| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Biology, Immunology, Cytokines |
SRI International, Menlo Park, California 94025 [K. R. L., M. D. B., J. M. C., A. M. K.], and Department of Radiation Oncology, Stanford University School of Medicine, Stanford, California 94305 [R. M. A., E. Y. C., Z. Y., N. C. D., A. J. G.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and
interleukin 8), fragments of circulatory system proteins
(e.g., angiostatin), and extracellular matrix components
[e.g., TSPs (9
, 12, 13, 14, 15, 16)
]. Presumably, this
diversity of angiogenic factors reflects a strict requirement for
controlling angiogenesis under normal physiological conditions and in
response to oncogenic events. Given that the switch to the angiogenic
phenotype requires a positive net increase in inducing activity, we
hypothesize that tumor hypoxia promotes angiogenesis by modulating the
expression of both angiogenic inducers and inhibitors. The present
study focuses on the influence of the hypoxic tumor microenvironment on
the expression of two critical angiogenic factors, the inhibitor TSP-1
and the inducer VEGF. TSP-1 is a secreted adhesive glycoprotein that has been shown to have antiangiogenic properties by an unknown mechanism (17, 18, 19) . Although paradoxical effects of TSP-1 on malignant progression have been reported (20 , 21) , these observations may reflect the wide range of potential interactions of the complex domain structure of this molecule with its microenvironment, as well as the variable distribution of TSP-1 receptors in tissues (22 , 23) . Compelling evidence of an antiangiogenic function for TSP-1 is provided by studies showing that activation of the angiogenic switch in Li-Fraumeni fibroblasts lacking wt p53 is associated with diminished TSP-1 expression (17 , 24) . Conversely, the introduction of wt p53 into BT549 human breast carcinoma cells generates an antiangiogenic activity involving up-regulated TSP-1 expression (15) . In addition, overexpression of TSP-1 is antiangiogenic in tumor models (25 , 26) . Reports that endogenous TSP-1 expression can be induced by wt p53, repressed by oncogenic signals such as c-jun overexpression, and silenced by DNA methylation imply that down-regulation of TSP-1 contributes to oncogenesis (24 , 27, 28, 29, 30) . Because both p53 and c-jun are inducible by hypoxia (4 , 31) , the regulation of TSP-1 expression by these proteins suggests that it is a hypoxia-responsive angiogenic inhibitor.
VEGF is a powerful angiogenic inducer that is activated by hypoxia at both the transcriptional and posttranscriptional levels (32, 33, 34, 35) . Inducible VEGF expression correlates with hypoxic regions present in tumor models in vitro and in experimental solid tumors in vivo (2 , 36) , suggesting a model of tumor angiogenesis in which VEGF isoforms generated in hypoxic regions stimulate host endothelial cells to assemble new vasculature by a feedback loop. VEGF expression in aerobic and hypoxic cells also appears to be enhanced by ras oncogenes, one of the most common oncogenic alterations in human cancer (10 , 37) . This response to oncogenic signals may explain, in part, why VEGF expression does not always correlate with evidence of tumor hypoxia, as reported for human cervical carcinomas exposed to the bioreductive hypoxia marker pimonidazole (38) . Transcription of the VEGF gene in response to hypoxia is critically dependent on an enhancer site in its 5'-regulatory region for the hypoxia-inducible transcription factor HIF-1 (34 , 39) . Considering recent reports that wt p53 interacts with HIF-1 and inhibits its transactivation activity (40 , 41) , the dependence of VEGF transcription on HIF-1 suggests that p53 could antagonize VEGF-dependent angiogenesis stimulated by tumor hypoxia.
