| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Article |
From the Molecular Pathology Unit, Massachusetts General Hospital, Boston; and the Department of Pathology, Harvard Medical School, Boston, Massachusetts
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Numerous classes of molecules are implicated in regulating angiogenesis, including growth factors and their receptors, a variety of proteases, adhesion receptors, and ECM components.4-6 Key endothelial cell-specific growth factor receptors include the vascular endothelial growth factor (VEGF) receptors 1 (Flt1)7,8 and 2 (Flk1)9 and the Tie receptor tyrosine kinases (RTK), Tie-1 and Tie-2.10-15 Tie-2, which is expressed in the normal vasculature16 and is up-regulated in tumor angiogenic vessels,17,18 has at least two known ligands, angiopoietin (Ang)-1 and -2.15,19,20 Stimulation of Flk1 and Flt1 by VEGF, respectively, promotes endothelial cell migration/proliferation and tube formation.21-23 Engagement of Tie-2 by Ang-1 promotes recruitment of pericytes and smooth muscle cells, thereby helping to establish and maintain vascular integrity and quiescence, induces endothelial sprouting in vitro,24 and stimulates endothelial cell migration in culture.25 Ang-2 is an antagonist of Ang-1, that competes for Tie-2, and induces the loosening of the interactions between endothelial and perivascular support cells and ECM, reducing vascular integrity and facilitating access to angiogenic inducers.26
Angiogenesis is dependent on a tightly regulated balance between angiogenic promoters and inhibitors,4 and it is becoming apparent that the production of VEGFs and angiopoietins must be coordinated both quantitatively and temporally to ensure appropriate angiogenesis.26 Thus, Ang-2-dependent loosening of endothelial cell interactions with support cells and ECM, that precedes vessel sprouting, requires the cooperation of VEGF that is thought to supply endothelial cells with critical survival signals that may, in part, substitute for those provided by cell-cell and cell-ECM interaction.18,27 In the absence of VEGF and/or Ang-1, Ang-2 production may cause irreversible loss of vascular structures, which, in the physiological situation, such as the menstrual cycle, corresponds to vascular involution.
Ang-1 and -2 are soluble 70-kd factors, which consist of an amino-terminal coiled-coil domain and a carboxy-terminal fibrinogen-like domain.19,20 Targeted disruption of Ang-1 and overexpression of Ang-2 both resulted in embryonic death with defects in the normal angiogenesis.15,20 Recent evidence indicates that both factors are expressed in tumor vasculature,18,27 but the role of angiopoietins in tumor angiogenesis remains to be determined. To directly address the effect of angiopoietins in tumor angiogenesis, we overexpressed Ang-2 and Ang-1 in Lewis lung carcinoma (LLC) and TA3 mammary carcinoma cells and tested the ability of the transfectants to form local and metastatic tumors in syngeneic mice. Our results indicate that overexpression of Ang-2, but not Ang-1, in both LLC and TA3 cells inhibits their tumor forming properties in both subcutaneous growth and experimental metastasis assays. Tumors derived from cells overexpressing Ang-2 exhibited aberrant angiogenesis in vivo, characterized by disorganized aggregates of endothelial cells with few or no associated smooth muscle cells, and massive apoptosis of endothelial cells and surrounding tumor cells. These observations provide evidence that Ang-2 may play a regulatory role in tumor angiogenesis and suggest that its potential use toward therapeutic ends may justify further studies.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
LLC (ATCC, Rockville, MD) and TA3 mammary carcinoma28 cells were maintained in Dulbecco’s minimal essential medium (DMEM) with 10% fetal bovine serum (Irvine Scientific, Santa Ana, CA). All LLC and TA3 transfectants were cultured in DMEM supplemented with 10% fetal bovine serum and 5 mg/ml blasticidin (Invitrogen, Carlsbad, CA). Anti-vWF, anti-smooth muscle actin, anti-CD34 mAb, anti-VEGF, and anti-v5 epitope tag antibodies were from DAKO (Carpinteria, CA), Pharmingen La Jolla, (CA), Upstate Biotechnology (Lake Placid, NY), and Invitrogen, respectively. The Apoptag kit was purchased from Oncor, Inc. (Gaithersburg, MD).
Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) and Expression Constructs
Total RNA was isolated from mouse placenta, B16F1 and B16 F10 melanoma cells, and LLC and TA3 carcinoma cells using TRIzol reagent (Life Technologies, Inc., Grand Island, NY) according to the manufacturer’s recommendations. cDNA was synthesized from 5 mg of total RNA using Superscript II RNase H- reverse transcriptase (Life Technologies, Inc.). PCR was performed as described28 using the following primer pairs: Ang-1: forward, 5'-ATCTACACTATTTATTTTAATAAT-3'; reverse, 5'-AAAGTCCAAGGGCCGGATCATCAT-3'; Ang-2: forward, 5'-GAGGGAGGACTGGTGACAGCCACGG-3'; reverse, 5'-GAAATCTGCTGGCCGGATCATCAT-3'; VEGF: forward, 5'-ACCATGAACTTTCTGCTCTCTTGG-3'; reverse, 5'-CCGCCTTGGCTTGTCACATCTGCA-3'.
Full-length Ang-1 and Ang-2 cDNAs were isolated by PCR from mouse placenta cDNA using Pfu DNA polymerase (Stratagene, La Jolla, CA) and primer pairs corresponding to the 5'- and 3'-most 24 nucleotides of the coding sequence of each molecule derived from GenBank accession numbers U83509 and AF004326. The stop codons were omitted from the reverse primers to ensure C-terminal v5 epitope tag for both Ang-1 and Ang-2 molecules. The resulting PCR fragments were inserted into pEF6/V5-His TOPO vector (Invitrogen), which contains an C-terminal v5-epitope tag and a blasticidine resistance gene (Invitrogen). Authenticity and correct orientation of the Ang-1 and Ang-2 inserts were confirmed by DNA sequencing using the dideoxy chain termination method.
Transfection
LLC and TA3 cells were transfected using Lipofectamine (Life Technologies, Inc.) with expression vectors containing cDNAs encoding Ang-1, Ang-2, or with expression vector alone. Stable transfectants were selected for growth in DMEM supplemented with 10% fetal bovine serum and blasticidine (5 µg/ml), and resistant colonies were picked after 2to 3 weeks of growth in the selection medium. The culture supernatants of the transfectants were tested by enzyme-linked immunosorbent assay and Western blot analysis using anti-v5 mAb (Invitrogen) for expression of the appropriate gene products.
Sample Preparation and Western Blot Analysis
Serum-free supernatants of cultured LLC and TA3 transfectants were collected, concentrated using Amicon filters (Millipore, Bedford, MA), and separated by sodium dodecyl sulfate-10% polyacrylamide gel electrophoresis. The proteins were transferred onto Hybond-ECL membranes (Amersham Corp., Arlington Heights, IL) and v5-tagged Ang-1 and Ang-2 were detected using anti-v5 mAb (Invitrogen).
Tumor Growth and Metastasis
Transfected LLC and TA3 cells (106 in 0.2 ml of Hanks’ balanced salt solution per mouse) were injected subcutaneously or intravenously into male syngeneic C57BL or A/Jax mice, respectively (Jackson Laboratory, Bar Harbor, ME). Three independent isolates of each transfectant were used. Six mice were injected for each transfectant and two independent experiments were performed, such that each transfectant was injected into a total of 12 mice. The animals were observed daily. The duration of survival of the mice injected intravenously was monitored, and animals that seemed to be symptom-free 3 months after injection were sacrificed and examined at autopsy for detectable tumor growth. The lungs from all intravenously injected mice were removed and fixed for histological analysis. Mice were sacrificed 4 weeks after subcutaneous injection, and the tumors were isolated, weighed, and sectioned to assess angiogenesis.
Histology and Immunocytochemistry
Solid tumors from the experimental animals and lungs derived from intravenously injected mice were dissected and fixed in 4% paraformaldehyde (Fisher Scientific, Pittsburgh, PA) in phosphate-buffered saline (PBS), washed with PBS, dehydrated through 30%, 70%, 95%, and 100% ethanol and xylene, and embedded in paraffin wax (Fisher). Five- to 10-mm sections were cut, mounted onto slides, and stained with either Gill-2 hematoxylin and eosin (Fisher) for histological analysis, anti-von Willebrand factor, anti-smooth muscle actin, anti-CD34, and anti-VEGF antibodies (DAKO, Pharmingen, and Upstate) to assess tumor angiogenesis, or Apoptag (Oncor, Inc.) to assess apoptosis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Unlike its more widely expressed antagonist, Ang-1, Ang-2
expression is thought to be primarily confined to endothelial
cells.20,25
Because malignant cells frequently generate
their own angiogenesis regulating factors, we addressed angiopoietin
expression in B16F1 and F10 melanoma, and LLC and TA3 mammary carcinoma
cells by RT-PCR analysis. Ang-2 transcripts were readily detected in
LLC, B16F1, and F10 melanoma cells (Figure 1A
; lanes 3, 9, 12, arrow), but not in
TA3 cells (Figure 1A
, lane 6). Ang-1 was only weakly expressed by LLC
cells as assessed by RT-PCR (Figure 1A
, lane 2). By contrast, VEGF was
expressed in all four tumor cell lines (Figure 1A
; lanes 1, 4, 7, 10).
