| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cell Growth & Differentiation |
Articles |
Department of Pathobiology [C. V. N., D. Y., A. T-T.], University of Pennsylvania, Philadelphia, Pennsylvania, 19104-6051 and Cell and Molecular Biology Graduate Program [M. G., A. T-T.], University of Pennsylvania, Philadelphia, Pennsylvania 19104-6051; Department of Zoology, [N. A., R. A.], University of Wisconsin, Madison, Wisconsin 53706; and Department of Microbiology and Immunology and R. H. Lurie Cancer Center, [O. V.], Northwestern University Medical School, Chicago, Illinois 60611
Abstract
The ability of neoplastic cells to recruit blood vasculature is crucial to their survival in the host organism. However, the evidence linking dominant oncogenes to the angiogenic switch remains incomplete. We demonstrate here that Myc, an oncoprotein implicated in many human malignancies, stimulates neovascularization. As an experimental model, we used Rat-1A fibroblasts that form vascular tumors upon transformation by Myc in immunocompromised mice. Our previous work and the use of neutralizing antibodies reveal that in these cells, the angiogenic switch is achieved via down-modulation of thrombospondin-1, a secreted inhibitor of angiogenesis, whereas the levels of vascular endothelial growth factor, a major activator of angiogenesis, remain high and unaffected by Myc. Consistent with this finding, overexpression of Myc confers upon the conditioned media the ability to promote migration of adjacent endothelial cells in vitro and corneal neovascularization in vivo. Furthermore, mobilization of estrogen-dependent Myc in vivo with the appropriate steroid provokes neovascularization of cell implants embedded in Matrigel. These data suggest that Myc is fully competent to trigger the angiogenic switch in vivo and that secondary events may not be required for neovascularization of Myc-induced tumors.
Introduction
Despite enormous interest in the Myc oncoprotein, one of the key molecules in many human cancers, the mechanisms of neoplastic transformation by Myc remain enigmatic. It is generally believed that these mechanisms can be deduced from the putative normal function of Myc, which is to promote entry into the cell cycle (1, 2, 3) . Indeed, c-myc is transcriptionally inactive in the majority of growth-arrested or terminally differentiated cells but undergoes rapid activation during the course of G0-G1 transition (4 , 5) . c-Myc is involved in mitogenic signaling (6 , 7) , and its ectopic activation causes quiescent cells to reenter the cell cycle (8) . Conversely, inhibition of c-myc expression using antisense oligonucleotides has been shown to impede cell proliferation (9 , 10) .
However, even uncontrollably dividing cells would not give rise to tumors unless their survival in the host organism was ensured. One crucial element of their survival is the ability to fight off hypoxia (11) . This is achieved via recruitment of the blood vasculature, which provides an unimpeded supply of oxygen (12 , 13) . Normal adult tissues, however, are not capable of undergoing neovascularization due to the fact that they generally produce more inhibitors of angiogenesis than activators (14) . In fibroblasts, for instance, this negative balance is achieved through high levels of Tsp-1, 3 a secreted inhibitor of angiogenesis (15) . These high levels are maintained in primary cells because the tsp-1 gene is up-regulated by p53, a tumor suppressor (16) . Consequently, should the cell lose both alleles of the p53 gene (which is often the case in tumors), levels of Tsp-1 plunge, and the cell acquires the angiogenic phenotype (17) . Biochemical (18) and gene knockout experiments (19) reveal that the related protein Tsp-2 is an inhibitor of angiogenesis as well.
One might assume that dominant oncogenes would also exert profound effects on the expression of angiogenic factors, both positive and negative (20) . The prime example of such regulation is activation by Ras of the VEGF (21 , 22) . There are data suggesting that Src also has the capacity to stimulate secretion of VEGF (23 , 24) . Src (25) and another oncoprotein, Jun (26) , were reported to down-regulate Tsp-1 and Myb-Tsp-2 (27) . In some systems, overexpression of Ras correlates with lower levels of Tsp-1 (28) and high levels of matrix metalloproteinases (29) that are required for endothelial cell migration. Neutralizing antibodies against the oncogenic members of the epidermal growth factor receptor family and their cognate ligands have been reported to reduce expression of VEGF and inhibit angiogenesis (30 , 31) . Finally, expression of several other less-studied oncogenes correlates with the abundant production of matrix metalloproteinases and other proteolytic enzymes that favor angiogenesis (reviewed in Ref. 14 ). However, in most cases, it is not known whether activation of the oncogene is sufficient for the angiogenic switch or whether a secondary genetic event (e.g., loss of p53) is still required.
