(Received for publication, October 7, 1996, and in revised form, February 19, 1997)
From the Tissue factor pathway inhibitor (TFPI) is a
potent inhibitor of blood coagulation factor Xa (fXa) and factor VIIa.
We have recently shown that fXa binding stimulates the uptake and
degradation of cell surface-bound 125I-TFPI (Ho, G.,
Toomey, J. R., Broze, G. J., Jr., and Schwartz, A. L. (1996)
J. Biol. Chem. 271, 9497-9502). In the present study we examined the role of cell surface glycosaminoglycans (GAGs) in this
process. Removal of cell surface GAG chains by treatment of cells with
heparinase or heparitinase but not chondroitinase markedly reduced
fXa-stimulated 125I-TFPI uptake and degradation. Inhibition
of GAG sulfation by growth of cells in chlorate-containing medium
similarly decreased fXa-stimulated 125I-TFPI degradation.
These results suggest that heparan sulfate proteoglycans (HSPGs) are
required for the uptake and degradation of 125I-TFPI·fXa
complexes. Chemical cross-linking/immunoprecipitation analyses revealed
that 125I-TFPI was directly associated with HSPGs on the
cell surface and that fXa binding increased the amount of
125I-TFPI bound. Of the several cell lines evaluated, bend
endothelial cells demonstrated the greatest fXa stimulation of
125I-TFPI uptake and degradation.
Cross-linking/immunoprecipitation analyses on bend cells also revealed
that HSPGs were specifically associated with TFPI and fXa. These data
suggest that HSPGs may directly act as the uptake and degradation
receptor for TFPI·fXa complexes.
Coagulation factor Xa (fXa),1 a serine
protease positioned at the convergence of the intrinsic and extrinsic
pathways of the coagulation cascade, plays a central role in
hemostasis. Activated fXa along with phospholipids and factor Va
converts prothrombin to thrombin, which in turn generates the fibrin
clot.
The nonthrombogenic properties of the endothelial cell surface are
maintained in part by protease inhibitors of the coagulation cascade.
Tissue factor pathway inhibitor (TFPI) is a potent direct inhibitor of
fXa and, in a fXa-dependent fashion, produces feedback inhibition of
the factor VIIa-tissue factor catalytic complex (1). The TFPI molecule
consists of three tandem Kunitz-type protease inhibitor domains and a
basic C-terminal region (2). Its second Kunitz domain is required for
binding to fXa, and its first Kunitz domain appears to bind factor VIIa
in the factor VIIa-tissue factor complex (1, 3). Unlike other
coagulation protease inhibitors, whose binding is irreversible, TFPI
binds to the proteases in a reversible manner (4, 5). Instead of
transferring fXa to other protease inhibitors for final inactivation, we have shown that cell surface-bound TFPI mediates
125I-fXa uptake and degradation and, reciprocally, fXa
binding stimulates the uptake and degradation of cell surface-bound
125I-TFPI (6). Unlike the uptake and degradation of
uncomplexed 125I-TFPI, which is mediated via the endocytic
receptor low density lipoprotein-related protein (LRP) (7), the uptake
and degradation of the fXa·TFPI complex is independent of LRP
(6).
Circumstantial evidence suggests that TFPI may be bound to
glycosaminoglycans (GAGs) on the endothelial cell surface. This notion
derives from the following facts. (a) TFPI binds to heparin agarose (8); (b) heparin and sulfated polysaccharides
enhance the anticoagulant activity of TFPI (9); (c) after
intravenous administration of heparin, plasma levels of TFPI increase
severalfold (10, 11); and (d) heparin competes for abundant
TFPI binding sites on the cell surface (7, 12, 13). In the present
investigation, we attempted to address whether GAGs play a role in the
uptake and degradation of the fXa·TFPI complex. Using PEA 13 fibroblasts, a cell line deficient in LRP (14), we show that enzymatic
removal of cell surface heparin sulfate or growth of cells in chlorate reduced fXa-stimulated 125I-TFPI degradation. Chemical
cross-linking of 125I-TFPI to cells coupled with
immunoprecipitation analyses showed an association of high molecular
weight species with 125I-TFPI and enhancement following fXa
binding. These data suggest that heparan sulfate proteoglycans (HSPGs)
may directly serve as the receptor for the uptake and degradation of
TFPI·fXa complexes.
