| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
* Hematology-Oncology Division and
Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA;
Department of Physiology, Center for Vascular Biology, University of Connecticut School of Medicine, Farmington, Connecticut, USA;
§ New York Medical College, Valhalla, New York, USA;
¶ D.E. Shaw & Co., Inc., New York, New York and Attenuon L.L.C, San Diego, California, USA;
Department of Chemotherapeutics, Pharmaceutical Product Division, Abbott Laboratories, Abbott Park, Illinois, USA; and
Center for Neurovirology and Neurooncology, Allegheny University of the Health Sciences, Philadelphia, Pennsylvania, USA
1Correspondence: Hematology-Oncology Division, BRB 3, Case Western Reserve University, School of Medicine, 10900 Euclid Ave., Cleveland, OH 44106-4937, USA. E-mail: kxm71{at}po.cwru.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10 nM HKa. This activity was Zn2+
dependent and not shared by either single-chain high molecular weight
kininogen (HK) or low molecular weight kininogen. HKa selectively
inhibited the proliferation of human umbilical vein and dermal
microvascular endothelial cells, but did not affect that of umbilical
vein or human aortic smooth muscle cells, trophoblasts,
fibroblasts, or carcinoma cells. Inhibition of endothelial
proliferation by HKa was associated with endothelial cell apoptosis and
unaffected by antibodies that block the binding of HK or HKa to any of
their known endothelial receptors. Recombinant HK domain 5 displayed
activity similar to that of HKa. In vivo, HKa inhibited
neovascularization of subcutaneously implanted Matrigel plugs, as well
as rat corneal angiogenesis. These results demonstrate that HKa is a
novel inhibitor of angiogenesis, whose activity is dependent on the
unique conformation of the two-chain molecule.—Zhang, J.-C., Claffey,
K., Sakthivel, R., Darzynkiewicz, Z., Shaw, D. E., Leal, J., Wang,
Y.-C., Lu, F. M., McCrae, K. R. Two-chain high molecular
weight kininogen induces endothelial cell apoptosis and inhibits
angiogenesis: partial activity within domain 5.
Key Words: angiogenesis • coagulation • neovascularization • tumor • endothelium
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Of
particular interest is the observation that the growth of tumors is
angiogenesis dependent (1
, 2)
. Primary or metastatic tumor
foci are unable to achieve a size of more than several millimeters in
the absence of neovascularization (1
, 3)
. Serial
evaluation of transgenic mice predisposed to develop neoplasms has
demonstrated that neovascularization of premalignant lesions precedes
their evolution into tumors (2)
and that inhibition of
angiogenesis delays the growth of such lesions (4)
. In
humans, increased microvessel density is associated with a poor
prognosis in a number of tumor types (5
6
7)
.
An emerging paradigm is that proteolytic fragments or
conformationally altered forms of specific plasma or extracellular
matrix proteins inhibit angiogenesis (2
, 8)
. Several of
these are members of the coagulation and/or fibrinolytic systems.
Examples include angiostatin, which contains kringles 1–4 of
plasminogen (9)
, prothrombin kringle 2 (10)
,
plasminogen activator inhibitor type 1 (PAI-1) (11)
, and
antithrombin (12)
. Thrombospondin, an abundant component
of platelet alpha granules, as well as peptides derived from its
procollagen domain and properdin-like type 1 repeats, also inhibit
angiogenesis (13)
. A partial list of other antiangiogenic
polypeptides includes endostatin, a 20 kDa carboxyl-terminal fragment
of collagen XVIII (14)
, the related collagen XV fragment,
restin (15)
, the hemopexin domain of matrix
metalloprotease 2 (16)
, an amino-terminal fragment of
prolactin (17)
, a 29 kDa fibronectin fragment
(18)
, a 24 kDa fragment of the 
2 chain of type IV
collagen (19)
, and vasostatin, an amino-terminal fragment
of calreticulin (20)
.
High molecular weight kininogen (HK) is a 120 kDa single-chain
glycoprotein that plays a central role in contact activation
(21)
. HK is comprised of heavy and light chains that
contain domains 1 through 3, and 5 and 6, respectively, and are linked
by domain 4, which contains the vasoactive nonapeptide bradykinin
(22)
(Fig. 1
). Cleavage of HK between
Lys362-Arg363 and
Arg371-Ser372 by kallikrein
(23)
, an event that may occur on the endothelial surface
(24)
, results in the release of bradykinin and generation
of two-chain high molecular weight kininogen (HKa). HKa contains a 62
kDa heavy chain and 56 kDa light chain linked by a single disulfide
bond (25)
. Kallikrein may further cleave HKa between
Arg419-Lys420, reducing the
size of the light chain to 46 kDa (Fig. 1)
(25)
.
Conversion of HK to HKa is accompanied by a dramatic conformational
rearrangement characterized by reorientation of HKa domains and
increased exposure of domain 5 (21)
. Another member of the
kininogen family, low molecular weight kininogen (LK), is comprised of
a heavy chain identical to that of HKa, bradykinin, and a unique 4 kDa
light chain (D5L) (Fig. 1)
(26)
. In
this report we demonstrate that HKa, but not HK or LK, induces
endothelial apoptosis and inhibits angiogenesis in vivo.