We have demonstrated previously that overexpression of Myc in avian and rodent fibroblasts results in rapid posttranscriptional down-regulation of the tsp-1 mRNA4 and diminished levels of the Tsp-1 protein in CM (32) . This raised the possibility that activation of Myc is sufficient to confer the angiogenic phenotype upon fibroblastoid cells. However, it remained to be seen whether Myc-transformed fibroblasts produced significant amounts of proangiogenic factors and whether the net balance of pro- and antiangiogenic molecules would favor angiogenesis. Furthermore, earlier studies on angiogenesis in reconstituted prostate have failed to detect a contribution of Myc to tumor vascularization in that organ, and this function has been ascribed to Ras (33) , an oncoprotein with which Myc is known to cooperate in neoplastic transformation (34) . Thus, although many Myc-overexpressing tumors (e.g., Burkitt’s lymphoma) are known to be highly vascular (35) , it remained a distinct possibility that their vascularization had been triggered not by Myc per se but rather by secondary mutations that followed the initial transforming event. We set out to determine whether activation of Myc would result in the establishment of the angiogenic phenotype directly, in the absence of an in vivo selection for additional mutations.
To this end, we infected Rat-1A fibroblasts with retroviruses expressing either constitutively active Myc protein (LMycSN) or the chimera between Myc and the mutated estrogen receptor (pBabePuroMycERTM). ERTM is refractory to endogenous steroids but can be activated by a synthetic ligand, 4-OHT (36) . Rat-1A is a pseudo-normal, immortalized, p53-positive (37) cell line which, unlike primary embryo fibroblasts, is readily transformed by Myc (38) . These Myc-expressing cultures were tested for their tumorigenic and angiogenic potentials in an effort to determine whether activation of Myc is sufficient to trigger the angiogenic switch in both in vitro and in vivo settings.
Results
Myc-induced Fibroblastoid Tumors Possess Extensive Vasculature.
It is widely accepted that the majority of malignant, aggressively
growing tumors develop new vasculature as a means to assure a constant
supply of oxygen and nutrients. Some tumors, however, can tolerate
surprisingly high levels of hypoxia and remain viable in a relatively
avascular environment (39)
. To confirm that Myc-induced
experimental mesenchymal tumors belong to the first group, we analyzed
the neoplastic and angiogenic potentials of Rat-1A cells infected with
the human c-Myc-encoding retrovirus LMycSN. Rat-1A/LMycSN or parental
(Rat-1A) cells (3 x 106) were injected into
the flanks of Rag-1-/- immunodeficient mice.
After 2 weeks, no palpable tumors developed from Rat-1A cells (Fig. 1A, left specimen)
; however, their Myc-transformed
counterparts formed large, aggressively growing neoplasms exceeding 1
cm in diameter. Gross examination revealed robust vascularization of
both the tumor per se and the surrounding fascia (Fig. 1A, right specimen)
. After histological staining with H&E,
more differences became apparent. Whereas the Rat-1A cell masses were
composed of fusiform, collagen-producing cells (Fig. 1B)
,
Rat-1A/LMycSN tumors contained poorly differentiated, polygonal,
irregularly arranged cells with high mitotic activity (Fig. 1C)
. Characteristically, these masses were interlaced by
very prominent capillaries and venules that were apparently well
perfused, as evidenced by the presence of intact RBCs. These blood
vessels could have been recruited either by a bulk of
Myc-overexpressing cells or by a small number of clones that had
acquired the angiogenic phenotype independently of Myc during selection
in vivo. To prove that the former scenario is correct,
we set out to demonstrate that cultures of un-selected
Myc-overexpressing fibroblasts, but not the parental cells, secrete
factors that promote the formation of blood vessels. To rule out the
contribution of random genetic events, all subsequent experiments were
performed with pools of LMycSN-infected Rat-1A cells.
