chemotherapy

It has long been known that the hypoxic nature of solid tumors limits the
efficacy of chemo- and radiotherapies. The decreased level of molecular oxygen within
the central portion of the tumor limits the ability of radiation therapy to create oxygen
free radicals, which is the basis of the pharmacological efficacy. It has been estimated
that hypoxic cells are approximately 3 times less susceptible to radiation-induced

damage (134).

234

The limitations of chemotherapy are influenced both by the decrease in available
oxygen at the level of the organism as well as site-specific loss, such as in the central
portion of the tumors. A decrease in tissue oxygen tension can influence therapies in at
least four major ways. First, as mentioned above, changing expression of the P4505
within hypoxic tissues can serve to increase or decrease a drug’s effectiveness. In
addition, low oxygen concentration can influence the catalytic activity of the P4503 by
removing a required cosubstrate. Second, distribution of the drug can be decreased
within the affected tissue due to inadequate vascularization and perfusion (95). For
example, within a tumor, the main cause of the localized hypoxia is an inability of the
angiogenic process to keep up with the fast-growing tumor cells. The rapid growth of
the tumor also leads to misforrned and leaky vessels that cannot support proper
function. Third, the decreased oxygen tension can lead to a decrease in cell cycle
progression. In fact, hypoxic conditions can lead to complete cell cycle arrest (44). A
wide range of cancer therapeutics require rapidly dividing cells for efficacy, and
hypoxia-induced removal of cells from the cell cycle will limit their effectiveness.
Finally, the changes in intermediate metabolism that occur as a result of hypoxia can
also influence drug metabolism and effectiveness. For example, as discussed above,
the HIF1 signaling system is a metabolic switch that can lead to tissue lactic acidosis,
especially within a tumor. Changes in local pH can influence drug delivery and

metabolism. Each of these four factors will influence a drug’s effectiveness in the

235

hypoxic microenvironment and, in the case of a tumor, they might ultimately
determine prognosis.

The hypoxic microenvironment within a tumor might also select for a more
aggressive phenotype. It has been shown that hypoxia increases the rate of mutations
within a cell (156). This increased mutation rate might promote an already partially
compromised cell to full transformation, or it might start a normal cell down the path
to full tumorigenic potential. It has also been suggested that hypoxia can select for
mutant forms of p53 (49). This selective pressure within the tumor could lead to tumor
cells that have removed themselves from one principal regulator of cellular
homeostasis. In addition, prolonged hypoxia can select for cells that are resistant to
pro-apoptotic signals, thus eliminating one potential mechanism of drug-induced cell
removal (152). Finally, the inability of therapy to kill all of the cells within a tumor due
partly to a combination of the systemic and localized effects of hypoxia could produce
cells refractory to later treatment. lndeed, each of these factors might play a role in the
ability of hypoxia to promote more aggressive tumors and ultimately increase the
chance of a poor prognosis.

4. Hypoxia and toxicity
It is well known that many toxicants act by producing tissue hypoxia. These
include agents that cause respiratory depression, such as CNS depressants or

convulsant chemicals, irritant gases that injure the respiratory tract and impair the

236

ability of the lungs to transfer 02 to the blood, and chemicals that cause tissue ischemia,
such as cocaine and ergot alkaloids. Chemicals that impair the ability of hemoglobin to
act as an oxygen carrier also lead to tissue hypoxia. These include agents such as
carbon monoxide, which competes with O; for hemoglobin binding, and
methemoglobin-forrning chemicals, such as nitrites. Although they act by different
mechanisms, all of these agents are toxic because they reduce tissue p02 and thereby
impair delivery of oxygen to the mitochondrial electron transport chain (50).

Other agents act at the level of the mitochondrion by interfering with electron
transport, inhibiting ATP synthase or dissipating mitochondrial membrane potential.
For example, cyanide is toxic because it binds to cytochrome oxidase, thereby
inhibiting the transfer of electrons to 02. All of these agents reduce ATP biosynthesis,
so that many of their clinical effects resemble those of hypoxia. For this reason, their
effects have often been called “chemical hypoxia.” This is an inappropriate term that
should be avoided, since tissue p02 during intoxication is not decreased. For example,
in cyanide intoxication, tissue p02 does not decrease, but rather increases due to the
inability of cells to consume oxygen as the terminal electron acceptor in the respiratory
chain. Accordingly, cyanide and other such agents could have markedly different
effects on hypoxia signaling than true hypoxia because there is no decrease in cellular
oxygen concentration to drive PHD-mediated stabilization of HIFs.

Although it is clear that severe hypoxia from toxicant exposure can cause tissue

237

injury by itself, nontoxic decreases in cellular p02 can promote or unveil harmful
effects of other agents. For example, panadiplon is a drug that causes idiosyncratic
liver injury in humans. In vitro, this drug was only weakly toxic to hepatocytes, unless
the hepatocytes were made hypoxic, in which case cytotoxicity became pronounced (5).
By now, many drugs and other chemicals have been shown to have hypoxia as a
component of their direct toxicity or as a progression factor that promotes worsening
of tissue lesions and functional impairment. Indeed, much has been written on hypoxia
as a factor in chemical intoxication. The remainder of this section will focus primarily
on the role of HIF signaling in toxic responses.

To analyze the roles of hypoxia and HIF1 in the toxicities of various agents,
careful consideration must be made of the balance that is established within the cell
with regard to hypoxia signaling and the life and death decision. As described above,
HIF1 mediated signaling can promote an adaptive response, which promotes cell and
tissue survival through the upregulation of genes such as glycolytic enzymes and
VEGF-induced angiogenesis. At the same time, HIF1 can promote cell injury by
upregulating cell death genes such as BNIP3 and NIX and by influencing p53-
mediated processes. Very little is known about how this balance is maintained within
the cell or how it is initially established. Any toxicant-induced upset of this balance
can lead to adverse cellular events by either overstimulation of the prodeath response

or inhibition of the adaptive response or both. In addition, toxicity might be a result of

238

direct toxicant-induced HIF1 activation or an indirect action that alters HIF signaling.
For example, a toxicant might stabilize HIF 101, leading to overstimulation of the
prodeath response. Alternatively, a toxicant might indirectly overstimulate the hypoxic
signaling system by influencing phosphorylation events or cofactors required for HIF
signaling.
4.1. Link between HIF1 activation, metal-induced toxicity and the prodeath
response

One example of how direct activation of hypoxia signaling can cause
cytotoxicity involves divalent metals. Three metals, cobalt, nickel and manganese, are
recognized as hypoxia mimics, and recent research has focused on the role of hypoxia
signaling in the ability of these metals to cause cellular injury (106, 108, 142). These
metals inhibit PHD function and lead to HIF10t stabilization. Currently, two different
mechanisms have been proposed for their inhibitory activity. First, it is suggested that
these metals directly compete for the iron binding site within the enzyme and render it
inactive (47). Second, it has been proposed that they chelate free ascorbate within the
cell, leading to PHD inhibition (107). Regardless of their mechanism of action, it is
clear that exposure to these metals leads to HIFIOL stabilization and to the regulation of
a battery of genes similar to the changes caused by hypoxia. The expression profile
following exposure to these metals does not completely overlap that of hypoxia (143).