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 25)
, as well as the presence of predominant 62 kDa
and 46 kDa bands upon analysis of HKa by 10% SDS-PAGE under
reducing conditions. LK was from American Research Products (Belmont,
Mass.). Bradykinin was from Peninsula (Belmont, Calif.). All HKa
preparations contained <0.05 ng/ml of endotoxin, determined using the
E-Toxate (Limulus amoebocyte lysate) assay (Sigma).
Cell culture and endothelial cell proliferation assays
Human umbilical vein endothelial cells (HUVEC), smooth muscle
cells, and fibroblasts, as well as human aortic smooth muscle cells
were cultured and characterized as described previously (27
, 28)
. Human dermal microvascular endothelial cells (HDMVEC) were
obtained from Technoclone (Vienna, Austria). Human trophoblast cells
(ED 27) (29)
were provided by Dr. Douglas Kniss,
Department of Obstetrics and Gynecology, Ohio State University School
of Medicine. HEK 293 cells, MDA MB-231 breast carcinoma cells and U937
cells were obtained from the ATCC (Rockville, Md.).
To assess the effect of HKa on endothelial cell proliferation, cells
were suspended at a concentration of 3 x
104 cells/ml in medium 199 (M199) containing 2%
fetal calf serum (FCS). One hundred microliters of this suspension was
plated in individual wells of a 96-well microplate precoated with 1%
gelatin (or, in selected experiments, other extracellular matrix
proteins). After incubation for 4 h to allow cells to adhere and
spread, medium was replaced with fresh M199 containing 2% FCS (or, in
selected experiments, 10% human AB serum), 10 ng/ml bFGF (or another
growth factor, as specified), 10 µM ZnCl2, and,
in experimental wells, HKa. Cells were then incubated for 48 h at
37°C, at which time the relative numbers of cells in each well were
determined using the Cell Titer AQueous cell
proliferation assay (Promega, Madison, Wis.). The percent inhibition of
proliferation by HKa was determined using the formula:
 |
To determine whether HK was converted to HKa during the course of
endothelial cell proliferation assays, HK was labeled with NHS-LC
biotin, as described (30)
. Labeled HK was subjected to
10% SDS-PAGE under reducing conditions and transferred to a
polyvinylidene fluoride membrane. The immobilized protein was detected
by incubation of the membrane with a 1:500 dilution of
streptavidin-peroxidase, followed by exposure to Super Signal
chemiluminescence reagent. HK remained > 99% single chain after
labeling. The ability of biotinylated HK to inhibit endothelial cell
proliferation was then assessed, with the extent of conversion of the
biotin-HK to biotin-HKa during the assay determined simultaneously by
analysis of conditioned medium and cell extracts obtained 24 and
48 h after addition of biotin-HK to cells. Parallel studies in
which biotin-HK was incubated in fresh, cell-free medium were also
performed. Bands observed on the autoradiograms were quantitated using
NIH Image v 1.59. In complementary studies, we determined that both the
heavy and light chains of HKa were labeled with NHS-LC biotin, that the
antiproliferative activity of HKa was not affected by biotinylation,
and that HKa was not further degraded during the course of
proliferation assays.
In selected experiments, we assessed the ability of antibodies that
block the binding of HK or HKa to specific endothelial cell receptors
to directly inhibit endothelial cell proliferation or to inhibit the
antiproliferative activity of HKa. Three antibodies were used. An
affinity-purified rabbit polyclonal antibody that inhibits the binding
of HKa to a site within domains 2 + 3 of the urokinase receptor has
been described (28)
. A monoclonal antibody that blocks the
binding of HK to the receptor for the globular heads of C1q (gC1qR) was
a kind gift from Dr. Berhane Ghebrehewit, SUNY, Stony Brook
(31)
. A rabbit antiserum that blocks the binding of HK to
an additional endothelial cell HK receptor, cytokeratin 1, as well as a
15 amino acid peptide that corresponds to the HK binding region within
cytokeratin 1 and inhibits HK binding to this receptor
(IC50 10 µM), were kindly provided by Dr.
Alvin Schmaier, University of Michigan (32
, 33)
.
Detection of endothelial cell apoptosis
Endothelial cell apoptosis after exposure of cells to HKa was
documented using several assays. First, endothelial cells plated on
gelatin-coated glass coverslips were cultured in medium 199 containing
2% FCS in the absence or presence of 30 nM HKa for 6–24 h. Cells were
then fixed in phosphate-buffered saline (PBS) containing 1%
formaldehyde and stained with a solution containing 1 µg/ml of 4',
6'-diamidino-2-phenylindole dihydrochloride (DAPI) and 10 µg/ml of
sulforhodamine 101 (Molecular Probes, Eugene, Oreg.). Stained cells
were visualized by UV illumination using a Nikon Microphot FXA
microscope (objective 40x, Neofluor). To assess DNA fragmentation
after exposure of cells to HKa, cytoplasmic DNA was isolated from
control and HKa-exposed cells using a buffer containing 0.4% Triton
X-100, and analyzed by 0.8% agarose gel electrophoresis and ethidium
bromide staining. Apoptosis was also assessed by flow cytometric
analysis of control and HKa-exposed cells labeled with fluorescein-dUTP
using the TUNEL reaction (In Situ Cell Death Detection Kit, Boehringer
Mannheim, Indianapolis, Ind.).