IODOGEN and dithiobis(sulfosuccinimidyl
propionate) (DTSSP) were purchased from Pierce.
[125I]Iodide was from Amersham Corp. Heparitinase (EC
4.2.2.8) and chondroitinase ABC (EC 4.2.2.4) were from ICN. Heparinase
(EC 4.2.2.7, heparinase I), bovine serum albumin, and normal rabbit (nonimmune) IgG were from Sigma. Sodium chlorate and sodium sulfate were purchased from Aldrich. Immobilized rProtein A beads were from
RepliGen (Cambridge, MA). Human factor Xa, goat-anti-human factor X
polyclonal antibody, and normal goat IgG were obtained from American
Diagnostica (Greenwich, CT). Anti-TFPI polyclonal antibodies were
described previously (15). Tissue culture media and plasticware were
obtained from Life Technologies, Inc.
Proteins (10-50 µg) were iodinated
using the IODOGEN methods (16). Specific radioactivities were typically
0.5-3 × 106 cpm/pmol of protein.
HepG2 cells (17), PEA 10 (LRP heterozygous),
PEA 13 (LRP homozygous deleted) cells (14), and mouse brain endothelial
cells (bend-3, gift of Dr. W. Frazier, Washington University at St. Louis) were cultured in Dulbecco's modified Eagle's medium (with glutamine) supplemented with 10% fetal calf serum, 100 units/ml penicillin, and 100 µg/ml streptomycin. Cells were incubated at 37 °C in humidified air containing 5% CO2.
Cells were seeded into
12-well dishes 1 day prior to assays. Cell monolayers were generally
used at 70-80% confluence. Assay buffers were Dulbecco's modified
Eagle's medium containing 3% bovine serum albumin. In general,
binding was carried out by washing cell monolayers with prechilled
assay buffer twice and binding was initiated by adding 0.5 ml of
4 °C assay buffer containing 125I-labeled proteins or
unlabeled proteins. After incubation at 4 °C for the time duration
specified in the figure legends, overlying buffer containing unbound
ligand was removed, and the cells were washed three times with 4 °C
assay buffer. The cells were then incubated for a second round of
binding as indicated in the figure legends.
Degradation assays were carried out at 37 °C for 3 h in 0.5 ml
of assay buffer. Radioligands were either prebound to the cell surface
or contained in the assay buffer as indicated in the figure legends.
Thereafter, the medium overlying the cell monolayers was removed and
proteins were precipitated by addition of bovine serum albumin to 5 mg/ml and trichloroacetic acid to 10%. Degradation of radioligand was
defined as the appearance of radioactive fragments in the overlying
medium that were soluble in 10% trichloroacetic acid. Degradation of
125I-ligand in parallel dishes that did not contain cells
was subtracted from the total degradation (17).
Cells grown in 12-well
dishes (~70% confluence) were digested at 37 °C for 1.5 h in
assay buffer containing the appropriate enzymes at concentrations
indicated in the figure legends.
Chemical cross-linking was performed
as described (18). Briefly, cells (70-80% confluence) were incubated
at 4 °C with assay buffer containing the desired concentrations of
TFPI. After removing unbound ligand by washing twice with assay buffer,
the cell monolayers were incubated at 4 °C with assay buffer alone
or containing the desired concentrations of fXa. The cells were then
washed with PBS supplemented with 1 mM CaCl2
and 0.5 mM MgCl2 (PBSc). Cross-linking of
ligands to the cells was performed by incubating cell monolayers with
PBSc containing 0.5 mM DTSSP. After 30 min at 4 °C, the
reaction was quenched by washing cells with Tris-buffered saline. Cells were then lysed in PBSc containing 1% Triton X-100 and 1 mM phenylmethylsulfonyl fluoride for 30 min at 4 °C with
intermittent votexing, followed by brief sonication at low power
output. The cell lysates were subsequently used for
immunoprecipitation.