In the case of nickel and cobalt, however, there is enough similarity with the hypoxia

239

profile to suggest that a significant portion of the toxic effects of these metals may
involve HIF1 signaling (106, 142). These experiments utilized engineered mouse
embryonic fibroblasts that were incapable of expressing HIFIOL to show that the
hypoxia signaling cascade plays a critical role in these processes. These experiments
further identified a set of gene products that might be responsible for this pheonotype,
including several pro-apoptotic factors (144). These results support the idea that
unregulated HIF1 signaling can promote cytotoxicity by altering the balance between
the adaptation and cell death responses. The established link between HIF1 over-
activation and toxicity suggests a need for caution when considering prolyl
hydroxylase or iron chelators for therapeutic purposes. These types of drugs could
have inherent toxicity because of their ability to promote a similar response to that of
cobalt and nickel.
4.2. Indirect toxicity and unregulated hypoxic signaling

A more common mode for hypoxia to influence toxicity is through secondary
mechanisms. Drugs or toxicants that produce tissue injury can cause disruption of
blood flow. This disruption can lead to tissue injury downstream of the initial insult
caused, in part, by local hypoxia. This type of injury is thought to play a central role in
a phenobarbital/carbon tetrachloride-induced model of liver cirrhosis (141). Moreover,
carbon tetrachloride-induced liver injury can be potentiated by hypoxia cotreatment

(120). Oxygen deprivation is also thought to play a role in ethanol-induced

240

hepatoxicity (137). Ethanol-induced hypoxia involves an increase in liver oxygen
consumption and metabolic activity (“hypermetabolic state”) and results in pericentral
liver hypoxia and necrosis (2). Interestingly, the mode of action involves Kupffer cell
activation, since pre-treatment ”with gadolinium chloride (GdClz) inhibits the
hepatocellular injury. GdClz, an agent known to destroy Kupffer cells, inhibits ethanol
induced changes in intracellular metabolism and oxygen utilization, suggesting that
Kupffer cell activation drives the localized hypoxia and is necessary for ethanol-
mediated liver injury (136). Endotoxin-induced liver damage also involves disruption
of blood flow. An animal model of sepsis-induced liver damage was studied and shown
to involve inadequate oxygenation of the tissue (62). Moreover, endotoxin
hepatotoxicity is potentiated by breathing a low pOz atmosphere (122).

The role of HIF signaling in drug toxicity seems to be emerging as an area of
interest. A recent study found that HIFlo: in liver increases during acetaminophen
hepatotoxicity in mice (61). This activation begins before the onset of hepatocellular
injury and also occurs in isolated hepatocytes incubated with acetaminophen in an
oxygen replete atmosphere. In vitro, inclusion of cyclosporine A, an inhibitor of the
mitochondrial permeability transition and the oxidative stress that occur in this model,
prevented the increase in hepatocellular HIFIOL.

The hepatoxicities of other medicinal agents also appear to involve hypoxia

and/or HIF-signaling. For example, pyrrolizidine alkaloids are hepatotoxic constituents

241

of certain herbal medicines. One of these, monocrotaline, caused vascular disruption,
hemostasis and hypoxia in livers of exposed rats (Copple. et al., 2006). Moreover,
expression of the HIF-regulated cell death factor, BNIP3, was increased (23).
Nitrofurantoin is an antibiotic that causes idiosyncratic hepatotoxicity in human
patients. In rats, treatment with this drug increased HIF-1 binding to DNA in liver,
suggesting that HIF signaling is activated during nitrofurantoin exposure (132).

Small, nontoxic doses of bacterial endotoxin (lipopolysaccharide, LPS) that
cause modest inflammation potentiate the toxicity of many hepatotoxicants and lower
the threshold for toxicity of several drugs (43). For example, ranitidine is a drug that
causes idiosyncratic hepatotoxicity in humans but is not hepatotoxic in rats. However,
rats develop liver injury when cotreated with ranitidine and LPS at doses that are not
hepatotoxic when given alone (78). Accordingly, a modest inflammatory stress appears
to be a susceptibility factor for drug hepatotoxicity, and this might be the basis for
some idiosyncratic adverse drug reactions in people (43). In livers of LPS/ranitidine-
cotreated rats, hemostasis occurs around the time of onset of liver injury, and this is
accompanied by liver hypoxia and increased expression of HIF 101 (Luyendyk et al.,
2004). Moreover, the hepatic expression of hypoxia-regulated cell death factors such
as BNIP3, EGLN3, Nr4a1 and RTP801 was selectively enhanced (Luyendyk et al.,
2006). Heparin administration reduced both the hepatocellular injury and liver hypoxia,

suggesting the possibility that hypoxia could be an important factor in the pathogenesis

242

(Luyendyk et al., 2005). These results suggest that certain drugs interact with an
inflammatory stress to precipitate liver hypoxia and HIF signaling that leads to
expression of proapoptotic factors and death of hepatocytes. Together, these studies
with drugs have revealed associations between hepatotoxcity and HIF activation in
livers of drug-treated animals; however, much remains to be learned about the role of
HIF signaling, if any, in the pathogenesis of adverse drug responses.

Drug-induced decreases in blood oxygenation can also affect development.
The toxicity of phenytoin, an antiepileptic drug and human teratogen, is linked to
hypoxia-induced malformations during development (28). Fetuses from phenytoin-
treated rats show a drastic decrease in heart rate and increased risk for various
abnormalities, including craniofacial defect and impaired growth. These effects could
be partially ameliorated by treatment with hyperoxia, suggesting that systemic hypoxia
is involved (86). Similar results were found with other antiepileptic drugs that affect
human ether-a-go-go related gene (hERG) channels, and this has led to the suggestion
that cardiac arythmias and hypoxia are the cause of teratogenicity of this class of
compounds (4, 29). These results emphasize the critical role of oxygen homeostasis
during development and suggest that disruption of the balance between adaptation and
cell death has severe consequences during fetal maturation.
5. Neuronal protection by stimulation of the hypoxia adaptive response

The balance between adaptation and cell death can be manipulated to offer

243

protection against subsequent hypoxic challenges. Most of this work has involved
neuronal protection against severe ischemic injury by pre-exposure to low-level
hypoxia (“hypoxia preconditioning”). In vitro and in vivo data suggest that
upregulation of HIFla signaling is involved in this protection (76). VEGF and EPO,
two primary HIFl target genes, play a critical role in this process (104, 151). In fact,
EPO offered some protection on its own. EPO is a growth factor that has been studied
for its hematopoietic regulatory function; however, recent studies have suggested a
broader role for this hypoxia target gene. Neuroprotection from ischemia by EPO
requires pretreatment, lasts for approximately three days and might involve activation
of phosphoinositide-3-kinase or mitogen activated protein kinases (149). In addition,
using primary neurons, Ruscher et al. have shown that EPO is critical to establishing a
paracrine signaling pathway that mediates this protection (104). These studies suggest
that prior adaptation to hypoxia can increase cell survival that likely involves HIFltx-
mediated transcription. These treatments have extended to iron chelation and prolyl
hydroxylase inhibition for neuroprotection (89, 126).

Inhibition of this adaptive response could limit a cell’s survival chances under
stress. In a neuronal model of ischemic damage, the glycolysis inhibitor, iodoacetate,
potentiated glutamate-induced damage (85). The action of iodoacetate, a
glyceraldehyde-3-phosphate dehydrogenase inhibitor, could be prevented by pyruvate

cotreatment, suggesting that maintenance of metabolic flux into the mitochondria is

244

important for this function (85). Though this model does not directly deal with hypoxic
stress, it does entail one downstream consequence of cerebral ischemia, i.e.
accumulation of excitatory amino acids. All of these results, when taken together,
demonstrate that inhibition of the hypoxia-induced adaptive response can have
detrimental consequences. Additionally, they support the notion that there is a delicate
balance between adaptation and death, and manipulation of this balance can influence
a cell’s survival under stress.
6. Conclusions

The ability to adapt to a hypoxic environment is critical to survival, and HIF1
signaling has been demonstrated to be important to this adaptation. HIF1-mediated
signaling drives the expression of a wide variety of genes essential to the adaptive
response, including VEGF, EPO and the glycolytic enzymes, but can also enhance cell
death signaling. The HIF1 signaling cascade, therefore, can promote either adaptation
or cell death. Which of these two outcomes happens in a given condition likely rests
with the control of HlFlot stability and thus its activity (Fig. B-6). Under normoxic
conditions and during normal cellular processes, a cell is capable of modulating HIF10t
activity to cope with small transient changes in oxygen availability. This adaptability is
also important for attempting to re-establish “normoxia” for cells within an organ
during hypoxic stress. Exposure to severe hypoxia is capable of driving the expression

of prodeath genes, such as BNIP3, as well as p53-mediated signaling which promotes

245

Figure B-6. Relationship between HIF1 activity and cell survival under stress

During normal cellular processes, HIF1 activity can be modulated to adapt depending
upon cellular conditions (hatched section). Under severe hypoxia, however, HIF1 can
promote cell death through upregulation of pro-death genes and p53 mediated
processes. Conversely, if the cell is unable to promote HIF1 activity and adaptation
during hypoxic stress, either through drug intervention or metabolic disruption, cell
death may also occur.