Inhibition of angiogenesis by HKa
The effect of HKa on angiogenesis in vivo was
determined using the Matrigel plug and corneal micropocket angiogenesis
assays. In the former, athymic Ncr nude mice (7–8 wk, female) were
injected subcutaneously (s.c.) with 0.25 ml of chilled Matrigel
containing 400 ng of bFGF and 25 µg heparin, to which either 25 µl
of PBS (left flank injection) or an equal volume of PBS containing 10
µg HKa (right flank injection) had been added. The Matrigel
solidified after injection and remained as an intact, s.c. plug for the
duration of the experiment. Mice were killed after 4 days, and digital
images of intact control and experimental plugs were obtained prior to
en bloc excision of the plugs and overlying skin for
histological analysis (21)
.
The corneal micropocket angiogenesis assay was performed as described
previously (34
, 35)
. Pellets were prepared using 12%
hydron. Control pellets containing 50 ng of bFGF were implanted in the
left eyes of 10 Sprague-Dawley rats. Test pellets containing 50 ng of
bFGF and 12 ng of HKa were implanted in the right eyes of the same
animals. Each pellet was implanted in a 2 mm pocket prepared in the
cornea, 1 mm from the limbus. Corneal neovascularization was assessed 7
days after implantation of the pellets, at which time digital images of
each eye were obtained using a Nikon NS-1 slit lamp. The total area of
neovessels in each digital image was then determined using a Leica-Qwin
(Northvale, N.J.) image analysis system (36)
.
Preparation of recombinant HK domain 5 (HK-D5)
The exposure of HKa domain 5, which contains high-affinity
endothelial cell binding sites (37)
, is enhanced after
cleavage of HK to HKa (38)
. To determine whether domain 5
might mediate the effects of HKa on endothelial cells, a 301 nucleotide
cDNA encompassing nucleotides 173–472 of kininogen exon 10 and
encoding amino acids 412–503 of kininogen was generated by reverse
transcription-polymerase chain reaction (RT-PCR) of human liver RNA.
Amplification was performed in a 100 µl reaction mixture containing
2.5 U Pfu DNA polymerase (Stratagene, La Jolla, Calif.), 10
µl 10x Pfu buffer, 0.8 µl of 100 mM dNTPs, 5 µl first
strand reaction mix, and 100 ng each of sense and antisense primers.
The sense primer was 5'-CAGGGATCCAAAATGGACTGGGGCCATGAAAAA-3', and the
antisense primer was 5'-GGCGAATTCAGAAGAGCTTGCCAAATG-3'. The PCR product
was cloned into pGEX-6p-1 (Pharmacia Biotech, Piscataway, N.J.). This
polypeptide encompassed a previously defined high-affinity endothelial
cell binding region (amino acids 479–498) within HK domain 5
(30)
. The glutathione-S-transferase (GST) domain 5 fusion
protein was isolated using glutathione-agarose, and GST-free domain
5 was purified by a second passage of the PreScission protease
(Pharmacia-Biotech) -cleaved fusion protein over the same column.
Purified domain 5 migrated with the expected
Mr of 11.1 when analyzed by 16%
tricine SDS-PAGE and yielded a single peak when analyzed by
reverse-phase high performance liquid chromatography (HPLC).
Amino-terminal amino acid sequencing of recombinant domain 5 was
ambiguous beyond 9 amino acids. However, the first 6 amino acids
corresponded to vector-encoded sequence carboxyl-terminal to the
PreScission protease cleavage site, whereas the next three amino acids
were identical to the kininogen sequence beginning at
Asp412. Recombinant domain 5 preparations were
routinely passed through a Detoxigel column (Pierce) before study,
after which they contained < 0.05 ng/ml of endotoxin as
determined using the E-Toxate assay (Sigma).
Free GST was prepared from a control pGEX-6p-1 vector and used as a
negative control in studies where the effect of free domain 5 on
endothelial cell proliferation was assessed. The 39 kDa low density
lipoprotein receptor-related
protein/2-macroglobulin receptor
receptor-associated protein (RAP) was also prepared as a GST fusion
protein. GST-free RAP was isolated as described (39)
and
used as an additional control for studies utilizing HKa domain 5.
Synthetic peptides
Synthetic 16 amino acid peptides, each overlapping by 8 amino
acids, were prepared based on the amino acid sequence of HK domain 5
(amino acid residues 384–503 of HK). Peptides were synthesized on a
Rainin Symphony peptide synthesizer, using standard solid-phase methods
(40)
. FMOC amino acids were purchased from Perseptive
Biosystems (Foster City, Calif.). All peptides were purified using
reverse-phase HPLC and purity was analyzed by MALDI-TOF mass
spectrometry.
