| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
and HIF-2 Expression to Vascular Endothelial Growth Factor Induction and Hypoxia Survival in Human Breast Cancer Cell Lines
Imperial Cancer Research Fund, Molecular Oncology Unit, Institute of Molecular Medicine, John Radcliffe Hospital, Oxford OX3 9DS, United Kingdom
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and HIF-2 and their effects on survival under
hypoxia were studied in six human breast cancer cell lines. We also
evaluated the basal and inducible expression of two hypoxically
regulated genes, vascular endothelial growth factor (VEGF) and lactate
dehydrogenase-A (LDH-A). All of the cell lines studied expressed
HIF-1 at various levels, but HIF-2 was low or absent from the
more aggressive cell lines. There was an inverse correlation between
HIF-1 and HIF-2 induction and clonogenic survival under hypoxia.
Thus, cell lines with reduced induction of HIF-1 or HIF-2 showed
high basal levels of VEGF and improved survival under hypoxia. A
reduction in HIF expression was also associated with a more aggressive
phenotype in vivo. To confirm these results, we carried
out stable transfection of the MDA 435 cell line with human HIF-2
cDNA. There was no change in the growth rate in monolayer culture.
However, in vitro growth as colonies and in
vivo tumor growth of the HIF-2 overexpressing cells were
significantly impaired compared with the control transfectants. Thus,
despite the fact that HIF proteins are necessary for optimal tumor
growth and angiogenesis in vivo, overexpression of these
molecules seems detrimental to tumor growth. A balance between the
angiogenic and tumor-inhibiting levels of HIF proteins may, therefore,
be necessary for optimal tumor growth. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and HIF-1ß subunits (4)
. HIF-1ß,
also known as aryl hydrocarbon receptor nuclear translocator, is
constitutively expressed (5)
, whereas HIF-1 is
protected from ubiquitination and proteasomal degradation under hypoxic
conditions (5, 6, 7, 8)
. Both HIF-1 subunits belong to a growing
family of basic helix-loop-helix periodic acid Schiff proteins
(4)
. Recently, another member of the family has been
reported that showed close sequence homology and similar properties to
HIF-1. This molecule was initially described as endothelial and
fetal specific and named endothelial PAS protein-1/HIF-related
factor/HIF-like factor (9, 10, 11)
. It was recently renamed
HIF-2 (2)
.
The ability of cancer cells to respond to and survive under hypoxia
through induction of VEGF and glycolytic enzymes relates to their
capacity to sense hypoxia and to activate the HIF pathway. Indeed, it
has been shown that in transformed cells, defective HIF-1 signaling
is associated with poor in vivo tumor growth and reduced
angiogenesis (12, 13, 14, 15, 16)
. It is therefore possible that
constitutive up-regulation or increased inducibility of HIF-1 may be
associated with increased VEGF induction and tumor growth. However,
Carmeliet et al. (17)
showed that growth of
HIF-1-/- tumors was not retarded but was
accelerated because of decreased hypoxia-induced apoptosis and
increased stress-induced proliferation.
To further evaluate the role of HIF-1 and HIF-2 in tumor biology,
we have analyzed a panel of six human breast cancer cell lines.
Measurements of oxygen levels in tumors have revealed that some tumors
contain regions that are down to 0.01–2% oxygen. In these regions,
cell proliferation was affected but not necessarily cell viability
(18)
. We have, therefore, also compared expression and
induction of HIF-1 and HIF-2 at a range of oxygen tensions and
after treatment with DFO, an iron chelator known to induce these
transcription factors under normoxia.
In this report, we show that HIF-2 has a broader spectrum of
expression than reported previously by Tian et al.
(9)
. We report that a reduced inducibility of HIF-1 and
HIF-2 is correlated with an increased survival of the breast cancer
cell lines under hypoxia.
To provide some insight on the function of HIF-2 in tumor response,
we stably transfected human MDA 435 breast cancer cells with the human
HIF-2 cDNA. We show that there was no difference between the
clonogenic survival of the HIF-2 transfectant and an empty vector
control transfectant. However, the size of the colonies was markedly
reduced, and in vivo growth rates of the MDA 435-HIF-2
transfectant xenografts were impaired significantly.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Survival Analysis.
Exponentially growing cells were trypsinized and seeded in 100-mm Petri
dishes. The plating density depended on the growth rate of the cell
line (100 and 200 cells/plate for faster growing cells, or 250 and 500
cells/plate for slower growing cells). The cells were allowed to adhere
for 4 h prior to starting the experiments. The plates were then
placed, for 24 h, in hypoxic conditions (1%
O2, 5% CO2), while their
counterparts was left in normoxia. After the incubation, cells were all
placed in normoxic conditions and allowed to grow for 2–3 weeks with
medium change every 5 days. Colonies were fixed with methanol:acetic
acid (3:1), stained with crystal violet, and counted. Colonies
containing 50 cells or more were counted as survivors.
Permanent Transfection of MDA 435 Cell Lines.
The transfectant cell lines MDA 435-HIF-2 and MDA 435-cont were
produced by electroporating MDA 435 cells with the human HIF-2 cDNA
cloned into pcDNA3-Neo mammalian expression vector or with the vector
alone, respectively. The HIF-2-pcDNA3 constructs were obtained from
Steven L. McKnight (Texas Southwest Medical Center, Dallas, TX)
(9)
. Cells were selected by growing for 2 weeks in DMEM
with 10% FCS and 3 mg/ml G418 sulfate (Life Technologies, Inc.,
Paisley, Scotland). Colonies of cells were isolated with pipette tips
and expanded by growing in the same medium. Clones were selected by
Western blot analysis using HIF-2 MoAb 190b, as described below.
RNA Analysis.
