| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 276, Issue 37, 34990-34998, September 14, 2001
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Pathobiology, School of Veterinary Medicine,
University of Pennsylvania, Philadelphia, Pennsylvania 19104
Received for publication, April 24, 2001, and in revised form, June 28, 2001
Angiopoietin-1 (Ang-1) and angiopoietin-2 (Ang-2)
affect angiogenesis differently during embryogenesis and tumorigenesis. In an attempt to understand the molecular basis underlying the distinct
roles of those two homologous molecules, we investigated the
association of Ang-1 and Ang-2 with the extracellular matrix (ECM). TA3
murine mammary carcinoma (TA3) and Lewis lung carcinoma cells
expressing v5 epitope-tagged Ang-1 and Ang-2 were used in our studies.
The results indicated that Ang-1 is secreted and incorporated into the
ECM of the tumor cells, whereas Ang-2 is not associated with the ECM.
The mutagenesis study indicated the domain that is responsible for the
ECM association of Ang-1 is the linker peptide region between the
coiled-coil and the fibrinogen-like domains. A weak binding between
the coiled-coil domain of Ang-1 and the ECM was observed.
Immunocytochemistry study revealed a distinct ECM distribution pattern
of Ang-1, which is quite different from that of fibronectin, laminin,
and collagen types I and IV. The ECM-associated Ang-1 proteins
are released, and Tie-2 receptors are phosphorylated upon the adhesion
of human umbilical vein endothelial cells. Implications of the
difference in the ECM association of Ang-1 and Ang-2, which are related
to the regulation of angiopoietin activity and their roles in local
versus distant angiogenesis during tumor metastasis, are discussed.
Angiogenesis plays an important role in embryogenesis and
tumorigenesis (1-5). It is a complicated multistep process, which includes the dynamic changes of cell-cell and cell-matrix interactions, endothelial cell proliferation and migration, recruitment of the peri-vascular supporting cells, and the maturation process. Numerous molecules are involved in those processes, including growth factors and
their receptors, proteases, adhesion receptors, and the
ECM1 components. VEGF and
angiopoietin families play special roles in angiogenesis due to the
restricted expression of their receptors (6-12).
Ang-1 and -2 are ~70 kDa with a considerable sequence homology, which
consists of a signal peptide, an N-terminal coiled-coil domain, a short
linker peptide region, and a C-terminal fibrinogen homology domain
(FHD) (13, 14). The coiled-coil region is responsible for
dimerization/multimerization of angiopoietins, and the fibrinogen
homology domain binds to Tie-2 receptor (15, 16). Both Ang-1 and Ang-2
form dimers and oligomers (13-15, 17).
Angiopoietin-1 and -2 are the unique antagonists. Ang-1 induces
tyrosine phosphorylation of Tie-2 receptor and promotes recruitment of
the pericytes and smooth muscle cells, thereby playing a role in
establishing and maintaining the vascular integrity and quiescence. As
an antagonist of Ang-1, Ang-2 competes with Ang-1 for binding of Tie-2,
blocks the phosphorylation of Tie-2 receptors induced by Ang-1, and
loosens the interactions between endothelial and peri-vascular support
cells and ECM (14).
Targeted disruption of Ang-1 and Tie-2 and overexpression of Ang-2
resulted in embryonic death with the similar vascular defects (7, 14,
18). Those mice have normal primary vascular development, but the
remodeling and maturation of the vasculature are defective (7, 14, 18,
19). The transgenic mice overexpressing Ang-1 displayed increased
vascularization and decreased adult vasculature leakage (20-22).
Together, those results indicated that Ang-1 plays an indispensable
role in the formation of blood vessels during mouse development by
recruiting and maintaining peri-endothelial support cells.
Several studies have offered possible mechanisms for the pro-angiogenic
effect of Ang-1. Although Ang-1 does not stimulate the proliferation of
endothelial cells, it stimulates endothelial cell migration (23),
induces the capillary-like tubule formation, and promotes survival of
endothelial cells (24-28). Ang-1 inhibits apoptosis of the endothelial
cells via phosphatidylinositol 3-kinase/Akt pathway (26, 27).
Angiogenesis is regulated by the precise balance between pro- and
anti-angiogenic factors (4). Ang-2 expression is often induced in the
endothelia undergoing active remodeling or regression and by hypoxia
and several growth factors, including VEGF (14, 29-33). Ang-2
destabilizes the vasculature. Thus, it initiates angiogenesis in the
presence of VEGF, which supplies endothelial cells with
necessary survival and proliferation signals (30, 34, 35), or induces
apoptosis of endothelial cells in the absence of the pro-angiogenic
factors (36).
Recent evidence indicates that Ang-1 and -2 are expressed by tumor
cells (17, 29). We found that overexpression of exogenous Ang-2, but
not Ang-1, inhibits growth and metastasis of Lewis lung carcinoma (LLC)
and TA3 murine mammary carcinoma cells. The tumors overexpressing Ang-2
exhibited aberrant and incomplete angiogenesis in vivo,
which is characterized by formation of the disorganized endothelial
cell aggregates, the lack of endothelial associated smooth muscle
cells, and massive apoptosis of the endothelial cells and surrounding
tumor cells (17). This result is consistent with the notion that Ang-2
inhibits the Ang-1-dependent recruitment of smooth muscle cells.
To reveal the molecular basis underlying the different roles of
angiopoietins in tumor angiogenesis, we investigated the relationship between angiopoietins and the ECM in the present study. We found that,
unlike Ang-2, Ang-1 is secreted and incorporated into the ECM via its
linker peptide region. The association between Ang-1 and the ECM is
strong, and the distribution of Ang-1 in the ECM is unique and
different from that of fibronectin, laminin, and collagen type I and
type IV. The releasing or incorporation of Ang-1 from or into the ECM
is regulated by different factors. Tie-2 phosphorylation was detected
in HUVECs seeded onto the ECM containing Ang-1, which indicates a
regulatory role of the ECM association of Ang-1 in tumor angiogenesis.
Ang-2 expression is regulated by hypoxia and growth factors (29-33).
Unlike Ang-2, little is known about the regulation of Ang-1 expression.
The finding reported herein offers a possible regulatory mechanism for
the availability of Ang-1 proteins, that is instead of regulating the
production of Ang-1, it may be regulated by its ECM association. The
strong ECM binding of Ang-1 implies the effect of Ang-1 is limited to
the local environment where it is produced, whereas Ang-2 can diffuse
to and affect angiogenesis in the distant sites.
Cell Culture and Reagents--
Lewis lung carcinoma (LLC, ATCC),
TA3 mammary carcinoma (37) cells, and the tumor cell transfectants were
maintained as described (17). Anti-laminin, -fibronectin (Sigma),
-collagen types I and IV (Biodesign international), -phosphotyrosine
(PY20, Calbiochem), -Tie-2 (C-20, Santa Cruz Biotechnology), -Ang-1
(C-19, Santa Cruz Biotechnology), and -v5 antibodies were used.
