Báo cáo khoa học: Intermittent hypoxia is a key regulator of cancer cell and endothelial cell interplay in tumours pot

REVIEW ARTICLE
Intermittent hypoxia is a key regulator of cancer cell and
endothelial cell interplay in tumours
S. Toffoli and C. Michiels
Laboratory of Biochemistry and Cellular Biology (URBC), University of Namur – FUNDP, Belgium
Introduction
Hypoxia is increasingly perceived as one of the
tumour microenvironment features favouring tumour
cell survival, and also resistance to chemotherapy and
radiotherapy. Hypoxia is defined as a decrease in oxy-
gen level within the tissue. However, recent studies
have shown that the time frame within which this
decrease occurs and, more importantly, its duration
may vary greatly from one tumour to another, or even
from one area to another within the same tumour.
These observations have led to the definition of two
kinds of hypoxia: chronic hypoxia and intermittent
hypoxia.
Intermittent and chronic hypoxia in
solid tumours
Chronic hypoxia in tumours, first described in 1955
[1,2], results from limitation of the diffusion of oxygen.
Oxygen diffuses to a distance of 100–150 lm from
blood vessels in normal and malignant tissues. At a
greater distance, the oxygen tension becomes close to
zero, and cells become hypoxic [1]. In parallel with
chronic hypoxia, it was suggested in 1979 that tran-
sient hypoxia or intermittent hypoxia could also
appear in tumours, due to the temporary ‘closure’ of
blood vessels [3]. The existence of acute hypoxia events
in tumours was shown a few years later, with the

Keywords
apoptosis; cancer; chemoresistance;
endothelial cell; hypoxia-inducible factor-1;
intermittent hypoxia; radioresistance;
reactive oxygen species; reoxygenation;
tumor cell
Correspondence
C. Michiels, Laboratory of Biochemistry and
Cellular Biology (URBC), University of
Namur – FUNDP, 61 rue de Bruxelles, 5000
Namur, Belgium
Fax: +32 81 72 41 35
Tel: +32 81 72 41 31
E-mail:
(Received 1 March 2008, accepted 9 April
2008)
doi:10.1111/j.1742-4658.2008.06454.x
Solid tumours are complex structures in which the interdependent relation-
ship between tumour and endothelial cells modulates tumour development
and metastasis dissemination. The tumour microenvironment plays an
important role in this cell interplay, and changes in its features have a
major impact on tumour growth as well as on anticancer therapy respon-
siveness. Different studies have shown irregular blood flow in tumours,
which is responsible for hypoxia and reoxygenation phases, also called
intermittent hypoxia. Intermittent hypoxia induces transient changes, the
impact of which has been underestimated for a long time. Recent in vitro
and in vivo studies have shown that intermittent hypoxia could positively
modulate tumour development, inducing tumour growth, angiogenic pro-
cesses, chemoresistance, and radioresistance. In this article, we review the
effects of intermittent hypoxia on tumour and endothelial cells as well as

its impacts on tumour development.
Abbreviations
AP-1, activator protein-1; ARNT, aryl hydrocarbon receptor nuclear translocator; EPR, electron paramagnetic resonance; HIF-1, hypoxia-
inducible factor-1; NF-jB, nuclear factor kappaB; ROS, reactive oxygen species; VEGF, vascular endothelial growth factor.
FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS 2991
demonstration that intermittent hypoxia resulted from
transient changes in blood flow [1,4,5]. Histological
analysis of tumour blood vessels showed that struc-
tural abnormalities were responsible for this irregular
blood flow. Indeed, tumour blood vessels are often tor-
tuous and dilated, with excessive branching and
numerous dead ends [6]. Moreover, compression of
these vessels by tumour cells, associated with the
immaturity of the tumour vascular network, which is
characterized by an absence of or a loose association
with mural cells, pericytes and vascular smooth muscle
cells, could also play a role in the heterogeneity of the
blood flow [7–9].
The blood flow stop periodicity, depending on the
architectural complexity and maturation level of the
tumour vascular network, is very variable from one
tumour to another, and also within the same tumour
[10,11]. Therefore, a precise duration for blood flow
interruption in tumours cannot be given. However,
studies of murine and human tumours have shown
that the blood flow fluctuations observed in these
tumours could vary from several minutes to more
than 1 h in duration [10–16]. These blood flow irregu-
larities in tumours can be demonstrated by different
methods. Direct real-time measurement in vivo of

tumour blood flow fluctuations can be performed by
the use of microprobes that are directly implanted in
tumours. Different microprobe systems can be used
to study the blood flow fluctuations. One of the most
used microprobes is the Eppendorf polarographic
needle electrode, which allows measurement of the
oxygen partial pressure (po
2
) within tissues [12,17].
Polarographic oxygen microelectrode functioning is
based on reduction of oxygen at the surface of a
cathode by applying a negative voltage between the
cathode and the anode. The reduction current mea-
sured with this kind of electrode is proportional to
the number of oxygen molecules being reduced, and
diminishes when blood flow is decreased or inter-
rupted [18,19]. Other microprobes, such as the Oxy-
Lite laser Doppler probe, allow monitoring of tumour
blood perfusion [14]. These probes illuminate the
tissue under observation with single-frequency light
from optical fibres coupled to a sensor. Mobile red
blood cells scatter the monochromatic light and gen-
erate a signal that is proportional to the mean eryth-
rocyte velocity multiplied by the number of moving
erythrocytes within the sampling volume [20,21]. This
signal decreases when the blood flow diminishes or
stops, and vice versa. However, the spatial resolution
of these techniques is low, and the use of polaro-
graphic or laser Doppler microprobes is restricted
to easily accessible tumours [22]. For less accessible

neoplasms, the direct real-time measurement in vivo of
oxygen tension is performed by the use of imaging
techniques [18,22,23], most of which are based on
magnetic resonance. Blood oxygen level-dependent
magnetic resonance imaging and electron paramag-
netic resonance (EPR) oxymetry are examples of such
techniques [22–24]. Blood flow modifications observed
with blood oxygen level-dependent magnetic reso-
nance imaging are based on the oxygenation status of
endogenous haemoglobin. This becomes paramagnetic
when it is deoxygenated, and it is then detectable by
magnetic resonance imaging. Changes in blood flow
modify the blood concentration of paramagnetic
deoxyhaemoglobin and hence induce variations in the
magnetic resonance signal [22,23,25,26]. On the other
hand, EPR oxymetry is based on the broadening of
the resonance spectrum of a paramagnetic material
by oxygen [27]. Modifications in the EPR signal are
directly correlated with the oxygen concentration,
which is linked to the blood flow [18]. One injection
of a paramagnetic agent, such as India ink or char-
coal, directly into a tumour is sufficient to allow
repeated measurements to be performed over a rela-
tively long period [18,24]. Indirect measurements
in vivo of tumour po
2
fluctuation can also be per-
formed by the use of a double hypoxia marker tech-
nique [11,28–30]. 2-Nitroimidazoles (e.g. misonidazole,
EF5, CCI-103F, and pimonidazole) are commonly