Current evidence indicates that the expression of TSP-1 and VEGF is genetically controlled by both tumor suppressor and proto-oncogene activity, providing molecular mechanisms that could contribute to the switch to the angiogenic phenotype when these controls are deregulated during oncogenesis (17) . However, because tumor angiogenesis is also physiologically controlled by environmental signals such as hypoxia, it is important to determine the contribution of this stress to the expression of these angiogenic factors. In the present study, we investigated the influence of pathophysiological hypoxia on TSP-1 and VEGF expression by using normal and transformed rodent and human cells, including isogenic rodent cells that are wt or null for p53.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.01% (relative to air at
pO2 
21%). Rodent cells were incubated for 3
days after plating (3 x 105 to 1.5 x
106 cells/60-mm-diameter plastic culture dish)
before exposure to hypoxia in aluminum gas-exchange chambers according
to a standard protocol (6)
. The HCFs, HCEs, HCE.E6E7
cells, NDKs, and the C33a, SiHa, and FaDu cells were plated at 5 x 106 cells/100-mm-diameter glass culture dish
and exposed to hypoxia for approximately 18 h in an anaerobic box
(Bactron X; Sheldon Manufacturing Inc., Cornelius, OR) containing an
atmosphere of 90% CO2, 5%
CO2, and 5% H2. Cells were
harvested immediately in air to prepare total RNA for Northern
analysis. Slightly different incubation conditions were used in the
transient transfection studies described below.
Growth of Tumors in SCID Mice.
The preparation of tumor xenografts in SCID mice from EH-MEFs has been
described elsewhere (3
, 43)
. Briefly, 8–10-week-old
female SCID mice were injected with 8 x 106
to 2 x 107 cells intradermally in the
midline of the back approximately 2 cm from the base of the tail. After
3–5 weeks, tumors were ready for excision and sectioning. To visualize
hypoxic regions in the tumors, mice were injected i.p. with 0.2 ml of
10 mM EF5 (3)
1 h before tumor excision.
Immunohistochemistry.
Tumors were embedded in OCT compound (Miles, Elkhart, IN) and frozen in
liquid nitrogen. Cryostat sections were cut at 8–10 µm in thickness,
placed on Fisher Plus slides (Fisher Scientific, Pittsburgh, PA), and
stored in air-tight boxes at -80°C. The sections were returned to
room temperature before opening the boxes. Sections were fixed in
Streck Tissue Fixative (Streck Laboratories, Inc., Omaha, NE) for 30
min at room temperature. The fixed sections were washed three times in
PBS (5 min/wash), incubated at room temperature for 30 min in a
background suppressing solution [BRS; 150 mM NaCl, 10
mM Tris-HCl (pH 8.0), 5 mM EDTA, 0.25% (w/v)
gelatin, and 0.05% (v/v) Tween 20], and washed as described above. To
detect TSP-1 expression, sections were incubated with an anti-TSP-1
immune complex solution at 4°C overnight in a humidified chamber
(45)
. The immune complex was formed by diluting the
primary anti-TSP-1 monoclonal antibody (Clone TSP-B7; Sigma
Immunochemicals, St. Louis, MO) 1:5000 with a solution of the secondary
antibody (FluorX-conjugated antimouse donkey serum; BDS, Inc.,
Pittsburgh, PA) diluted 1:100 in BRS. After rocking at 4°C overnight,
heat-inactivated normal mouse serum (Sigma) was added to the immune
complex solution to a final concentration of 0.1% (v/v), and the
solution was rocked at 4°C for 2 h before applying 200 µl to
each section. To detect VEGF, a rabbit antihuman polyclonal IgG
antibody raised against a peptide fragment (amino acids 4–24) of human
VEGF (Santa Cruz Biotechnology, Santa Cruz, CA) was used as the primary
antibody. This antibody was diluted to a final concentration of 5
µg/ml in PBS-HS and added to the sections (100 µl/section) for a
2-h incubation at room temperature. The slides were then washed three
times in PBS-HS and incubated for 2 h at room temperature with
FITC-conjugated antirabbit goat serum diluted in PBS-HS (200
µl/section). To visualize hypoxic regions of tumor sections by using
EF5 immunostaining, an anti-EF5 monoclonal antibody (ELK3-51)
conjugated with the indocarbocyanine dye Cy3 (obtained from Dr. Cameron
Koch, University of Pennsylvania, Philadelphia, PA) was diluted
to 30 µg/ml in PBS-HS and applied to each section (200 µl/section)
for incubation at 4°C overnight. The slides were washed three times
in PBS-HS (5 min/wash) and mounted in 2% n-propyl gallate
in 70% glycerol-0.03 M Tris-HCl (pH 9.0). To
quantitatively compare hypoxic regions on tumor sections with VEGF
expression, fields containing both oxygenated (i.e.,
EF5-negative) and hypoxic (i.e., EF5-positive) areas were
chosen randomly and photographed by using a Zeiss fluorescence
microscope (Zeiss Axioskop). The number of immunofluorescent regions
per unit area corresponding to EF5 binding and VEGF protein was then
calculated to estimate the degree of overlap of VEGF expression with
regions of hypoxia, as described previously for p53 protein expression
(3)
.