|
, FGM-2-fed Rat-1A and Rat-1A/LMycSN
cultures remained viable and were morphologically similar to cultures
maintained in a serum-containing growth medium (DMEM; Fig. 2, A and B
). In contrast, in serum-free DMEM, only
Rat-1A cultures remained viable (Fig. 2C)
, but virtually all
Rat-1A/LMycSN cells rounded off and detached from the plate (Fig. 2D)
. To confirm that no significant apoptosis took place in
FGM-2 medium, we performed TUNEL assays that detect breaks in the
genomic DNA characteristic of apoptosis. As illustrated by the flow
cytometry profiles in Fig. 2
, Rat-1A cultures remain largely apoptosis
free under any conditions (Fig. 2, A`, C`, and E`
), as evidenced by low percentages of FITC-labeled
apoptotic cells (the fraction designated M1) and high
percentages of unlabeled viable cells (the fraction designated
M2). In contrast, Rat-1A/LMycSN cells only remained viable
in DMEM + 10% FCS and FGM-2 media (Fig. 2, B` and F`
) where less than 10% of cells were in the M1 fraction.
The same culture deteriorated rapidly in serum-free DMEM (Fig. 2, D`
) where 64% of cells were in the M1 fraction 24 h
after the switch. We thus concluded that CM from FGM-2-fed cells would
represent mainly live cells and can be used in angiogenesis
experiments.
|
. We hypothesized that the angiogenic phenotype of Rat-1A/LMycSN cells
was due to down-regulation of Tsp-1. One way to corroborate this idea
was to demonstrate that the addition of a neutralizing anti-Tsp-1
antibody to the Rat-1A CM makes the medium angiogenic, as does
activation of Myc. To ensure that the antibody does not diffuse out
during the course of the experiment, we elected to perform a more
expeditious endothelial cell chemotaxis assay because substances known
to promote angiogenesis in vivo are usually able to promote
the migration of endothelial cells comprising blood vessels
(42)
.
|
CM from Rat-1A cells (
,
Rat-1A cells) does not promote migration any better than the
control medium containing only BSA (dotted reference line).
This is due to the presence of an inhibitor because bFGF, a strong
angiogenic factor in itself (
, no cells), was not active
when mixed with Rat-1A CM (, Rat-1A cells). This
inhibitor is likely to be Tsp-1 because the inclusion of a neutralizing
anti-Tsp-1 antibody, either alone (
) or in combination with bFGF
(
) fully restored migration. The same effect was achieved by
substituting CM from Rat-1A cells with CM from Rat-1A/LMycSN cells
(Fig. 4A, right panel
). Thus, down-regulation of
Tsp-1 is likely to underlie the angiogenic phenotype of Rat-1A/LMycSN
cells.
|
]. Even
before concentrating, these levels exceed the
ED50 for VEGF (43)
and thus can
account for the angiogenic potential of Rat-1A cells deprived of Tsp-1.
Importantly, no significant differences between Rat-1 and Rat-1A/LMycSN
cells were observed. In contrast, levels of Tsp-1 were sharply reduced
in Myc-transformed cells, as evidenced by the immunoprecipitation
experiment (Fig. 4C)
. This down-regulation was apparent in
cells maintained in either complete medium or FGM-2. All these findings
were consistent with the idea that it is down-regulation of the
tsp-1 gene by Myc that triggers the angiogenic switch in
Rat-1A cells.
Transient Activation of Myc in Vivo Results in the
Acquisition of the Angiogenic Phenotype.
Having demonstrated that overexpression of Myc in cultured fibroblasts
renders them angiogenic, we set out to determine whether activation of
Myc in vivo has the same effect and whether Myc is all that
is necessary to induce angiogenesis. Both questions could be addressed
by transient activation of the conditional mutant of Myc (MycER), whose
activity can be regulated in vivo by treatment with 4-OHT
(44)
. To facilitate the delivery of 4-OHT to
MycER-expressing cells, we chose to perform the assay where test cells
are embedded in Matrigel and injected between the dermis and the
peritoneum of the mouse (45)
, a site that, unlike the
cornea, is rich in preexisting blood vessels. Furthermore, because
Matrigel is derived from the extracellular matrix of a murine sarcoma
originating from the basement membrane (46
, 47)
, it more
or less faithfully simulates the normal extracellular environment of
the connective tissue (48)
. Additionally, the analysis of
vascularization is convenient because normally fluid Matrigel, along
with embedded cells, solidifies at 37°C and forms easily excisable
pellets. In a typical experiment, parental or Myc-expressing Rat-1A
cells were mixed with chilled Matrigel at 4°C and injected into
C57BL/6 mice. After 10 days, the animals were sacrificed, and the
Matrigel pellets were excised, examined visually, reliquified at 4°C,
and further assayed for hemoglobin content as described in "Materials
and Methods" or subjected to histological staining.