used as hypoxia markers. These molecules are
reduced by cellular nitroreductases at po
2
levels below
10 mmHg to intermediates that covalently bind to
cellular macromolecules [31–33]. Hypoxic markers are
administered in vivo separately at different times
according to a pre-established timing schedule.
Tumour areas stained only by one marker show tran-
sient changes in hypoxia during the time interval
between the injections of the two hypoxia markers
[22,29]. Reduced 2-nitroimidazoles can be detected
by immunohistochemistry or immunofluorescence
staining after the tumour resection. Using radio-
labelled 2-nitroimidazoles (e.g. [
18
F]fluoromisonida-
zole), transiently hypoxic areas can also be detected
in vivo by positron emission tomography, which is
based on the detection of electromagnetic radiation
emitted indirectly by the positron-emitting radio-
isotope [22,34]. Modifications in blood flow are
shown by performing scans after each hypoxia mar-
ker injection [18,22,23,35]. The use of these techniques
and their combination allow a better understanding
of spatial and temporal changes in hypoxia in solid
tumours, and also allow the linkage of these changes
with other tumour microenvironmental parameters
[18,22,23,35].
Intermittent hypoxia in cancer S. Toffoli and C. Michiels

2992 FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS
Hypoxia-inducible factor-1 (HIF)
a-subunit stabilization and HIF-1
activation under intermittent hypoxia
Hypoxia induces numerous changes in gene expression
in normal and tumour cells [36]. This adaptive response
to hypoxia is orchestrated by a family of transcription
factors induced by hypoxia. The most important and
best-studied member of this family is hypoxia-inducible
factor-1 (HIF-1). HIF-1 is a heterodimeric transcription
factor composed of the HIF-1a (120 kDa) and aryl
hydrocarbon receptor nuclear translocator (ARNT,
94 kDa; also called HIF-1b) subunits. These two subun-
its belong to the Per-ARNT-Sim basic–helix–loop helix
family [37,38]. HIF-1a and ARNT are constitutively
expressed [39], but the formation of HIF-1 transcription
factor in the nucleus depends on HIF-1a stabilization,
which is principally O
2
-dependent [40]. Under nor-
moxia, HIF-1a is hydroxylated on proline 402 in the
N-terminal domain and proline 564 in the C-terminal
domain by prolyl-4-hydroxylases [41]. These hydroxyla-
tions allow the binding of von Hippel–Lindau tumour
suppressor protein on the oxygen-dependent degrada-
tion (ODD) domain of HIF-1a [42]. von Hippel–Lindau
tumour suppressor protein acts as the substrate recogni-
tion protein of the E3 ubiquitin ligase complex [43], and
induces the ubiquitination of HIF-1a on its N-terminal
and C-terminal domains (amino acids 390–417 and

549–582, respectively) [41]. This ubiquitination targets
HIF-1a for proteasomal degradation. On the other
hand, under hypoxic conditions, the prolyl hydroxylase
activity decreases and the degradation pathway
described above is interrupted [44]. HIF-1a therefore
rapidly accumulates and translocates into the nucleus,
where, after dimerization with ARNT, it induces the
transcription of target genes involved, notably, in
glycolysis (e.g. the glyceraldehyde-3-phosphate dehydro-
genase gene) and angiogenesis [e.g. the vascular endo-
thelial growth factor (VEGF) gene] [45], thus allowing
cells to adapt to hypoxia [46].
The stabilization of HIF-1 a and activation of HIF-1
have been widely studied under chronic hypoxia. The
new interest in intermittent hypoxia in recent years has
led us to consider again this point: can the succession
of short hypoxia and reoxygenation phases, typical
of intermittent hypoxia, also stabilize HIF-1a and
activate HIF-1?
In the absence of oxygen, HIF-1a is rapidly stabi-
lized, and short, intermittent hypoxia periods can be
sufficient to induce HIF-1. Indeed, Yuan et al. showed,
in vitro, HIF-1a stabilization during intermittent
hypoxia (cycles of 30 s of hypoxia followed by 4 min
of reoxygenation) [47]. This increase in HIF-1a abun-
dance was dependent on the number of intermittent
hypoxia cycles. The kinetics used by Yuan et al.
undoubtedly demonstrate that short hypoxia–reoxy-
genation cycles can induce HIF-1a stabilization. How-
ever, considering these kinetics, the increase in

abundance of HIF-1a during intermittent hypoxia
cycles could be due to an accumulation of HIF-1a sub-
unit during each cycle, and not to an increase in its
stabilization. Indeed, although HIF-1a may be extre-
mely rapidly degraded when cells are reoxygenated, its
degradation after 4 min of reoxygenation was not
assayed by Yuan et al. Furthermore, Berra et al.
showed that HIF-1a could still be detected after 5 min
of reoxygenation in HeLa cells incubated for 1 h or or
8 h under hypoxia. They showed that the half-life of
HIF-1a is inversely proportional to the duration of
hypoxic stress [48], suggesting that long hypoxia peri-
ods could decrease HIF-1a stability. Other recent stud-
ies have also shown an increase in HIF-1a abundance
in the course of intermittent hypoxia cycles, using
longer cycles of 1 h of hypoxia followed by 30 min of
reoxygenation [49,50]. The times used in these studies
allowed the demonstration of complete HIF-1a degra-
dation after each cycle of 30 min of reoxygenation,
showing that HIF-1a had not accumulated in the
course of intermittent hypoxia cycles, and therefore
that it is its stabilization that is increased in these
conditions [50].
HIF-1a stabilization does not always translate into
HIF-1 activity. One can therefore ask whether hypoxia
periods interrupted by reoxygenation periods can be
sufficient to induce the transcription of HIF-1 target
genes. HIF-1a degradation after each reoxygenation
makes HIF-1 inactive. In these circumstances, HIF-1
can only be transcriptionally active during the hypoxia

phases, which can be short. Reporter assays showed a
significant gradual increase in hypoxia response element
(HRE) promoter activity in PC12 cells incubated under
intermittent hypoxia, in the course of hypoxia–reoxy-
genation cycles [47]. Interestingly, with the same incu-
bation time, HIF-1 transcriptional activity observed
under intermittent hypoxia was almost equal to HIF-1
transcriptional activity observed under chronic hypoxia
[50]. Moreover, with the same incubation time under
hypoxia (duration of reoxygenation under intermittent
hypoxia was not considered in this case), Yuan et al.
showed higher HIF-1 transcriptional activity in com-
parison to chronic hypoxia [47].
Although intermittent hypoxia can induce, like
chronic hypoxia, HIF-1a stabilization as well as HIF-1
transcriptional activity, some differences can be seen
between these two kinds of hypoxia. It was shown in
PC12 cells and EAhy926 endothelial cells that, under
S. Toffoli and C. Michiels Intermittent hypoxia in cancer
FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS 2993
transient hypoxia, extracellular signal-related
kinase 1 ⁄ 2 mitogen-activated protein kinases and phos-
phoinositide-3-kinase are not required for HIF-1 sta-
bilization and transcriptional activity [47,50], whereas
the inhibition of these kinases under chronic hypoxia
impaired HIF-1 target gene expression [51,52]. On the
other hand, at least in endothelial cells, protein kina-
se A is involved in HIF-1a phosphorylation under
intermittent hypoxia but not under chronic hypoxia,
and protein kinase A inhibition decreased the tran-