246

 

 

Cell Survival

CELL DEATH

Inability to
promote an
adapfive
response
leading to
cytotoxicity

 

NORMOXIA

Cells can adapt

to small changes

in oxygen and
metabolites

8\

         

CELL DEATH

Increased
pro-apoptotic
factors and p53
mediated cellular

injury

 

HIF1 Activity

247

cell death (Fig. B-6). Under these conditions, such as during cerebral ischemia and
severe drug-induced tissue damage, cells within an affected region undergo necrotic or
apoptotic cell death. Finally, if the HIF-mediated adaptive process is inhibited
through physiological or exogenous means, then even small stresses can promote cell
injury through an inability to cope with the energy debt that results from hypoxia. The
balance between cell survival through adaptation and cell death is probably determined
differently for various cellular conditions. The boundaries between too much and not
enough tissue oxygen and oxygen’s influence on HIFIOL activity are important
determinants of our ability to cope with pathological conditions that alter oxygen
homeostasis and our ability to design drugs that target HIF1 signaling.

HH31 signaling and hypoxia are components of the pathological progression of
various diseases, including stroke, cardiovascular disease, and cancer. Targeting
hypoxia signaling for therapeutic purposes, therefore, has garnered considerable
interest. Current therapeutic strategies involve modulating hypoxia signaling through
four distinct mechanisms: directly inhibiting HIF, indirectly modulating HIF activity
through secondary signaling, specifically influencing the expression or activity of
HIF1 target genes and utilizing HRE-mediated therapies to influence the hypoxic cell
(Table B-l). Each new approach utilizes our expanding understanding of hypoxia
biochemistry and signaling. This understanding has also raised consciousness about

cautions needed when targeting this system for drug design and determining what role

248

these drugs will have on the HIF1-regulated balance betweencell and tissue survival
and programmed cell death. Exploiting HIF1 signaling for therapeutics will require
an understanding of how this balance will be shifted in the various tissues of the body
by drugs and under various pathological conditions. For example, increases in HIF10t
activity might promote neuronal protection, but a similar increase in another tissue
might prompt cellular transformation or cell death. In addition, as recent research has
shown, upsetting this balance can alter the toxicity profile of drugs, both directly and
indirectly. A complete understanding of the metabolic flux, normal oxygen levels,
growth factor exposure and endogenous HIF-signaling will be critical to our ability to

develop therapies that are both efficacious and safe.

249

REFERENCES

1. An, W. G., Kanekal, M., Simon, M.C., Maltepe, E., Blagosklonny, M.V., and
Neckers, L.M. 1998. Stabilization of wild-type p53 by hypoxia-inducible factor
lalpha. Nature 392:405-408.‘

2. Arteel, G E., Y. Iimuro, M. Yin, J. A. Raleigh, and R. G Thurman. 1997.
Chronic enteral ethanol treatment causes hypoxia in rat liver tissue in vivo.
Hepatology 25:920-926.

3. Asikainen, T. M., B. K. Schneider, N. S. Waleh, R. I. Clyman, W. B. Ho, L. A.
Flippin, V. Gunzler, and C. W. White. 2005. Activation of hypoxia-inducible
factors in hyperoxia through prolyl 4-hydroxylase blockade in cells and explants
of primate lung. Proc Natl Acad Sci USA 102:10212-10217.

4. Azarbayjani, F., and B. R. Danielsson. 2002. Embryonic Arrhythmia by
Inhibition of HERG Channels: A Common Hypoxia-related Teratogenic
Mechanism for Antiepileptic&nbsp;Drugs? Epilepsia 43:457-468.

5. Bacon, J., C. Cramer, D. Petrella, E. Sun, and R. Ulrich. 1996. Potentiation of
hypoxic injury in cultured rabbit hepatocytes by the quinoxalinone anxiolytic,
panadiplon. Toxicology 108:9-16.

6. Beck, 1., Ramirez, S., Weinmann, R., and Care, J. 1991. Enhancer element at
the 3'-flanking region controls transcriptional response to hypoxia in the human
erythropoietin gene. J. Biol. Chem. 266:15563-15566.

7. Berra, E., E. Benizri, A. Ginouves, V. Volmat, D. Roux, and J. Pouyssegur.
2003. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state
levels of HIF-lalpha in normoxia. EMBO J 22:4082-4090.

8. Briere, J. J ., J. Favier, P. Benit, V. E. Ghouzzi, A. Lorenzato, D. Rabier, M. F.
Di-Renzo, A. P. Gimenez-Roqueplo, and P. Rustin. 2005. Mitochondrial
succinate is instrumental for HlFlalpha nuclear translocation in SDHA-mutant
fibroblasts under normoxic conditions. Human Mol Genet 14:3263-3269.

9. Brown, J. M., and A. J. Giaccia. 1998. The unique physiology of solid tumors:
opportunities (and problems) for cancer therapy. Cancer Res 58:1408-1416.

250

10.

ll.

12.

13.

14.

15.

16.

17.

18.

19.

Bruick, R. K., and S. L. McKnight. 2001. A conserved family of prolyl-4-
hydroxylases that modify HIF. Science 294:1337-1340.

Brunelle, J. K., E. L. Bell, N. M. Quesada, K. Vercauteren, V. Tiranti, M.

Zeviani, R. C. Scarpulla, and N. S. Chandel. 2005. Oxygen sensing requires
mitochondrial ROS but not oxidative phosphorylation. Cell Metab 1:409-414.

Bunn, H. F., and R. O. Poyton. 1996. Oxygen sensing and molecular adaptation
to hypoxia. Physiol Rev 76:839-880.

Burke, B., G, A. iannoudis, K. P. Corke, D. Gill, M. Wells, L. Ziegler-
Heitbrock, and C. E. Lewis. 2003. Hypoxia-induced gene expression in human
macrophages: implications for ischemic tissues and hypoxia-regulated gene
therapy. Am J Pathol 163: 1233-1243.

Carmeliet, P., Y. Dor, J. M. Herbert, D. Fukumura, K. Brusselmans, M.
Dewerchin, M. Neeman, Bono, F., R. Abramovitch, P. Maxwell, C. J. Koch, P.
Ratcliffe, L. Moons, R. K. Jain, D. Collen, and E. Keshertk. 1998. Role of
HIF-lalpha in hypoxia-mediated apoptosis, cell proliferation and tumour
angiogenesis. Nature 394:485-490.

Chandel, N. S., E. Maltepe, E. Goldwasser, C. E. Mathieu, M. C. Simon, and
P. T. Schumacker. 1998. Mitochondrial reactive oxygen species trigger hypoxia-
induced transcription. Proc Natl Acad Sci USA 95: 1 1715-11720.

Chandel, N. S., D. S. McClintock, C. E. Feliciano, T. M. Wood, J. A.
Melendez, A. M. Rodriguez, and P. T. Schumacker. 2000. Reactive oxygen
species generated at mitochondrial complex III stabilize hypoxia-inducible factor-
lalpha during hypoxia: a mechanism of O2 sensing. J Biol Chem 275:25130-

25138.

Chandel, N. S., and P. T. Schumacker. 2000. Cellular oxygen sensing by
mitochondria: old question and new insight. J Appl Physiol 88:1880-1889.