In the first experiment, we determined that Myc-overexpressing Rat-1A
cells are positive in this assay and that 4-OHT, an antiestrogen
(49)
, does not interfere with the assay by further
promoting or suppressing angiogenesis. To this end, we implanted
mice with "empty" or Rat-1A/LMycSN-containing Matrigels and treated
the animals with either a suspension of 4-OHT in olive oil or vehicle
alone. The drug and the vehicle were introduced directly into the
alimentary canals using gastric gavage, as described in "Materials
and Methods." We observed that "empty" Matrigels were avascular
in either drug- or vehicle-treated mice (Fig. 5A
, ), but Matrigels with Myc-transformed Rat-1A cells
(Rat-1A/LMycSN) were undergoing robust neovascularization (Fig. 5A
, ). On average, they contained five times more
hemoglobin than "empty" Matrigels or Matrigels containing parental
Rat-1A cells (Fig. 5B
; data not shown). Importantly,
neovascularization was not affected by 4-OHT because the slight
difference between -4-OHT and +4-OHT was neither reproducible nor
statistically significant. Hence, any effect that 4-OHT might have on
neovascularization of Matrigels containing Rat-1A/MycER cells would not
be due to changes in the hormone levels of the host.
|
). Rat-1A/MycER-expressing
cells, however, were avascular in control animals, but the majority of
pellets in 4-OHT-treated mice were vascularized rather heavily (Fig. 5B, two right bars
). Consequently, although some pellets
exhibited poor neovascularization (presumably due to insufficiently
high levels of functional Myc), resulting in average hemoglobin levels
slightly lower than those observed with LMycSN cells, the difference
between vehicle- and 4-OHT-treated Matrigels was reproducible and
statistically significant (P < 0.03). When individual
Rat-1A/MycER clones were tested, consistent neovascularization was also
observed in 4-OHT-treated animals but not in control animals.
Hemoglobin content data for one of these clones (MER-9) are shown in
Fig. 5C
. Interestingly, gross examination of Matrigel
pellets revealed that activation of Myc also results in increased
neovascularization of surrounding fascia that develops a network of
small vessels in 4-OHT-treated mice (Fig. 5D, right specimen
) but not in control (left specimen) mice.
To confirm that MycER-expressing cells promote the growth of bona
fide blood vessels, MER-9 pellets were also embedded in paraffin
blocks and subjected to standard H&E histological staining. In
vehicle-treated animals, we observed a small number of blood vessels in
the connective tissue surrounding the Matrigel (white
arrow); however, no blood vessels have traversed the Matrigel
pellets, as evidenced by the lack of RBCs (Fig. 5E)
. In
contrast, in the Matrigels from 4-OHT-treated mice (Fig. 5F)
, rather large vessels that contained numerous RBCs were
readily detectable. Thus, transient activation of Myc in rodent
fibroblasts can stimulate neovascularization in the absence of an
in vivo selection for angiogenic clones and independently of
tumor development.
Discussion
According to the paradigm, the transforming potential of the Myc oncoprotein is based essentially on its ability to promote unwarranted proliferation of quiescent cells, which are either deprived of stimulatory growth factors or exposed to inhibitory growth factors. Consistent with this idea, the majority of its putative target genes control entry into and progression through the cell cycle (1, 2, 3) , with cdc25 phosphatase being the prime example (50) . All those genes are up-regulated by Myc through interaction with Max (51) and subsequent binding to the so-called E-box DNA element (52, 53, 54) . Myc is also known to act as a transcriptional repressor (55) , and some of those targets also relate to cell cycle (56) . To complicate the matter, there is a host of "orphan" targets whose relationship to neoplastic transformation by Myc remains obscure (57) . Our initial identification of Tsp-1, a known inhibitor of angiogenesis, as a Myc-repressed gene afforded new insights into the role of Myc in neoplastic transformation (32) . Specifically, it implicated Myc overexpression as a possible reason for the acquisition of the angiogenic phenotype by the target cell. In this study, we demonstrate that this indeed is the case and that activation of the myc oncogene in untransformed cells is sufficient to render them angiogenic.