scription of HIF-1 target genes [50]. These results sug-
gest that the pathways regulating HIF-1 activity under
chronic or intermittent hypoxia are different. Figure 1
shows a brief comparison of HIF-1a stabilization and
HIF-1 activation under intermittent hypoxia and
chronic hypoxia.
Tumour resistance induced by
intermittent hypoxia
The effects of chronic hypoxia have been extensively
studied, and it has been clearly demonstrated that
chronic hypoxia protects tumour cells from apoptosis
induced by radiotherapy and chemotherapy [53–60].
Recent studies have shown that intermittent hypoxia
could also protect tumour cells from anticancer treat-
ments.
Martinive et al. showed, in vivo, a decrease in
tumour cell apoptosis in transplantable liver tumour
implanted in mice subjected to cycles of intermittent
hypoxia before irradiation (10 Gy) with respect to mice
kept under normoxia [49]. This inhibition of apoptosis
under transient hypoxia was also observed in vitro in
FsaII fibrocarcinoma cells and B16 melanoma cells
[49]. Moreover, Dong & Wang demonstrated the possi-
bility of death-resistant cell selection by the repetition
of hypoxia episodes [61]. Such selected cells were
shown to be resistant to cell death induced by different
types of molecules, such as azide, cisplatin and stauro-
sporine [61].
Transient hypoxia could also render tumours more
invasive. Cairns et al. observed a highly significant

increase in the number of lung micrometastases in
KHT tumour-bearing mice exposed to 12 cycles per
day (for 8–15 days) of 10 min of hypoxia followed by
10 min of reoxygenation, in comparison to control
mice. Interestingly, no increase in lung micrometastasis
was observed in mice exposed to chronic hypoxia [15],
suggesting again that intermittent hypoxia has different
effects from chronic hypoxia.
In addition, it was shown by Durand & Aquino-
Parsons that blood flow decreases could transiently
arrest the division of tumour cells in S-phase [62].
These cells are the main targets of chemotherapy, and
the arrest of their cell cycle during S-phase reduces
considerably their sensitivity to antiproliferative drugs,
but it also implies a more rapid initiation of tumour
cell repopulation when the blood flow restarts [62].
More generally, the transient cessation of tumour
blood flow reduces tumour cell exposure to the most
highly diffusible anticancer agents, but also reduces
their sensitivity to radiotherapy because of the decrease
in oxygen supply during tumour irradiation [57,62].
The protection against antitumour treatment can also
be linked to a particular phenotype acquired by the
cells in the course of intermittent hypoxia phases. An
Fig. 1. Effects of intermittent hypoxia and
chronic hypoxia on HIF-1a stabilization and
HIF-1 target gene transcription.
Intermittent hypoxia in cancer S. Toffoli and C. Michiels
2994 FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS
example of such acquired resistance is described by

Dong & Wang [61]: upregulation of Bcl-X
L
has been
observed in immortalized rat kidney epithelial cells
exposed to repeated periods of hypoxia. It was shown
in these cells that Bcl-X
L
could directly interact with
the proapoptotic molecule Bax at the mitochondrial
level, impeding Bax oligomerization and cytochrome c
release, and hence preventing cell apoptosis [61].
Genetic instability due to abnormal DNA meta-
bolism linked to impaired activity of enzymes such as
topoisomerases, helicases and ligases is often observed
in hypoxic tumours [63]. Strand breaks, translocations,
transversions and other chromosomal rearrangements
observed in these conditions can also be responsible
for tumour resistance. Reynolds et al. showed that
hypoxia could induce a 3–4-fold elevation in mutation
frequency, and higher levels of mutagenesis were
observed in cells exposed multiple times to hypoxia
[63], suggesting that exposure of cells to transient
hypoxia could also induce resistance to antitumour
treatments by this mechanism. Moreover, oxidative
injuries generated by reoxygenation in the course of
intermittent hypoxia phases can also be responsible for
DNA damage through an increase in 8-oxoguanine,
which has been shown to miscode for A and lead to
C:G to A:T transversions [64].
Reactive oxygen species (ROS) generated during the

reoxygenation periods can also play an important role,
modifying gene expression through the regulation of
the activity of some transcription factors, such as
activator protein-1 (AP-1) or nuclear factor kappaB
(NF-jB).
AP-1 is known to play a pivotal role in tumorigen-
esis, regulating the expression and function of cell
cycle regulators such cyclin D1, p53, p21, p19, and
p16. Moreover, its activity was shown to increase in
multiple human tumour types, and its inhibition can
block tumour promotion, transformation, progression,
and invasion [65]. AP-1 activation was shown in
PC12 cells under intermittent hypoxia, and was
clearly associated with ROS production and, more
particularly, with superoxide (O
2
Æ
–
) anion generation
[66]. Furthermore, it was shown that AP-1 activation
involved c-fos, the activation of which persisted for
several hours after the intermittent hypoxia ‘stimulus’
[66]. Deregulation of c-fos and c-jun proteins can
induce transformation in vivo [67], and c-fos upregu-
lation was shown in tumour formation and, more
particularly, in liver tumour development. In this kind
of tumour, AP-1 and c-fos were shown to be able to
downregulate tumour suppressor genes and favour
angiogenesis and tumour invasiveness [68]. Therefore,
AP-1 activation under intermittent hypoxia, associated

with c-fos upregulation, could promote tumour devel-
opment.
NF-jB can also be activated by ROS [69]. ROS pro-
duction during the reoxygenation periods [70] might
also be able to activate NF-jB. Ryan et al. showed in
HeLa cells and bovine aortic endothelial cells that
transient hypoxia activated NF-jB in a number of
hypoxia–reoxygenation cycles in an ROS-dependent
manner [71]. Despite the potential production of ROS
during reoxygenation concomitant with NF-jB activa-
tion, these authors suggested that NF-jB activation
under intermittent hypoxia was not linked to ROS
production, because no decrease in NF-jB activation
in the presence of the ROS scavenger N-acetyl-l-cyste-
ine was observed [71]. However, inhibition of NF-jB
activation by N-acetyl-l-cysteine has been shown to
occur not through ROS-dependent mechanisms, but
rather through inhibition of tumour necrosis factor-
stimulated signal transduction by lowering tumour
necrosis factor receptor affinity [72,73] or through inhi-
bition of its DNA-binding activity [74]. Therefore,
involvement of ROS in NF-jB activation under inter-
mittent hypoxia cannot be completely excluded. It has
to be noted that NF-jB activation by ROS is extre-
mely cell type-dependent. Beyond the question of the
regulation mechanisms of NF-jB, its activation under
intermittent hypoxia remains a critical point, because
NF-jB plays an important role in tumour development
through its ability to induce the transcription of genes
coding for apoptosis inhibitor factors (cIAPs, Bcl-X