|
and data not shown). Monomeric
Ang-1 and Ang-2 are 
70 kd (Figure 1B)
, and as reported, have a
tendency to form aggregates (Figure 1B)
. Multiple independent
transfectants were selected for subsequent experiments to minimize the
possibility of observing behavior that reflects peculiarities of
individual clones. Comparison for growth rate in vitro
failed to show any effect of either Ang-1 or Ang-2 overexpression on
LLC or TA3 cell proliferation (data not shown). Ang-2 Expression by LLC and TA3 Tumor Cells Inhibits Local Tumor Growth
LLC and TA3 transfectants expressing Ang-1 or Ang-2 were injected
subcutaneously into syngeneic mice and tumor development was compared
to that derived from cells transfected with the expression vector
alone. Three independent isolates of each transfectant were used for
each experiment, and two independent experiments were performed. In
each experiment six syngeneic mice were injected with each
transfectant, such that a total of 36 mice were used to assess the
effect of each angiopoietin in each cell type. Ang-1 expression did not
significantly alter the rate of growth of either LLC- or TA3-derived
tumors (Figure 2, A and B)
. However,
expression of Ang-2 in both cell types resulted in marked reduction of
tumor growth (Figure 2, A and B)
. The observed reduction in tumor
growth varied according to the level of Ang-2 expression, and
transfectants with the highest expression formed tumor nodules of 1 to
3 mm that failed to grow to a larger size even after several weeks
(data not shown). In the cases illustrated in Figure 2
, mock-, Ang-1-
and Ang-2-transfectant tumors were palpable 2 weeks after injection,
but whereas mock transfectants formed rapidly growing tumors from that
point on, Ang-2 transfectants displayed consistently slower growth.
Thus, the difference in tumor size shown does not reflect delayed onset
but rather slower and more limited growth. Histologically, these tumors
displayed several distinct features. First, viable tumor tissue was
found in a peripheral cuff, whereas the center appeared pale,
consistent with necrotic and/or apoptotic tissue, and contained
multiple hemorrhagic areas. By contrast, tumors derived from Ang-1 and
vector-only transfectants seemed more homogeneous (Figure 3, A and B)
. Second, whereas Ang-1 and
vector-only transfectant-derived tumors displayed well-defined blood
vessels surrounded by smooth muscle cells (Figure 3
; C, E, and G, and
data not shown), the central areas of Ang-2 transfectant-derived tumors
stained positive for von Willebrand factor (vWF) and CD34, both of
which are endothelial cell markers, with CD34 being more restricted to
endothelium of angiogenic vessels, but the vessels appeared as
cord-like structures with rare or absent lumina (Figure 3, D and F
, and
data not shown), and few disorganized smooth muscle cells (Figure 3H)
.
No detectable difference in VEGF expression was observed among Ang-2
and Ang-1 overexpressing and parental tumors (data not shown). Massive
apoptosis of vascular endothelial cells and surrounding tumor cells, as
indicated by terminal dUTP nick-end labeling (TUNEL) staining, was
observed within the Ang-2 transfectant-derived tumor centers (Figure 3, J and L)
. In contrast, the LLC carcinoma cells transfected with Ang-1
and expression vector showed a low percentage of apoptotic cells that
were evenly distributed within the solid tumor (Figure 3, I and K
, and
data not shown).
|
|
As in the above experiments, a total of 36 mice were injected
intravenously with three independent isolates of each angiopoietin
transfectant (12 mice per isolate per cell type). Both TA3 and LLC
cells form lung tumor nodules when injected into the tail vein of
syngeneic mice, and death because of tumor growth typically occurred
within 4 to 6 weeks after injection of 1 x
106
tumor cells. Ang-1 overexpression neither
hindered nor promoted TA3 and LLC tumor formation in the lungs and did
not affect animal survival (Figure 2, C and D
, and data not shown).
However, Ang-2 expression reduced and even abrogated tumor nodule
formation in the lungs by both types of cells, and most of the animals
survived to the end of the study period (Figure 2, C and D)
. Gross
examination of lung tissue at autopsy revealed no apparent tumor
nodules (data not shown).