Statistics
Error bars in all figures represent the standard deviation of
quadruplicate points. Each experiment was performed a minimum of three
times. The significance of differences in the numbers of endothelial
cells remaining in each well at the end of proliferation assays, as
well as that between the total vessel area in rat corneas that received
hydron pellets containing bFGF or bFGF and HKa, was determined using
the Student’s two-tailed t test for paired samples.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Since ligand binding to uPAR has
been associated with activation of signaling events
(41
42
43)
and stimulation of cell proliferation
(44)
, we wished to determine the effect of HKa binding on
the proliferation of endothelial cells. Contrary to expectations, HKa
potently inhibited the proliferation of HUVEC and HDMVEC. This activity
was dependent on supplementation of the medium with
ZnCl2 (Fig. 2
), perhaps reflecting the requirement for Zn2+ for
specific binding of HKa to endothelial cells (45)
. Maximal
effects occurred at a Zn2+ concentration of 10
µM, demonstrating that HKa inhibits endothelial proliferation at
physiological (<15 µM) Zn2+ concentrations. In
the absence of HKa, Zn2+ did not affect
endothelial cell proliferation.
|
Inhibition of endothelial cell proliferation is mediated
selectively by HKa
HKa inhibited endothelial cell proliferation by 50% at a
concentration of 10 nM (Fig. 3
). In contrast, LK, bradykinin, or anti-bradykinin antibodies did not
affect endothelial cell proliferation. HK (single chain) inhibited
proliferation only modestly at concentrations approaching 100 nM (Fig. 3)
, with 50% inhibition evident at a concentration of 320 nM.
However, studies using biotinylated HK revealed that 2% of the
added HK was converted to HKa during the course of the proliferation
assays (not shown). Since biotinylation did not affect the
antiproliferative activity of HKa, it is likely that the newly
generated HKa contributed to the observed activity of HK. Hence, these
studies suggest a critical role for the unique conformation of the
two-chain HKa molecule in mediating the antiproliferative activity of
high molecular weight kininogen.
|
Characterization of the antiproliferative activity of HKa
The antiproliferative activity of HKa was apparent within 6 h
of its addition to cells and was maintained in the presence of 10%
human serum. Extensive dialysis (25,000
Mr cutoff) did not affect the activity
of HKa. HKa completely inhibited endothelial proliferation in response
to several growth factors, including bFGF, VEGF, HGF, and PDGF (not
shown). Since the mitogenic activity of each of these is mediated
through interactions with distinct receptors, we hypothesize that the
mechanism(s) by which HKa inhibits endothelial proliferation is
unlikely to depend entirely on inhibition of growth factor binding.
HKa inhibited the proliferation of human umbilical vein and
microvascular endothelial cells with similar potency, though it did not
affect the proliferation of several other cell types, including primary
cultures of human aortic smooth muscle and trophoblast cells
(Table 1
). Moreover, in studies in which endothelial cells, smooth muscle cells
and fibroblasts were isolated from the same umbilical cord, we
confirmed that the effects of HKa were endothelial cell specific (Table 1)
. The ability of HKa to inhibit endothelial cell proliferation was
inversely proportional to cell density, with no effect on confluent or
near-confluent endothelial cells.
|
In selected experiments, we assessed the ability of HKa to inhibit the proliferation of endothelial cells on different extracellular matrix (ECM) proteins. HKa (10 nM) potently inhibited the proliferation of HUVEC cultured on gelatin, laminin, and Matrigel though a slightly less potent inhibition, overcome by a higher concentration of HKa (50 nM), occurred when cells were cultured on fibronectin or vitronectin. Intermediate effects were observed when cells were cultured on fibrinogen, and cells cultured on collagen types I or IV were relatively resistant to the antiproliferative activity of HKa.
To define the endothelial cell receptors that mediate the effects of
HKa, we assessed whether its antiproliferative activity was inhibited
by antibodies that inhibit the binding of HK or HKa to specific
endothelial cell receptors (Table 2
). Neither an affinity-purified polyclonal antibody that blocks the
binding of HKa to endothelial cell urokinase receptors
(28)
, a monoclonal antibody that blocks the binding of HK
to endothelial cell gC1q receptors (31
, 46)
, nor a
polyclonal antibody that blocks HK binding to endothelial cell CK-1
(32
, 33)
affected the antiproliferative activity of HKa.
Moreover, a 15 amino acid peptide (PGG-15) from within the HK binding
region of cytokeratin 1, which inhibits the binding of HK to
cytokeratin-1 by 50% at a concentration of 9–10 µM
(33)
, did not affect the antiproliferative activity of HKa
even when used at concentrations exceeding 250 µM (not shown).
|
The antiproliferative activity of HKa is associated with
endothelial cell apoptosis
To determine whether inhibition of endothelial cell proliferation
by HKa was associated with endothelial cell apoptosis, we stained
control and HKa-exposed cells with DAPI, observing morphological
changes characteristic of apoptosis (nuclear condensation,
hyperchromaticity, and fragmentation) in 30–50% of HKa-exposed cells
within 6 h of HKa addition (Fig. 4
). These changes were apparent in all of the cells at later time points.