L
,
FLICE), proproliferation molecules (interleukin-2, G1
cyclins), proangiogenic factors (VEGF, interleukin-8),
and enzymes that lead to extracellular matrix degra-
dation (matrix metalloproteases) [75–78]. In addition,
NF-jB activation was reported as an early event in
malignant transformation in vitro [79], and continuous
activation of NF-jB was also shown in many kinds of
solid tumours [80]. Conversely, NF-jB inhibition
impairs tumour development. NF-jB inhibition in
prostate cancer in mice led to a marked reduction in
the growth of tumour, demonstrating again the impor-
tant role played by this transcription factor in tumour
development [80].
HIF-1 activation in tumours is often associated with
a poor prognosis. It allows the tumour cells to survive
in the absence of oxygen, regulating their cell metabo-
lism and inducing the production of prosurvival mole-
cules, but also inducing the formation of new blood
vessels, favouring metastasis [81]. HIF-1 activation is
always associated with hypoxia, but it was shown that
the production of ROS under normoxia was also able
to stabilize HIF-1a subunit and contribute to HIF-1
activation [82]. The production of ROS during the
S. Toffoli and C. Michiels Intermittent hypoxia in cancer
FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS 2995
reoxygenation periods under intermittent hypoxia
could then also influence HIF-1 activity. However,
HIF-1a subunit degradation is always observed during

reoxygenation after incubation under chronic or tran-
sient hypoxia. Therefore, HIF-1 activation during
reoxygenation after a hypoxia period should be
impaired in this case. Paradoxically, it was shown that
reoxygenation could stimulate HIF-1 signalling.
Increases in the translation of HIF-1 target genes and
HRE–green fluorescent protein construction transcripts
were observed after reoxygenation, despite the com-
plete degradation of HIF-1a [83,84]. This peculiar
observation was explained by Moeller et al., who
showed that reoxygenation could enhance downstream
HIF-1 signalling by depolymerizing stress granules.
They showed that a pool of HIF-1-regulated tran-
scripts were kept untranslated in the course of hypoxia
in stress granules that were depolymerized during reox-
ygenation, allowing the rapid translation of seques-
trated transcripts under normoxia [83]. Interestingly,
Moeller et al. also observed stress granule formation in
tumour cells under hypoxia as well as their degrada-
tion during reoxygenation. Hence, they suggested that
this post-transcriptional regulation process could help
cancer cells to recover from a hypoxic shock and pre-
pare the cells for a future insult. This mechanism, the
regulation of which could involve ROS, as suggested
again by Moeller et al., could also explain, at least in
part, the cancer cell resistance to anticancer treatment
observed under intermittent hypoxia. It would be inter-
esting to investigate the involvement of stress granules
in the gradual increase in the abundance of HIF-1a
observed after each hypoxia step in the course of

hypoxia–reoxygenation cycles.
Effects of intermittent hypoxia on
tumour vasculature
Tumour blood vessel formation is essential for tumour
development. As well as comprising a tumour cell dis-
semination pathway in the body, tumour blood vessels
supply to cancer cells the oxygen and nutrients essen-
tial for their survival and proliferation. In the absence
of angiogenesis and new blood vessel formation,
tumour growth is restricted, and the tumour size
remains ‘microscopic’, generally not increasing beyond
0.5 mm, even in the case of a highly proliferative
tumour, in which cell division is balanced by cell apop-
tosis induced by unfavourable survival conditions [85–
88]. In these circumstances, in situ tumours can remain
dormant ‘indefinitely’ in the absence of angiogenesis
[89]. Indeed, antiangiogenic treatments were shown to
be able to impair or slow down tumour development
and to reduce the volume of some solid tumours [90–
92]. One of the main targets of these treatments com-
prises the endothelial cells. One endothelial cell can
control the survival of approximately 50–100 tumour
cells [93]. Therefore, the destruction of a few endothe-
lial cells may induce the death of a large number of
tumour cells. Moreover, it was shown that endothelial
cell suppression could also mediate apoptosis in drug-
resistant tumour cells [94,95]. The role played by the
tumour vascular network is thus critical in the devel-
opment of a tumour, and therefore the effects of the
tumour microenvironment on the cells constituting this

network, i.e. the endothelial cells, must also be consid-
ered. Indeed, the tumour environment induces faster
endothelial cell proliferation than in normal tissue, and
the turnover of endothelial cells in tumours was
estimated to be 20–2000 times faster [96]. Moreover,
significant differences have been shown in the tran-
scriptome of tumour endothelial cells in comparison to
endothelium in surrounding normal tissue [97–99]. In
addition, tumour cells can favour endothelial cell
survival within tumours by the production of VEGF,
and particularly after irradiation [100]. As described
previously in this review, intermittent hypoxia influ-
ences tumour cell behaviour. Transient hypoxia also
affects endothelial cells. It was shown in vivo that
intermittent hypoxia had a proangiogenic effect. An
increase in capillary density in mouse brains was
observed after the repetition of cycles of 4 min of
hypoxia followed by 4 min of reoxygenation for
2 weeks [101]. Moreover, in vitro, an increase in endo-
thelial cell migration and formation of tubes was also
reported under intermittent hypoxia [49]. Therefore,
transient hypoxia could increase angiogenic processes
also in tumours. Furthermore, it was observed that
endothelial cells become, like tumour cells, radio-
resistant after an intermittent hypoxia preconditioning.
In vitro, an increase in the survival of endothelial cells
was observed after irradiation (2 Gy) when intermit-
tent hypoxia preconditioning was performed [49].
This protective effect of intermittent hypoxia against
radiotherapy on endothelial cells was shown to be