Histological examination of lung tissue sections from animals injected
with LLC carcinoma cells expressing Ang-1 or vector only, revealed
large confluent tumor nodules that invaded most of the lung parenchyma
(Figure 4A)
. Histological examination of
lungs from animals injected with LLC carcinoma cells expressing Ang-2,
on the other hand, revealed either absence of detectable tumor tissue
(in 50% of mice) or between one and 10 microscopic tumor nodules that
were typically centered by bronchioles or pre-existing peribronchial
blood vessels (Figure 4B)
. Remarkably, and in contrast to Ang-1 and
vector transfectant-derived tumors, the majority of angiogenic vessels
within these small tumor nodules, as identified by anti-vWF and
anti-CD34 antibody staining, displayed an aberrant structure,
characterized by aggregates of endothelial cells, with a narrow or
absent lumen (Figure 4, C–F
, and data not shown). A notable feature
was the paucity or absence of smooth muscle cells surrounding the
endothelial cells, in contrast to the normal angiogenic vessels in
tumors derived from Ang-1 and vector-only transfected cells (Figure 4, G–J
, and data not shown). TUNEL staining revealed that both
endothelial and surrounding tumor cells were undergoing apoptosis in
the Ang-2-expressing tumors whereas only occasional scattered
TUNEL-positive cells were observed in tumors derived from Ang-1 and
vector-only transfected cells (Figure 4, K–M
, and data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Earlier studies had suggested a possible association between high levels of Ang-2 expression, in the absence of VEGF, and tumor cell death.18,27 In rat gliomas, Ang-2, VEGF, and Tie-2 expression were observed to vary in location and magnitude as a function of tumor size. In small tumors, Ang-2 transcripts were consistently observed in the vessels whereas VEGF expression was minimal. Tie expression was up-regulated in larger tumors, whereas VEGF induction was observed only in very large tumors and appeared in tumor cuffs surrounding the peripheral vessels,27 where it coincided with Ang-2 expression. Areas where massive tumor cell apoptosis occurred were associated with Ang-2 expression in the absence of VEGF. In a separate study, Ang-2 expression was observed in endothelial cells of a subset of glioblastoma blood vessels that were typically associated with few periendothelial support cells.18 Our present observations provide strong functional support for these correlations and indicate that disrupting the balance between VEGF (found to be expressed at comparable levels in vector-only, Ang-1, and Ang-2 transfectants), Ang-1 and Ang-2 by tumor cell overexpression of Ang-2 has a potent inhibitory effect on tumor growth and dissemination.
The mechanism of Ang-2-associated endothelial cell death remains to be elucidated, but the absence of smooth muscle cells, that normally associate with endothelial cells to form mature vascular structures may provide an important lead. The inability of the endothelial cells to recruit smooth muscle cells may on the one hand lead to the absence of fully formed blood vessels, resulting in leakage and/or hemorrhage, and on the other, induce endothelial cell apoptosis because of the lack of signals generated by the physical interaction between smooth muscle and endothelial cells that ensure endothelial survival.29 In addition, recent observations suggest that Tie-2 receptor engagement by Ang-1 may induce survival signals in vascular endothelial cells by activating the phosphatidyl 3'-inositol kinase/Akt signal transduction pathway.30,31 It seems reasonable to predict that absence of Tie-2-mediated Akt activation coupled to ineffective interaction with smooth muscle cells, is responsible, at least in part, for the observed endothelial cell apoptosis.
The observed effect of Ang-2 on tumor growth seems to be the result of its interaction with endothelial cell receptors. We have not observed Ang-2 interaction with the tumor cell surface (data not shown) rendering unlikely the possibility that Ang-2 might have a direct effect on tumor survival and growth by engaging tumor cell surface receptors other than Tie-2. This notion is supported by the observation that expression of Ang-2 failed to alter tumor cell growth in vitro. Furthermore, histological examination of lung tissue at early time points in animals injected with Ang-2 overexpressing tumor cells, revealed focal host vascular endothelial apoptosis and hemorrhage (data not shown), consistent with the possibility that tumor-derived Ang-2 may trigger pre-existing vascular endothelial cell death. Taken together, our results provide evidence that Ang-2 overexpression in tumor cells inhibits tumor angiogenesis and may induce apoptosis of host endothelial cells to which tumor cells adhere. The hypoxia and lack of nutrients that accompany these effects may explain the extensive apoptosis observed in Ang-2 overexpressing tumors as well as the microscopic nature of tumor nodules in the lungs of animals injected with Ang-2 transfectants.