Construction of the Expression Vectors--
Full-length Ang-1
and Ang-2 cDNAs were generated by RT-PCR as described (17, 40). The
coiled-coil domain and the coiled-coil plus the linker peptide region
were generated by RT-PCR using the following pairs of primers. The
forward primer for the coiled-coil domain is
5'-ACAATGACAGTTTTCCTTTCCTTT-3' and the reverse primer is
5'-TGTGTCCATGAGCTCCAGTTGTTG-3'. The coiled-coil plus the linker peptide
region was amplified using the same forward primer as that of the
coiled-coil domain and a reverse primer,
5'-AAATGGTTTCTCTTCTTCTCTTTT-3'. The stop codons were omitted from the
reverse primers so that the v5 and 6× histidine tags in the
pEF6/V5-His expression vector (Invitrogen) can be attached to the C
terminus of the Ang-1 fragments. To fuse the FHD of Ang-1 to the signal
peptide of Ang-1, we used the full-length Ang-1 in the expression
vector (17) as a template and the ExSite PCR-based site-directed
mutagenesis kit (Stratagene) together with a forward primer derived
from the beginning of the FHD, 5'-CGAGACTGTGCAGATGTATATCAA-3', and a
reverse primer derived from the end of the signal peptide,
5'-TCTTCTCCCTCCGTTTTCTGGATT-3'. The extracellular domain of Tie-2 was
obtained by RT-PCR using the following pair of primers. The forward
primer is 5'-AGTATGGACTCTTTAGCCGGCTTA-3' and the reverse primer is
5'-CATCTTTCCCCCTCCGAGGTCTGC-3'. The PCR products were inserted in frame
into 5'-end of the Fc fragments of human IgG in pEF6/V5-His expression
vectors to generate the Tie-2-Fc fusion construct (40). The
authenticity and orientation of the cDNA inserts were confirmed by
DNA sequencing.
Transfection--
LLC and TA3 carcinoma cells were transfected
with the expression constructs containing Ang-1, Ang-2, or the
expression vector alone. The transfected cells expressing Ang-1 and
Ang-2 were identified as described (17). COS-7 cells were used
in the transient transfection of Ang-1 and -2, the coiled-coil domain,
the coiled-coil plus linker region, and the FHD of Ang-1 using
LipofectAMINE (Life Technologies, Inc.) as described (17).
Preparation of the Secreted and the ECM-associated Proteins and
Western Blot Analysis--
The cell culture supernatants of LLC and
TA3 transfectants or the transiently transfected COS-7 cells were
collected. The confluent cells layers were released from the culture
dishes by incubating with PBS containing 5 mM EDTA. The ECM
components remaining on the culture dishes were washed and extracted
with 1× SDS Laemmli sample buffer with or without 5%
To determine the affinity of the ECM association of Ang-1, the
confluent LLC cells expressing Ang-1 were lifted as described. The
remaining ECM components were extracted with 0.5 M PBS and 1 M sodium chloride (NaCl) and 0.5 and 1% deoxycholate
(DOC) at room temperature for 10 min. The remaining insoluble materials were solubilized by 1× SDS Laemmli sample buffer and subjected to
Western blot analysis.
Immuocytochemistry--
LLC cells expressing Ang-1 or Ang-2 or
transfected with the expression vectors were cultured in 35-mm dishes
until subconfluence or confluence. They were fixed with methanol at
Extraction of the ECM Components and Affinity
Purification--
The ECM materials derived from the cultured LLC
cells were extracted overnight at 4 °C in 2 M urea and
0.05 M Tris-HCl (pH 7.4). The soluble materials were
dialyzed against PBS and used to coat the enzyme-linked immunosorbent
assay plates. The ECM derived from the LLC transfectants expressing
Ang-1 was extracted in the same way. The ECM extracts containing Ang-1
proteins and the serum-free culture media containing secreted Ang-2,
both of which contain v5 and 6× histidine tags at their C-terminal
ends, were loaded onto Ni+ Probond affinity columns
(Invitrogen) and purified following the manufacturer's instructions.
Purified Ang-1v5/His proteins were loaded onto a heparin-Sepharose
CL-6B column to test their affinity to heparin. The flow-through fraction was collected, and the column was washed with different concentrations of NaCl from 0.15 to 1 M. The eluted
fractions were collected and, along with the flow-through fractions,
were subjected to Western blot analysis.
Solid Phase Binding Assay--
96-Well enzyme-linked
immunosorbent assay plates were coated overnight at 4 °C with
different ECM components. The components were fibronectin, laminin,
type I and IV collagens, Matrigel (20 µg/ml, Becton Dickinson),
vitronectin (10 µg/ml, Sigma), fibrinogen (20 µg/ml, Sigma),
heparin, chondroitin sulfate, hyaluronic acid (200 µg/ml, Sigma), and
the whole ECM extracts derived from LLC cells (100 µg/ml). The coated
plates were washed and blocked with 0.5% bovine serum albumin. The
affinity-purified Ang-1v5/His and Ang-2v5/His were added into the
coated plates (100 ng/ml) for overnight incubation at 4 °C. After
extensive washing, the bound Ang-1 and Ang-2 were detected. The assays
were performed in triplicate.
Tie-2 Tyrosine Phosphorylation Assay--
LLC carcinoma cells
expressing Ang-1 or Ang-2 were cultured in 100-mm dishes until
confluence and were lifted from the dishes as described. The dishes
containing the ECM components deposited by the cultured cells were used immediately.
HUVECs (ATCC) were cultured until subconfluence, and switched into the
serum-free medium overnight. The cells were then lifted with 0.53 mM EDTA in Hanks' balanced buffer (Life Technologies, Inc.) and washed. 1 × 106 HUVECs were seeded onto a
plastic dish or one of the ECM-coated dishes freshly generated as
described above. They were cultured at 37 °C in serum-free medium
with or without soluble Ang-1 (200 ng/ml) or Tie-2-Fc fusion proteins
(2 µg/ml) for 30 min. The cells were then lysed at 4 °C with the
lysis buffer (50 mM Tris-HCl (pH 7.4), 50 mM
NaCl, 1% Triton X-100, 2 mM EDTA, 2 mM sodium orthovanadate, 2 mM sodium fluoride, 2 mM
phenylmethylsulfonyl fluoride, 1 mM leupeptin, 1 mM pepstatin A, and 10 µg/ml aprotinin). Tie-2 proteins
were immunoprecipitated using anti-Tie-2 polyclonal antibody (Santa
Cruz Biotechnology). The immunoprecipitated proteins were subjected
Western blot analysis using anti-phosphotyrosine antibody (Y20, Calbiochem).
Angiopoietin-1, but Not Angiopoietin-2, Binds to the Extracellular
Matrix--
In the process of studying the role of angiopoietins in
tumor angiogenesis, we experienced difficulty obtaining the transfected tumor cells that secrete high levels of Ang-1, whereas the
transfectants that secrete high levels of Ang-2 were easily obtained.