Chen, C., N. Pore, A. Behrooz, F. Ismail-Beigi, and A. Maity. 2001. Regulation
of glutl mRNA by Hypoxia-inducible Factor-1. Interaction between H-ras and
hypoxia. J Biol Chem 276:9519-9525.

Chowdhury, G, D. Kotandeniya, J. S. Daniels, C. L. Barnes, and K. S. Gates.

2004. Enzyme-activated, hypoxia-selective DNA damage by 3-amino-2-
quinoxalinecarbonitrile 1,4-di-N-oxide. Chem Res Toxicol 17:1399—1405.

251

20.

21.

22.

23.

24

25.

26.

27.

28.

Cioffi, C. L., X. Qin Liu, P. A. Kosinski, M. Garay, and B. R. Bowen. 2003.
Differential regulation of HIF-lalpha prolyl-4-hydroxylase genes by hypoxia in
human cardiovascular cells. Biochem Biophys Res Co 303:947-953.

Coelho, A., A. Ferraz, R. Neto, A. Souza, and E. Ferraz. 2000.
Subdiaphragmatic venous stasis and tissular hypoperfusion as source of metabolic
acidosis during passive portal-jugular and cacal-jugular bypasses in dogs. Acta
Cir Bras 15.

Coleman, C., N., Mitchell, J.B., and Camphausen, K. 2001. Tumor hypoxia:
chicken, egg, or a piece of the farm? Journal of Clinical Oncology 20:610-615.

Copple, B., C. Rondelli, J. Maddox, N. Hoglen, P. Ganey, and R. Roth. 2004.
Modes of cell death in rat liver after monocrotaline exposure. Toxicol Sci 77: 172-
1 82.

. Cowen, R. L., K. J. Williams, E. C. Chinje, M. Jaffar, F. C. D. Sheppard, B. A.

Telfer, N. S. Wind, and I. J. Stratford. 2004. Hypoxia targeted gene therapy to
increase the efficacy of tirapazamine as an adjuvant to radiotherapy: reversing
tumor radioresistance and effecting cure. Cancer Res 64:1396-1402.

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. 2003. HIF-lalpha is essential for
myeloid cell-mediated inflammation. Cell 112:645-657.

Cusack, B. J., J. J. Crowley, G D. Mercer, N. B. Charan, and R. E. Vestal.
1986. Theophylline clearance in patients with severe chronic obstructive
pulmonary disease receiving supplemental oxygen and the effect of acute
hypoxemia. Am Rev Respir Dis 133:1110-1114.

Dahia, P. L. M., K. N. Ross, M. E. Wright, C. e. s. Y. Hayashida, S. Santagata,
M. Barontini, A. L. Kung, G Sanso, J. F. Powers, A. S. Tischler, R. Hodin, S.
Heitritter, F. Moore, R. Dluhy, J. A. Sosa, I. T. Ocal, D. E. Benn, D. J. Marsh,
B. G Robinson, K. Schneider, J. Garber, S. M. Arum, M. a. r. Korbonits, A.
Grossman, P. Pigny, S. e. r. P. A. Toledo, V. Nos&eacute;, C. Li, and C. D.
Stiles. 2005. A HIF1-alpha; Regulatory Loop Links Hypoxia and Mitochondrial
Signals in Pheochromocytomas. PLoS Genet 1:e8.

Dan-ielsson, B., A.-C. Skold, F. Azarbayjani, I. Ohman, and W. Webster. 2000.

252

29.

30.

31.

32.

33.

34.

35.

36.

Pharmacokinetic data support pharmacologically Induced embryonic dysrhythmia
as explanation to fetal hydantoin syndrome in rats. Toxicol Appl Pharmacol
163:164-175.

Danielsson, B. R., A.-C. Skold, A. Johansson, B. Dillner, and B. Blomgren.
2003. Teratogenicity by the hERG potassium channel blocking drug almokalant:
use of hypoxia marker gives evidence for a hypoxia-related mechanism mediated
via embryonic arrhythmia. Toxicol Appl Pharmacol 193:168-176.

Dayan, F., D. Roux, M. C. Brahimi-Horn, J. Pouyssegur, and N. M. Mazure.
2006. The Oxygen Sensor Factor-Inhibiting Hypoxia-Inducible Factor-l Controls
Expression of Distinct Genes through the Bifunctional Transcriptional Character
of Hypoxia-Inducible Factor-1 . Cancer Res 66:3688-3698.

Demougeot, C., M. Van Hoecke, N. Bertrand, A. Prigent-Tessier, C. Mossiat,
A. Beley, and C. Marie. 2004. Cytoprotective efficacy and mechanisms of the
liposoluble iron chelator 2,2'-Dipyridyl in the rat photothrombotic ischemic stroke
model. J Pharmacol Exp Ther 311:1080-1087.

Denko, N. C., L. A. Fontana, K. M. Hudson, P. D. Sutphin, S. Raychaudhuri,
R. Altman, and A. J. Giaccia. 2003. Investigating hypoxia tumor physiology
through gene expression pattern. Oncogen 22:5907-5914.

Duong, T. Q., and S.-G Kim. 2000. Quantitative MR measurements of
interstitial oxygen tension in intact rat brain; hyperoxia and hypercapnia. Proc Int]
Soc Mag Reson Med 8:440.

Epstein, A. C., J. M. Gleadle, L. A. McNeil], K. S. Hewitson, J. O'Rourke, D.
R. Mole, M. Mukherji, E. Metzen, M. 1. Wilson, A. Dahnda, 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. Ratcloffe. 2001. C.elegans EGL-9 and
mammalian homologs define a family of dioxygenases that regulate HIF by prolyl
hydroxylation. Cell 107:43-54.

Erez, N., P. Stambolsky, I. Shats, M. Milyavsky, T. Kachko, and V. Rotter.
2004. Hypoxia-dependent regulation of PHD1: cloning and characterization of the
human PHD1/EGLN2 gene promoter. FEBS Lett 567:311-5.

Fang, J., C. Xia, Z. Cao, J. Z. Zheng, E. Reed, and B. H. Jiang. 2005.

Apigenin inhibits VEGF and HIF-1 expression via PI3K/AKT/p7OS6K1 and
HDM2/p53 pathways. FASEB J 19:342-53.

253

 

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

F errara, N. 1999. Molecular and biochemical properties of vascular endothelial
growth factor. J Mol Med 77:527-543.

F errara, N. 2005. VEGF as a therapeutic target in cancer. Oncology 69 Suppl
3:11-16.

F errara, N., K. J. Hillan, H. P. Gerber, and W. Novotny. 2004. Discovery and
development of bevacizumab, an anti-VEGF antibody for treating cancer. Nature
Reviews. Drug Discovery 3:391 -400.

Fisher, T. S., S. D. Etages, L. Hayes, K. Crimin, and B. Li. 2005. Analysis of
ARDl function in hypoxia response using retroviral RNA interference. J Biol
Chem 280: 17749-17757.

Forsythe, J. A., B. H. Jiang, N. V. Iyer, F. Agani, S. W. Leung, R. D. Koos, and

G L. Semenza. 1996. Activation of vascular endothelial growth factor gene
transcription by hypoxia-inducible factor 1. Mol Cell Biol 16:4604-4613.

Fradette, C., and P. Du Souich. 2004. Effect of hypoxia on cytochrome P450
activity and expression. Curr Drug Metab 5:257-271.

Ganey, P. E., J. P. Luyendyk, J. F. Maddox, and R. A. Roth. 2004. Adverse
hepatic drug reactions: inflammatory episodes as consequence and contributor.
Chem Biol Interact 150:35-51.

Gardner, L. B., Q. Li, M. S. Park, W. M. Flanagan, G L. Semenza, and C. V.
Dang. 2001. Hypoxia inhibits G1/S transition through regulation of p27
expression. J Biol Chem 276:7919-7926.