Northern Analysis.
Purification of total cellular RNA for Northern analysis was performed
by using a standard guanidinium isothiocyanate method, and
CsCl-purified RNA was resolved in 1% denaturing agarose gels and
blotted onto nylon membranes, as described elsewhere (46)
.
Alternatively, total RNA was purified by the RNeasy method (Qiagen
Inc., Santa Clarita, CA). Probes included cDNA fragments of human TSP-1
(1.3 kb from the 5' end; obtained from American Type Culture
Collection; cross-reactive with rodent TSP-1 mRNA), and human
VEGF165 (0.6 kb). The probes were labeled with
[-32P]dCTP by the random primer method
(Amersham Pharmacia Biotech, Arlington Heights, IL). As controls for
RNA loading/transfer, ethidium bromide fluorescence from the 28S rRNA
band of total RNA was photographed, blotted rRNA bands were visualized
by staining with methylene blue, or blots were probed with a labeled
2.0-kb cDNA probe for human ß-actin that recognizes the corresponding
mouse mRNA (Clontech Laboratories, Palo Alto, CA).
Message Stability.
The half-life of TSP-1 mRNA was determined according to a protocol
described elsewhere (4)
. Briefly, Rat-1 cells were exposed
to hypoxia for 12 h, as described above. Hypoxic cells were
removed from the aluminum hypoxia chambers in the anaerobic box held at
37°C. The hypoxic cells were given 5 µg/ml actinomycin D (Sigma)
for 10 min (time 0), and then cells were harvested for total RNA at
various times afterward. Total RNA was processed for Northern analysis,
and TSP-1 mRNA signals on autoradiographs of the blots were measured by
using a video densitometer system (Lynx Molecular Biology Work Station,
Santa Clara, CA). The half-lives of TSP-1 mRNA were calculated from
plots of the natural log (intensity) against the time of actinomycin D
exposure starting at time 0.
Western Analysis.
Cells were washed twice in ice-cold PBS and lysed by scraping in an
ice-cold detergent buffer [50 mM Tris-HCl (pH 7.4), 250
mM NaCl, 0.5% NP40, 50 mM NaF, 15
mM Na PPi, 1 mM
Na3VO4, 20 µg/ml
aprotinin, 5 µg/ml leupeptin, and 1 mM
phenylmethylsulfonyl fluoride]. After centrifugation at 12,000 x
g for 20 min at 4°C, portions of the supernatants were
diluted with equal volumes of 2x SDS sample buffer and boiled for 5
min. The protein concentrations of the supernatants were determined by
a bicinchoninic acid assay (Pierce, Rockford, IL). Equal protein
samples (5 µg) were used for gel electrophoresis. Proteins were
resolved in a discontinuous 10% SDS-polyacrylamide gel and
electroblotted in a buffer containing 25 mM
Tris-HCl (pH 8.3), 192 mM glycine, 0.1% SDS, and
15% methanol onto Immobilon P membranes (Millipore, Marlborough, MA)
by using a TE 22 Mighty Small Transphor Tank Transfer Unit (Hoefer
Pharmacia Biotech Inc., San Francisco, CA). Blots were blocked in 5%
nonfat dried milk in PBS containing 0.1% Tween 20 at 4°C overnight.