This result argues that the oncogenic potential of the Myc protein has a dual basis. Not only does Myc trigger unwarranted proliferation of the preneoplastic cells, but it also, at least in fibroblasts, allows them to undergo neovascularization, which is known to promote survival and metastatic spread of neoplastic cells. Down-regulation of Tsp-1 turns out to be sufficient to tilt the balance of secreted growth factors toward inducers of angiogenesis, which are now able to realize their potential. Indeed, in rodent fibroblast, we observed abundant expression of VEGF, one of the major proangiogenic molecules, on which activation of Myc had no considerable effect. This suggests that high levels of Tsp-1 expression in parental Rat-1A cells are required to counteract the angiogenic potential of VEGF. Although we cannot rule out the possibility that some other yet-to-be-identified proangiogenic factors are positively influenced by Myc in Rat-1A fibroblasts, in our in vitro assays, neutralizing anti-Tsp-1 antibodies exert exactly the same effect as activation of Myc. We are thus inclined to believe that down-regulation of Tsp-1 by Myc, just like its down-regulation via loss of p53 (16) , is sufficient to explain the angiogenic switch during neoplastic transformation. This effect of Myc would be of particular importance in cells retaining functional p53, such as Rat-1A cells (37) . Admittedly, during normal development the outcome of tsp-1 loss can be less dramatic because the tsp-1 knockout mice do not exhibit exaggerated angiogenesis (58) . Thus, regulation by Myc of other gene products might contribute to angiogenesis. One interesting possibility is that Myc affects expression of the closely related Tsp-2 gene, whose disruption in the germ line can provoke exaggerated vascularization (19) . In any event, it appears likely that the incipient Myc-overexpressing tumor would not need to undergo selection for angiogenic clones.
Myc therefore has the potential to act as a "complete" oncogene capable of inducing tumors rapidly and without relying on propitious genetic events, e.g., loss of p53. This idea is supported by a recent finding that activation of MycER chimera in keratinocytes leads not only to uncontrolled proliferation but also to appreciable angiogenesis (59) . It remains to be seen whether similar events occur in other cell types and whether they involve Tsp. The tsp-1 mRNA is not expressed in either parental or Myc-transformed B lymphocytes,5implicating other angiogenic factors in Myc-induced tumors of B lineage. Furthermore, overexpression of N-Myc in human neuroblastoma cells has been reported to down-modulate several inhibitors of endothelial cell proliferation, none of which has the biochemical properties of Tsp-1 (60) . However, in colon cancers, in which Myc is often overexpressed due to the loss of the APC tumor suppressor (61) , inactivation of Tsp-1 is thought to play an important role in the angiogenic switch (62) . Although the patterns of regulation of the tsp-1 gene in primary and Myc-overexpressing colonocytes has not as yet been established, the ability of Myc to function as an angiogenic trigger is likely to be exploited in these cells and in other neoplasias.
Materials and Methods
Generation and Propagation of Myc-expressing Clones.
Rat-1A cells and Rat-1A-derived clones were maintained in either DMEM
supplemented with 10% FCS (Life Technologies, Inc., Gaithersburg, MD)
or in fibroblast growth medium (FGM-2; Clonetics). The Rat-1A/LMycSN
cells have been described previously (32)
. The pBabePuro
vector (63)
expressing the MycERTM protein
(36)
was introduced into the Rat-1A cells via
electroporation, which was performed using the Bio-Rad apparatus and
the following conditions: 320 V and 960 µF. Selection of Rat-1A
clones producing MycER was performed in puromycin-containing media (1
mg/ml; Sigma, St. Louis, MO). 4-OHT was purchased from Research
Biochemicals International (Natick, MA) or Sigma (catalogue number
H6278). Cells to be used as a source of CM were grown in complete
medium on 10-cm dishes to approximately 90% confluence. They were then
washed twice with Ca2+-,
Mg2+-free PBS and placed in serum-free FGM-2
medium for 24–36 h. CM were collected and chilled on ice. The protease
inhibitor phenylmethylsulfonyl fluoride (Sigma) was added to the final
concentration of 0.1 mM. CM were then spun in Ultrafree
concentrators (Millipore, Bedford, MA) with a cutoff size of 5 kDa for
approximately 2 h, until the volume of the retentate decreased to
0.25 ml. The retentate was then diluted with 15 ml of
Ca2+-,Mg2+-free PBS, and
the centrifugation was repeated. The final protein concentrations in CM
were in the range of 1–1.5 mg/ml.