HIF-1-dependent. Indeed, a decrease in endothelial cell
survival after a low level of irradiation (2 Gy) on cells
previously incubated under intermittent hypoxia was
shown when HIF-1a was silenced by small interfering
RNA [49]. Endothelial cell radioprotection through the
repetition of hypoxia–reoxygenation cycles was also
observed in vivo. Terminal dUTP nick-end labelling
assays showed a decrease in the number of apoptotic
cells in the vasculature of transplantable liver tumour
borne by mice when rodents were submitted to
three cycles of 1 h of hypoxia followed by 30 min of
Intermittent hypoxia in cancer S. Toffoli and C. Michiels
2996 FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS
reoxygenation before tumour irradiation (10 Gy) [49].
Interestingly, chronic hypoxia incubation before the
irradiation did not protect endothelial cells against
apoptosis: in contrast to intermittent hypoxia, it even
drastically increased cell apoptosis [49]. Moeller et al.
also showed in vivo a radioprotective effect of
hypoxia ⁄ reoxygenation in endothelial cells after irradi-
ation [83]. They suggested that this radioprotection
was induced by the secretion of endothelial cell-radio-
protective cytokines by tumour cells after reoxy-
genation. They showed that tumour cell-conditioned
medium recovered after incubation under hypoxia fol-
lowed by reoxygenation was more radioprotective for
endothelial cells than conditioned medium from
tumour cells incubated under normoxia, normoxia
with radiation, or hypoxia without reoxygenation [83].
Moreover, it was shown that this endothelial cell

radioprotection mediated by tumour cells after
hypoxia–reoxygenation was also HIF-1-dependent.
Indeed, no significant endothelial cell radioprotective
Fig. 2. Schematic representation of the
effects of intermittent hypoxia on cancer
cells and endothelial cells within a tumour.
Fig. 3. Schematic representation of the
effects of HIF-1 activation under intermittent
hypoxia on cancer cells and endothelial cells
within a tumour.
S. Toffoli and C. Michiels Intermittent hypoxia in cancer
FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS 2997
effect was observed when conditioned medium was
taken from HIF-1-incompetent tumour cells [83].
Therefore, these results suggest that intermittent
hypoxia protects endothelial cells in a direct manner
by acting directly on the endothelial cell phenotype, as
observed by Martinive et al. in vitro [49], and also by
indirect pathways involving secreted molecules released
from tumour cells, as suggested by Moeller et al. [83].
Conclusion
Until now, most attention has been paid to chronic
hypoxia. However, during the last few years, a new
concept has arisen, showing first that changes in po
2
level are not always sustained in tumours but that they
can be transient, and second that intermittent hypoxia
can exert effects that are different from those induced
by chronic hypoxia. Both tumour cells and endothelial
cells are affected by intermittent hypoxia, which can be

perceived as the consequence of different stresses
resulting from repeated combinations of hypoxia and
reoxygenation periods, which may induce different cell
responses. In contrast, chronic hypoxia causes a pro-
longed and unique modification of the cell environ-
ment. Figures 2 and 3 schematically summarize the
effects of intermittent hypoxia. The major conclusion
drawn from these observations is the intricate interplay
between tumour cells and endothelial cells, each
favouring the survival of the other. This delicate ballet
has to be understood in detail in order to allow the
design of new therapies targeting these processes.
Acknowledgements
Se
´
bastien Toffoli is recipient of a FNRS-Te
´
le
´
vie grant.
Carine Michiels is research director of FNRS (Fonds
National de la Recherche Scientifique, Belgium). This
article presents results of the Belgian Programme on
Interuniversity Poles of Attraction initiated by the
Belgian State, Prime Minister’s Office, Science Policy
Programming. The responsibility is assumed by its
authors.
References
1 Brown JM (1990) Tumor hypoxia, drug resistance, and
metastases. J Natl Cancer Inst 82, 338–339.

2 Thomlinson RH & Gray LH (1955) The histological
structure of some human lung cancers and the possible
implications for radiotherapy. Br J Cancer 9, 539–549.
3 Brown JM (1979) Evidence for acutely hypoxic cells in
mouse tumours, and a possible mechanism of reoxy-
genation. Br J Radiol 52, 650–656.
4 Chaplin DJ, Durand RE & Olive PL (1986) Acute
hypoxia in tumors: implications for modifiers of radia-
tion effects. Int J Radiat Oncol Biol Phys 12, 1279–
1282.
5 Chaplin DJ, Olive PL & Durand RE (1987) Intermit-
tent blood flow in a murine tumor: radiobiological
effects. Cancer Res 47, 597–601.
6 Bergers G & Benjamin LE (2003) Tumorigenesis and
the angiogenic switch. Nat Rev Cancer 3, 401–410.
7 Morikawa S, Baluk P, Kaidoh T, Haskell A, Jain RK
& McDonald DM (2002) Abnormalities in pericytes on
blood vessels and endothelial sprouts in tumors. Am J
Pathol 160, 985–1000.
8 Padera TP, Stoll BR, Tooredman JB, Capen D, di
Tomaso E & Jain RK (2004) Pathology: cancer cells
compress intratumour vessels. Nature 427, 695.
9 Baudelet C, Cron GO, Ansiaux R, Crokart N, DeW-
ever J, Feron O & Gallez B (2006) The role of vessel
maturation and vessel functionality in spontaneous
fluctuations of T2*-weighted GRE signal within
tumors. NMR Biomed 19, 69–76.
10 Chaplin DJ & Hill SA (1995) Temporal heterogeneity
in microregional erythrocyte flux in experimental solid
tumours. Br J Cancer 71, 1210–1213.

11 Bennewith KL & Durand RE (2004) Quantifying tran-
sient hypoxia in human tumor xenografts by flow
cytometry. Cancer Res 64, 6183–6189.
12 Braun RD, Lanzen JL & Dewhirst MW (1999) Fourier
analysis of fluctuations of oxygen tension and blood
flow in R3230Ac tumors and muscle in rats. Am J
Physiol 277, H551–H568.
13 Hill SA, Pigott KH, Saunders MI, Powell ME, Arnold
S, Obeid A, Ward G, Leahy M, Hoskin PJ & Chaplin
DJ (1996) Microregional blood flow in murine and
human tumours assessed using laser Doppler microp-
robes. Br J Cancer Suppl 27, S260–S263.
14 Pigott KH, Hill SA, Chaplin DJ & Saunders MI
(1996) Microregional fluctuations in perfusion within
human tumours detected using laser Doppler flowme-
try. Radiother Oncol 40, 45–50.
15 Cairns RA, Kalliomaki T & Hill RP (2001) Acute (cyc-
lic) hypoxia enhances spontaneous metastasis of KHT
murine tumors. Cancer Res 61, 8903–8908.
16 Cardenas-Navia LI, Yu D, Braun RD, Brizel DM,
Secomb TW & Dewhirst MW (2004) Tumor-dependent
kinetics of partial pressure of oxygen fluctuations dur-
ing air and oxygen breathing. Cancer Res 64, 6010–
6017.
17 Nordsmark M, Bentzen SM & Overgaard J (1994) Mea-
surement of human tumour oxygenation status by a
polarographic needle electrode. An analysis of inter-
and intratumour heterogeneity. Acta Oncol 33, 383–389.
18 Foo SS, Abbott DF, Lawrentschuk N & Scott AM
(2004) Functional imaging of intratumoral hypoxia.