Consistent with these results, a specific ‘laddering’ pattern of DNA
fragmentation, characteristic of apoptosis, was apparent on
electrophoretic analysis of DNA from cells exposed to HKa
(47)
(Fig. 5
); flow cytometric analysis of control and HKa-exposed cells labeled
with fluorescein-dUTP using the TUNEL reaction revealed increased
labeling of HKa-exposed cells. HKa-exposed cells did not stain with
trypan blue, demonstrating that HKa did not cause cell lysis or
cytotoxicity.
|
|
HKa inhibits angiogenesis
The observations that HKa specifically induced apoptosis of
subconfluent, proliferative endothelial cells suggested that it might
inhibit angiogenesis. Therefore, we assessed the ability of HKa to
inhibit the neovascularization of s.c.-implanted Matrigel plugs
containing bFGF. As depicted in Fig. 6A
(left panels), Matrigel plugs that contained bFGF induced
exuberant vessel ingrowth within 4 days after implantation. In
contrast, little neovascularization occurred in plugs that contained
bFGF and HKa (Fig. 6A
, right panels). Histological analysis
of control and HKa-containing plugs confirmed that the vessel density
within the latter was markedly reduced (Fig. 6B
).
|
The antiangiogenic activity of HKa was also assessed by determining its
effect on rat corneal angiogenesis. In 10 control corneas,
bFGF-containing hydron pellets induced a robust angiogenic response
(Fig. 7A
, left panel). In comparison, the length and density of
neovessels were significantly reduced in 10 corneas in which the
implanted pellets contained bFGF and HKa (Fig. 7B
, right
panel). Computer analysis of digital images revealed that the mean
vessel area within corneas that received HKa-containing pellets (53,931
µm2) was reduced by 82% compared to those in
which pellets contained bFGF only (293,807 µm2)
(P<0.000005).
|
HKa domain 5 inhibits endothelial cell proliferation and induces
endothelial apoptosis
Previous studies have demonstrated the presence of endothelial
cell binding regions within domain 5 of kininogen (30
, 37)
. Since the exposure of domain 5 is enhanced on cleavage of
HK (38)
, we hypothesized that domain 5 might mediate the
antiangiogenic activity of HKa. To address this hypothesis, we
determined the effect of a recombinant domain 5 polypeptide
encompassing amino acids 412–503 of kininogen on endothelial
proliferation. Recombinant domain 5 potently inhibited proliferation
(IC50 40 nM) in a Zn2+
and concentration-dependent manner (Fig. 8A
, B
, respectively) and, like HKa, induced endothelial cell
apoptosis (not shown). Neither the isolated GST fusion partner from the
GST domain 5 polypeptide nor recombinant 39 kDa RAP affected
endothelial cell proliferation (Fig. 8A
). HKa domain 5 also
inhibited bFGF and VEGF-induced angiogenesis in the chick
chorioallantoic membrane (not shown).
|
To further define the active regions within HK domain 5, synthetic 16
amino acid peptides encompassing domain 5 (amino acids 384–503), each
overlapping by 8 amino acids, were prepared and tested for their
ability to inhibit endothelial cell proliferation (Table 3
). These studies revealed activity within peptides from the
carboxyl-terminal region of domain 5, with two peptides (H5–13 and
H5–14) that encompass the high-affinity endothelial cell binding
region within domain 5 (amino acids 479–498 of kininogen)
(30)
causing near-complete inhibition of proliferation
when used at a concentration of 50 µM. In subsequent studies, we
determined that these peptides inhibited endothelial cell proliferation
by 50% at concentrations of 8 µM and 14 µM, respectively.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
49
50)
, endostatin (14)
, and thrombospondin
1 (13)
. Hence, when considering the abundance of HK in
plasma (670 nM), it is likely that HKa may be a physiologically
important regulator of angiogenesis.
The mechanisms by which HKa causes endothelial cell apoptosis and
inhibits angiogenesis have not been defined. Previous reports have
demonstrated that HKa inhibits spreading of several cell types on
vitronectin (51)
and fibrinogen (52)
. It has
been proposed that these effects may result from the binding of HKa to
vitronectin, thereby inhibiting the interaction between the RGD site
within vitronectin and integrin
vß3 (51)
.
Based on this hypothesis, it is possible that HKa might induce
endothelial cell apoptosis by causing disadhesion of cells from the
extracellular matrix and disrupting integrin-mediated signaling
(anoikis) (53
, 54)
. For several reasons, however, we do
not believe that this mechanism is likely to be solely responsible for
the results obtained in this study. First, HKa concentrations
10-fold higher than those that induce apoptosis (10 nM) are required
to inhibit endothelial cell adhesion and spreading (51
;
J.-C. Zhang, and K. R. McCrae, unpublished results). Second, in
addition to vitronectin, HKa causes apoptosis of endothelial cells
cultured on several ECM substrates, including laminin, Matrigel,
fibronectin, and fibrinogen. Though the ability of
vß3 to bind several
ECM ligands (55
56
57)
makes it difficult to exclude the
possibility that the effects of HKa reflect disruption of
vß3-substratum
interactions, adhesion to some of the purified ECM components used in
these studies (fibrinogen, laminin) is mediated primarily by integrins
other than vß3.
Furthermore, neither fibrinogen (an
vß3 ligand) nor an
anti-ß3 integrin monoclonal antibody, 7E3,
inhibit the binding of HKa to endothelial cells (28)
.