Tumor Growth.
Rat-1A or Rat-1A/LMycSN cells (1 x 106)
harvested from semiconfluent plates were injected s.c. into the flanks
of immunodeficient Rag-1-/- mice. Implanted
cell masses were excised 10 days later, weighed, photographed, fixed in
formalin, and subjected to standard histological staining with H&E.
TUNEL Apoptosis Assay.
Cells were plated in DMEM supplemented with 10% FCS at 
60% density
(6.5 x 105 cells/35-mm plate) and allowed
to attach overnight. They were then washed twice with
Ca2+-, Mg2+-free PBS and
refed with either DMEM + 10% FCS, serum-free DMEM, or FGM-2
(Clonetics). Twenty-four h later, both adherent and floating cells were
collected, counted, washed, and resuspended in cold 1% BSA in PBS at a
concentration of 107 cells/ml. Aliquots (100 ml)
were used in the TUNEL assay with the kit and the protocol from
Boehringer Mannheim. DNA breaks characteristic of apoptosis were
labeled with FITC-dUTP, followed by detection of labeled cells using a
flow cytometer (Becton Dickinson).
Corneal Neovascularization Assay.
Polyvinyl sponges preirradiated with 2000 Gy of 
-irradiation (cesium
source) were cut into 0.4 x 0.4 x 0.2 mm pieces, and test
material (0.2–0.5 µl) was introduced into each sponge. The loaded
sponges were dried, covered with a layer of 12% Hydron, and then dried
under vacuum. Sponges were then introduced into a surgically created
micropocket created in the avascular cornea of adult mice as described
previously (64)
. For positive control, sponges were loaded
with 100 ng of bFGF. Mouse eyes were examined for neovascularization
daily, using an ophthalmic microscope. On day 7 after implantation, 300
µl of FITC-conjugated high molecular weight dextran
(Mr 1,000,000–2,000,000) were
injected into the tail vein, and the animal was sacrificed after 5 min.
The eyeball was enucleated from the orbital cavity and fixed for 5 min
in 4% paraformaldehyde. The cornea with the adjacent limbus was
dissected, rinsed in PBS, and mounted on a glass slide in 10%
glycerol. Phase-contrast microscopy and fluorescence microscopy were
used to visualize the general layout of the cornea and the presence of
perfused blood vessels, respectively.
Endothelial Cell Chemotaxis Assay.
The bovine adrenal capillary endothelial cells were maintained in
DMEM supplemented with 10% donor calf serum, 2 mM
glutamine, and 100 µg/ml endothelial cell mitogen. For the
endothelial cell chemotaxis assay, serum-starved cells were plated in
serum-free medium supplemented with 0.1% BSA on the bottom side of
gelatinized microporous membrane (pore size, 5 µm; Nucleopore Corp.)
in the inverted modified Boyden chamber (Neuroprobe Science). The cells
were allowed to attach for 1–1.5 h, and then the chambers were
reinverted, and test substances in media (20 mg/ml) were added to the
top part of the wells and incubated for 3–4 h. The chambers were then
disassembled, the membranes were fixed and stained, and the number of
cells that migrated to the top part of the filter was determined in 10
high-power fields.
VEGF Assay.
Rat-1A or Rat-1A/LMycSN cells (5 x 105
cells) were seeded 24 h before the experiment in duplicate in
35-mm plates. Cells were refed with regular growth medium or FGM-2 and
incubated for an additional 24 h. CM (500 µl) were harvested to
determine VEGF concentrations. The mouse VEGF ELISA kit from R&D
was used according to the manufacturer’s recommendations. Optical
densities were measured using a plate reader (Fisher).
Radioimmunoprecipitation.