To detect HIF-1 protein, blots were washed once in PBS-0.1% Tween
20 and then incubated with rocking at room temperature for 2 h
with a monoclonal antihuman HIF-1 antibody (Novus Biologicals, Inc.,
Littleton, CO) diluted 1:1,000 in PBS/0.1% Tween 20/5% nonfat dried
milk. A secondary antimouse IgG antibody conjugated with horse radish
peroxidase (IgG-HRP; Santa Cruz Biotechnology) diluted 1:5,000 in
PBS/0.1% Tween 20 was added, and the blot was incubated at room
temperature for 2 h. After washing three times in PBS/0.1% Tween
20, primary antibody binding was detected and visualized by using the
Enhanced Chemiluminescence Plus Western blotting detection system
(Amersham Pharmacia Biotech, Piscataway, NJ).
Plasmid Constructs and Transfections.
Luciferase reporter plasmids containing fragments of the human
TSP-1 gene proximal promoter region [TSP (-2033), (-548),
and (-330)] were obtained from Dr. Paul Bornstein (Department of
Biochemistry, University of Washington, Seattle, WA). They were
constructed by using the pGL3-Enhancer plasmid (Promega Corp., Madison,
WI) as described elsewhere (47)
. The constructs were
verified by restriction digestion. The numerical designations of the
plasmids are the same as those described previously (48)
.
For the analysis of TSP-1 promoter activities, Rat-1 cells
were plated in 6-well tissue culture dishes at 2.5 x
105 cells/well and incubated at 37°C for
24 h. Cells were then transfected with a reporter plasmid using
the Lipofection Plus transfection agent (Promega) according to the
procedure recommended by the manufacturer. Briefly, cells were
incubated with plasmid DNA (0.5 µg/well) for 4 h in serum-free
medium and allowed to recover for 24 h in complete medium. Before
exposure to hypoxia, cells received fresh complete medium. After
10 h in the anaerobic box (pO2 0.02%),
cells (and control aerobic cells) were placed on ice, and cell lysates
were prepared by adding 150 µl Reporter Lysis Buffer (Promega)/well.
Promoter activities were analyzed by measuring luciferase activity in a
Monolight 2010 luminometer (Analytical Luminescence Laboratory, Ann
Arbor, MI).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and
Fig. 2
). In addition, exposure of
p53-/- primary MEFs to
identical hypoxic conditions suppressed TSP-1 mRNA expression with
essentially the same kinetics as those for the primary
p53+/+ MEFs (Fig. 1A
). Together, these
findings demonstrate that prolonged hypoxia can physiologically
override the signal for the induction or sustained expression of
TSP-1 mRNA in primary or immortalized rodent fibroblasts containing wt
p53. In contrast to primary MEFs, TSP-1 mRNA was barely detectable in
total RNA from the transformed p53+/+ and
p53-/- MEFs (EH-MEFs)
under either aerobic or hypoxic conditions (data not shown). This low
level of TSP-1 mRNA could be an effect of oncogenic ras
activity on TSP-1 expression (49)
. Two lines of evidence
support this possibility: (a) TSP-1 mRNA levels in primary
MEFs (p53+/+ and
p53-/-) are abundant
(Fig. 1)
; and (b) stable transfection of Rat-1 fibroblasts
with oncogenic H-ras (Rat-1R cells) almost eliminated
detectable TSP-1 mRNA expression under aerobic conditions (Fig. 2)
. The
potential interaction between hypoxia and transformation on TSP-1
expression is considered in more detail below.
|
|
). Thus, the induction of
VEGF mRNA in primary or transformed MEFs exposed to stringent hypoxia
was not significantly influenced by the presence or absence of wt p53.