One possible reason for the phenomena is that Ang-1 and Ang-2 may
associate with the ECM differently. To investigate that, we generated
the stably transfected Lewis lung carcinoma (LLC) and TA3 murine
mammary carcinoma (TA3) cells expressing v5 epitope-tagged Ang-1 or
Ang-2 (Fig. 1, A, a and
b). As reported previously (15, 17), under non-reducing
conditions both Ang-1 and Ang-2 tend to aggregate with each other to
form dimers and oligomers (Fig. 1, A and B). The
patterns of the aggregation are distinct between Ang-1 and Ang-2. Ang-1
tends to form higher order oligomers (Fig. 1A), whereas Ang-2 forms dimers, trimers, and oligomers. The molecular weight of the
monomer of Ang-1 and Ang-2 is ~70 kDa. Because of the C-terminal v5
and 6× histidine epitope tags, Ang-1v5 and Ang-2v5 migrate a little
slower.
The aggregation of Ang-1 and Ang-2 is sensitive to the reducing agents,
such as
To confirm that the endogenously expressed Ang-1 is associated with the
ECM as the transfected Ang-1, the polyclonal anti-Ang-1 antibody (C19,
Santa Cruz Biotechnology) was first purified through an
Ang-1v5-conjugated protein A affinity column. The purified antibody was
used in Western blot analysis of the ECM components derived from
HUVECs. The results indicated that Ang-1 produced by HUVECs is
incorporated into the ECM (data not shown). Due to the superior quality
of the anti-v5 monoclonal antibody compared with that of the polyclonal
anti-Ang-1 antibody, even after the affinity purification (Santa Cruz
Biotechnology), most of the following experiments were performed using
the transfected tumor cells expressing v5-tagged Ang-1 and Ang-2.
The Biochemical Characters of the ECM-associated Ang-1--
The
strength of the ECM association of Ang-1 was tested by extracting the
cell-free ECM deposited on the culture dishes by LLC cells expressing
Ang-1v5 with different concentrations of sodium chloride (NaCl) and
deoxycholate (DOC). The insoluble materials after the extractions were
solubilized with 1× SDS Laemmli buffer and subjected to Western blot
analysis. As shown in Fig. 2, most of the
ECM-incorporated Ang-1 resisted 0.15 M NaCl extraction (Fig. 2, lane 2); a fair amount of Ang-1 remains in the ECM
after 1 M NaCl extraction (Fig. 2, lane 4), and
a fraction of Ang-1 is in the ECM fraction even after 1% DOC
extraction (Fig. 2, lane 6), which indicated a strong ECM
association and implicated a gradual assembly process of Ang-1 into the
ECM (Fig. 2).
The Binding Affinity of Ang-1 to Different ECM Components--
The
C terminus of Ang-1 shares the sequence homology with fibrinogen, which
binds to heparin. To investigate whether Ang-1 binds to the ECM via its
interaction with the sulfated glycosaminoglycans, the purified
Ang-1v5/His (Fig. 3A, lane 1)
was applied to a heparin-Sepharose CL-6B affinity column (Amersham
Pharmacia Biotech). The unbound flow-through was collected, and the
column was washed with 0.15, 0.3, 0.5, 1 M NaCl (Fig.
3A, lanes 3-6). The eluted fractions were collected, and
along with the flow-through fraction, they were subjected to the
Western blot analysis by using anti-v5 monoclonal antibody to detect
the presence of Ang-1v5/His in each fraction. The results indicated
that Ang-1v5/His does not bind to the heparin affinity column and was
fully recovered in the flow-through fraction (Fig. 3A, lane
2). To confirm the above finding and avoid the possibility that
the purified soluble Ang-1v5/His may be modified and different from the
ECM-associated Ang-1 and thereby unable to bind to heparin, soluble
sulfated glycosaminoglycans, heparin, and chondroitin sulfate (200 µg/ml) were added into serum-free cell culture medium (SFM) of LLC
carcinoma cells expressing Ang-1. After 2, 12, and 24 h of
incubation, the SFM and ECM fractions were collected and analyzed. The
Western blot results indicated that neither heparin nor chondroitin
sulfate releases Ang-1 from the ECM of the transfected LLC cells (Fig.
3B, and data not shown).
In an attempt to identify the ECM component(s) that bind(s) to Ang-1,
solid phase binding assays were performed to assess the binding
affinity of Ang-1 to several ECM components. The assays were performed
in triplicate, and the results are listed in the Fig.
4. The purified Ang-1v5 binds to whole
ECM extracts derived from LLC carcinoma cells with high affinity (Fig.
4, column 8). The weak bindings to Matrigel, fibrinogen, and
vitronectin were observed, which can't account entirely for the high
affinity binding between Ang-1 and the ECM extracts. Ang-1 displayed no
affinity to fibronectin, laminin, collagen type I and type IV, heparin, chondroitin sulfate, and hyaluronic acid (Fig. 4). This result offered
the possibility that Ang-1 binds to the ECM via an unidentified ECM
protein(s).
Immunocytochemistry Studies Revealed a Distinct ECM Distribution
Pattern of Angiopoietin-1--
The distribution patterns of Ang-1 and
Ang-2 were investigated and compared with fibronectin, laminin, and
type I and IV collagens by performing immunocytochemistry on LLC
carcinoma cells expressing Ang-1v5 or Ang-2v5 using anti-v5 antibody or
antibodies against the appropriate ECM proteins. The cell-free
ECM deposited by LLC cells expressing Ang-1 or Ang-2 was also examined.
The immunocytochemistry studies uncovered a distinct Ang-1 distribution
pattern in the ECM of LLC carcinoma cells, which is different from the
distribution of fibronectin, laminin, and types I and IV collagen (Fig.
5, A and B). Ang-1
is incorporated into the ECM as small granule-like depositions, which
are more or less evenly distributed beneath the cells (Fig. 5A,
a and b). Some cell-free spaces, which are left behind
by the migrating cells, are positive for the similar Ang-1 depositions
(Fig. 5A, a, arrow). The distribution pattern of Ang-1 was
preserved even after the cells were lifted by the treatment of EDTA
(Fig. 5A, b), which indicated that instead of loosely
binding to the tumor cells, those Ang-1 proteins are incorporated into
the ECM of the tumor cells. On the contrary, there is no trace of Ang-2
in the cell-free ECM deposited by the cells expressing Ang-2 (Fig.
5A, d). Ang-2 was only detected in the cytoplasm of the
transfected cells, which presumably reflects the presence of Ang-2 in
the secretory pathway of the cells (Fig. 5A, c,
arrowheads).
Angiopoietin-1 Is Released in the Response to Phorbol 12-Myristate
13-Acetate (PMA)--
The association of Ang-1 with the ECM of tumor
cells led us to explore the regulatory mechanisms of its incorporation
and releasing, which may modulate its activity. The confluent LLC carcinoma cells expressing Ang-1 were cultured overnight in the presence of different growth factors or PMA in serum-free cell culture
medium. After releasing the cells from the culture dishes, the
remaining ECM components were extracted with 1× SDS Laemmli buffer and subjected to Western blot analysis using anti-v5 antibody. The results indicated that PMA stimulates the releasing of the ECM-associated Ang-1 (Fig. 6, lane
2). Transforming growth factor- The ECM-associated Ang-1 Does Not Bind to Tie-2-Fc Fusion
Protein--
To determine whether the ECM-associated Ang-1 proteins
bind to Tie-2-Fc fusion proteins, the confluent LLC cells expressing Ang-1 or the ECM deposited by the LLC cells expressing Ang-1 were fixed
and incubated with anti-v5 antibody or the purified Tie-2-Fc fusion
proteins. Anti-v5 antibody and FITC-conjugated rabbit anti-mouse antibody (Sigma) revealed that many Ang-1 proteins were deposited into
the ECM of the tumor cells (Fig.