Mol Imaging Biol 6, 291–305.
Intermittent hypoxia in cancer S. Toffoli and C. Michiels
2998 FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS
19 Nozue M, Lee I, Yuan F, Teicher BA, Brizel DM,
Dewhirst MW, Milross CG, Milas L, Song CW, Tho-
mas CD et al. (1997) Interlaboratory variation in oxy-
gen tension measurement by Eppendorf ‘Histograph’
and comparison with hypoxic marker. J Surg Oncol
66, 30–38.
20 Jakobsson A & Nilsson GE (1993) Prediction of sam-
pling depth and photon pathlength in laser Doppler
flowmetry. Med Biol Eng Comput 31, 301–307.
21 Nilsson GE, Tenland T & Oberg PA (1980) Evaluation
of a laser Doppler flowmeter for measurement of tissue
blood flow. IEEE Trans Biomed Eng 27, 597–604.
22 Ljungkvist AS, Bussink J, Kaanders JH & van der
Kogel AJ (2007) Dynamics of tumor hypoxia measured
with bioreductive hypoxic cell markers. Radiat Res
167, 127–145.
23 Padhani AR, Krohn KA, Lewis JS & Alber M (2007)
Imaging oxygenation of human tumours. Eur Radiol
17, 861–872.
24 Gallez B, Baudelet C & Jordan BF (2004) Assessment
of tumor oxygenation by electron paramagnetic reso-
nance: principles and applications. NMR Biomed 17,
240–262.
25 Landuyt W, Hermans R, Bosmans H, Sunaert S,
Beatse E, Farina D, Meijerink M, Zhang H, Van Den
Bogaert W, Lambin P et al. (2001) BOLD contrast
fMRI of whole rodent tumour during air or carbogen

breathing using echo-planar imaging at 1.5 T. Eur
Radiol 11, 2332–2340.
26 Ogawa S, Lee TM, Kay AR & Tank DW (1990) Brain
magnetic resonance imaging with contrast dependent
on blood oxygenation. Proc Natl Acad Sci USA 87,
9868–9872.
27 Mahy P, De Bast M, Gallez B, Gueulette J, Koch CJ,
Scalliet P & Gregoire V (2003) In vivo colocalization
of 2-nitroimidazole EF5 fluorescence intensity and elec-
tron paramagnetic resonance oximetry in mouse
tumors. Radiother Oncol 67, 53–61.
28 Dewhirst MW, Braun RD & Lanzen JL (1998) Tempo-
ral changes in PO2 of R3230AC tumors in Fischer-344
rats. Int J Radiat Oncol Biol Phys 42, 723–726.
29 Ljungkvist AS, Bussink J, Rijken PF, Raleigh JA,
Denekamp J & Van Der Kogel AJ (2000) Changes in
tumor hypoxia measured with a double hypoxic mar-
ker technique. Int J Radiat Oncol Biol Phys 48, 1529–
1538.
30 Rijken PF, Bernsen HJ, Peters JP, Hodgkiss RJ,
Raleigh JA & van der Kogel AJ (2000) Spatial rela-
tionship between hypoxia and the (perfused) vascular
network in a human glioma xenograft: a quantitative
multi-parameter analysis. Int J Radiat Oncol Biol Phys
48, 571–582.
31 Arteel GE, Thurman RG & Raleigh JA (1998) Reduc-
tive metabolism of the hypoxia marker pimonidazole
is regulated by oxygen tension independent of the
pyridine nucleotide redox state. Eur J Biochem 253,
743–750.

32 Joseph P, Jaiswal AK, Stobbe CC & Chapman JD
(1994) The role of specific reductases in the intracellu-
lar activation and binding of 2-nitroimidazoles. Int J
Radiat Oncol Biol Phys 29, 351–355.
33 Melo T, Ballinger JR & Rauth AM (2000) Role of
NADPH:cytochrome P450 reductase in the hypoxic
accumulation and metabolism of BRU59-21, a techne-
tium-99m-nitroimidazole for imaging tumor hypoxia.
Biochem Pharmacol 60, 625–634.
34 Saleem A, Charnley N & Price P (2006) Clinical molec-
ular imaging with positron emission tomography. Eur
J Cancer 42, 1720–1727.
35 Koh WJ, Bergman KS, Rasey JS, Peterson LM, Evans
ML, Graham MM, Grierson JR, Lindsley KL, Lewel-
len TK, Krohn KA et al. (1995) Evaluation of oxygen-
ation status during fractionated radiotherapy in human
nonsmall cell lung cancers using [F-18]fluoromisonida-
zole positron emission tomography. Int J Radiat Oncol
Biol Phys 33, 391–398.
36 Chi JT, Wang Z, Nuyten DS, Rodriguez EH, Schaner
ME, Salim A, Wang Y, Kristensen GB, Helland A,
Borresen-Dale AL et al. (2006) Gene expression pro-
grams in response to hypoxia: cell type specificity and
prognostic significance in human cancers. PLoS Med 3,
e47.
37 Wang GL, Jiang BH, Rue EA & Semenza GL (1995)
Hypoxia-inducible factor 1 is a basic-helix-loop-helix-
PAS heterodimer regulated by cellular O2 tension.
Proc Natl Acad Sci USA 92, 5510–5514.
38 Wang GL & Semenza GL (1995) Purification and

characterization of hypoxia-inducible factor 1. J Biol
Chem 270, 1230–1237.
39 Salceda S & Caro J (1997) Hypoxia-inducible fac-
tor 1alpha (HIF-1alpha) protein is rapidly degraded
by the ubiquitin–proteasome system under normoxic
conditions. Its stabilization by hypoxia depends on
redox-induced changes. J Biol Chem 272, 22642–22647.
40 Semenza GL (2001) HIF-1, O(2), and the 3 PHDs: how
animal cells signal hypoxia to the nucleus. Cell 107,
1–3.
41 Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh
M, Salic A, Asara JM, Lane WS & Kaelin WG Jr
(2001) HIFalpha targeted for VHL-mediated destruc-
tion by proline hydroxylation: implications for O2
sensing. Science 292, 464–468.
42 Cockman ME, Masson N, Mole DR, Jaakkola P,
Chang GW, Clifford SC, Maher ER, Pugh CW, Ratc-
liffe PJ & Maxwell PH (2000) Hypoxia inducible fac-
tor-alpha binding and ubiquitylation by the von
Hippel–Lindau tumor suppressor protein. J Biol Chem
275, 25733–25741.
43 Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert
J, Gaskell SJ, Kriegsheim A, Hebestreit HF, Mukherji
S. Toffoli and C. Michiels Intermittent hypoxia in cancer
FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS 2999
M, Schofield CJ et al. (2001) Targeting of HIF-alpha
to the von Hippel–Lindau ubiquitylation complex by
O2-regulated prolyl hydroxylation. Science 292, 468–
472.
44 Jewell UR, Kvietikova I, Scheid A, Bauer C, Wenger