Because it has been proposed that HIF-1 stabilizes wt p53 in hypoxic
cells (40
, 41)
, we also determined whether HIF-1
protein was induced in primary p53+/+ MEFs under
the same low oxygen conditions. Fig. 1C
shows that HIF-1
protein accumulated well above aerobic levels in the primary
p53+/+ MEFs and that this expression was
sustained even during prolonged stress (50)
. Therefore,
the presence of both HIF-1 protein and wt p53 did not prevent or
inhibit the hypoxic induction of VEGF mRNA in these primary rodent
fibroblasts. This dominant effect of hypoxia on VEGF expression was
also confirmed in vivo using tumor-bearing mice injected
before sacrifice with the hypoxia-specific marker EF5. EF5
immunofluorescence from tumor sections or hypoxic cells has been shown
to correlate well with other established indicators of tumor
oxygenation, such as in situ pO2
measurements (51)
and the determination of radiobiological
hypoxia (52
, 53)
. Fig. 3
shows that VEGF immunofluorescence overlapped with tumor hypoxia (as
detected by EF5 immunofluorescence) on the same sections of tumor
xenografts derived from p53+/+ and
p53-/- EH-MEFs.
Quantitative analysis of the EF5 and VEGF signals on randomly chosen
sections demonstrated that the ratio of EF5:VEGF immunofluorescence was
not significantly different in tumors derived from either p53 genotype
(Table 1)
. Because the activation and
binding of EF5 under low oxygen conditions and the hypoxic induction of
VEGF expression have different dependencies, an exact overlap is not
expected. These findings indicate that hypoxia can induce VEGF protein
expression in tumor cells in vivo independently of p53
status, similar to its effect on VEGF mRNA accumulation observed
in vitro (Fig. 1B
). Finally, in accord with the
low level of TSP-1 expression found in the p53+/+
and p53-/- EH-MEFs
in vitro, TSP-1 immunofluorescence was low or undetectable
on sections of either of the corresponding EH-MEF tumors (data not
shown).
|
|
shows that compared with
aerobic controls, hypoxia (pO2 0.02% for
10 h) weakly inhibited but did not suppress the basal activity of
the proximal TSP-1 promoter region in Rat-1 cells. Because
these differences are statistically significant, this finding indicates
that the decline of TSP-1 mRNA accumulation observed in hypoxic Rat-1
cells (Fig. 2)
could involve cis-acting elements within the
proximal promoter region. Moreover, these studies indicate that the
intronic p53 binding sites are not sufficient to transactivate the
TSP-1 proximal promoter region in hypoxic rodent cells
containing wt p53. The decreased basal response of the truncated TSP-1
promoter constructs TSP (-548) and TSP (-330) relative to that of TSP
(-2033) may be a consequence of the elimination of serum-responsive
cis-acting elements from these constructs (47
, 54)
. To determine whether hypoxia also affects the half-life of
TSP-1 mRNA, we used actinomycin D to block transcription in aerobic
Rat-1 cells and in Rat-1 cells exposed to hypoxia for 12 h. Fig. 4B
shows a representative Northern blot of total RNA
obtained from aerobic and hypoxic Rat-1 cells at 0, 15, 30, 45, 60, and
75 min after a 10-min incubation with actinomycin D. Analysis of plots
of the natural log (signal intensity) versus time from two
independent Northern blotting experiments gave a value of 87 ± 4
and 108 ± 4 min for the half-life of TSP-1 mRNA in hypoxic and
aerobic Rat-1 cells, respectively. Taken together, these findings
involving the effect of hypoxia on reporter gene expression regulated
by the human TSP-1 proximal promoter region and on the
half-life of endogenous TSP-1 mRNA suggest that both transcriptional
and posttranscriptional effects contribute to the decline of
steady-state TSP-1 mRNA levels in Rat-1 cells exposed to prolonged
stress. However, it should be recognized that the relative contribution
of these mechanisms to the maintenance of TSP-1 mRNA levels in response
to hypoxia could depend on several variables, including the degree of
stress and the cell type. The effect of cell type is considered in more
detail below.