Total RNA was extracted using TRI Reagent (Sigma Chemical Co., Poole,
United Kingdom). RNase protection assays were performed essentially as
described by Petersen et al. (19)
, with
parallel hybridization using 10–30 µg of total RNA for HIF-1,
HIF-2, VEGF121, and LDH-A.
32P-labeled riboprobes were generated to the
highest specific activity possible using SP6, T3, or T7 RNA polymerase.
To attenuate the signal of the highly abundant loading control, U6
small nuclear RNA, a riboprobe of significantly lower specific
activity, was prepared (20)
. The templates used yielded
protected fragments as follows: 517 bp for
VEGF121 (accession no. M95200), 420 bp for LDH-A
(nucleotides 1070–1490, accession no. X02152), 255 bp for HIF-1
(nucleotides 764-1018, accession no. U22431), 210 bp for HIF-2
(nucleotides 2542–2762, accession no. U81984), and 106 bp for U6
(nucleotides 1–107, accession no. X01366). After resolution on 8%
polyacrylamide gels, quantification was performed using a
PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
Protein Extraction and Immunoblot Analysis.
For whole-cell extracts, cells were washed with ice-cold PBS and
collected by scraping. Cell pellets were homogenized in extraction
buffer [8 M urea, 10% glycerol, 10 mM
Tris-HCl (pH 6.8), 1% SDS, 5 mM DTT, 0.5 mM
phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, 10 µg/ml
pepstatin, and 10 µg/ml leupeptin], using an IKA Ultra-Turrax T8
homogenizer (Janke and Kunkel, Staufen, Germany) for 15 s at full
speed. Protein levels in the extracts were quantified using the Bio-Rad
DC protein assay (Bio-Rad, Hemel Hempstead, United Kingdom).
For immunoblotting, whole-cell extracts (50–100 µg/lane) were
resolved in SDS 6% polyacrylamide gels and transferred with a semidry
blotter Imm-2 (W.E.P. Co., Concord, CA) to Immobilon P membrane
(Millipore, Bedford, MA) overnight in 25 mM Tris-base, 190
mM glycine, and 15% methanol. Membranes were blocked with
5% fat-free milk, 0.05% Tween 20, in PBS. For HIF-1 detection,
MoAb 28b was used at 4 µg/ml and for HIF-2, MoAb 190b was
used at 2 µg/ml. Detection was performed with horseradish
peroxidase-conjugated goat antimouse immunoglobulins at 1:1000 (Dako,
Ely, United Kingdom) and enhanced chemiluminescence (ECL; Amersham
Corp., Buckinghamshire, United Kingdom). After analysis, membranes were
stained with Ponceau S solution (Sigma, Poole, United Kingdom) to
verify equal protein loading and transfer. HIF-1 and HIF-2
expression was quantified from a representative autoradiograph, using
densitometer analysis, and normalized to a nonspecific cross-reacting
protein.
Immunostaining for HIF-2.
Subconfluent cells were grown on chamber slides (Labtek; Nunc) and
incubated for 16 h either in air or under 0.1% hypoxia. After
washing in ice-cold PBS, cells were fixed in formaldehyde (3.7% in
PBS) for 10 min at room temperature. Cells were then washed twice in
PBS and permeabilized by incubating in 0.2% Triton X-100 in PBS for 10
min at room temperature. Slides were then incubated with MoAb 190b
hybridoma supernatant at 5 µg/ml for 1 h at room temperature,
washed in PBS for 5 min, and followed by 30 min incubation with
horseradish peroxidase-conjugated goat antimouse immunoglobulins at
1:200 (Dako). The detection was performed using 3'3'-diaminobenzidine,
and cells were counterstained with hematoxylin.
Determination of VEGF Protein Levels in Cell Culture Medium.
Subconfluent cells were grown in 100-mm Petri dishes with 10 ml of
fresh medium for 16 h, either in air or under 0.1% oxygen. Cell
supernatants were collected, clarified by centrifugation at 2000 rpm
for 5 min, and stored at -20°C. Concomitantly, cell pellets were
harvested by trypsinization, and cell number was determined with an
automated cell counter (Beckman Coulter, High Wycombe, Buckinghamshire,
United Kingdom). The amount of VEGF in the supernatant was determined
with an ELISA kit (VEGF-ELISA; R&D Systems, Minneapolis, MN) according
to the manufacturer’s instructions. VEGF was expressed as pg of VEGF
protein/ml of medium and per mg of total protein or
105 cells.
Determination of LDH-A Activity in Cells.
Subconfluent cells were grown and collected as described for the
determination of VEGF protein level. Cell pellets were homogenized in
0.1 M potassium phosphate buffer (pH 7.6), using an IKA
Ultra-Turrax T8 homogenizer (Janke and Kunkel) for 15 s at full
speed. Extracts were quantified using the Bio-Rad DC protein assay
(Bio-Rad, Hemel Hempstead, United Kingdom). The measurement of LDH-A
activity was carried out by the Clinical Biochemistry Unit of the John
Radcliffe Hospital (Oxford, United Kingdom). The enzymatic method used
is based on the recommended method of Scandinavian Society for Clinical
Chemistry and Clinical Physiology, following the pyruvate to lactate
reaction (21)
. The assay was performed by monitoring the
disappearance of NADH at 340 nm, bichromatically corrected at 380 nm
following the Bayer-Axon method used on an Axon spectrophotometer, as
recommended by the manufacturer (Bayer Diagnostics, Halstead, United
Kingdom).
Growth of Xenografts in Nude Mice.