Approximately 5 x 106 cells grown in either
DMEM/10% FCS or FGM-2 were labeled for 4 h with 50 µCi of
[35S]methionine (DuPont New England Nuclear,
Boston, MA) in DMEM/5% dialyzed FCS. Cells were lysed in the
Ab buffer (65)
. The A4.1 anti-Tsp-1 monoclonal
antibody (2.5 µl; Life Technologies, Inc.) was added to cell lysates
and incubated overnight at 4°C. Secondary goat antimouse IgM
antiserum (2.5 µl; TAGO Inc., Camarillo, CA) was added, along with 50
µl of Gamma-Bind G-Sepharose (Pharmacia Biotech, Piscataway, NJ), and
incubation continued at room temperature for another 90 min.
Standard collection and washing techniques were applied, and the
immunoreactive proteins were run on 7.5% denaturing PAGE.
Matrigel Angiogenesis Assay.
Logarithmically growing Rat1A, Rat1A/LMycSN, and Rat1A/MycER cells were
harvested by treatment with 1 mM EDTA in
Ca2+-, Mg2+-free PBS. Cell
pellets were resuspended in cold Ca2+-,
Mg2+-free PBS at a concentration of
108 cells/ml. Five x
106 cells in 50 µl were combined with 0.5 ml of
thawed Matrigel (Collaborative Biomedical Products, Bedford, MA) and
placed briefly on ice. Using a 27-gauge needle, cells were then
injected s.c. between the dermis and the peritoneum of 5–7-week-old
C57BL/6 mice (Jackson Laboratory, Bar Harbor, ME). Aliquots (50 mg) of
4-OHT powder (Sigma) were resuspended in 2.5 ml of pharmacy-grade olive
oil using a Model 50 Sonic Dismembrator (Fisher Scientific, Hampton,
NH). 4-OHT was delivered p.o. (5 mg in 250 µl of olive oil) every 2
days using a 20-gauge intubation needle (Popper and Sons, Inc., New
Hyde Park, NY). Animals in the control groups received the same dose of
pure oil. Matrigel pellets were excised after 10 days and reliquefied
for 24 h at 4°C in 0.3 ml of either PBS or the Matrisperse
solution (Collaborative Biomedical Products). When a significant
fraction of Matrigel-embedded cells remained in the solid phase (due to
vigorous cell proliferation), pellets were further disrupted using an
ultrasonic dismembrator. Hemoglobin content was determined using the
Drabkin reagent (Sigma). Alternatively, solid Matrigel pellets, along
with the surrounding connective tissue, were photographed and fixed in
10% phosphate-buffered formalin, embedded in paraffin blocks, cut into
thin sections, and stained with H&E.
Acknowledgments
We thank Drs. William M. F. Lee (University of Pennsylvania, Philadelphia, PA), Maxine Linial (Fred Hutchinson Cancer Research Center, Seattle, WA), and Noel Bouck (Northwestern University, Chicago, IL) for advice and many helpful discussions. We thank Dr. Michael Goldschmidt (University of Pennsylvania) for help with histological analyses. The plasmid pBabePuroMycER was a kind gift of Martin Eilers (European Molecular Biology Laboratory, Heidelberg, Germany). The bovine adrenal capillary endothelial cells were a kind gift of Dr. J. Folkman (Harvard Medical School, Boston, MA).
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 Supported by National Cancer Institute Grant R29
CA 71881 and a grant from the University of Pennsylvania Research
Foundation (to A. T-T.). During the preliminary phase of this work,
A. T-T. was a Special Fellow of the Leukemia Society of America. 
2 To whom requests for reprints should be
addressed, at University of Pennsylvania, 3800 Spruce Street, 368E,
Philadelphia, PA 19104-6051. Phone: (215) 573-5138; Fax: (215)
898-0719; E-mail: andreit{at}vet.upenn.edu
3 The abbreviations used are: Tsp, thrombospondin;
VEGF, vascular endothelial growth factor; bFGF, basic fibroblast growth
factor; 4-OHT, 4-hydroxytamoxifen; CM, conditioned media; TUNEL,
terminal deoxynucleotidyl transferase-mediated nick end labeling.
4 A. Janz, C. Sevignani, K. Kenyon, C. V. Ngo,
and A. Thomas-Tikhonenko. Activation of the Myc oncoprotein leads
to increased turnover of the thrombospondin-1 mRNA, submitted for
publication.