Third, the ability of HKa to induce apoptosis is exquisitely specific
for proliferating endothelial cells, though HKa inhibits the spreading
of a number of cell types, including osteosarcoma and melanoma cells,
monocytes, and platelets (52)
. These results raise the
possibility that HKa might interact with a receptor selectively
expressed on proliferating endothelial cells, perhaps inducing
signaling responses that disrupt or circumvent common integrin
signaling pathways required for maintenance of cell viability
(53)
. However, our studies do not suggest a role for any
of the known endothelial cell HK or HKa receptors in mediating these
effects, since antibodies that inhibit the binding of HK or HKa to
these receptors do not affect the antiproliferative activity of HKa
(Table 2)
.
The effects of HKa demonstrate remarkable molecular specificity, since the two-chain molecule inhibited endothelial proliferation and induced apoptosis far more potently than HK or LK. Though our studies do not exclude the possibility that HK itself may inhibit endothelial cell proliferation, our data suggest that the small amount of HKa generated during the 48 h proliferation assay may be sufficient to account for the apparent activity of HK. Definitive resolution of this issue will require expression and functional testing of HK mutants resistant to kallikrein-mediated cleavage.
Our results support the paradigm in which conformational alteration of
specific proteins, often belonging to the coagulation cascade
(58)
, is thought to lead to exposure of antiangiogenic
neoepitopes (2
, 8)
. The observation that cleavage of HK to
HKa is accompanied by a conformational change characterized by
increased exposure of domain 5 (38)
suggests that the
endothelial cell binding regions within this domain (21
, 30)
may mediate the activity of HKa. This hypothesis is
supported by our studies with recombinant domain 5 and domain 5-derived
peptides (Table 3)
, with peptides most active in inhibition of
endothelial cell proliferation occurring within the previously
described high-affinity endothelial binding region of this domain
(30)
. We speculate that the conformational change that
occurs after HK cleavage allows this region to interact in a novel
manner with endothelial cells, perhaps by binding to a receptor that
does not recognize HK or does so with only low affinity. Indeed, like
thrombospondin 1, which binds to at least 10 distinct cell surface
molecules (59)
, HKa may have yet undefined endothelial
receptors that mediate these effects.
Several additional issues concerning the antiangiogenic properties of
HKa warrant discussion. First, the dependence of the antiangiogenic
activity of HKa on Zn2+ suggests potential
overlap between the mechanisms of HKa and endostatin. Endostatin exists
as a zinc-dependent dimer, which coordinates Zn2+
through interactions with three amino-terminal histidines and an
aspartic acid (60)
. The presence of
Zn2+ within endostatin has been documented by
physicochemical methods, and site-directed mutagenesis of the
endostatin Zn2+ coordination ligands has
confirmed their importance in its biological activity and/or stability
(14)
. HKa domain 5 contains at least two potential
Zn2+ coordination sites, and the binding of HK to
endothelial cells is Zn2+ dependent
(45)
. Taken together, these results suggest that HKa and
endostatin might share a common Zn2+-dependent
pharmacophore that mediates their antiangiogenic activity or stabilizes
the active conformation of the molecules.
Colman et al. have recently reported that domain 5 of kininogen
inhibits endothelial cell proliferation and migration, as well as
angiogenesis in the chick chorioallantoic membrane (61)
.
These authors reported that a recombinant GST domain 5 fusion protein
encompassing HK residues 420–513 inhibited endothelial cell
proliferation by 50% at a concentration of 50 nM, whereas a
synthetic peptide corresponding to amino acids 440–455 of HK domain 5
inhibited proliferation to a similar extent at a concentration of
110 nM (61)
. Our studies extend these observations, as
we have demonstrated that cleavage of naturally occurring single-chain
high molecular weight kininogen to its two-chain form yields a potent
antiangiogenic molecule, and that the activity of HKa, as well as that
of domain 5 (Figs. 2
and 8)
and the domain 5-derived peptides (not
shown), is Zn2+ dependent. We have also
demonstrated that HKa inhibits angiogenesis in mice. However, our
studies do not support a role for the interaction between HKa or HK
domain 5 and the urokinase receptor as contributing significantly to
the antiproliferative activity of HKa, as proposed by Colman et al.,
since an antibody that blocks the binding of HKa to this receptor did
not diminish the effects of HKa on endothelial cell proliferation.
Moreover, we observed minimal inhibition of endothelial cell
proliferation by the G440-H445 peptide, instead detecting much more
potent activity in two peptides (H5–13, H5–14) from the
carboxyl-terminal region of domain 5 previously reported to mediate
high-affinity binding of HK to endothelial cells (Table 3)
(30)
.