Tumors were initiated by injection of 106 cells
in 200 µl of PBS under the flank skin of NuNu mice. Two
independent series of experiments were performed in which aliquots of
the two transfectant cell lines were implanted in parallel in five
mice. Every second day, tumor size was measured using calipers. Tumors
were excised when ulceration occurred or when they reached 1.44
cm2
. They were then frozen in liquid nitrogen or
fixed in formaldehyde. Growth rates were calculated by subtracting the
initial from the final volume of the tumor and dividing the total by
the number of days for the xenograft growth.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
,
HIF-2, and two hypoxia-responsive targets in a panel of six breast
cell lines of epithelial origin. These cell lines had been chosen to
reflect the heterogeneity of the human breast cancer progression
pathway. Five of the six cell lines represent different stages and
phenotypes of breast carcinoma, and Table 1
summarizes the cell characteristics
(22
, 23)
. T47 D is an ER-positive cell line that is
responsive to estrogens (24)
, MDA 468 shows amplification
of the epidermal growth factor receptor (25)
, and SKBr3
similarly for c-erbB2 (26)
. MDA 468 presents a
mutation of the PTEN tumor suppressor gene (27)
,
and MDA 231 similarly for c-ras (28)
. MDA 231 and 435 show
the highest in vivo tumor growth and take rates in xenograft
transplantation (29)
. MDA 435 alone shows spontaneous
metastases in vivo (29)
. HBL 100 is a
spontaneously immortalized, nontransformed breast cell line
(30)
.
|
and HIF-2 mRNA Expression. and HIF-2 gene expression was analyzed by RNase protection
assay using total RNA from the breast carcinoma cell lines. Under
normoxic conditions, both HIF-1 and HIF-2 mRNA were detected at
varying levels in all of the cell lines studied (Fig. 1)
. The abundance of HIF-2 mRNA
varied between cell lines over a wide range (>40-fold), with SKBr3
showing the highest and MDA 435 showing the lowest levels of
expression. The level of HIF-1 mRNA was less variable, ranging
4-fold from the lowest expressing cell line MDA 435 to the highest
expressing cell line MDA 468. HIF-1 mRNA was more abundant than
HIF-2 mRNA in five of six cell lines examined, ranging between 5-
and 34-fold difference. SKBr3, however, expressed similar levels of
HIF-1 and HIF-2 mRNA. The levels of HIF-1 and HIF-2 mRNA
after 16 h of exposure to hypoxia (0.1% oxygen) remained
unaltered in all of the cell lines studied (data not shown).
|
and HIF-2 Protein Expression. and HIF-2 protein levels and their
regulation by hypoxia across the panel of breast carcinoma cell lines.
Whole-cell extracts were prepared following parallel incubation under
normoxia, hypoxia, or with DFO for 6 or 16 h. The level of hypoxia
used (0.1 and 3%) reflected the heterogeneous oxygen concentration
found within tumors.
All of the cell lines studied showed some degree of induction of both
HIF-1 and HIF-2 at 0.1% oxygen, but there was a wide variation
in the extent of up-regulation and the ratio of HIF-1:HIF-2 (Fig. 2)
. Some of the cell lines showed an
increased ability to respond to mild hypoxia because HIF-1 and
HIF-2 proteins were detectable at 3% oxygen (Fig. 2B)
.
SKBr3 was the only cell line expressing a similar amount of HIF-1
and HIF-2 at both 3 and 0.1% oxygen. There was a 30–50% induction
in HIF-1 and HIF-2 proteins at 3% compared with 0.1% oxygen in
T47 D, MDA 468, and MDA 435 but only a very small increase was detected
in HBL 100 and MDA 231. In addition, exposure to 3% oxygen resulted in
a slightly greater increase in HIF-2 compared with HIF-1, as
reported previously for HeLa cells by Wiesener et al.
(31)
.
|
was expressed in all of the breast cell lines exposed to 0.1%
oxygen, with MDA 468 showing the greatest induction and MDA 231 and 435
showing the least induction (Fig. 2A)
. HIF-2 was not
detected in MDA 435, and the smallest induction under hypoxia was
demonstrated in MDA 231. Conversely, the protein was highly expressed
in SKBr3 when stimulated by hypoxia. In all cell lines, HIF-1 and
HIF-2 expression was fully reversible within 1 h of returning
to normoxic conditions (data not shown).
To compare the absolute amount of HIF-1 and HIF-2 proteins,
respectively, the comparison of a standardized amount of protein has
shown that MoAb 28b gives 4-fold less signal than detection with
MoAb 190b (31)
. HIF-1 protein was, therefore, more
abundant than HIF-2 protein under hypoxia (0.1% oxygen) in all cell
lines studied except SKBr3, which showed a higher level of HIF-2.
There was a highly significant correlation between the level of
HIF-2 mRNA expression and the level of HIF-2 protein under
hypoxia (r = 0.53, P = 0.41, linear regression analysis) but not for HIF-1 proteins
(r = 0.95, P = 0.003, linear regression analysis; Fig. 3
). MDA 231 and 435 were the least
inducible cell lines under hypoxia and showed the lowest expression of
HIF-1 protein and very low or no HIF-2 expression.
|
and
HIF-2 under hypoxia (31)
. Human breast cancer cell
lines were analyzed for expression of VEGF and LDH-A mRNA under tissue
culture conditions simulating the hypoxic tumor microenvironment.
After exposure to 0.1% oxygen for 16 h, all of the cell lines
studied exhibited an increased expression of these two genes. However,
the abundance and extent of induction of VEGF mRNA under hypoxia varied
between cell lines. Induction varied from 1.7- to 7.9-fold (Fig. 4A)
. The least tumorigenic
cell lines, T47 D and HBL 100, had the lowest basal level and showed
the highest induction; MDA 468 and SKBr3 had a high basal level but
showed only moderate induction; and the most tumorigenic cell lines,
MDA 231 and 435, had the highest basal level but the lowest
induction of VEGF mRNA under hypoxia.
|
.
Expression of VEGF and LDH-A Protein.