|
and 2)
, we investigated the effect of hypoxia on TSP-1 as well as VEGF
mRNA levels in a panel of human cells obtained from different normal
tissues and relevant tumors. Potentially, these normal cells could also
be exposed to hypoxia in a heterogeneous tumor. Fig. 5A
indicates that the
inhibition of TSP-1 expression by stringent hypoxia can be influenced
by cell type. For example, hypoxia strongly decreased the level of
TSP-1 mRNA in HCEs, had less of an inhibitory effect on the basal level
in NDKs, and had no detectable effect on the level of TSP-1 mRNA in
HCFs. Interestingly, it has been reported that TSP-1 expression is
posttranscriptionally induced in hypoxic endothelial cells, supporting
a cell type-specific effect on the regulation of this gene by hypoxia
(55)
. Fig. 5B
shows that although there
appeared to be a small induction of TSP-1 mRNA accumulation in normal
HMVECs by 6 h of hypoxia, in general there was no detectable
decrease in this message, even after 24 h of prolonged stress.
Among the immortalized/tumor cells, hypoxia strongly decreased TSP-1
mRNA levels in HCE.E6E7, C33a, and SiHa cells but had little or no
effect on TSP-1 mRNA levels in FaDu cells. However, the basal
level of TSP-1 expression was already relatively low in FaDu cells. In
this context, examination of aerobic TSP-1 mRNA levels in the panel of
cells in Fig. 5A
suggests that basal TSP-1 expression
decreases in response to transformation: relative aerobic TSP-1 mRNA
levels in either HCEs or their immortalized/minimally transformed
counterparts (HCE.E6E7 cells) were found to be substantially higher
than those in the human cervical carcinoma cell lines (C33a and SiHa).
Similarly, aerobic NDKs were found to have higher basal TSP-1
expression than aerobic FaDu cells, which were obtained from an
epidermoid carcinoma (56)
. Because the SiHa cell line is
HPV positive (containing E6 and E7), whereas the C33a and FaDu cell
lines are HPV negative, the relative basal TSP-1 expression in
these cells does not seem to depend critically on how transformation
was achieved. Together, these findings suggest that hypoxia could
interact with oncogenic signals to further inhibit or suppress TSP-1
expression in the tumor microenvironment.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
:HIF-1ß; reviewed in Refs. 62
and
63
) is a critical regulator of tumor angiogenesis and the
associated growth of a tumor mass (61
, 64, 65, 66)
. It has
been proposed that these pathological effects of HIF-1 are mediated in
part by the direct transcriptional activation of VEGF and components of
glucose metabolism (64)
, similar to its function in normal
tissues exposed to hypoxia (62
, 63)
. Recent reports that
wt p53 interacts directly with HIF-1 at the protein level (40
, 41)
suggest a more complex model for HIF-1 in tumorigenesis. In
this model, loss of wt p53 would remove a check on HIF-1 activation in
response to hypoxia (41)
, favoring VEGF expression but
discriminating against the expression of angiogenic inhibitors such as
TSP-1 (67)
. However, analysis of the effect of wt p53 on
the induction of VEGF expression has produced conflicting results
(68
, 69)
, and the generality of the conclusion that p53
and HIF-1 interact in hypoxic cells is not established
(70)
. As part of this study, we investigated the
relationship between wt p53 and its effect on the simultaneous
expression of TSP-1 and VEGF in response to pathophysiological hypoxia
in both rodent and human cells. There are three major findings of the
present study. First, hypoxia acted in a dominant manner to suppress
TSP-1 expression in normal/immortalized rodent fibroblasts
independently of their p53 genotype. In normal human cells, hypoxia
also inhibited TSP-1 expression, but this effect was dependent on the
cell type. Second, in both rodent and human cells, the level of
basal/aerobic TSP-1 expression was decreased in response to
transformation or to the degree of transformation. In general, where
basal TSP-1 expression was detectable by Northern analysis of the
transformed cells, it could be inhibited by stringent hypoxia. Third,