The present work extends previous observations by others that Ang-2 expression correlates with vascular regression and provides functional evidence that overexpression of Ang-2 has an anti-angiogenic effect in tumors.18,27 The observation that Ang-2 can effectively block disseminated tumor growth raises the possibility that it may have a role to play in therapeutic anti-angiogenic intervention. Thus far, multiple synthetic reagents and ECM degradation products have been shown to have anti-angiogenic action,32 but their therapeutic value remains to be fully elucidated. Creating an imbalance among angiogenic and anti-angiogenic factors by targeting Ang-2 or synthetic Ang-2 mimetics that bind Tie-2 and displace Ang-1, to the tumor vasculature, may provide a means to control local and disseminated tumor growth. Such an approach offers the advantage of modulating the activity of a naturally occurring receptor whose physiological function is precisely the regulation of angiogenesis.
|
Footnotes |
|---|
This work was supported by National Institutes of Health grants CA55735 and GM54176.
Accepted for publication October 19, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  W. S.N. Shim, I. A.W. Ho, and P. E.H. Wong Angiopoietin: A TIE(d) Balance in Tumor Angiogenesis Mol. Cancer Res., July 1, 2007; 5(7): 655 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Cao, P. Sonveaux, S. Liu, Y. Zhao, J. Mi, B. M. Clary, C.-Y. Li, C. D. Kontos, and M. W. Dewhirst Systemic Overexpression of Angiopoietin-2 Promotes Tumor Microvessel Regression and Inhibits Angiogenesis and Tumor Growth Cancer Res., April 15, 2007; 67(8): 3835 - 3844. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Kunz, J. Hoffend, A. Altmann, A. Dimitrakopoulou-Strauss, D. Koczan, M. Eisenhut, G. A. Bonaterra, T. J. Dengler, W. Mier, U. Haberkorn, et al. Angiopoietin-2 Overexpression in Morris Hepatoma Results in Increased Tumor Perfusion and Induction of Critical Angiogenesis-Promoting Genes J. Nucl. Med., September 1, 2006; 47(9): 1515 - 1524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Nakayama, N. Mutsuga, L. Yao, and G. Tosato Prostaglandin E2 promotes degranulation-independent release of MCP-1 from mast cells J. Leukoc. Biol., January 1, 2006; 79(1): 95 - 104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Machein, A. Knedla, R. Knoth, S. Wagner, E. Neuschl, and K. H. Plate Angiopoietin-1 Promotes Tumor Angiogenesis in a Rat Glioma Model Am. J. Pathol., November 1, 2004; 165(5): 1557 - 1570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Xu, Y.-j. Liu, and Q. Yu Angiopoietin-3 Is Tethered on the Cell Surface via Heparan Sulfate Proteoglycans J. Biol. Chem., September 24, 2004; 279(39): 41179 - 41188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Xu, Y.-j. Liu, and Q. Yu Angiopoietin-3 Inhibits Pulmonary Metastasis by Inhibiting Tumor Angiogenesis Cancer Res., September 1, 2004; 64(17): 6119 - 6126. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Cerimele, L. F. Brown, F. Bravo, G. M. Ihler, P. Kouadio, and J. L Arbiser Infectious Angiogenesis: Bartonella bacilliformis Infection Results in Endothelial Production of Angiopoetin-2 and Epidermal Production of Vascular Endothelial Growth Factor Am. J. Pathol., October 1, 2003; 163(4): 1321 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zhang, N. Yang, J.-W. Park, D. Katsaros, S. Fracchioli, G. Cao, A. O'Brien-Jenkins, T. C. Randall, S. C. Rubin, and G. Coukos Tumor-derived Vascular Endothelial Growth Factor Up-Regulates Angiopoietin-2 in Host Endothelium and Destabilizes Host Vasculature, Supporting Angiogenesis in Ovarian Cancer Cancer Res., June 15, 2003; 63(12): 3403 - 3412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Tanaka, S. Ishikawa, K. Yanagihara, R. Miyahara, Y. Kawano, M. Li, Y. Otake, and H. Wada Expression of Angiopoietins and Its Clinical Significance in Non-Small Cell Lung Cancer Cancer Res., December 1, 2002; 62(23): 7124 - 7129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Chavakis and S. Dimmeler Regulation of Endothelial Cell Survival and Apoptosis During Angiogenesis Arterioscler. Thromb. Vasc. Biol., June 1, 2002; 22(6): 887 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hawighorst, M. Skobe, M. Streit, Y.-K. Hong, P. Velasco, L. F. Brown, L. Riccardi, B. Lange-Asschenfeldt, and M. Detmar Activation of the Tie2 Receptor by Angiopoietin-1 Enhances Tumor Vessel Maturation and Impairs Squamous Cell Carcinoma Growth Am. J. Pathol., April 1, 2002; 160(4): 1381 - 1392. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Xu and Q. Yu Angiopoietin-1, Unlike Angiopoietin-2, Is Incorporated into the Extracellular Matrix via Its Linker Peptide Region J. Biol. Chem., September 7, 2001; 276(37): 34990 - 34998. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