RH & Gassmann M (2001) Induction of HIF-1alpha
in response to hypoxia is instantaneous. FASEB J 15,
1312–1314.
45 Jiang BH, Rue E, Wang GL, Roe R & Semenza GL
(1996) Dimerization, DNA binding, and transactiva-
tion properties of hypoxia-inducible factor 1. J Biol
Chem 271, 17771–17778.
46 Wenger RH (2000) Mammalian oxygen sensing, signal-
ling and gene regulation. J Exp Biol 203, 1253–1263.
47 Yuan G, Nanduri J, Bhasker CR, Semenza GL & Pra-
bhakar NR (2005) Ca
2+
⁄ calmodulin kinase-dependent
activation of hypoxia inducible factor 1 transcriptional
activity in cells subjected to intermittent hypoxia.
J Biol Chem 280, 4321–4328.
48 Berra E, Richard DE, Gothie E & Pouyssegur J (2001)
HIF-1-dependent transcriptional activity is required for
oxygen-mediated HIF-1alpha degradation. FEBS Lett
491, 85–90.
49 Martinive P, Defresne F, Bouzin C, Saliez J, Lair F,
Gregoire V, Michiels C, Dessy C & Feron O (2006)
Preconditioning of the tumor vasculature and tumor
cells by intermittent hypoxia: implications for antican-
cer therapies. Cancer Res 66, 11736–11744.
50 Toffoli S, Feron O, Raes M & Michiels C (2007) Inter-
mittent hypoxia changes HIF-1alpha phosphorylation
pattern in endothelial cells: unravelling of a new PKA-
dependent regulation of HIF-1alpha. Biochim Biophys
Acta 1773, 1558–1571.

51 Mottet D, Dumont V, Deccache Y, Demazy C, Ninane
N, Raes M & Michiels C (2003) Regulation of
hypoxia-inducible factor-1alpha protein level during
hypoxic conditions by the phosphatidylinositol
3-kinase ⁄ Akt ⁄ glycogen synthase kinase 3beta pathway
in HepG2 cells. J Biol Chem 278, 31277–31285.
52 Minet E, Arnould T, Michel G, Roland I, Mottet D,
Raes M, Remacle J & Michiels C (2000) ERK activa-
tion upon hypoxia: involvement in HIF-1 activation.
FEBS Lett 468, 53–58.
53 Graeber TG, Osmanian C, Jacks T, Housman DE,
Koch CJ, Lowe SW & Giaccia AJ (1996) Hypoxia-
mediated selection of cells with diminished apoptotic
potential in solid tumours. Nature 379, 88–91.
54 Cosse JP, Sermeus A, Vannuvel K, Ninane N, Raes M
& Michiels C (2007) Differential effects of hypoxia on
etoposide-induced apoptosis according to the cancer
cell lines. Mol Cancer 6, 61–76.
55 Lee SM, Lee CT, Kim YW, Han SK, Shim YS & Yoo
CG (2006) Hypoxia confers protection against apopto-
sis via PI3K ⁄ Akt and ERK pathways in lung cancer
cells. Cancer Lett 242, 231–238.
56 Merighi S, Benini A, Mirandola P, Gessi S, Varani K,
Leung E, Maclennan S, Baraldi PG & Borea PA
(2007) Hypoxia inhibits paclitaxel-induced apoptosis
through adenosine-mediated phosphorylation of bad in
glioblastoma cells. Mol Pharmacol 72, 162–172.
57 Moeller BJ, Richardson RA & Dewhirst MW (2007)
Hypoxia and radiotherapy: opportunities for improved
outcomes in cancer treatment. Cancer Metastasis Rev

26, 241–248.
58 Piret JP, Lecocq C, Toffoli S, Ninane N, Raes M &
Michiels C (2004) Hypoxia and CoCl2 protect HepG2
cells against serum deprivation- and t-BHP-induced
apoptosis: a possible anti-apoptotic role for HIF-1.
Exp Cell Res 295, 340–349.
59 Schnitzer SE, Schmid T, Zhou J & Brune B (2006)
Hypoxia and HIF-1alpha protect A549 cells from drug-
induced apoptosis. Cell Death Differ 13, 1611–1613.
60 Song X, Liu X, Chi W, Liu Y, Wei L, Wang X & Yu
J (2006) Hypoxia-induced resistance to cisplatin and
doxorubicin in non-small cell lung cancer is inhibited
by silencing of HIF-1alpha gene. Cancer Chemother
Pharmacol 58, 776–784.
61 Dong Z & Wang J (2004) Hypoxia selection of death-
resistant cells. A role for Bcl-X(L). J Biol Chem 279,
9215–9221.
62 Durand RE & Aquino-Parsons C (2001) Non-constant
tumour blood flow – implications for therapy. Acta
Oncol 40, 862–869.
63 Reynolds TY, Rockwell S & Glazer PM (1996) Genetic
instability induced by the tumor microenvironment.
Cancer Res 56, 5754–5757.
64 Cheng KC, Cahill DS, Kasai H, Nishimura S & Loeb
LA (1992) 8-Hydroxyguanine, an abundant form of
oxidative DNA damage, causes G- - - -T and A- - - -C
substitutions. J Biol Chem 267
, 166–172.
65 Matthews CP, Colburn NH & Young MR (2007) AP-1
a target for cancer prevention. Curr Cancer Drug

Targets 7, 317–324.
66 Yuan G, Adhikary G, McCormick AA, Holcroft JJ,
Kumar GK & Prabhakar NR (2004) Role of oxidative
stress in intermittent hypoxia-induced immediate early
gene activation in rat PC12 cells. J Physiol 557, 773–
783.
67 Shaulian E & Karin M (2001) AP-1 in cell prolifera-
tion and survival. Oncogene 20, 2390–2400.
68 Eferl R & Wagner EF (2003) AP-1: a double-edged
sword in tumorigenesis. Nat Rev Cancer 3, 859–868.
69 Gloire G, Legrand-Poels S & Piette J (2006) NF-kap-
paB activation by reactive oxygen species: fifteen years
later. Biochem Pharmacol 72, 1493–1505.
70 Prabhakar NR (2001) Oxygen sensing during intermit-
tent hypoxia: cellular and molecular mechanisms.
J Appl Physiol 90, 1986–1994.
71 Ryan S, McNicholas WT & Taylor CT (2007) A criti-
cal role for p38 map kinase in NF-kappaB signaling
Intermittent hypoxia in cancer S. Toffoli and C. Michiels
3000 FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS
during intermittent hypoxia ⁄ reoxygenation. Biochem
Biophys Res Commun 355, 728–733.
72 Hayakawa M, Miyashita H, Sakamoto I, Kitagawa M,
Tanaka H, Yasuda H, Karin M & Kikugawa K (2003)
Evidence that reactive oxygen species do not mediate
NF-kappaB activation. EMBO J 22, 3356–3366.
73 Yang J, Su Y & Richmond A (2007) Antioxidants
tiron and N-acetyl-L-cysteine differentially mediate
apoptosis in melanoma cells via a reactive oxygen spe-
cies-independent NF-kappaB pathway. Free Radic Biol