7B, a and b);
however, Ang-1 proteins in the ECM exhibited no affinity to Tie-2-Fc
fusion proteins (Fig. 7B, c and d). In order to
eliminate the possibility that the fixation procedure could affect the
binding between the ECM-associated Ang-1 and Tie-2-Fc fusion protein,
Tie-2-Fc fusion proteins were directly added into the cultured LLC
carcinoma cells expressing Ang-1 or into the ECM freshly derived from
the LLC cells expressing Ang-1. No binding was detected in both cases
(data not shown). Those ECM-associated Ang-1 proteins were extracted by
2 M urea/Tris-HCl buffer (pH 7.4), dialyzed in PBS, and
immunoprecipitated using Tie-2-Fc fusion proteins and protein A beads.
The precipitated proteins were subjected to Western blot analysis using
anti-v5 antibody. Ang-1 proteins were found to be precipitated by
Tie-2-Fc fusion proteins (Fig. 7A, lane 2). Together, those
results suggested that after their secretion from the tumor cells,
Ang-1 proteins are incorporated into the ECM, and their Tie-2-binding
sites are no longer accessible; upon the release from the ECM, the
solubilized Ang-1 proteins regained their ability to bind to Tie-2.
Under physiologic conditions, this may serve as an efficient regulatory mechanism for Ang-1 activity.
Tie-2 Phosphorylation Is Achieved by Adhering HUVECs to the ECM
Containing Ang-1 Proteins--
To study whether the ECM-associated
Ang-1 plays a role in angiogenesis, the ability of the ECM-associated
Ang-1 to promote Tie-2 receptor phosphorylation was evaluated using
HUVECs. The subconfluent HUVECs were serum-starved for about 14 h
and then lifted from the culture dishes. 1 × 106
cells were seeded onto a plastic culture dish with (Fig.
8, lane 3) or without (Fig. 8,
lane 1) purified soluble Ang-1 (200 ng/ml) or the
culture dish containing the ECM components deposited by the confluent
LLC cells expressing Ang-1 (Fig. 8, lane 4) or Ang-2 (Fig.
8, lane 2). The HUVECs were incubated on those cell culture dishes for 30 min and then lysed at 4 °C with the lysis buffer. The
amount of the Tie-2 proteins in the lysates was determined by Western
blot analysis of 50 µg of proteins from each lysate using anti-Tie-2
antibody (C-20, Santa Cruz Biotechnology, Fig. 8B). The
results indicated that the endothelial cells used in the experiments
express a similar amount of Tie-2 receptors (Fig. 8B). The
phosphorylated Tie-2 proteins in the lysates were assessed by
performing immunoprecipitation using anti-Tie-2 polyclonal antibody
(C-20, Santa Cruz Biotechnology) and analyzing the precipitated proteins on Western blot using anti-phosphotyrosine antibody (Y20, Calbiochem). The results indicated that Tie-2 receptors on HUVECs are
phosphorylated by Ang-1 derived from the ECM, and the phosphorylation is inhibited by the addition of the excess amount of Tie-2-Fc fusion
proteins (2 µg, Fig. 8, lane 5).
To assess whether HUVECs release Ang-1 proteins from the ECM, HUVECs
cells (1 × 106 cells/100-mm dish) or the serum-free
culture media (SFM) alone were placed on the ECM deposited by LLC cells
expressing Ang-1 for 30 min. The cell culture dishes with or without
HUVECs were treated with 5 mM EDTA in PBS for 5 min to
release the adhering cells. The remaining ECM materials on the culture
dishes were extracted and subjected to Western blot analysis, and the
results indicated that the reduced amount of Ang-1 is retained in the ECM after the adhesion of HUVECs compared with that retained in the ECM
incubated with SFM alone (data not shown). Thus, HUVECs promote the
release of Ang-1 from the ECM.
The Domain That Mediates the ECM Binding of Ang-1 Is Mapped to Its
Linker Peptide Region--
To determine which domain of Ang-1 mediates
its ECM association, several expression constructs were made as
indicated in Fig. 9A. They
were the coiled-coil domain, the coiled-coil plus linker peptide
region, and the FHD of Ang-1. The cDNA sequence encoding the signal
peptide of Ang-1 was constructed into the N terminus of the above
cDNA fragments so those fragments can be secreted properly (Fig.
9A). The full-length Ang-1 and -2 and the fragments of
Ang-1, which contain C-terminal v5 epitope tags, were used to transfect
COS-7 cells. 72 h after the transient transfection, the cell
culture supernatants and the ECM materials derived from the transfected
cells were either collected or extracted from the cell culture dishes
and subjected to Western blot analyses using anti-v5 antibody to
determine the distribution patterns of Ang-1, Ang-2, and the different
fragments of Ang-1. The results indicated that the linker peptide
region of Ang-1 between the coiled-coil and the fibrinogen-like
domains, which contains 27 amino acids, is mainly responsible for the
ECM association of Ang-1 (Fig. 9, B and C, lanes
5 and 6). A weak interaction between the coiled-coil
domain of Ang-1 and the ECM was also detected (Fig. 9, B and
C, lanes 7 and 8).
To investigate whether the ECM binding domain in the linker peptide
region of Ang-1 is also present in Ang-2, we first compared the
sequence homology between the domains of Ang-1 and Ang-2. We found that
the percentages of the identical amino acids are 59, 19, and 64%,
respectively, in the coiled-coil domain, the linker peptide region, and
the fibrinogen-like domains of Ang-1 and Ang-2. No significant homology
in the linker peptide region implied that the ECM binding domain in
this region of Ang-1 is likely absent in the Ang-2 molecule.
To confirm that the absence of the ECM binding of Ang-2 is due to the
lack of the ECM-binding site(s) in the molecule, but not the blockage
of the binding site(s) by possible steric restraints in the full-length
Ang-2 molecule, we made three additional deletions of Ang-2, which are
similar to those of Ang-1 deletions (Fig. 9A). They are the
coiled-coil domain, the coiled-coil plus the linker peptide region, and
the fibrinogen-like domain of Ang-2. All the deletion constructs
contain the N-terminal signal peptides of Ang-2 for their proper
expression and secretion and the C-terminal v5 epitope tags derived
from the pEF/6/v5-His expression vectors for their detection. After
confirming the authenticities of the expression constructs by DNA
sequencing, they were used to transfect COS-7 cells. After the
transient transfection, the distributions of Ang-2 fragments in the
cell culture media and the ECM fractions were examined by Western blot
analysis. Our results indicated that all three fragments of Ang-2, the
coiled-coil domain, the coiled-coil plus the linker peptide region, and
the fibrinogen-like domain, are secreted as the full-length Ang-2; none
of them associates with the ECM of the transfected COS-7 (data not
shown). This result indicated clearly that unlike Ang-1, Ang-2 does not
contain the ECM binding domain.