in contrast to TSP-1, hypoxia induced VEGF expression in both
normal/immortalized and transformed rodent and human cells, again
independent of the presence of wt p53. The finding that hypoxia can suppress TSP-1 expression in normal/immortalized rodent fibroblasts containing wt p53 contrasts with reports indicating that p53 positively regulates this gene in angiogenic human fibroblasts (17 , 24) . One explanation for this finding is that prolonged or stringent hypoxia may override normal physiological controls on the TSP-1 gene. For example, the effect of hypoxia on TSP-1 expression could be the outcome of a general metabolic response, the inhibition of macromolecule synthesis during prolonged hypoxic stress (71 , 72) . Alternatively, because hypoxia did not suppress basal transcriptional activity of the TSP-1 proximal promoter region in this study, it may affect TSP-1 expression by a specific mechanism such as the inhibition of mRNA stabilization. Although mRNA stabilization was postulated to account for the accumulation rather than the decline of TSP-1 mRNA in hypoxic endothelial cells (55) , it is possible that the posttranscriptional control of TSP-1 expression in normal cells is cell type specific. In accord with this possibility, we observed that the same in vitro hypoxic conditions that suppressed TSP-1 expression in rodent fibroblasts had little or no effect on the level of TSP-1 mRNA in HCFs and microvascular endothelial cells but suppressed or inhibited TSP-1 mRNA levels in HCEs and NDKs. Interestingly, the expression of TSP-2 mRNA in NIH 3T3 cells appears to involve message stability (73) . Further research is necessary to determine the degree to which transcriptional and posttranscriptional controls regulate TSP-1 mRNA expression in response to tumor hypoxia. The generally low levels of basal TSP-1 expression found in transformed rodent and human cells in this study may be the consequence of oncogenic changes leading to TSP-1 silencing or down-regulation (17 , 30) . Inhibition of basal TSP-1 expression has been reported for various oncogenes or oncogenic signals, including oncogenic ras, v-src, v-myc, polyoma middle T antigen, and overexpressed c-jun (27, 28, 29 , 49 , 74, 75, 76) . As mentioned above, the finding that oncogenic ras almost eliminated detectable TSP-1 expression in Rat-1 fibroblasts supports this idea. We hypothesize that tumor hypoxia can synergize with oncogenic signals to suppress TSP-1 expression in the tumor microenvironment, analogous to its cooperation with ras to induce VEGF expression in transformed cells (10) .
In summary, this study demonstrates that hypoxia can act in a dominant manner to inhibit the expression of TSP-1 and induce that of VEGF in diverse rodent and human cells independently of any p53-associated controls on these genes. Because hypoxia can select against transformed cells containing wt p53 in vivo (3) , these findings indicate that tumor hypoxia could activate the angiogenic switch in a population of apoptotically resistant tumor cells (11) , providing a powerful stimulus for malignant progression. Establishing the relative contributions of hypoxia-dependent and -independent mechanisms to the expression of angiogenic factors such as TSP-1 and VEGF in human tumors is an important area for further research.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by NIH/National Cancer Institute
Grants CA67166 (to K. R. L. and A. J. G.), CA73807 (to K. R. L.),
and CA73832 (A. J. G.). 
2 To whom requests for reprints should be
addressed, at SRI International, Pharmaceutical Discovery Division, 333
Ravenswood Avenue, Menlo Park, CA 94025. Phone: (650) 859-3080; Fax:
(650) 859-5816; E-mail: keith.laderoute{at}sri.com
3 The abbreviations used are: VEGF, vascular
endothelial growth factor; TSP, thrombospondin; wt, wild-type; HIF-1,
hypoxia-inducible factor 1; MEF, mouse embryonic fibroblast; HCF, human
cervical fibroblast; HCE, human cervical epithelial cell; NDK, normal
dermal keratinocyte; HMVEC, human microvascular endothelial cell;
SCID, severe combined immunodeficient; PBS-HS, PBS-5% horse serum;
HPV, human papillomavirus.