Gavalakis, J ., P. du Souich, and M. Sharkawi. 1999. Acute moderate hypoxia
reduces ethanol elimination in the conscious rabbit. Toxicology 137:109-116.

Geoerger, B., K. Kerr, C. B. Tang, K. M. Fung, B. Powell, L. N. Sutton, P. C.
Phillips, and A. J. Janss. 2001. Antitumor activity of the rapamycin analog CCI-
779 in human primitive neuroectoderrnal tumor/medulloblastoma models as
single agent and in combination chemotherapy. Cancer Research 61: 1527 -1532.

Goldberg, M. A., S. P. Dunning, and H. F. Bunn. 1988. Regulation of the
erythropoietin gene: evidence that the oxygen sensor is a heme protein. Science
242:1412-1415.

254

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

Gong, Y., and Agani, EH. 2005. Oligimycin inhibits HIFlalpha expression in
hypoxic tumor cells. Am J Physiol Cell Physiol 288:1023-1029.

Graeber, T. G, C. Osmanian, T. Jacks, D. E. Housman, C. J. Koch, S. W.
Lowe, and A. J. Giaccia. 1996. Hypoxia-mediated selection of cells with
diminished apoptotic potential in solid tumours. Nature 379:88-91.

Gregus, Z., and C. Klaassen. 2001. Mechanisms of toxicity. In: Cassarett and
Doull's Toxicology: The Basic Science of Poisons (C.D. Klaassen, ed.) New
York: McGraw-Hill, 6th edition.59.

Guzy, R. D., B. Hoyos, E. Robin, H. Chen, L. Liu, K. D. Mansfield, M. C.
Simon, U. Hammerling, and P. T. Schumacker. 2005. Mitochondrial complex
III is required for hypoxia-induced ROS production and cellular oxygen sensing.
Cell Metab 1:401-8.

Guzy, R. D., B. Hoyos, E. Robin, H. Chen, L. Liu, K. D. Mansfield, M. C.
Simon, U. Hammerling, and P. T. Schumacker. 2005. Mitochondrial complex
III is required for hypoxia-induced ROS production and oxygen sensing. Cell
Metab 1:401-408.

Harris, A. L. 2002. Hypoxia - a key regulatory factor in tumor growth. Nat Rev
Cancer 2:38-47.

Hockel, M., and P. Vaupel. 2001. Tumor hypoxia: Definitions and current
clinical, biologic, and molecular aspects. J Natl Cancer Inst 98:266-276.

Hogenesch, J. B., W. C. Han, V. H. Jackiw, R. C. Brown, Y. Z. Gu, M. Pray-
Grant, G H. Perdew, and C. A. Bradfield. 1997. Characterization of a subset of
the basic-helix-loop-helix-PAS superfamily that interact with components of the
dioxin signaling pathway. J Biol Chem 272:8581-8593.

Huang, Z. J., I. Edery, and M. Rosbash. 1993. PAS is a dimerization domain
common to drosophila period and several transcription factors. Nature 364:259-
262.

Ibrahim, N. O., T. Hahn, C. Franke, D. P. Stiehl, R. Wirthner, R. H. Wenger,
and D. M. Katschinski. 2005. Induction of the hypoxia-inducible factor system
by low levels of heat shock protein 90 inhibitors. Cancer Res 65:11094-11100.

Isaacs, J. S., Y. J. Jung, E. G Mimnaugh, A. Martinez, F. Cuttitta, and L. M.

255

59

60

61.

62.

63.

64.

65.

66.

67.

68.

Neckers. 2002. Hsp90 regulates a von Hippel Lindau-independent hypoxia-
inducible factor-1 alpha-degradative pathway. J Biol Chem 277:29936-19944.

. Isaacs, J. S., Y. J. Jung, D. R. Mole, S. Lee, C. Torres-Cabala, Y.-L. Chung, M.

Merino, J. Trepel, B. Zbar, and J. Tom. 2005. HIF overexpression correlates
with biallelic loss of fumarate hydratase in renal cancer: Novel role of fumarate in
regulation of HIF stability. Cancer Cell 8: 143-153.

. 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. 1998.
Cellular and developmental control of 02 homeostasis by hypoxia-inducible
factor 1 alpha. Genes Dev 12:149-162.

James, L. P., B. Donahower, A. S. Burke, S. McCullough, and J. A. Hinson.
2006. Induction of the nuclear factor HIF-lalpha in acetaminophen toxicity:
evidence for oxidative stress. Biochem Biophys Res Co 343: 171-176.

James, P. E., M. Madhani, W. Roebuck, S. K. Jackson, and H. M. Swartz.
2002. Endotoxin-induced liver hypoxia: defective oxygen delivery versus oxygen
consumption. Nitric Oxide 6:18-28.

Jeong, J. W., M. K. Bae, M. Y. Ahn, S. H. Kim, T. K. Sohn, M. H. Bae, M. A.
Yoo, E. J. Song, K. J. Lee, and K. W. Kim. 2002. Regulation and destabilization
of HIF-1a by ARD-mediated acetylation. Cell 111:709-720.

Jiang, J ., T. Nakashima, k. Liu, F. Goda, T. Shima, and H. and Swartz. 1996.
Measurement of P02 in liver using EPR oximetry. J Appl Physiol 80:552-558.

Kasper, L. H., F. Boussouar, K. Boyd, W. Xu, M. Biesen, J. Rehg, T. A.
Baudino, J. L. Cleveland, and P. K. Brindle. 2005. Two transactivation
mechanisms cooperate for the bulk of HIF-l-responsive gene expression. EMBO
J 24:3846-58.

Kiaer, T., and K. D. Kristensen. 1988. Intracompartmental pressure, P02, PC02
and blood flow in the human skeletal muscle. Arch Orthop Trauma Surg 107:114-
6.

Kim, J.-w., I. Tchernyshyov, G L. Semenza, and C. V. Dang. 2006. HIF-1-
mediated expression of pyruvate dehydrogenase kinase: A metabolic switch
required for cellular adaptation to hypoxia. Cell Metabolism 3: 177-185.

Kong, D., E. J. Park, A. G Stephen, M. Calvani, J. H. Cardellina, A. Monks,

256

69

70.

71.

72.

73.

74.

75.

76.

77.

78

R. J. Fisher, R. H. Shoemaker, and G Melillo. 2005. Echinomycin, a small-
molecule inhibitor of hypoxia-inducible factor-1 DNA-binding activity. Cancer
Res 65:9047-9055.

. Kunz, M., S. Ibrahim, D. Koczan, H. J. Thiesen, H. J. Koehler, T. Akcer, K. H.

Plate, S. Ludwig, U. R. Rapp, and E. B. Broecker. 2001. Activation of c-jun
NH2-terminal kinase/stress-activated protein kinase (J NK/SAPK) is critical for

hypoxia-induced apoptosis of human malignant melanoma. Cell growth Differ
12:137-145.

Laderoute, K., P. Wardman, and A. M. Rauth. 1988. Molecular mechanisms
for the hypoxiadependent activation of 3-amino-1,2,4-benzotriazine-l, 4-dioxide
(SR 4233). Biochem Pharmacol 37 : 1487-1495.

Land, S. C. 2003. Oxygen sensing pathway and the development of mammalian
gas exchange. Redox Rep 8:325-340.

Lando, D. 2002. Asparagine hydroxylation of the HIF transactivation domain a
hypoxia switch. Science 295:856-861.

LaPres, J. J., E. Glover, E. E. Dunham, M. K. Bunger, and C. A. Bradfield.
2000. ARA9 modifies agonist signaling through an increase in cytosolic aryl
hydrocarbon receptor. J Biol Chem 275:6153-6159.