Studies have shown that angiopoietin-1 and -2 are unique
antagonists. Ang-2 blocks tyrosine phosphorylation of Tie-2 induced by
Ang-1 and disrupts angiogenesis in vivo (14). Our previous study (17) showed that Ang-1 and -2 play different roles in tumor angiogenesis. Ang-2 inhibits tumor angiogenesis by blocking recruitment of smooth muscle cells to the newly formed blood vessels and causing apoptosis to the endothelial cells (17).
Functional Significance of the ECM Association of Ang-1--
In
this study, we examined the potential relationship of Ang-1 and -2 with
the ECM, which is known to modulate activities of many growth factors,
including VEGF and bFGF (41, 42). Our results indicated that unlike
Ang-2, Ang-1 is secreted and incorporated into and sequestered by the
ECM. The different ECM association capacity enables those two
antagonists to regulate local and distant angiogenesis differently.
Ang-1 is most likely to regulate angiogenesis in the vicinity of its
secretion sites, whereas Ang-2 is likely to diffuse through
interstitial tissues and blood vessels to the distant organs.
The ability to distribute Ang-1 and -2 to local environment and/or
distant sites may play an important role in regulating tumor dormancy
in some situations (43-45). Both Ang-1 and -2 are expressed by
endothelial and tumor cells (17, 29). During the growth of primary
tumors, balanced expression of local Ang-1 and -2 and expression of
other pro-angiogenic factors, such as VEGF, may ensure tumor
angiogenesis and growth in the primary sites. When Tie-2 receptors on
local endothelial cells are saturated, it is likely that the excess
amount of Ang-1 produced by the primary tumors would be incorporated
into and sequestered by the surrounding ECM, whereas the
excessive amount of Ang-2 may diffuse to distant organs. The
micrometastases seeded and developed in the distant organs, such as
lung and liver, have microenvironments that contain pro- and
anti-angiogenic factors produced by local micrometastatic tumors and
surrounding host stromal tissues and are more or less similar to the
microenvironments of primary tumors. However, on top of the
microenvironments, Ang-2, not Ang-1, produced by the predominant
primary tumors, which is stable and has a long half-life (data
not shown), could travel to the distant organs and change the balance
in favor of inhibiting tumor angiogenesis, which may contribute to the
dormancy observed in some secondary tumors. More detailed studies are
required to confirm this hypothesis.
The ECM Association of Ang-1 Provides a Different Regulatory
Mechanism for the Availability of Ang-1--
Ang-1 is wildly expressed
both in embryo and in adult and exhibits a more uniform expression
pattern compared with that of Ang-2 (13, 14). Ang-2 has a wild
expression pattern in embryo. However, in adult tissues, its expression
is restricted in the tissues where vascular remodeling is ongoing, such
as ovary, uterus, and placenta (14). Ang-2 is up-regulated by hypoxia
and several different cytokines including VEGF and tumor necrosis
factor-
The association of Ang-1 with the ECM reported herein offered a
different type of regulation of the availability of Ang-1. Instead of
regulating its production, as Ang-2, free Ang-1 in microenvironment can
be regulated by changing its ECM association status. The factors that
promote or inhibit its ECM incorporation or releasing affect the
interaction between Ang-1 and Tie-2 receptor, sequential signal
transduction, and angiogenesis. This type of regulation provides a
quicker response to the changes in the microenvironment without the
requirement of mRNA and proteins syntheses. It is well established
that the activities of some growth factors are regulated in that way,
including TGF- The Biochemical Characters of the ECM Association of Ang-1--
An
extensive effort was made to try to identify the ECM component(s) that
mediate(s) the ECM association of Ang-1. Matrigel, fibrinogen, and to a
less extent vitronectin exhibited weak affinities to Ang-1, which is
much lower than that of unpurified ECM extracts derived from LLC cells
and cannot account entirely for the ECM association of Ang-1 (Fig. 4).
The immunocytochemistry studies indicated that the ECM distribution
pattern of Ang-1 is unique and unlike that of fibronectin, laminin, and
collagen types I and IV (Fig. 5).
To determine the effect of the ECM association on the function of
Ang-1, we assessed whether Tie-2-Fc fusion protein binds to the
ECM-associated Ang-1. The result indicated that the ECM-associated Ang-1 is not accessible for Tie-2, and the ECM serves as a storage and
sequestration site for Ang-1. The incubation of HUVECs on the ECM
containing Ang-1 induced the release of Ang-1 from the ECM and tyrosine
phosphorylation of Tie-2, which indicated that HUVECs cells are able to
respond to Ang-1 originally sequestered in the ECM (Fig. 8). This may
reflect the in vivo situations and indicate that activity of
Ang-1 is restricted in local environment and regulated more tightly
than that of Ang-2.
The Domain of Ang-1 That Mediates Its ECM Association Is the Linker
Peptide Region--
The deletion analysis indicated that the linker
peptide region of Ang-1 between the coiled-coil domain and the FHD
mediates the ECM association of Ang-1 (Fig. 9). The identification of
the ECM association domain of Ang-1 will allow us to study the
importance of the ECM association on Ang-1 function in the future.
In summary, we have discovered that Ang-1 is incorporated into the ECM
after its secretion, whereas Ang-2 is secreted and does not associate
with the ECM. The domain that mediates the ECM association of Ang-1 is
mapped to the linker peptide region between the coiled-coil and the
fibrinogen homology domains. The different ECM association capacity of
Ang-1 and Ang-2 offers a possible mechanism for the distinct
regulations of local and distant tumor angiogenesis by two antagonistic factors.
*
This work was supported by a Start-up fund from University
of Pennsylvania, School of Veterinary Medicine.The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
Published, JBC Papers in Press, July 10, 2001, DOI 10.1074/jbc.M103661200
The abbreviations used are:
ECM, extracellular
matrix;
Ang-1, angiopoietin-1;
Ang-2, angiopoietin-2;
TA3, TA3 murine
mammary carcinoma;
LLC, Lewis lung carcinoma;
HUVEC, human umbilical
vein endothelial cell;
FHD, fibrinogen homology domain;
PBS, phosphate-buffered saline solution;
BSA, bovine serum albumin;
PCR, polymerase chain reaction;
RT-PCR, reverse transcriptase-polymerase
chain reaction;
FITC, fluorescein isothiocyanate;
DOC, deoxycholate;
SFM, serum-free medium;
PMA, phorbol 12-myristate 13-acetate;
bFGF, basic fibroblast growth factor;
TGF-
Angiopoietin-1, Unlike Angiopoietin-2, Is
Incorporated into the Extracellular Matrix via Its Linker Peptide
Region*
ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
-mercaptoethanol. Western blot analysis was performed as
described (17).
20 °C for 15 min. Antibodies against v5 epitope, fibronectin,
laminin, type I or IV collagens were used to detected Ang-1v5, Ang-2v5,
fibronectin, laminin, types I or IV collagen, respectively, in those
fixed cells.
RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
View larger version (47K):
[in a new window]
Fig. 1.