The expression of VEGF protein in the cell culture medium was measured
by ELISA. Human breast cancer cell lines were cultured under normoxia,
hypoxia (0.1% oxygen), or in presence of DFO for 16 h (Fig. 5A)
. The least
tumorigenic cell lines that showed the lowest basal level of VEGF
expression (T47 D, HBL 100, and SKBr3) also showed the largest
induction under hypoxia. MDA 231, on the other hand, showed the highest
basal level but the smallest induction of VEGF under hypoxia.
|
. T47 D showed the highest
inducibility for LDH-A, and MDA 231 and 435 showed the highest basal
level of LDH-A activity. However, in general the cell lines studied
showed less variability in their LDH-A inducibility than VEGF, with the
induction ranging only between 1.1- and 1.6-fold.
Survival under Hypoxia.
The viability of the breast cancer cell lines after 24 h of
continuous exposure to 0.1% oxygen was analyzed by clonogenic survival
assay (Table 2)
. There was between 20 and
75% cell death in the least tumorigenic cell lines that showed
induction of VEGF under hypoxia. However, there was either no death or
a small increase in survival under hypoxia in the most tumorigenic cell
lines, MDA 231 and 435, that showed <2-fold induction of VEGF at the
mRNA level.
|
and HIF-2 Protein. and HIF-2 are expressed at different levels in the breast
cell lines, and both proteins are involved in the regulation of the
hypoxic pathway. Therefore, we have considered the requirement to
compare between cell lines the cumulative induction of both of these
transcription factors.
There was an inverse correlation between the clonogenic survival and
the cumulative induction of HIF-1 and HIF-2 proteins for each of
the breast cancer cell lines after exposure to hypoxia
(r = 0.896, P = 0.016, linear regression analysis; Fig. 6
). Indeed, the cells with the best
survival under hypoxia showed the smallest induction of HIF-1 and
HIF-2 proteins. Thus, despite the fact that HIF proteins are
necessary for optimal tumor growth and angiogenesis in vivo,
overexpression of these molecules could also be detrimental to tumor
growth.
|
cDNA. overexpression, the breast cancer
cell line MDA 435, which does not express this transcription factor,
was stably transfected with the human HIF-2 cDNA. The transfectant
MDA 435-HIF 2 showed a high level of HIF-2 mRNA and protein under
normoxic condition, compared with an empty-vector control (MDA
435-control). Expression of HIF-2 was further raised in 0.1% oxygen
(Fig. 7, A and B)
. HIF-2 was localized by immunostaining of the
transfectants using MoAb 190b. HIF-2 was strongly expressed in the
nucleus of a subset of MDA 435-HIF 2 transfectants both in normoxia
and hypoxia (data not shown). No staining was detected in the control
transfectant cell line.
|
overexpression to induce the
hypoxic pathway. The expression of VEGF mRNA and protein in the
transfectants were analyzed (Fig. 7, C and D)
. At
both mRNA and protein levels, the basal and induced VEGF expression
were significantly increased in the MDA 435-HIF 2 compared with the
empty-vector transfectant.
Growth rates of the two transfectant cell lines were compared in
vitro in monolayer tissue culture. The doubling times during the
logarithmic phase of growth were similar for MDA 435-HIF-2 (40.2
h ± 8; n = 8) and the control
transfectant (40.9 h ± 7.4; n = 8). We measured the hypoxic viability of the two transfectant
cell lines by clonogenic survival after 24 h of continuous
exposure to 0.1% oxygen, followed by 2 or 3 weeks incubation in normal
conditions. The percentage of cells surviving under hypoxia compared
with normoxia was not significantly different between the MDA
435-HIF-2 transfectant (112% ± 6; n = 4) and the control transfectant (103% ± 8;
n = 5). However, the size of the colonies
formed by HIF-2 transfectant cells after exposure to normoxia or
hypoxia (0.84 ± 0.19 and 0.82 ± 0.16
mm, respectively) was significantly smaller (40% reduction) than the
colonies formed by the empty-vector control after exposure to normoxia
or hypoxia (1.38 ± 0.29 and 1.47 ± 0.33
mm, respectively; paired samples t test,
P = 0.0001; n = 40).
We investigated further the effect of HIF-2 overexpression on tumor
growth. Both transfectant cell lines were injected s.c. in nude mice,
and growth rates of the tumor xenografts were measured. The results
from two independent series of experiments showed that 3 of 10 MDA
435-HIF-2 implantations did not form a detectable tumor. The growth
rates for the MDA 435-HIF-2 xenografts ranged between 0 and 0.172
cm3/day (median, 0.03) and for the MDA
435-control xenografts between 0.01 and 0.61
cm3/day (median, 0.19). Overall, the growth rate
of the MDA 435-HIF-2 tumors was significantly less than the control
transfectant xenograft (Student’s unpaired t test,
P = 0.002; n = 10).
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and
HIF-2 in a panel of human breast cancer cell lines and the effect of
their expression on tumor growth in vivo. HIF-1 is known
to be a key modulator of angiogenesis through induction of VEGF
transcription (2)
. HIF-2 is a closely related molecule
that was described as an endothelial-specific factor homologous to
HIF-1 (9, 10, 11)
. However, our studies showed that
HIF-2 is commonly expressed in breast cancer cell lines and is not
therefore endothelial specific, an observation that is in accordance
with our previous results (31
, 32)
. Indeed, most cells
express two independent hypoxia transcription factors that regulate
VEGF expression.
The cell line panel was chosen to reflect the heterogeneity of human
breast cancer progression pathways, involving for example the
expression of EGF-R, erbB2, ER, and PTEN. This allowed us to analyze
whether, despite this variability, there is a correlation among
phenotypes between induction of the hypoxia response pathway and the
expression of HIF-1 and HIF-2. The end points chosen were
in vitro survival under hypoxia, expression of VEGF mRNA and
protein, and the level of LDH-A mRNA and enzyme activity. Tumor growth
in vivo was also correlated with the induction of HIF-1
and HIF-2.