Several considerations suggest that HKa may contribute to the
physiological regulation of angiogenesis, particularly under
inflammatory conditions. HK is a substrate for other proteases in
addition to kallikrein, such as neutrophil elastase and mast cell
tryptase, which in combination efficiently release bradykinin from HK
(62)
. Plasmin promotes the proteolysis of HK through
direct cleavage between
Lys420-His421 (one amino
acid carboxyl-terminal to the kallikrein cleavage site) and
Arg371-Ser372
(63)
, as well as through activation of plasma kallikrein
(63)
. Though HK is the most abundant species of high
molecular weight kininogen in normal human plasma, HKa and other
kininogen degradation products predominate in pathological conditions
such as disseminated intravascular coagulation (64)
,
trauma (64
, 65)
, hereditary angioedema (66)
,
and thrombotic thrombocytopenic purpura (67)
. Contact
activation, accompanied by generation of HKa, also occurs in acute
myocardial syndromes (68)
and is stimulated by plasmin
during pharmacological thrombolysis (69)
. We are currently
investigating the possibility that HKa and/or other kininogen-derived
peptides circulate at increased levels in the plasma of patients with
malignancy.
In summary, our studies demonstrate that HKa induces endothelial apoptosis and inhibits angiogenesis. Further characterization of the mechanism(s) by which HKa exerts these effects should provide new information concerning the physiological role(s) of this intriguing protein as well as insight into the regulation of angiogenesis.
|
ACKNOWLEDGMENTS |
|---|
Received for publication December 8, 1999.
Revision received June 2, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2-macroglobulin receptor regulates cell surface plasminogen activator activity on human trophoblast cells. J. Biol. Chem. 273,32273-32280vß3: angiogenesis and apoptosis. Cell Adhesion Commun 3,367-374[Medline]
This article has been cited by other articles:
|
|
 |
|
  L. G. Coffman, D. Parsonage, R. D'Agostino Jr., F. M. Torti, and S. V. Torti Regulatory effects of ferritin on angiogenesis PNAS, January 13, 2009; 106(2): 570 - 575. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wang, W.-M. Yu, W. Zhang, K. R. McCrae, B. G. Neel, and C.-K. Qu Noonan Syndrome/Leukemia-associated Gain-of-function Mutations in SHP-2 Phosphatase (PTPN11) Enhance Cell Migration and Angiogenesis J. Biol. Chem., January 9, 2009; 284(2): 913 - 920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Merkulov, W.-M. Zhang, A. A. Komar, A. H. Schmaier, E. Barnes, Y. Zhou, X. Lu, T. Iwaki, F. J. Castellino, G. Luo, et al. Deletion of murine kininogen gene 1 (mKng1) causes loss of plasma kininogen and delays thrombosis Blood, February 1, 2008; 111(3): 1274 - 1281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sakai, K. Balasubramanian, S. Maiti, J. B. Halder, and A. J. Schroit Plasmin-Cleaved {beta}-2-Glycoprotein 1 Is an Inhibitor of Angiogenesis Am. J. Pathol., November 1, 2007; 171(5): 1659 - 1669. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Gondi, S. S. Lakka, D. H. Dinh, W. C. Olivero, M. Gujrati, and J. S. Rao Intraperitoneal Injection of a Hairpin RNA Expressing Plasmid Targeting Urokinase-Type Plasminogen Activator (uPA) Receptor and uPA Retards Angiogenesis and Inhibits Intracranial Tumor Growth in Nude Mice Clin. Cancer Res., July 15, 2007; 13(14): 4051 - 4060. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Sun and K. R. McCrae Endothelial-cell apoptosis induced by cleaved high-molecular-weight kininogen (HKa) is matrix dependent and requires the generation of reactive oxygen species Blood, June 15, 2006; 107(12): 4714 - 4720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. Nordahl, V. Rydengard, M. Morgelin, and A. Schmidtchen Domain 5 of High Molecular Weight Kininogen Is Antibacterial J. Biol. Chem., October 14, 2005; 280(41): 34832 - 34839. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Tertemiz, U. A. Kayisli, A. Arici, and R. Demir Apoptosis Contributes to Vascular Lumen Formation and Vascular Branching in Human Placental Vasculogenesis Biol Reprod, March 1, 2005; 72(3): 727 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Song, I. M. Sainz, S. C. Cosenza, I. Isordia-Salas, A. Bior, H. N. Bradford, Y.-L. Guo, R. A. Pixley, E. P. Reddy, and R. W. Colman Inhibition of tumor angiogenesis in vivo by a monoclonal antibody targeted to domain 5 of high molecular weight kininogen Blood, October 1, 2004; 104(7): 2065 - 2072. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Donate, J. C. Juarez, X. Guan, N. V. Shipulina, M. L. Plunkett, Z. Tel-Tsur, D. E. Shaw, W. T. Morgan, and A. P. Mazar Peptides Derived from the Histidine-Proline Domain of the Histidine-Proline-Rich Glycoprotein Bind to Tropomyosin and Have Antiangiogenic and Antitumor Activities Cancer Res., August 15, 2004; 64(16): 5812 - 5817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Cao, Y.-L. Guo, and R. W. Colman Urokinase-Type Plasminogen Activator Receptor Is Involved in Mediating the Apoptotic Effect of Cleaved High Molecular Weight Kininogen in Human Endothelial Cells Circ. Res., May 14, 2004; 94(9): 1227 - 1234. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Mahdi, Z. Shariat-Madar, A. Kuo, M. Carinato, D. B. Cines, and A. H. Schmaier Mapping the Interaction between High Molecular Mass Kininogen and the Urokinase Plasminogen Activator Receptor J. Biol. Chem., April 16, 2004; 279(16): 16621 - 16628. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-K. Olsson, H. Larsson, J. Dixelius, I. Johansson, C. Lee, C. Oellig, I. Bjork, and L. Claesson-Welsh A Fragment of Histidine-Rich Glycoprotein Is a Potent Inhibitor of Tumor Vascularization Cancer Res., January 15, 2004; 64(2): 599 - 605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kawasaki, T. Maeda, K. Hanasawa, I. Ohkubo, and T. Tani Effect of His-Gly-Lys Motif Derived from Domain 5 of High Molecular Weight Kininogen on Suppression of Cancer Metastasis Both in Vitro and in Vivo J. Biol. Chem., December 5, 2003; 278(49): 49301 - 49307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Daly, A. Makris, M. Reed, and C. E. Lewis Hemostatic Regulators of Tumor Angiogenesis: A Source of Antiangiogenic Agents for Cancer Treatment? J Natl Cancer Inst, November 19, 2003; 95(22): 1660 - 1673. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Schmaier The kallikrein-kinin and the renin-angiotensin systems have a multilayered interaction Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2003; 285(1): R1 - R13. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Skidgel, F. Alhenc-Gelas, and W. B. Campbell Regulation of Cardiovascular Signaling by Kinins and Products of Similar Converting Enzyme Systems: Prologue: Kinins and related systems. New life for old discoveries Am J Physiol Heart Circ Physiol, June 1, 2003; 284(6): H1886 - H1891. [Full Text] [PDF]
|
|
|
|
|
|
S. Wang, M. G. Hasham, I. Isordia-Salas, A. Y. Tsygankov, R. W. Colman, and Y.-L. Guo Regulation of Cardiovascular Signaling by Kinins and Products of Similar Converting Enzyme Systems: Upregulation of Cdc2 and cyclin A during apoptosis of endothelial cells induced by cleaved high-molecular-weight kininogen Am J Physiol Heart Circ Physiol, June 1, 2003; 284(6): H1917 - H1923. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Krijanovski, V. Proulle, F. Mahdi, M. Dreyfus, W. Muller-Esterl, and A. H. Schmaier Characterization of molecular defects of Fitzgerald trait and another novel high-molecular-weight kininogen-deficient patient: insights into structural requirements for kininogen expression Blood, June 1, 2003; 101(11): 4430 - 4436. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-C. Zhang, F. Donate, X. Qi, N. P. Ziats, J. C. Juarez, A. P. Mazar, Y.-P. Pang, and K. R. McCrae The antiangiogenic activity of cleaved high molecular weight kininogen is mediated through binding to endothelial cell tropomyosin PNAS, September 17, 2002; 99(19): 12224 - 12229. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Juarez, X. Guan, N. V. Shipulina, M. L. Plunkett, G. C. Parry, D. E. Shaw, J.-C. Zhang, S. A. Rabbani, K. R. McCrae, A. P. Mazar, et al. Histidine-Proline-rich Glycoprotein Has Potent Antiangiogenic Activity Mediated through the Histidine-Proline-rich Domain Cancer Res., September 15, 2002; 62(18): 5344 - 5350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Chavakis, N. Boeckel, S. Santoso, R. Voss, I. Isordia-Salas, R. A. Pixley, E. Morgenstern, R. W. Colman, and K. T. Preissner Inhibition of Platelet Adhesion and Aggregation by a Defined Region (Gly-486-Lys-502) of High Molecular Weight Kininogen J. Biol. Chem., June 21, 2002; 277(26): 23157 - 23164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Mahdi, Z. S. Madar, C. D. Figueroa, and A. H. Schmaier Factor XII interacts with the multiprotein assembly of urokinase plasminogen activator receptor, gC1qR, and cytokeratin 1 on endothelial cell membranes Blood, May 15, 2002; 99(10): 3585 - 3596. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. CHAVAKIS, S. M. KANSE, R. A. PIXLEY, A. E. MAY, I. ISORDIA-SALAS, R. W. COLMAN, and K. T. PREISSNER Regulation of leukocyte recruitment by polypeptides derived from high molecular weight kininogen FASEB J, November 1, 2001; 15(13): 2365 - 2376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-L. Guo, S. Wang, and R. W. Colman Kininostatin, an Angiogenic Inhibitor, Inhibits Proliferation and Induces Apoptosis of Human Endothelial Cells Arterioscler. Thromb. Vasc. Biol., September 1, 2001; 21(9): 1427 - 1433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Takahashi, D. R. Rhodes, K. A. Furge, H.-o. Kanayama, S. Kagawa, B. B. Haab, and B. T. Teh Gene expression profiling of clear cell renal cell carcinoma: Gene identification and prognostic classification PNAS, August 1, 2001; (2001) 171209998. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Takahashi, D. R. Rhodes, K. A. Furge, H.-o. Kanayama, S. Kagawa, B. B. Haab, and B. T. Teh Gene expression profiling of clear cell renal cell carcinoma: Gene identification and prognostic classification PNAS, August 14, 2001; 98(17): 9754 - 9759. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