Angiopoietin-1 is incorporated into the
extracellular matrix. The expression of Ang-1 and Ang-2 in the
serum-free conditioned media (A) and in the ECM
(B) by the transfected LLC and TA3 cells were determined by
Western blot analyses using anti-v5 antibody (Invitrogen). The Western
blot analyses were performed under non-reducing conditions using the
concentrated serum-free media (A) or the ECM extracts
(B) derived from two independent isolates of the transfected
LLC cells expressing Ang-1v5 (A, a, lanes 1 and 2, and
B, lanes 1 and 2) or Ang-2v5 (A, a,
lanes 3 and 4, B, and lanes 3 and
4); two independent isolates of the transfected TA3 cells
expressing Ang-1v5 (A, b, lanes 1 and 2, B, lanes
6 and 7) or Ang-2v5 (A, b, lanes 3 and
4, B, lanes 8 and 9), and the tumor cells
transfected with the expression vector only (LLC, carcinoma cells,
A, a, lane 5, B, lane 5; and TA3 cells, A, b, lane 5, B, lane 10). C, the same protein samples in the
B were subjected to -mercaptoethanol (5%) treatment.
Molecular mass markers are as indicated. kd indicates
kilodalton.
-mercaptoethanol. After boiling the protein samples in 1×
Laemmli SDS sample buffer containing 5% -mercaptoethanol, the
aggregated Ang-1 and Ang-2 are dissociated into monomers (data not
shown). It was noted that the amount of the secreted Ang-1 is often
lower than that of Ang-2 derived from LLC and TA3 transfectants (Fig.
1A, a and b). To examine the possibility that the
lack of Ang-1 secretion is due to incorporation of Ang-1 into the ECM of the tumor cells, the transfectants expressing Ang-1 and Ang-2 were
grown until confluence, and the cells were then detached from the
culture dishes by the EDTA treatment, which is a standard procedure to
release cultured cells and leave the ECM components behind on the
culture dishes (38). The remaining ECM components were then extracted
from the culture dishes by 1× SDS Laemmli buffer with or without
-mercaptoethanol. Western blot analyses were performed using anti-v5
monoclonal antibody to detect v5-tagged Ang-1 or Ang-2. The results
indicated that Ang-1 is present in the ECM fraction of the transfected
tumor cells, whereas Ang-2 is absent (Fig. 1B). The
ECM-associated Ang-1 is highly aggregated to form oligomers, and no
monomer is detected (Fig. 1B, lanes 1 and 2, and
6 and 7). The aggregated Ang-1 oligomers were
dissociated into the monomers by the treatment with -mercaptoethanol
(Fig. 1C), which indicates the role of the cysteine residues
in the aggregations. Thus, we have established that Ang-1, but not
Ang-2, is incorporated into the ECM of those carcinoma cells.
View larger version (63K):
[in a new window]
Fig. 2.
The ECM association of angiopoietin-1.
The Ang-1-containing ECM was extracted at room temperature for 10 min
with 0.15, 0.5, and 1 M NaCl and 0.5 and 1% DOC,
respectively. The insoluble ECM components were then extracted with 1×
SDS Laemmli buffer and subjected to Western blot analysis using anti-v5
antibody (lanes 2-6, respectively). The proteins in the
lane 1 were derived from the ECM without prior extraction
with any reagents. Lanes 7 and 8 are the ECM
extracts derived from LLC cells expressing Ang-2 or transfected with
the expression vector alone, respectively. The molecular mass
markers are as indicated.
View larger version (27K):
[in a new window]
Fig. 3.
Ang-1 does not bind to heparin.
A, the purified v5-tagged Ang-1 proteins (500 ng/ml) were
loaded onto a heparin-Sepharose CL-6B affinity column. The flow-through
was collected, and the column was washed with 0.15, 0.3, 0.5, and 1 M sodium chloride (NaCl, A, lanes 3-6), and the
eluted proteins were collected. All the collected fractions were
subjected to Western blot analysis using anti-v5 antibody. The results
indicated that Ang-1v5 does not bind to heparin-Sepharose, and all the
v5-tagged Ang-1 was in the flow-through fraction (A, lane
2). Lane 1 in A represents the starting
materials. B, the ECM proteins were extracted, and Western
blot analysis was performed after the incubation of the confluent LLC
cells expressing Ang-1 with serum-free cell culture media
containing heparin (200 µg/ml, B, lane 2), chondroitin
sulfate (200 µg/ml, B, lane 3), or SFM alone (B,
lane 1) for 12 h. Molecular mass markers are as
indicated.
View larger version (35K):
[in a new window]
Fig. 4.
Binding of Ang-1 to the different ECM
components. The binding affinities of Ang-1 to several different
ECM components were determined in the solid phase binding assays. Ang-1
exhibited no affinity to fibronectin, laminin, collagen types I and IV
(columns 1-4, respectively), heparin, chondroitin sulfate,
and hyaluronic acid (columns 10 and 12,
respectively). The weak affinities of Ang-1 to vitronectin and
fibrinogen were observed (columns 5 and 6). Ang-1
displayed a moderate affinity to Matrigel (column 7) and a
strong affinity to the 2 M urea ECM extracts derived from
LLC carcinoma cells (column 8). All the experiments were
performed in triplicate.
View larger version (59K):
[in a new window]
Fig. 5.
Ang-1 displayed a distinct ECM
distribution pattern. The distribution of Ang-1 in LLC cells
expressing Ang-1v5 was investigated by immunocytochemistry using
anti-v5 antibody (A, a). The distribution was compared with
that of fibronectin (B, a), laminin (B, b),
collagen type I (B, c) and type IV (B, d), and
Ang-2 in LLC cells expressing Ang-2v5 (A, c). The cell-free
ECM derived from the LLC carcinoma cells expressing Ang-1 (A,
b) or Ang-2 (A, d) was also analyzed by
immunocytochemistry using anti-v5 antibody. LLC carcinoma cells stained
with the FITC-conjugated rabbit anti-mouse secondary antibody only are
shown in B, e. Bar, 40 µm.
1 (TGF-1) promotes slightly
the incorporation of Ang-1 into the ECM, which may reflect the positive
effect of TGF-1 on synthesis of the ECM components (39) (Fig. 6,
lane 3).
View larger version (70K):
[in a new window]
Fig. 6.
The ECM-associated Ang-1 is released in the
response to PMA stimulation. Several factors were tested for their
abilities to release the ECM-sequestered Ang-1. After the incubation of
the LLC carcinoma cells expressing Ang-1 with SFM alone (lane
1) or SFM containing PMA (0.5 µg/ml, lane 2),
TGF- 1 (0.5 ng/ml, lane 3), bFGF (1 ng/ml, lane
4), epidermal growth factor (10 ng/ml, lane 5),
heparin-binding epidermal growth factor (10 ng/ml, lane 6),
and TGF-2 (0.5 ng/ml, lane 7), respectively, for 14 h, the cells were released from the cell culture dishes by the
treatment of EDTA, and the remaining ECM components were extracted with
1× SDS Laemmli buffer and subjected to the Western blot analysis using
anti-v5 antibody. Molecular mass markers are as indicated.
View larger version (80K):
[in a new window]
Fig. 7.
The ECM-associated Ang-1 exhibited no
affinity to Tie-2-Fc fusion protein. A, Tie-2-Fc fusion
protein is capable of precipitating Ang-1 proteins from the protein
extracts, which were derived from the 2 M urea extraction
of the ECM of LLC carcinoma cells expressing Ang-1 (A, lane
2). The starting materials for the immunoprecipitation were loaded
in lane 1 (A), and the control for the
immunoprecipitation contains everything except Tie-2-Fc fusion proteins
(A, lane 3). Molecular mass markers are as indicated.