HIF-1 mRNA is expressed constitutively in a large number of human
tissues and cell lines, and these steady-state expression levels are
not further up-regulated by hypoxia (5
, 33, 34, 35)
, as has
also been shown for HIF-2 mRNA (31)
. Although not
regulated at the RNA level, Jiang et al. (36)
showed that up-regulation of basal mRNA could lead to higher HIF-1
protein expression. Our study showed that basal levels of HIF-2
mRNA, but not HIF-1, are highly related to the inducible level of
the respective HIF protein. Some macrophage and endothelial cell lines
have high basal levels of HIF-2 mRNA (32)
; therefore,
in these cell lines this factor could operate as the key hypoxic
pathway. It would be of interest to elucidate the regulation of
HIF-2 mRNA expression. The relative role of HIF-2 compared with
HIF-1 in cancer is unknown, but it appears to be of significant
importance that in this study, both MDA 435 and 231, the two most
aggressive cell lines in vivo, had the lowest level of
HIF-2 expression.
The cumulative induction of HIF-1 and HIF-2 was inversely related
to the clonogenic survival under hypoxia; cells that showed the best
survival under hypoxia (MDA 435 and 231) also showed the least
induction in both HIF proteins. This result was surprising because
aggressive cancer cells would be expected to show stronger induction of
these transcription factors. Although it has been demonstrated in
knockout experiments that lack of HIF-1 has a detrimental effect on
embryonic vessel growth, its effect on tumor growth is less clear.
Several studies have shown that cell lines deficient in the HIF-1
pathway, through mutations in HIF-1 or aryl hydrocarbon receptor
nuclear translocator/HIF-1ß, exhibit a poor induction of glycolytic
enzymes under hypoxia and reduced angiogenesis in vivo. In
addition, these cells showed poor in vivo tumor growth
properties (12, 13, 14, 15, 16)
. However, Carmeliet et al.
(17)
showed in their HIF-1 knockout embryonic stem
cells, that although induction of VEGF and glycolytic enzymes were
attenuated, survival under hypoxia in vitro and tumor growth
in vivo were enhanced, although angiogenesis was impaired.
These conflicting results suggest that the relationship between
HIF-1 and apoptosis under hypoxia depends on cellular context
(37)
. It has been shown recently that loss of function
mutations in tumor suppressor genes encoding PTEN
(38)
, VHL (39)
and p53
(40
, 41)
result in increased expression of HIF-1 and
VEGF. Gain of function mutations in oncogenes also induce HIF-1
expression, as demonstrated for v-src (36)
. It
is possible, therefore, that low HIF-2 expression is of survival
value if it normally couples to an apoptotic pathway under hypoxia.
In this study, we showed that induction of HIF-1 is qualitatively
different from HIF-2. Oxygen at the 0.1% level was equivalent to
DFO for HIF-2 induction but was weaker than DFO for HIF-1
induction in five of six cell lines. These results suggest that DFO,
although mimicking hypoxia, does not activate HIF-1 and HIF-2 to
the same extent.
We showed that all of the cell lines studied express HIF-1 protein,
which was maximally inducible under more severe conditions (0.1%
oxygen) than HIF-2. This suggests that HIF-1 is the key factor
for the regulation of VEGF. However, HIF-2 also induced
transcription from the VEGF promoter in transfection studies with CHO
cells deficient in HIF-1 (31)
. Thus, to allow analysis
of the effects of these pathways, we correlated the total induction of
HIF-1 and HIF-2 with biological end points. The sensitivity of
the pathways to hypoxia also varied between cell lines in that HIF-1
or HIF-2 was detectable at 3% oxygen in some cell lines
(e.g., SKBr3), whereas others usually required 0.1%
(e.g., MDA 231). Similar results were reported for prostate
cancer cell lines, and in some cell lines, HIF-1 is even seen in
normoxia (42)
. The implication for human tumors in
vivo is that some tumors may be more able to respond to small
changes in the microenvironment with less severe hypoxia. It will be of
interest to analyze the expression of these pathways in
vivo, in correlation with the tumor angiogenesis status and in
presence of bioreductive markers of the hypoxic areas, such as
pimonidazole (43)
.
The results for VEGF need to be considered separately for baseline and
inducible expression. In the three cell lines with high basal levels of
VEGF mRNA, this high level of mRNA correlated with high protein
expression (MDA 231, 435, and 468). In these cell lines, the
protein:RNA ratio was substantially greater than in the others. There
was no correlation, therefore, between VEGF baseline expression and
HIF-1 or HIF-2 inducibility. It is therefore likely that the
differences in baseline expression of VEGF are a result of oncogene
signaling. In addition, posttranslational mechanisms such as a high
level of eIF4E expression could modulate the protein:mRNA ratio
(44, 45, 46, 47, 48, 49)
. However, the hypoxic inducibility of VEGF, both
at the mRNA level and protein level, was correlated with cumulative
expression of HIF-1 and HIF-2. The two most aggressive cell lines
with the least HIF-1 and HIF-2 induction (MDA 231 and 435) had
the smallest increase in VEGF mRNA (<2-fold) and the lowest induction
of VEGF protein. At less extreme levels of hypoxia (3% oxygen), there
was little induction of VEGF in MDA 231 and 435 cell lines compared
with T47 D, where up to 90% of the maximal VEGF protein was expressed
(data not shown).
These results show an association between a poor inducibility of
HIF-1 and HIF-2 expression, high basal levels of VEGF, improved
survival under hypoxia, and enhanced tumor growth in vivo.
Thus, it is possible that oncogenes that up-regulate basal VEGF
expression could attenuate hypoxia signaling as part of a protective
mechanism against hypoxia-induced apoptosis (36
, 44
, 50)
.