Received 11/ 1/99; revised 3/28/00; accepted 3/30/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Nature (Lond.), 392: 405-408, 1998.[CrossRef][Medline]
B. Eur. J. Biochem., 234: 632-640, 1995.[Medline]
in hypoxia-mediated apoptosis, cell proliferation and tumor angiogenesis. Nature (Lond.), 394: 485-490, 1998.[CrossRef][Medline]
is required for solid tumor formation and embryonic vascularization. EMBO J., 17: 3005-3015, 1998.[CrossRef][Medline]
is not sufficient for hypoxic/anoxic p53 induction. Cancer Res., 58: 5678-5680, 1998.This article has been cited by other articles:
|
|
 |
|
  F. Nishimoto, M. Sakata, R. Minekawa, Y. Okamoto, A. Miyake, A. Isobe, T. Yamamoto, T. Takeda, E. Ishida, K. Sawada, et al. Metal Transcription Factor-1 Is Involved in Hypoxia-Dependent Regulation of Placenta Growth Factor in Trophoblast-Derived Cells Endocrinology, April 1, 2009; 150(4): 1801 - 1808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Brostjan, K. Gebhardt, B. Gruenberger, V. Steinrueck, H. Zommer, H. Freudenthaler, S. Roka, and T. Gruenberger Neoadjuvant Treatment of Colorectal Cancer with Bevacizumab: The Perioperative Angiogenic Balance Is Sensitive to Systemic Thrombospondin-1 Levels Clin. Cancer Res., April 1, 2008; 14(7): 2065 - 2074. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Becker, N. Rohwer, T. Funakoshi, T. Cramer, W. Bernhardt, A. Birsner, J. Folkman, and R. J. D'Amato 2-Methoxyestradiol Inhibits Hypoxia-Inducible Factor-1{alpha} and Suppresses Growth of Lesions in a Mouse Model of Endometriosis Am. J. Pathol., February 1, 2008; 172(2): 534 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Short, A. Derrien, R. P. Narsimhan, J. Lawler, D. E. Ingber, and B. R. Zetter Inhibition of endothelial cell migration by thrombospondin-1 type-1 repeats is mediated by {beta}1 integrins J. Cell Biol., February 14, 2005; 168(4): 643 - 653. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Colombel, S. Filleur, P. Fournier, C. Merle, J. Guglielmi, A. Courtin, A. Degeorges, C. M. Serre, R. Bouvier, P. Clezardin, et al. Androgens Repress the Expression of the Angiogenesis Inhibitor Thrombospondin-1 in Normal and Neoplastic Prostate Cancer Res., January 1, 2005; 65(1): 300 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Evans, K. D. Judy, I. Dunphy, W. T. Jenkins, W.-T. Hwang, P. T. Nelson, R. A. Lustig, K. Jenkins, D. P. Magarelli, S. M. Hahn, et al. Hypoxia Is Important in the Biology and Aggression of Human Glial Brain Tumors Clin. Cancer Res., December 15, 2004; 10(24): 8177 - 8184. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. M. Fenton, S. F. Paoni, and I. Ding Pathophysiological Effects of Vascular Endothelial Growth Factor Receptor-2-Blocking Antibody plus Fractionated Radiotherapy on Murine Mammary Tumors Cancer Res., August 15, 2004; 64(16): 5712 - 5719. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Douglas and G. G. Haddad Genetic Models in Applied Physiology: Invited Review: Effect of oxygen deprivation on cell cycle activity: a profile of delay and arrest J Appl Physiol, May 1, 2003; 94(5): 2068 - 2083. [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]
|
|
|
|
 |
|
 E. Fosslien Molecular Pathology of Cyclooxygenase-2 in Cancer-induced Angiogenesis Ann. Clin. Lab. Sci., October 1, 2001; 31(4): 325 - 348. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Green, P. Lichtlen, N. T. Huynh, M. Yanovsky, K. R. Laderoute, W. Schaffner, and B. J. Murphy Placenta Growth Factor Gene Expression Is Induced by Hypoxia in Fibroblasts: A Central Role for Metal Transcription Factor-1 Cancer Res., March 1, 2001; 61(6): 2696 - 2703. [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 |