B, the LLC cells expressing Ang-1 or the ECM derived from
LLC carcinoma cells expressing Ang-1 were incubated with Tie-2-Fc
fusion proteins (B, c and d) or
anti-v5 antibody (B, a and b), and
FITC-conjugated rabbit anti-human Fc or anti-mouse secondary antibodies
were used, respectively. Bar, 30 µm.
View larger version (17K):
[in a new window]
Fig. 8.
Tie-2 receptors on HUVECs is phosphorylated
upon the adherence of HUVECs to the Ang-1-containing ECM. The
serum-starved HUVECs were lifted by the treatment of EDTA/Hanks'
balanced salt solution, and 1 × 106 of the HUVECs
were seeded into the plastic dish containing SFM with (200 ng/ml,
lane 3) or without soluble Ang-1 (lane 1) or the
dishes coated with the ECM derived from LLC cells expressing Ang-2
(lane 2) or Ang-1 in the presence (lane 5, 2 µg/ml) or absence of Tie-2-Fc fusion proteins (lane 4) for
30 min. The HUVECs were lysed, and 50 µg of proteins from each lysate
were subjected to Western blot analysis using anti-Tie-2 antibody
(Santa Cruz Biotechnology, B). The rest of the proteins were
used in the immunoprecipitation using anti-Tie-2 antibody (Santa
Cruz Biotechnology). The immunoprecipitated proteins were subjected to
Western blot analysis using anti-phosphotyrosine (A).
Molecular mass markers are as indicated.
View larger version (41K):
[in a new window]
Fig. 9.
The linker peptide region of Ang-1 is
responsible for its ECM association. Several cDNA expression
constructs were made including the full-length of Ang-1 or Ang-2, the
coiled-coil region, the coiled-coil plus the linker peptide region, the
fibrinogen-like region of Ang-1 (A). All the Ang-1 fragments
contain N-terminal signal peptides of Ang-1 and the C-terminal v5
epitope tags. These expression constructs were used to transfect
COS-7 cells transiently. 72 h after the transfection, the cell
culture supernatants (B) and the ECM proteins (C)
were collected or extracted and subjected to Western blot analyses to
determine the distribution of Ang-1, Ang-2, and the Ang-1 fragments
using anti-v5 antibody. Lanes 1 and 2,
full-length Ang-1; lanes 3 and 4, full-length
Ang-2; lanes 5 and 6, the coiled-coil domain plus
the linker peptide region of Ang-1; lanes 7 and
8, the coiled and coil region of Ang-1; lanes 9 and 10, the fibrinogen homology domain of Ang-1; lane
11, COS-7 cells transfected with the expression vector
alone. Molecular mass markers are as indicated.
DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
(31-33), whereas the expression of Ang-1 is not affected.
(39).
FOOTNOTES
To whom correspondence should be addressed: 372E, 3800 Spruce St.,
Dept. of Pathobiology, School of Veterinary Medicine, University of
Pennsylvania, Philadelphia, PA 19104. Tel.: 215-898-2967; Fax: 215-898-0719; E-mail: qyu@vet.upenn.edu.
ABBREVIATIONS
, transforming growth
factor-;
VEGF, vascular endothelial growth factor.
REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
1.
Folkman, J.
(1971)
N. Engl. J. Med.
285,
1182-1186
2.
Kim, K. J.,
Li, B.,
Winer, J.,
Armanini, M.,
Gillett, N.,
Philips, H. S.,
and Ferrara, N.
(1993)
Nature
362,
841-844
3.
Millauer, B.,
Shawver, L. K.,
Plate, K. H.,
Risau, W.,
and Ullrich, A.
(1994)
Nature
367,
576-579
4.
Hanahan, D.,
and Folkman, J.
(1996)
Cell
86,
353-364
5.
Risau, W.
(1997)
Nature
386,
671-674
6.
Ingber, D. E.,
and Folkman, J.
(1989)
Cell
58,
803-805
7.
Dumont, D. J.,
Grawohl, G.,
Fong, G. H.,
Puri, M. C.,
Gertsenstein, M.,
Auerbach, A.,
and Breitman, M. L.
(1994)
Genes Dev.
8,
1897-1909
8.
Mustonen, T.,
and Alitalo, K.
(1999)
J. Cell Biol.
129,
895-898
9.
Fong, G. H.,
Rossant, J.,
Gertenstein, M.,
and Breitman, M. L.
(1995)
Nature
376,
66-70
10.
Gale, N. W.,
and Yancopoulos, G. D.
(1999)
Genes Dev.
13,
1055-1066
11.
Yancopoulos, G. D.,
Davis, S.,
Gale, N. W.,
Rudge, J. S.,
Wiegand, S. J.,
and Holash, J.
(2000)
Nature
407,
242-248
12.
Hanahan, D.
(1997)
Science
277,
48-50
13.
Davis, S.,
Aldrich, T. H.,
Jones, P. F.,
Acheson, A.,
Compton, D. L.,
Jain, V.,
Ryan, T. E.,
Bruno, J.,
Radziejewski, C.,
Maisonpierre, P.,
and Yancopoulos, G. D.
(1996)
Cell
87,
1161-1169
14.
Maisonpierre, P. C.,
Suri, C.,
Jones, P. F.,
Bartunkova, S.,
Wiegand, S. J.,
Radziejewski, C.,
Compton, D.,
McClain, J.,
Aldrich, T. H.,
Papadopoulos, N.,
Daly, T. J.,
Davis, S.,
Sato, T. N.,
and Yancopoulos, G. D.
(1997)
Science
277,
55-60
15.
Procopio, W. N.,
Pelavin, P. I.,
Lee, W. M. F.,
and Yeilding, N. M.
(1999)
J. Biol. Chem.
274,
30196-30201
16.
Valenzuela, D. M.,
Griffiths, J. A.,
Rojas, J.,
Aldrich, T. H.,
Jones, P. F.,
Zhou, H.,
McClain, J.,
Copeland, N. G.,
Gilbert, D. J.,
Jenkins, N. A.,
Huang, T.,
Papadopoulos, N.,
Maisonpierre, P. C.,
Davis, S.,
and Yancopoulos, G. D.
(1999)
Proc. Natl. Acad. Sci. U. S. A.
96,
1904-1909
17.
Yu, Q.,
and Stamenkovic, I.
(2001)
Am. J. Pathol.
158,
563-570
18.
Suri, C.,
Jones, P. F.,
Patan, S.,
Bartunkova, S.,
Maisonpierre, P. C.,
Davis, S.,
Sato, T. N.,
and Yancopoulos, G. D.
(1996)
Cell
87,
1171-1180
19.
Sato, T. N.,
Tozawa, Y.,
Deutsch, U.,
Wolburg-Buchholz, K.,
Fujiwara, Y.,
Gendron-Magurire, M.,
Gridley, T.,
Wolbrug, H.,
Risau, W.,
and Qin, Y.
(1995)
Nature
376,
70-74
20.