Lee, Y. m., Jeong, C.H., Koo, S.Y., Son, M.J., Song, H.S., Bae. S.K., Raleigh,
J .A., Chung, H.Y., Yoo, M.A., and Kim, K.W. 2001. Determination of hypoxic
region by hypoxic markerin developing mouse embryo in vivo: a possible signal
for vessel development. Dev.Dyn. 220: 175-186.

Li, L., X. Lin, M. Staver, A. Shoemaker, D. Semizarov, S. W. Fesik, and Y.

Shen. 2005. Evaluating hypoxia-inducible factor-l {alpha} as a cancer therapeutic
target via inducible RNA interference in viva. Cancer Res 65:7249-7258.

Liu, J., P. Narasimhan, F. Yu, and P. H. Chan. 2005. Neuroprotection by
hypoxic preconditioning involves oxidative stress-mediated expression of

hypoxia-inducible factor and erythropoietin. Stroke 36: 1264-1269.

Lu, H., R. A. Forbes, and A. Verma. 2002. Hypoxia-inducible factor 1
activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J
Biol Chem 277:23111-23115.

. Luyendyk, J., J. Maddox, G. Cosma, P. Ganey, G. Cockerell, and R. RA. 2003.

257

79.

80.

81.

82.

83.

84.

85.

86.

87.

Ranitidine treatment during a modest inflammatory response precipitates
idiosyncrasy-like liver injury in rats. J Pharmacol Exp Ther 307 :9-16.

Mabjeesh, N. J., D. E. Post, M. T. Willard, B. Kaur, E. G Van Meir, J. W.
Simons, and H. Zhong. 2002. Geldanamycin induces degradation of hypoxia-
inducible factor lalpha protein via the proteosome pathway in prostate cancer
cells. Cancer Res 62:2478-2482.

Mabjeesh, N. J., M. T. Willard, W. B. Harris, H.-Y. Sun, R. Wang, H. Zhong,
J. N. Umbreit, and J. W. Simons. 2003. Dibenzoylmethane, a natural dietary
compound, induces HIF-1{alpha} and increases expression of VEGF. Biochem
Biophys Res Co 303:279-286.

Macpberson, G R., and W. D. Figg. 2004. Small molecule-mediated anti-cancer
therapy via hypoxia-inducible factor-1 blockade. Cancer Biology Therapy 3:503-
504.

Mahon, P. C., K. Hirota, and G L. Semenza. 2001. FIH-1: a novel protein that
interact with HIF-lalpha and VHL to mediate repression of HIF-l transcriptional
activity. Gene Dev 15:2675-2686.

Mansfield, K. D., R. D. Guzy, Y. Pan, R. M. Young, T. P. Cash, P. T.
Schumacker, and M. C. Simon. 2005. Mitochondrial dysfunction resulting from
loss of cytochrome c impairs cellular oxygen sensing and hypoxic HIF-alpha
activation. Cell Metab 1:393-399.

Marti, H. H., H. H. Jung, J. Pfeilschifter, and C. Bauer. 1994. Hypoxia and
cobalt stimulate lactate dehydrogenase (LDH) activity in vascular smooth muscle
cells. Pflugers Arch 429:216-222.

Massieu, L., N. Gomez-Roman, and T. Montiel. 2000. In vivo potentiation of
glutamate-mediated neuronal damage after chronic administration of the
glycolysis inhibitor iodoacetate. Exp Neurology 165:257-267.

Millicovsky, G., and M. C. Johnston. 1981. Maternal hyperoxia greatly reduces
the incidence of phenytoin-induced cleft lip and palate in Al] mice. Science
212:671-672.

Minet, E., D. Mottet, G Michel, 1. Roland, M. Raes, J. Remacle, and C.

Michiels. 1999. Hypoxia-induced activation of HIF-1: role of HIF-lalpha-Hsp90
interaction. FEBS Lett 460:251-256.

258

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

Mizukami, Y., Jo.W-S., E.-M. Duerr,, and M. Gala, L. J. 2005. Induction of
interleukin-8 preserves the angiogenic response in HlFlalpha-deficiency colon
cancer cells. Nat Med 9:992-997.

Mu, D., Y. S. Chang, Z. S. Vexler, and D. M. Ferriero. 2005. Hypoxia-inducible
factor 1a and erythropoietin upregulation with deferoxamine salvage after

neonatal stroke. Exp Neurology 195:407-415.

Nies, A., D. Shand, and G Wilkinson. 1976. Altered hepatic blood flow and
drug disposition. Clin Pharrnacokinet 1: 135-155.

Osada, M., S. Imaoka, and Y. Funae. 2004. Apigenin suppresses the expression
of VEGF, an important factor for angiogenesis, in endothelial cells via
degradation of HIF-lalpha protein. FEBS Lett 575:59-63.

Papadopoulou, M. V., and W. D. Bloomer. 2003. NLCQ-l (NSC 709257):
Exploiting hypoxia with a weak DNA-intercalating bioreductive drug. Clin
Cancer Res 9:5714-5720.

Piret, J. P., D. Motter, M. Raes, and C. Michiels. 2002. Is HIF-lalpha a pro- or
an anti-apoptotic protein? Biochem Pharmacol. 64:889-892.

Poole, D. C., P. D. Wagner, and D. F. Wilson. 1995. Diaphragm microvascular
plasma P02 measured in vivo. J Appl Physiol 79:2050-2057.

Primeau, A. J., A. Rendon, D. Hedley, L. Lilge, and I. F. Tannock. 2005. The
Distribution of the Anticancer Drug Doxorubicin in Relation to Blood Vessels in
Solid Tumors. Clin Cancer Res 11:8782-8788.

Pugh, C. W., Tan, C.C., Jones, R.W., and Ratcliffe, PJ. 1991. Functional
analysis of an oxygen regulated transcriptional enhancer lying 3' to the mouse
erythropoietin gene. Proc. Natl. Acad. Sci. USA. 88:10553-10557.

Richard, D. E., E. Berra, E. Gothic, D. Roux, and J. Poussegur. 1999. p42/p44
mitogen-activated protein kinases phosphorylate hypoxia-induced factor lalpha
and enhance the transcriptional activity of HIF-l. J Biol Chem 274:32631-32637.

Ricker, J. L., Z. Chen, X. P. Yang, V. S. Pribluda, G M. Swartz, and C. Van

Waes. 2004. 2-Methoxyestradiol Inhibits Hypoxia-Inducible Factor 1{alpha},
Tumor Growth, and Angiogenesis and Augments Paclitaxel Efficacy in Head and

259

Neck Squamous Cell Carcinoma. Clin Cancer Res 10:8665-8673.

99. Riddick, D. S., C. Lee, S. Ramji, E. C. Chinje, R. L. Cowen, K. J. Williams, A.
V. Patterson, I. J. Stratford, C. S. Morrow, A. J. Townsend, Y. Jounaidi, C.-S.
Chen, T. Su, H. Lu, P. S. Schwartz, and D. J. Waxman. 2005. Cancer
chemotherapy and drug metabolism. Drug Metab Dispos 33: 1083-1096.

100.Rosen, L. S. 2002. Clinical experience with angiogenesis signaling inhibitors:
focus on vascular endothelial growth factor (VEGF) blockers. Cancer Control
9:36-44.

101.Roth, R., and R. Rubin. 1976. Comparison of the effect of carbon monoxide and
of hypoxic hypoxia. 1. In vivo metabolism, distribution and action of hexobarbital.
J Pharmacol Exp Ther. 199:53-60.

102.Roth, R., and R. Rubin. 1976. Role of blood flow in carbon monoxide- and
hypoxic hypoxia-induced alterations in hexobarbital metabolism in rats. Drug
Metab Dispos 4:460-467.