These results are consistent with a model linking HIF-1 induction to
both angiogenesis and apoptosis. Maintenance of hypoxia signaling has
been shown to be important for enabling angiogenesis to occur in
vivo (12
, 17)
. For a tumor to sustain a growth
advantage, the basal level of angiogenic factors are often
up-regulated; attenuation of the hypoxic pathway will allow in
vivo growth through reduction of the proapoptotic effect of HIF
proteins.
Glycolysis, another hypoxia-regulated pathway aberrantly regulated in
cancer, was evaluated by measuring the key glycolytic enzyme LDH-A
(51)
. The results showed that LDH-A mRNA was inducible in
all cell lines studied. It was observed that the cell line that showed
the strongest induction of VEGF also displayed the best inducibility
for LDH-A enzyme activity, suggesting coordinated hypoxic regulation of
both mRNAs. However, because the induction of LDH-A was less variable
than for VEGF, it appears that this pathway is regulated by hypoxia,
but it is not modulated by oncogenes to the same extent as VEGF. It is,
therefore, possible that a 2-fold induction of all of the steps of the
glycolytic pathway is sufficient to confer survival advantage
(52
, 53)
. In experiments using CHO cells defective in
HIF-1 signaling, HIF-1 and HIF-2 were shown to be equivalent
in their contribution to regulation of the LDH-A promoter
(54)
.
To provide some insight on the function of HIF-2 in tumor response,
we transfected the MDA 435 cell line with human HIF-2 cDNA. The MDA
435-HIF 2 stable transfectants showed an increased basal VEGF mRNA
and protein expression, as well as an enhanced maximal induction under
hypoxia. However, in vitro growth as colonies and in
vivo tumor growth were impaired, although no differences in the
tissue culture growth rate of the two cell lines was observed. Similar
results have been reported for other cell lines defective in HIF
signaling (12
, 17)
. In vitro growth as colonies
is a model associated with more severe stress, with an area of hypoxia
present near the center of the colonies, and a lack of attachment to an
extracellular matrix. There was no difference in clonogenic survival
under hypoxia between the two transfectant cell lines; however, the
level of HIF-2 in normoxia was already much greater in the MDA
435-HIF-2 than in the control transfectant (Fig. 7, A and B)
.
The transfection results support the concept of competing pathways for
selection of tumor cells with differences in HIF-1 or HIF-2
expression under normoxia and hypoxia. One pathway that will confer a
survival advantage is enhanced angiogenesis, and both Ha-ras and V-src
expression have been shown to up-regulate VEGF expression under hypoxia
(36
, 50) . However, a competing pathway is enhanced
apoptosis. Thus, mutations in proapoptotic genes (e.g.,
p53) or up-regulation of antiapoptotic pathways would favor
maintained induction of HIF-1 or HIF-2 (38)
. A
minimal level of HIF protein induction is important for angiogenesis,
because transformed cell lines have much poorer angiogenesis in
vivo in the absence of HIF-1 (12
, 17)
. The end
effect, therefore, depends on the balance between these pathways as
shown by the inverse correlation between hypoxia inducibility and
survival in this study.
To initially assess the mechanism for the inverse correlation of
HIF-2 expression and tumorigenicity, we studied the expression of
some apoptosis-related genes in the HIF-2 transfectants and showed
an enhanced basal expression of the growth inhibitor insulin-like
growth factor binding protein-3 in HIF-2 transfectants (data not
shown).
This study has several implications for human breast cancer biology. It
suggests that escape from HIF-1- or HIF-2-induced apoptosis
is important for the aggressive behavior of tumors, but a minimal level
of inducibility is still important for tumor growth. Thus, despite the
suggestion that inhibition of the hypoxia pathway would be of
therapeutic value, particularly in suppressing angiogenesis, because of
the potential interaction of the hypoxia response pathway with
apoptosis, it could also promote tumor cell survival.
Immunohistochemical assessment of HIF-1 or HIF-2 induction in
tumors may give a more valid measure of the overall population
experiencing hypoxia, which is relevant to induction of angiogenesis
and to radioresistance that occurs in this population.
This study has highlighted the complexity of the system with regard to
cell growth in vitro and in vivo. Escape from the
proapoptotic effect of HIF-1/2 may be an important selective
pressure promoting oncogene transformation and an "angiogenic
switch" allowing the pro-VEGF effect of the HIF pathway.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 To whom requests for reprints should be
addressed, at Molecular Oncology Unit, Imperial Cancer Research Fund,
Institute of Molecular Medicine, University of Oxford, John Radcliffe
Hospital, Oxford, OX3 9DS, United Kingdom. Phone: (44) 1865-222457;
Fax: (44) 1865-222431; E-mail: aharris.lab{at}icrf.icnet.uk 
2 The abbreviations used are: VEGF, vascular
endothelial growth factor; LDH-A, lactate dehydrogenase A; HIF,
hypoxia-inducible factor; HRE, hypoxia response element; DFO,
desferrioxamine; ER, estrogen receptor; MoAb, monoclonal antibody.