Cell lysates from chemical
cross-linking experiments were mixed with equal volumes of PBSc
containing 1% Triton X-100, 0.5% sodium deoxycholate, 1% SDS, and 1 mM phenylmethylsulfonyl fluoride (immunomix). The mixture
was aliquoted, and appropriate antibodies were added (10 µg of IgG in
a volume of ~0.4 ml). After rocking overnight at 4 °C, the
immunomixture was incubated with 30 µl of Protein A bead slurry at
room temperature for 1 h. The Protein A beads were collected by
spinning for 30 s at top speed in a Microfuge, washed twice with
immunomix and twice with PBSc. The immunoprecipitates bound to Protein
A beads were released by boiling for 5 min in 62.5 mM
Tris-HCl, pH 6.8, 2% SDS, 10% glycerol (Laemmli SDS sample buffer)
(19) without 2- SDS samples were analyzed by SDS-polyacrylamide
gel electrophoresis (19) under nonreducing conditions. A set of
prestained molecular weight standards (myosin, We have previously shown that cell surface-bound
125I-TFPI was degraded at very low rates by PEA 13 cells
but that this basal degradation rate is enhanced 5-fold upon binding of
125I-TFPI to fXa (6). Since PEA 13 cells are devoid of LRP
(14), this observation suggests that the fXa-stimulated
125I-TFPI degradation is mediated via a non-LRP endocytic
pathway. TFPI has been proposed to bind GAGs on the cell surface
(10-12). To investigate the role of GAGs in TFPI·fXa degradation,
PEA 13 cells were incubated with heparinase or heparitinase (1 unit/ml), which cleave GAG heparan sulfate moieties at distinct sites
(20). The enzyme-treated cells were then incubated at 4 °C with
125I-TFPI to allow for binding, followed by incubation at
4 °C in the presence or absence of fXa. Thereafter, degradation of
125I-TFPI was assessed at 37 °C. As shown in Table
I, degradation of fXa-stimulated 125I-TFPI,
measured as the difference in 125I-TFPI degraded in the
presence and absence of fXa, was reduced to ~48% of control (Table
I). Removal of GAG chondroitin sulfate and dermatan sulfate by treating
cells with chondroitinase ABC (21), however, did not affect
fXa-stimulated 125I-TFPI degradation (Table I). Since
heparan sulfate and chondroitin sulfate represent the two major types
of GAGs (22), these data suggest that HSPGs participate in the uptake
and degradation of TFPI·fXa complexes. The maximal effect of
heparinase on fXa-stimulated 125I-TFPI degradation was
determined, as shown in Fig. 1. The amount of
fXa-stimulated 125I-TFPI degradation decreased with
increasing concentrations of heparinase, reaching a maximal effect that
was 20% of the control, at 4 units/ml heparinase. These data thus
suggest that HSPGs play a major role in the uptake and degradation of
fXa·TFPI complexes.
Table I.
Effect of GAG-degrading enzymes on fXa-stimulated 125I-TFPI
degradation
The C-terminal heparin-binding domain of TFPI consists of a cluster of
positively charged residues, truncation of which abolishes the binding
of TFPI to the cell surface (23). To determine whether the negatively
charged sulfate groups on HSPGs were critical for TFPI·fXa binding
and degradation, PEA 13 cells were cultured in medium containing
chlorate, a competitive inhibitor of ATP sulfurylase (24). Cells grown
in media containing various concentrations of chlorate were then
analyzed for degradation of 125I-TFPI in the presence or
absence of fXa as described above. As shown in Fig. 2,
growth of cells for more than four doublings in the presence of
chlorate reduced the level of fXa-stimulated 125I-TFPI
degradation in a dose-dependent fashion. At 30 mM chlorate, the level of degradation was reduced to 9% of
the control value. To assure that the inhibition by chlorate was not
due to cytotoxicity but was a result of competitive inhibition of
sulfation, cells were cultured in medium containing chlorate (20 mM or 30 mM) as well as sulfate (10 mM). As shown in Fig. 2, degradation of
125I-TFPI was fully rescued by simultaneous addition of
sulfate in the medium. These data indicate that degradation of
fXa·TFPI complexes requires sulfate groups on HSPGs, and support the
observation from Fig. 1 that HSPGs play a key role in the uptake and
degradation of the TFPI·fXa complex by cells.
In addition to serving as a binding site for fXa on the cell surface
(25), we have shown that TFPI also mediates the cellular uptake and
degradation of fXa (6). To determine whether that fXa degradation was
similarly affected by HSPGs, fXa was radiolabeled and its degradation
was analyzed in PEA 13 cells that were either treated with 1 unit/ml
heparinase or cultured in 10 mM chlorate. The respective
cells were then incubated at 4 °C with TFPI to allow for binding.