103.Ruggeri, B., J. Singh, D. Gingrich, T. Angeles, M. Albom, H. Chang, C.
Robinson, K. Hunter, P. Dobrzanski, S. Jones-Bolin, L. Aimone, A. Klein-
Szanto, J .-M. Herbert, F. Bono, P. Schaeffer, P. Casellas, B. Bourie, R. Pili, J.
Isaacs, M. Ator, R. Hudkins, J. Vaught, J. Mallamo, and C. Dionne. 2003.
CBP-7055: A novel, orally active pan inhibitor of vascular endothelial growth
factor receptor tyrosine kinases with potent antiangiogenic activity and antitumor
efficacy in preclinical models. Cancer Res 63:5978-5991.

104.Ruscher, K., D. Freyer, M. Karsch, N. Isaev, D. Megow, B. Sawitzki, J. Priller,
U. Dirnagl, and A. Meisel. 2002. Erythropoietin is a paracrine mediator of
ischemic tolerance in the brain: Evidence from an in vitro model. J Neurosci
22:10291-10301.

105.Sabia, P. J., Powers, E.R., Ragosta, M., Sarembock, I.J., Burwell, L.R., and
Kaul, S. 1992. An association between collateral blood flow and myocardial
viability in patients with recent myocardial infaction. N. ENGL, J. Med.
327:1825-1831.

106.Salnikow, K., T. Davidson, and M. Costa. 2002. The role of hypoxia-inducible

signaling pathway in nickel carcinogenesis. Environ Health Perspect 110 Suppl
5:831-4.

260

 

107.Salnikow, K., S. P. Donald, R. K. Bruick, A. Zhitkovich, J. M. Phang, and K.
S. Kasprzak. 2004. Depletion of intracellular ascorbate by the carcinogenic
metals nickel and cobalt results in the induction of hypoxic stress. J Biol Chem.

108.Salnikow, K., W. Su, M. V. Blagosklonny, and M. Costa. 2000. Carcinogenic
metals induce hypoxia-inducible factor-stimulated transcription by reactive
oxygen species-independent mechanism. Cancer Res 60:3375-3378.

109.Sang, N., Stiehl. D.P., Bohensky. J., Leshchinsky. V., and Srinivas. J.C. 2003.
MAPK signaling up-regulated the activity of hypoxia-inducible factors by its
effects on p300. J Biol Chem 278:14013-14019.

110.Schultz, A., Lavie, L., Hochberg, 1., Beyar, R., Stone, T., Skorecki, K., Lavie,
P., Roguin, A., and Levy, A.P. 1999. Inter-individual heterogeneity in the
hypoxia regulation of VEGF: Significance for the development of the coronary
artery collateral circulation. Circulation 100:547-552.

111.Selak, M. A., S. M. Armour, E. D. Mackenzie, H. Boulahbel, D. G Watson, K.
D. Mansfield, Y. Pan, C. Simon, C. B. Thompson, and E. Gottlieb. 2005.
Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha
prolyl hydroxylase. Cancer Cell 7:77-85.

112.Selak, M. A., S. M. Armour, E. D. MacKenzie, H. Boulahbel, D. G Watson, K.
D. Mansfield, Y. Pan, M. C. Simon, C. B. Thompson, and E. Gottlieb. 2005.
Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-[alpha]
prolyl hydroxylase. Cancer Cell 7:77-85.

113.Semenza, G L. 2006. HIF1 and human disease: one highly involved factor. Genes
& Dev. 14:1983-1991.

114.Semenza, G L. 2000. HIF1: mediator of physiological and pathophysiological
response to hypoxia. J Appl Physiol 88: 1474-1480.

115.Semenza, G L. 2004. Hydroxylation of HIF-1: Oxygen sensing at the molecular
level. Physiology 19: 176-182.

116.Semenza, G L. 1998. Hypoxia-inducible factor 1: master regulator of Oz
homeostasis. Current Opinion in Genetics & Development 8:588-94.

117.Semenza, G L. 2003. Targeting HIF-l for cancer therapy. Nature 3:721-731.

261

118.Seow, H. A., P. G Penketh, K. Shyam, S. Rockwell, and A. C. Sartorelli. 2005.
1 ,2-Bis(methylsulfonyl)- 1-(2-chloroethyl)-2-[[ 1-(4-itrophenyl )ethoxy] carbonyl]
hydrazine: An anticancer agent targeting hypoxic cells. Proc Natl Acad Sci USA
102:9282-9287.

119.Shannon, A. M. B.-H., D.J., Condron, GM. and Toomey, D. 2003. Tumor
hypoxia, chemotherapeutic resistance and hypoxia-related therapies. Cancer

treatment review 29:297-307.

120.Shen, E. S., V. F. Garry, and M. W. Anders. 1982. Effect of hypoxia on carbon
tetrachloride hepatotoxicity. Biochem Pharmacol 31:3787-3793.

121.Shiba,ta, T., A. J. Giaccia, and J. M. Brown. 2002. Hypoxia-inducible regulation
of a prodrug-activating enzyme for tumor-specific gene therapy. Neoplasia 4:40-
48.

122.Shibayama, Y. 1987. Enhanced hepatotoxicity of endotoxin by hypoxia. Pathol
Res Pract 182:390-395.

123.Shimizu, S., Eguchi, Y., Kamiike, W., Itoh, Y., Hasegawa, J., Yamabe, K.,
Otsuki, Y., Matsud, H., and Tsujimoto, Y. 1996. Induction of apoptosis as well

as necrosis by hypoxia and predominant prevention of apoptosis by Bcl-l AND
bCL-XL. Cancer Res 56:2161-2166.

124.Shinkai, T., Shinkai, M., Pirker, M.A., Montedonico,S., and Puri, P. 2005. The
role of oxygen tension in the regulation of embryonic lung development. Journal
of pediatric surgery. 40:32-35.

125.Shweiki, D., Itin, A., Soffer, D., and Keshet, E. 1992. Vascular endothelial
growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis.
Nature 359:843-845.

126.Siddiq, A., I. A. Ayoub, J. C. Chavez, L. Aminova, S. Shah, J. C. LaManna, S.
M. Patton, J. R. Connor, R. A. Chemy, I. Volitakis, A. 1. Bush, I. Langsetmo,
T. Seeley, V. Gunzler, and R. R. Ratan. 2005. Hypoxia-inducible factor Prolyl
4-Hydroxylase inhibition: A Target for neuroprotection in the central nerve system.
J Biol Chem 280:41732-41743.

127.Sonna, L. A., M. L. Cullivan, H. K. Sheldon, R. E. Pratt, and C. M. Lilly. 2003.

Effect of hypoxia on gene expression by human hepatocytes (HepG2). Physiol
Genomics 12:195-207.

262

128.Sotaniemi, E., P. Arvela, and E. Huhti. 1971. Increased clearance of tolbutamide
from the blood of asthmatic patients. Ann Allergy 29:139-141.

129.Sotaniemi, E., P. Arvela, E. Huhti, and O. Koivisto. 1971. Half-life of
tolbutamide in patients with chronic respiratory failure. Eur J Clin Pharmacol.
4:29-31.

130.Stolze, I. P., Y.-M. Tian, R. J. Appelhoff, H. Thrley, C. C. Wykoff, J. M.
Gleadle, and P. J. Ratcliffe. 2004. Genetic analysis of the role of the asparaginyl
hydroxylase factor inhibiting hypoxia-inducible factor (HIF) in regulating HIF
transcriptional target genes. J Biol Chem 279:42719-42725.

131.Su, H., J. Arakawa-Hoyt, and Y. W. Kan. 2002. Adeno-associated viral vector-
mediated hypoxia response element-regulated gene expression in mouse ischemic
heart model. Proc Natl Acad Sci USA 99:9480-9485.

132.Tacchini, L., D. Fusar-Poli, and A. Bernelli-Zazzera. 2002. Activation of
transcription factors by drugs inducing oxidative stress in rat liver. Biochem
Pharmacol. 63: 139-148.