5 C. V. Ngo and A. Thomas-Tikhonenko, unpublished observations.
Received for publication 2/ 9/99. Revision received 12/27/99. Accepted for publication 2/ 8/00.
References
This article has been cited by other articles:
 |
|
 |
|
  K. Shchors and G. Evan Tumor Angiogenesis: Cause or Consequence of Cancer? Cancer Res., August 1, 2007; 67(15): 7059 - 7061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Shchors, E. Shchors, F. Rostker, E. R. Lawlor, L. Brown-Swigart, and G. I. Evan The Myc-dependent angiogenic switch in tumors is mediated by interleukin 1beta Genes & Dev., September 15, 2006; 20(18): 2527 - 2538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Mizukami, K. Fujiki, E.-M. Duerr, M. Gala, W.-S. Jo, X. Zhang, and D. C. Chung Hypoxic Regulation of Vascular Endothelial Growth Factor through the Induction of Phosphatidylinositol 3-Kinase/Rho/ROCK and c-Myc J. Biol. Chem., May 19, 2006; 281(20): 13957 - 13963. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Ricci, Z. Jin, M. Dews, D. Yu, A. Thomas-Tikhonenko, D. T. Dicker, and W. S. El-Deiry Direct Repression of FLIP Expression by c-myc Is a Major Determinant of TRAIL Sensitivity Mol. Cell. Biol., October 1, 2004; 24(19): 8541 - 8555. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. E. Knies-Bamforth, S. B. Fox, R. Poulsom, G. I. Evan, and A. L. Harris c-Myc Interacts with Hypoxia to Induce Angiogenesis In vivo by a Vascular Endothelial Growth Factor-Dependent Mechanism Cancer Res., September 15, 2004; 64(18): 6563 - 6570. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Thomas-Tikhonenko, I. Viard-Leveugle, M. Dews, P. Wehrli, C. Sevignani, D. Yu, S. Ricci, W. el-Deiry, B. Aronow, G. Kaya, et al. Myc-Transformed Epithelial Cells Down-Regulate Clusterin, Which Inhibits Their Growth in Vitro and Carcinogenesis in Vivo Cancer Res., May 1, 2004; 64(9): 3126 - 3136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ruddell, P. Mezquita, K. A. Brandvold, A. Farr, and B. M. Iritani B Lymphocyte-Specific c-Myc Expression Stimulates Early and Functional Expansion of the Vasculature and Lymphatics during Lymphomagenesis Am. J. Pathol., December 1, 2003; 163(6): 2233 - 2245. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. Christian, J. Pilch, M. E. Akerman, K. Porkka, P. Laakkonen, and E. Ruoslahti Nucleolin expressed at the cell surface is a marker of endothelial cells in angiogenic blood vessels J. Cell Biol., November 24, 2003; 163(4): 871 - 878. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Yi, T. Sakai, R. Fassler, and E. Ruoslahti Antiangiogenic proteins require plasma fibronectin or vitronectin for in vivo activity PNAS, September 30, 2003; 100(20): 11435 - 11438. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. C. Fernandez, S. R. Frank, L. Wang, M. Schroeder, S. Liu, J. Greene, A. Cocito, and B. Amati Genomic targets of the human c-Myc protein Genes & Dev., May 1, 2003; 17(9): 1115 - 1129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. UDAGAWA, A. FERNANDEZ, E.-G. ACHILLES, J. FOLKMAN, and R. J. D'AMATO Persistence of microscopic human cancers in mice: alterations in the angiogenic balance accompanies loss of tumor dormancy FASEB J, September 1, 2002; 16(11): 1361 - 1370. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Hunter, D. Yu, M. Gee, C. V. Ngo, C. Sevignani, M. Goldschmidt, T. V. Golovkina, S. Evans, W. F. Lee, and A. Thomas-Tikhonenko Cutting Edge: Systemic Inhibition of Angiogenesis Underlies Resistance to Tumors During Acute Toxoplasmosis J. Immunol., May 15, 2001; 166(10): 5878 - 5881. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Janz, C. Sevignani, K. Kenyon, C. V. Ngo, and A. Thomas-Tikhonenko Activation of the Myc oncoprotein leads to increased turnover of thrombospondin-1 mRNA Nucleic Acids Res., June 1, 2000; 28(11): 2268 - 2275. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||