Med 42, 1369–1380.
74 Qanungo S, Wang M & Nieminen AL (2004) N-Ace-
tyl-L-cysteine enhances apoptosis through inhibition of
nuclear factor-kappaB in hypoxic murine embryonic
fibroblasts. J Biol Chem 279, 50455–50464.
75 Bond M, Fabunmi RP, Baker AH & Newby AC
(1998) Synergistic upregulation of metalloproteinase-9
by growth factors and inflammatory cytokines: an
absolute requirement for transcription factor NF-
kappa B. FEBS Lett 435, 29–34.
76 Guttridge DC, Albanese C, Reuther JY, Pestell RG &
Baldwin AS Jr (1999) NF-kappaB controls cell growth
and differentiation through transcriptional regulation
of cyclin D1. Mol Cell Biol 19, 5785–5799.
77 Huang S, Robinson JB, Deguzman A, Bucana CD &
Fidler IJ (2000) Blockade of nuclear factor-kappaB sig-
naling inhibits angiogenesis and tumorigenicity of
human ovarian cancer cells by suppressing expression
of vascular endothelial growth factor and interleukin 8.
Cancer Res 60, 5334–5339.
78 Karin M, Cao Y, Greten FR & Li ZW (2002) NF-kap-
paB in cancer: from innocent bystander to major
culprit. Nat Rev Cancer 2, 301–310.
79 Kim DW, Sovak MA, Zanieski G, Nonet G, Romieu-
Mourez R, Lau AW, Hafer LJ, Yaswen P, Stampfer
M, Rogers AE et al. (2000) Activation of NF-kap-
paB ⁄ Rel occurs early during neoplastic transformation
of mammary cells. Carcinogenesis 21, 871–879.
80 Rayet B & Gelinas C (1999) Aberrant rel ⁄ nfkb genes
and activity in human cancer. Oncogene 18, 6938–

6947.
81 Zhong H, De Marzo AM, Laughner E, Lim M, Hilton
DA, Zagzag D, Buechler P, Isaacs WB, Semenza GL
& Simons JW (1999) Overexpression of hypoxia-induc-
ible factor 1alpha in common human cancers and their
metastases. Cancer Res 59, 5830–5835.
82 Kietzmann T & Gorlach A (2005) Reactive oxygen
species in the control of hypoxia-inducible factor-
mediated gene expression. Semin Cell Dev Biol 16,
474–486.
83 Moeller BJ, Cao Y, Li CY & Dewhirst MW (2004)
Radiation activates HIF-1 to regulate vascular
radiosensitivity in tumors: role of reoxygenation,
free radicals, and stress granules. Cancer Cell 5, 429–
441.
84 Vordermark D, Shibata T & Brown JM (2001) Green
fluorescent protein is a suitable reporter of tumor
hypoxia despite an oxygen requirement for chromo-
phore formation. Neoplasia 3, 527–534.
85 Brem S, Brem H, Folkman J, Finkelstein D & Patz A
(1976) Prolonged tumor dormancy by prevention of
neovascularization in the vitreous. Cancer Res 36 ,
2807–2812.
86 Folkman J (2003) Angiogenesis and apoptosis. Semin
Cancer Biol 13, 159–167.
87 Gimbrone MA Jr, Leapman SB, Cotran RS & Folk-
man J (1972) Tumor dormancy in vivo by prevention
of neovascularization. J Exp Med 136, 261–276.
88 Holmgren L, O’Reilly MS & Folkman J (1995) Dor-
mancy of micrometastases: balanced proliferation and

apoptosis in the presence of angiogenesis suppression.
Nat Med 1, 149–153.
89 Folkman J & Kalluri R (2004) Cancer without disease.
Nature 427, 787.
90 Wood JM, Bold G, Buchdunger E, Cozens R, Ferrari
S, Frei J, Hofmann F, Mestan J, Mett H, O’Reilly T
et al. (2000) PTK787 ⁄ ZK 222584, a novel and potent
inhibitor of vascular endothelial growth factor receptor
tyrosine kinases, impairs vascular endothelial growth
factor-induced responses and tumor growth after oral
administration. Cancer Res 60, 2178–2189.
91 Willett CG, Boucher Y, di Tomaso E, Duda DG,
Munn LL, Tong RT, Chung DC, Sahani DV, Kalva
SP, Kozin SV et al. (2004) Direct evidence that
the VEGF-specific antibody bevacizumab has anti-
vascular effects in human rectal cancer. Nat Med 10,
145–147.
92 Goto H, Yano S, Matsumori Y, Ogawa H, Blakey DC
& Sone S (2004) Sensitization of tumor-associated
endothelial cell apoptosis by the novel vascular-target-
ing agent ZD6126 in combination with cisplatin. Clin
Cancer Res 10, 7671–7676.
93 Folkman J (2002) Looking for a good endothelial
address. Cancer Cell 1 , 113–115.
94 Baguley BC, Holdaway KM, Thomsen LL, Zhuang L
& Zwi LJ (1991) Inhibition of growth of colon 38
adenocarcinoma by vinblastine and colchicine:
evidence for a vascular mechanism. Eur J Cancer 27,
482–487.
95 Browder T, Butterfield CE, Kraling BM, Shi B,

Marshall B, O’Reilly MS & Folkman J (2000) Antian-
giogenic scheduling of chemotherapy improves efficacy
against experimental drug-resistant cancer. Cancer Res
60, 1878–1886.
96 Hobson B & Denekamp J (1984) Endothelial prolifera-
tion in tumours and normal tissues: continuous label-
ling studies. Br J Cancer 49, 405–413.
97 St Croix B, Rago C, Velculescu V, Traverso G,
Romans KE, Montgomery E, Lal A, Riggins GJ,
Lengauer C, Vogelstein B et al. (2000) Genes
S. Toffoli and C. Michiels Intermittent hypoxia in cancer
FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS 3001
expressed in human tumor endothelium. Science 289,
1197–1202.
98 Hewett P, Popplewell A, Finney H & Murray JC
(1999) Changes in microvessel endothelial cell gene
expression in an in vitro human breast tumour endo-
thelial cell model. Angiogenesis 3, 221–229.
99 Lu C, Bonome T, Li Y, Kamat AA, Han LY, Sch-
mandt R, Coleman RL, Gershenson DM, Jaffe RB,
Birrer MJ et al. (2007) Gene alterations identified by
expression profiling in tumor-associated endothelial
cells from invasive ovarian carcinoma. Cancer Res 67,
1757–1768.
100 Gorski DH, Beckett MA, Jaskowiak NT, Calvin DP,
Mauceri HJ, Salloum RM, Seetharam S, Koons A,
Hari DM, Kufe DW et al. (1999) Blockage of the vas-
cular endothelial growth factor stress response
increases the antitumor effects of ionizing radiation.
Cancer Res 59, 3374–3378.

101 Kanaan A, Farahani R, Douglas RM, Lamanna JC &
Haddad GG (2006) Effect of chronic continuous or
intermittent hypoxia and reoxygenation on cerebral
capillary density and myelination. Am J Physiol Regul
Integr Comp Physiol 290, R1105–R1114.
Intermittent hypoxia in cancer S. Toffoli and C. Michiels
3002 FEBS Journal 275 (2008) 2991–3002 ª 2008 The Authors Journal compilation ª 2008 FEBS