Suri, C.,
McClain, J.,
Thurston, G.,
McDonald, D. M.,
Zhou, H.,
Oldmixon, E. H.,
Sato, T. N.,
and Yancopoulos, G. D.
(1998)
Science
282,
468-471
21.
Thurston, G.,
Suri, C.,
Smith, K.,
McClain, J.,
Sato, T. N.,
Yancopoulos, G. D.,
and McDonald, D. M.
(1999)
Science
286,
2511-2514
22.
Thurston, G.,
Rudge, J. S.,
Ioffe, E.,
Zhou, H.,
Ross, L.,
Croll, S. D.,
Glazer, N.,
Holash, J.,
McDonald, D. M.,
and Yancopoulos, G. D.
(2000)
Nat. Med.
6,
460-463
23.
Witzenbichler, B.,
Maisonpierre, P. C.,
Jones, P.,
Yancopoulos, G. D.,
and Isner, J. M.
(1998)
J. Biol. Chem.
273,
18514-18521
24.
Koblizek, T. L.,
Weiss, C.,
Yancopoulos, G. D.,
Deutsch, U.,
and Risau, W.
(1998)
Curr. Biol.
8,
529-532
25.
Hayes, A. J.,
Huang, W. Q.,
Mallah, J.,
Yang, D.,
Lippman, M. E.,
and Li, L. Y.
(1999)
Microvasc. Res.
58,
224-237
26.
Papapetropoulos, A.,
Fulton, D.,
Mahboubi, K.,
Kalb, R. G.,
O'Connor, D. S.,
Li, F.,
Altieri, D. C.,
and Sessa, W. C.
(2000)
J. Biol. Chem.
275,
9102-9105
27.
Kim, I.,
Kim, H. G.,
So, J. N.,
Kim, J. H.,
Kwak, H. J.,
and Koh, G. Y.
(2000)
Circ. Res.
86,
24-29
28.
Kwak, H. J.,
Lee, S. J.,
Lee, Y. H.,
Ryu, C. H.,
Koh, K. N.,
Choi, H. Y.,
and Koh, G. Y.
(2000)
Circulation
101,
2317-2324
29.
Stratmann, A.,
Risau, W.,
and Plate, K. H.
(1998)
Am. J. Pathol.
153,
1459-1466
30.
Holash, J.,
Maisonpierre, P. C.,
Compton, D.,
Boland, P.,
Alexander, C. R.,
Zagzag, D.,
Yancopoulos, G. D.,
and Weigand, S. J.
(1999)
Science
284,
1994-1998
31.
Oh, H.,
Takagi, H.,
Suzuma, K.,
Otani, A.,
Matsumura, M.,
and Honda, Y.
(1999)
J. Biol. Chem.
274,
15732-15739
32.
Mandriota, S. J.,
and Pepper, M. S.
(1998)
Circ. Res.
83,
852-859
33.
Kim, I.,
Kim, J. H.,
Ryu, Y. S.,
Liu, M.,
and Koh, G. Y.
(2000)
Biochem. Biophy. Res. Commun.
269,
361-365
34.
Asahara, T.,
Chen, D.,
Takahashi, T.,
Fujikawa, K.,
Kearney, M.,
Magner, M.,
Yancopoulos, G. D.,
and Isner, J. M.
(1998)
Circ. Res.
83,
233-240
35.
Holash, J.,
Wiegand, S. J.,
and Yancopoulos, G. D.
(1999)
Oncogene
18,
5356-5362
36.
Lauren, J.,
Gunji, Y.,
and Alitalo, K.
(1998)
Am. J. Pathol.
153,
1333-1339
37.
Yeo, T. K.,
Nagy, J. A.,
Yeo, K. T.,
Dvorak, H. F.,
and Toole, B. P.
(1996)
Am. J. Pathol.
148,
1733-1740
38.
Kuno, K.,
and Matsushima, K.
(1998)
J. Biol. Chem.
273,
13912-13917
39.
Massague, J.
(1990)
Annu. Rev. Cell Biol.
6,
597-641
40.
Yu, Q.,
and Stamenkovic, I.
(1999)
Genes Dev.
13,
35-48
41.
Ortega, N.,
L'Faqihi, F. E.,
and Plouet, J.
(1998)
Biol. Cell
90,
381-390
42.
Klagsbrun, M.
(1992)
Semin. Cancer Biol.
3,
81-87
43.
Fidler, I. J.,
and Ellis, L. M.
(1994)
Cell
79,
185-188
44.
O'Reilly, M. S.,
Holmgren, L.,
Shing, Y.,
Chen, C.,
Rosenthal, R. A.,
Moses, M.,
Lane, W. S.,
Cao, Y.,
Sage, E. H.,
and Folkman, J.
(1994)
Cell
79,
315-328
45.
Holmgren, L.,
O'Reilly, M. S.,
and Folkman, J.
(1995)
Nat. Med.
1,
149-153
Copyright © 2001 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. M. Dallabrida, N. S. Ismail, E. A. Pravda, E. M. Parodi, R. Dickie, E. M. Durand, J. Lai, F. Cassiola, R. A. Rogers, and M. A. Rupnick Integrin binding angiopoietin-1 monomers reduce cardiac hypertrophy FASEB J, August 1, 2008; 22(8): 3010 - 3023. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
O.-H. Lee, J. Xu, J. Fueyo, G. N. Fuller, K. D. Aldape, M. M. Alonso, Y. Piao, T.-J. Liu, F. F. Lang, B. N. Bekele, et al. Expression of the Receptor Tyrosine Kinase Tie2 in Neoplastic Glial Cells Is Associated with Integrin {beta}1-Dependent Adhesion to the Extracellular Matrix Mol. Cancer Res., December 1, 2006; 4(12): 915 - 926. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cazes, A. Galaup, C. Chomel, M. Bignon, N. Brechot, S. Le Jan, H. Weber, P. Corvol, L. Muller, S. Germain, et al. Extracellular Matrix-Bound Angiopoietin-Like 4 Inhibits Endothelial Cell Adhesion, Migration, and Sprouting and Alters Actin Cytoskeleton Circ. Res., November 24, 2006; 99(11): 1207 - 1215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. M. Kim, K. E. Kim, G. Y. Koh, Y.-S. Ho, and K.-J. Lee Hydrogen peroxide produced by angiopoietin-1 mediates angiogenesis. Cancer Res., June 15, 2006; 66(12): 6167 - 6174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Ramsden, M. A. Buchanan, S. Egginton, J. C. Watkinson, V. Mautner, and M. C. Eggo Complete Inhibition of Goiter in Mice Requires Combined Gene Therapy Modification of Angiopoietin, Vascular Endothelial Growth Factor, and Fibroblast Growth Factor Signaling Endocrinology, July 1, 2005; 146(7): 2895 - 2902. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Tammela, A. Saaristo, M. Lohela, T. Morisada, J. Tornberg, C. Norrmen, Y. Oike, K. Pajusola, G. Thurston, T. Suda, et al. Angiopoietin-1 promotes lymphatic sprouting and hyperplasia Blood, June 15, 2005; 105(12): 4642 - 4648. [Abstract] [Full Text] [PDF]
|
|