133.Tan, C., R. G de Noronha, A. J. Roecker, B. Pyrzynska, F. Khwaja, Z. Zhang,
H. Zhang, Q. Teng, A. C. Nicholson, P. Giannakakou, W. Zhou, J. J. Olson,
M. M. Pereira, K. C. Nicolaou, and E. G Van Meir. 2005. Identification of a
novel small-molecule inhibitor of the hypoxia-inducible factor 1 pathway. Cancer
Res 65:605-612.

134.Teicher, B. A. 1995. Physiologic mechanisms of therapeutic resistance. Blood
flow and hypoxia. Hematol Oncol Clin North Am 9:475-506.

135.Thomas, G V., C. Tran, I. K. Mellinghoff, D. S. Welsbie, E. Chan, B. Fueger, J.
Czemin, and C. L. Sawyers. 2006. Hypoxia-inducible factor determines
sensitivity to inhibitors of mTOR in kidney cancer. Nature Medicine 12:122-7.

136.Thurman, R. G. 1998. II. Alcoholic liver injury involves activation of Kupffer
cells by endotoxin. Am J Physiol 275:G605-611.

137.Thurman, R. G, B. U. Bradford, Y. Iimuro, M. V. Frankenberg, K. T. Knecht,
H. D. Connor, Y. Adachi, C. Wall, G E. Arteel, J. A. Raleigh, D. T. Forman,
and R. P. Mason. 1999. Mechanisms of alcohol-induced hepatotoxicity: studies
in rats. Front Biosci 4:e42-46.

263

138.Tsuzuki, Y., Fukumura,D., Oosthuyse, B., Koike, C., Carmeliet, P., and Jain.
R.K. 2000. Vasculae endothelial growth factor (VEGF) modulatin by targeting
Hypoxia-Inducible Factor-lalpha -> Hypoxia Response Element -> VEGF
cascade. Differentially regulates vascular reponse and growth rate in tumors.
Cancer Res 60:6248-6252.

139.Tuckerman, J. R., Y. Zhao, K. S. Hewitson, Y.-M. Tian, C. W. Pugh, P. J.
Ratcliffe, and D. R. Mole. 2004. Determination and comparison of specific
activity of the HIF-prolyl hydroxylases. FEBS Lett 576:145-150.

140.'Itryl, M., Liu, J., Wang, J., Kuliszewski, M., Tibboel, D., and Post, M. 2005.
Role of oxygen and vascular development in epitherlial branching morphogenesis
of the developing mouse lung. Am. J. Physiol. Lung Cell Mol Physiol. 288:167-
178.

141.Varin, F., and P. M. Huet. 1985. Hepatic microcirculation in the perfused
cirrhotic rat liver. J Clin Invest 76: 1904-1912.

142.Vengellur, A., and J. J. LaPres. 2004. The role of hypoxia inducible factor
1{alpha} in cobalt chloride induced cell death in mouse embryonic fibroblasts.
Toxicol Sci 82:638-646.

143.Vengellur, A., Phillips, J .M., Hogenesch, J .B. and LaPres, J .J . 2005. Gene
expression profiling of hypoxia signaling in human hepatocellular carcinoma cells.
Phys Genomics 22:308—3 18.

144.Vengellur, A., B. G Woods, H. E. Ryan, R. S. Johnson, and J. J. LaPres. 2003.
Gene expression profiling of the hypoxia signaling pathway in hypoxia inducible
factor 1a null mouse embryonic fibroblasts. Gene Expression 11:181-197.

145.Verna McErlane, A. Y., Helen O. McCarthy, Ciara M. Hughes, Laurence H.
Patterson, David G Hirst, Tracy Robson, Stephanie R. McKeown,. 2005. A
cytochrome P450 2B6 meditated gene therapy strategy to enhance the effects of
radiation or cyclophosphamide when combined with the bioreductive drug AQ4N.
The Journal of Gene Medicine 7 :851-859.

146ng, G L., B. H. Jiang, E. A. Rue, and G L. Semenza. 1995. Hypoxia-
inducible factor 1 is a basic-he]ix-loop-helix-PAS heterodimer regulated by

cellular 02 tension. Proc. Natl. Acad. Sci. USA 92:5510-5514.

147.Wang, W.-d., Z.-t. Chen, R. Li, D.-z. Li, Y.-z. Duan, and Z.-h. Cao. 2005.

264

 

Enhanced efficacy of radiation-induced gene therapy in mice bearing lung
adenocarcinoma xenografts using hypoxia responsive elements. Cancer Sci
96:918-924.

148.Warburg, O. 1956. On the origin of cancer cells. Science 123:309-314.

149.Weber, A., M. Dzietko, M. 'Berns, U. Felderhoff-Mueser, U. Heinemann, R. F.
Maier, M. Obladen, C. Ikonomidou, and C. Buhrer. 2005. Neuronal damage
after moderate hypoxia and erythropoietin. Neurobiology of Disease 20:594-600.

150.Welsh, S., R. Williams, L. Kirkpatrick, G Paine-Murrieta, and G Powis. 2004.
Antitumor activity and pharrnacodynamic properties of PX-478, an inhibitor of
hypoxia-inducible factor-lalpha. Molecular Cancer Therapeutics 3:233-244.

151.Wick, A., W. Wick, J. Waltenberger, M. Weller, J. Dichgans, and J. B. Schulz.
2002. Neuroprotection by hypoxic preconditioning requires sequential activation
of vascular endothelial growth factor receptor and Akt. J Neurosci 22:6401-6407.

152.Yao, K., J. Gietema, S, S. hida, M. Selvakumaran, X. Fonrose, N. Haas, J.
Testa, and P. O'Dwyer. 2005. In vitro hypoxia-conditioned colon cancer cell
lines derived from HCT116 and HT 29 exhibit altered apoptosis susceptibility and
a more angiogenic profile in vivo. Br J Cancer. 93:1356-63.

153.Yu, A. Y., M. G Frid, L. A. Shimoda, C. M. Wiener, K. Stenmark, and G L.
Semenza. 1998. Temporal, spatial, and oxygen-regulated expression of hypoxia-
inducible factor-1 in the lung. American Journal of Physiology 275:L818-26.

154.Yu, A. Y., Frid, M.G, Shimoda, L.A., Wiener, C.M., Stenmark, K., and
Semenza, GL. 1998. Temporal, spatial and oxygen-regulated expression of
hypoxia-inducible factor 1 in the lung. Am. J. Physiol. 275:L818-L826.

155.Yu, A. Y., Shimoda, L.A., Iyer, N.V., Huso, D.L., Sun, X., McWilliams, R.,
Beaty, T., Sham, J.S.K., Wiener, C.M., Sylvester, J.T., and Semenza, GL.
1999. Impaired physiological responses to chronic hypoxia in mice partially
deficient for hypoxia-inducible factor la. Journal of Clinical Investigation.

103:691-696.
156.Yuan, J ., L. Narayanan, S. Rockwell, and P. M. Glazer. 2000. Diminished DNA

repair and elevated mutagenesis in mammalian cells exposed to hypoxia and low
pH. Cancer Res 60:4372-4376.

265

 

157.Yue, X., and Tomanek R.J. 1999. Stimulation of coronary vasculogenesis/
angiogenesis by hypoxia in cultured embryonic heart. Dev. Dyn. 216:28-36.

158.Zaman, K., H. Ryu, D. Hall, K. O'Donovan, K.-I. Lin, M. P. Miller, J. C.
Marquis, J. M. Baraban, G L. Semenza, and R. R. Ratan. 1999. Protection
from Oxidative Stress-Induced Apoptosis in Cortical Neuronal Cultures by Iron
Chelators Is Associated with Enhanced DNA Binding of Hypoxia-Inducible
Factor-1 and ATF-l/CREB and Increased Expression of Glycolytic Enzymes,
p21waf1/cip1, and Erythropoietin. J Neurosci 19:9821-9830.

266

 

llllllllllllllllllllllllll

lllllllllllllllllllllllllllllllll/l/lllll

3 1293 02956 3404