Received 2/ 3/00. Accepted 10/ 9/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
subunit. J. Biol. Chem., 271: 32253-32259, 1996. (HIF-1-) 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, 1997.-subunit. J. Biol. Chem., 272: 11205-11214, 1997. is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway. Proc. Natl. Acad. Sci. USA, 95: 7987-7992, 1998. and developmentally expressed in blood vessels. Mech. Dev., 63: 51-60, 1997.[Medline]
regulates the VEGF expression and is potentially involved in lung and vascular development. Proc. Natl. Acad. Sci. USA, 94: 4273-4278, 1997.-subunit (HIF-1-)–characterization of HIF-1--dependent and HIF-1--independent hypoxia-inducible gene expression. J. Biol. Chem., 273: 8360-8368, 1998.. Genes Dev., 12: 149-162, 1998. is required for solid tumor formation and embryonic vascularization. EMBO J., 17: 3005-3015, 1998.[Medline]
in hypoxia-mediated apoptosis, cell-proliferation and tumor angiogenesis. Nature (Lond.), 394: 485-490, 1998.[Medline]
. Blood, 92: 2260-2268, 1998.–posttranscriptional regulation and conformational change by recruitment of the ARNT transcription factor. Proc. Natl. Acad. Sci. USA, 94: 5667-5672, 1997. expression by the epidermal growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP pathway in human prostate cancer cells: implications for tumor angiogenesis and therapeutics. Cancer Res., 60: 1541-1545, 2000. in common human cancers and their metastases. Cancer Res., 59: 5830-5835, 1999.. Genes Dev., 14: 34-44, 2000. in rat and human prostate cancer. Cancer Res., 58: 5280-5284, 1998.This article has been cited by other articles:
|
|
 |
|
  K. Helczynska, A.-M. Larsson, L. Holmquist Mengelbier, E. Bridges, E. Fredlund, S. Borgquist, G. Landberg, S. Pahlman, and K. Jirstrom Hypoxia-Inducible Factor-2{alpha} Correlates to Distant Recurrence and Poor Outcome in Invasive Breast Cancer Cancer Res., November 15, 2008; 68(22): 9212 - 9220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Stoletov, V. Montel, R. D. Lester, S. L. Gonias, and R. Klemke High-resolution imaging of the dynamic tumor cell vascular interface in transparent zebrafish PNAS, October 30, 2007; 104(44): 17406 - 17411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Niemela, P. Hynninen, J.-P. Mecklin, T. Kuopio, A. Kokko, L. Aaltonen, A.-K. Parkkila, S. Pastorekova, J. Pastorek, A. Waheed, et al. Carbonic Anhydrase IX Is Highly Expressed in Hereditary Nonpolyposis Colorectal Cancer Cancer Epidemiol. Biomarkers Prev., September 1, 2007; 16(9): 1760 - 1766. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Winner, A. C. Koong, B. E. Rendon, W. Zundel, and R. A. Mitchell Amplification of Tumor Hypoxic Responses by Macrophage Migration Inhibitory Factor-Dependent Hypoxia-Inducible Factor Stabilization Cancer Res., January 1, 2007; 67(1): 186 - 193. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. T. Jones, C. W. Pugh, S. Wigfield, M. F.G. Stevens, and A. L. Harris Novel Thioredoxin Inhibitors Paradoxically Increase Hypoxia-Inducible Factor-{alpha} Expression but Decrease Functional Transcriptional Activity, DNA Binding, and Degradation. Clin. Cancer Res., September 15, 2006; 12(18): 5384 - 5394. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. T. Jones and A. L. Harris Identification of novel small-molecule inhibitors of hypoxia-inducible factor-1 transactivation and DNA binding. Mol. Cancer Ther., September 1, 2006; 5(9): 2193 - 2202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. McMahon, M. Charbonneau, S. Grandmont, D. E. Richard, and C. M. Dubois Transforming Growth Factor beta1 Induces Hypoxia-inducible Factor-1 Stabilization through Selective Inhibition of PHD2 Expression J. Biol. Chem., August 25, 2006; 281(34): 24171 - 24181. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Zheng, J. L. Ruas, R. Cao, F. A. Salomons, Y. Cao, L. Poellinger, and T. Pereira Cell-Type-Specific Regulation of Degradation of Hypoxia-Inducible Factor 1{alpha}: Role of Subcellular Compartmentalization Mol. Cell. Biol., June 15, 2006; 26(12): 4628 - 4641. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Manka, Z. Spicer, and D. E. Millhorn Bcl-2/Adenovirus E1B 19 kDa Interacting Protein-3 Knockdown Enables Growth of Breast Cancer Metastases in the Lung, Liver, and Bone Cancer Res., December 15, 2005; 65(24): 11689 - 11693. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Sun, K. Guo, J. J. Rusche, and L. H. Hurley Facilitation of a structural transition in the polypurine/polypyrimidine tract within the proximal promoter region of the human VEGF gene by the presence of potassium and G-quadruplex-interactive agents Nucleic Acids Res., October 20, 2005; 33(18): 6070 - 6080. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Hose, G. Kaur, E. A. Sausville, and A. Monks Transcriptional Profiling Identifies Altered Intracellular Labile Iron Homeostasis as a Contributing Factor to the Toxicity of Adaphostin: Decreased Vascular Endothelial Growth Factor Secretion Is Independent of Hypoxia-Inducible Factor-1 Regulation Clin. Cancer Res., September 1, 2005; 11(17): 6370 - 6381. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Covello, M. C. Simon, and B. Keith Targeted Replacement of Hypoxia-Inducible Factor-1{alpha} by a Hypoxia-Inducible Factor-2{alpha} Knock-in Allele Promotes Tumor Growth Cancer Res., March 15, 2005; 65(6): 2277 - 2286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kanematsu, S. Osada, N. Amaoka, S. Goshima, H. Kondo, H. Kato, H. Nishibori, R. Yokoyama, H. Hoshi, and N. Moriyama Expression of Vascular Endothelial Growth Factor in Hepatocellular Carcinoma and the Surrounding Liver and Correlation with MRI Findings Am. J. Roentgenol., March 1, 2005; 184(3): 832 - 841. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-H. Li, Z.-H. Miao, W.-F. Tan, J.-M. Yue, C. Zhang, L.-P. Lin, X.-W. Zhang, and J. Ding Pseudolaric Acid B Inhibits Angiogenesis and Reduces Hypoxia-Inducible Factor 1{alpha} by Promoting Proteasome-Mediated Degradation Clin. Cancer Res., December 15, 2004; 10(24): 8266 - 8274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. J. Hutchison, H. R. Valentine, J. A. Loncaster, S. E. Davidson, R. D. Hunter, S. A. Roberts, A. L. Harris, I. J. Stratford, P. M. Price, and C. M. L. West Hypoxia-Inducible Factor 1{alpha} Expression as an Intrinsic Marker of Hypoxia: Correlation with Tumor Oxygen, Pimonidazole Measurements, and Outcome in Locally Advanced Carcinoma of the Cervix Clin. Cancer Res., December 15, 2004; 10(24): 8405 - 8412. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hoeben, B. Landuyt, M. S. Highley, H. Wildiers, A. T. Van Oosterom, and E. A. De Bruijn Vascular Endothelial Growth Factor and Angiogenesis Pharmacol. Rev., December 1, 2004; 56(4): 549 - 580. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kanematsu, S. Osada, N. Amaoka, S. Goshima, H. Kondo, H. Nishibori, H. Kato, M. Matsuo, R. Yokoyama, H. Hoshi, et al. Expression of Vascular Endothelial Growth Factor in Hepatocellular Carcinoma and the Surrounding Liver: Correlation with Angiographically Assisted CT Am. J. Roentgenol., December 1, 2004; 183(6): 1585 - 1593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Acker and H. Acker Cellular oxygen sensing need in CNS function: physiological and pathological implications J. Exp. Biol., August 15, 2004; 207(18): 3171 - 3188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Giatromanolaki, M. I. Koukourakis, H. M. Sowter, E. Sivridis, S. Gibson, K. C. Gatter, and A. L. Harris BNIP3 Expression Is Linked with Hypoxia-Regulated Protein Expression and with Poor Prognosis in Non-Small Cell Lung Cancer Clin. Cancer Res., August 15, 2004; 10(16): 5566 - 5571. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Arsham, D. R. Plas, C. B. Thompson, and M. C. Simon Akt and Hypoxia-Inducible Factor-1 Independently Enhance Tumor Growth and Angiogenesis Cancer Res., May 15, 2004; 64(10): 3500 - 3507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. M. Sowter, R. Raval, J. Moore, P. J. Ratcliffe, and A. L. Harris Predominant Role of Hypoxia-Inducible Transcription Factor (Hif)-1{alpha} versus Hif-2{alpha} in Regulation of the Transcriptional Response to Hypoxia Cancer Res., October 1, 2003; 63(19): 6130 - 6134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Favier, P.-F. Plouin, P. Corvol, and J.-M. Gasc Angiogenesis and Vascular Architecture in Pheochromocytomas : Distinctive Traits in Malignant Tumors Am. J. Pathol., October 1, 2002; 161(4): 1235 - 1246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Engelhardt, R. F. Schneider, S. H. Seeholzer, C. C. Stobbe, and J. D. Chapman The Synthesis and Radiolabeling of 2-Nitroimidazole Derivatives of Cyclam and Their Preclinical Evaluation as Positive Markers of Tumor Hypoxia J. Nucl. Med., June 1, 2002; 43(6): 837 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Pellizzaro, D. Coradini, and M. G. Daidone Modulation of angiogenesis-related proteins synthesis by sodium butyrate in colon cancer cell line HT29 Carcinogenesis, May 1, 2002; 23(5): 735 - 740. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. J. P. Beasley, R. Leek, M. Alam, H. Turley, G. J. Cox, K. Gatter, P. Millard, S. Fuggle, and A. L. Harris Hypoxia-inducible Factors HIF-1{alpha} and HIF-2{alpha} in Head and Neck Cancer: Relationship to Tumor Biology and Treatment Outcome in Surgically Resected Patients Cancer Res., May 1, 2002; 62(9): 2493 - 2497. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. N. Coleman, J. B. Mitchell, and K. Camphausen Tumor Hypoxia: Chicken, Egg, or a Piece of the Farm? J. Clin. Oncol., February 1, 2002; 20(3): 610 - 615. [Full Text] [PDF]
|
|
|
|
|
|
T. Onita, P. G. Ji, J. W. Xuan, H. Sakai, H. Kanetake, P. H. Maxwell, G.-H. Fong, M. Y Gabril, M. Moussa, and J. L. Chin Hypoxia-induced, Perinecrotic Expression of Endothelial Per-ARNT-Sim Domain Protein-1/Hypoxia-inducible Factor-2{alpha} Correlates with Tumor Progression, Vascularization, and Focal Macrophage Infiltration in Bladder Cancer Clin. Cancer Res., February 1, 2002; 8(2): 471 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Li, N. W. Shworak, and M. Simons Increased responsiveness of hypoxic endothelial cells to FGF2 is mediated by HIF-1{alpha}-dependent regulation of enzymes involved in synthesis of heparan sulfate FGF2-binding sites J. Cell Sci., January 5, 2002; 115(9): 1951 - 1959. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Blancher, J. W. Moore, N. Robertson, and A. L. Harris Effects of ras and von Hippel-Lindau (VHL) Gene Mutations on Hypoxia-inducible Factor (HIF)-1{alpha}, HIF-2{alpha}, and Vascular Endothelial Growth Factor Expression and Their Regulation by the Phosphatidylinositol 3'-Kinase/Akt Signaling Pathway Cancer Res., October 1, 2001; 61(19): 7349 - 7355. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Qin, C. Wilson, C. Blancher, M. Taylor, S. Safe, and A. L. Harris Association of ARNT Splice Variants with Estrogen Receptor-negative Breast Cancer, Poor Induction of Vascular Endothelial Growth Factor under Hypoxia, and Poor Prognosis Clin. Cancer Res., April 1, 2001; 7(4): 818 - 823. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
20% in all cases.