Following washing to remove unbound TFPI, the cell monolayers were then
incubated with various concentrations of 125I-fXa and the
degradation was evaluated upon incubation at 37 °C as above. As
shown in Fig. 3A, heparinase treatment
reduced TFPI-mediated 125I-fXa degradation by ~50% at
each concentration used. Growing cells in chlorate also markedly
decreased 125I-fXa degradation (Fig. 3B), and
this inhibitory effect was abrogated in cells cultured with an
additional 5 mM sulfate (Fig. 3B).
We have shown previously that the cellular
degradation of uncomplexed 125I-TFPI is
inhibited >80% both by antibodies directed against LRP and by the
39-kDa protein, an LRP-associated protein that inhibits all ligand
interactions with LRP (7). LRP, however, is not the major cell surface
receptor, as the vast majority of 125I-TFPI binding is not
inhibitable by the 39-kDa protein (7, 12). HSPGs have been proposed as
the initial binding site for several LRP ligands in addition to TFPI
(7, 12) including apoE-enriched remnant lipoproteins (26), hepatic
lipase (27), and thrombospondin (28), whose uptake and degradation by
LRP requires HSPGs. To further determine the role of HSPGs in the LRP-mediated uptake and degradation of TFPI, degradation of
125I-TFPI was analyzed in PEA 10 cells, which express LRP
(14). The cells were incubated with 2 units/ml heparinase or grown in medium containing 10 mM chlorate as described in Fig. 3.
The cells were then incubated with 125I-TFPI at 37 °C to
assess degradation. As shown in Fig. 4, neither treatment of PEA 10 cells with heparinase nor culture of cells in
chlorate affected 125I-TFPI degradation.
We showed previously that fXa-stimulated
125I-TFPI degradation results from an enhanced uptake of
125I-TFPI by cells (6). One potential explanation for this
notion is that the interaction of TFPI for its endocytic receptor is increased upon association with fXa, resulting in the enhanced cellular
uptake. To address this question as well as to define candidate
receptors for the fXa·TFPI complex, chemical cross-linking coupled
with immunoprecipitation analyses was performed. Initially fXa was
radioiodinated and cross-linking was performed on PEA 13 cells. Cells
were incubated with or without TFPI at 4 °C, followed by incubation
with 125I-fXa to allow for binding. Chemical cross-linking
of the bound ligands to the cells was performed with DTSSP, a
thio-cleavable, water-soluble, and membrane-impermeable reagent (29).
After cross-linking, the cells were lysed and immunoprecipitated with anti-TFPI or anti-fX polyclonal antibodies. To characterize membrane proteins cross-linked to 125I-fXa, the immunoprecipitates
were resolved on 7.5% SDS gels under nonreducing conditions. As shown
in Fig. 5A, in the presence of TFPI
prebinding and DTSSP, a high molecular weight (high
Mr) band at the top of the gel was
immunoprecipitated by both anti-fX and anti-TFPI antibodies
(lanes 3 and 6), suggesting its association with
125I-fXa as well as TFPI. The specificity of association is
affirmed by showing that this high Mr band is
absent without the cross-linking reagent (lanes 4 and
7) and is not precipitated by normal rabbit IgG (lane
1). The high Mr band is undetectable when
cells were not prebound with TFPI (lane 2), suggesting that
binding of 125I-fXa to the high Mr
band is mediated by TFPI. Species at size ~90 kDa are free
cross-linked 125I-fXa·TFPI complexes (lanes 3 and 6). Intense bands at 50 kDa are free
125I-fXa (lanes 3 and 4).
Since 125I-fXa binding to the high
Mr band is mediated through TFPI, we next
examined the effect of fXa on 125I-TFPI binding to the
receptor by cross-linking/immunoprecipitation analyses. PEA 13 cells
were incubated at 4 °C with 125I-TFPI and thereafter
with or without fXa at 4 °C. Subsequent cross-linking and
immunoprecipitation were performed as above. As shown in Fig.
5B, in the absence of fXa addition but presence of DTSSP,
the same high Mr band was immunoprecipitated
with anti-TFPI antibodies (lane 4). The band intensity,
however, was enhanced when cells were incubated with fXa prior to
cross-linking (lane 5). Densitometric analysis showed a
2.7-fold enhancement. These data indicate that fXa binding increases
the interaction of TFPI with the high Mr band
species. The high Mr band was precipitated by
anti-fX antibodies in the presence of fXa and DTSSP (lane 2) but not in the absence of either cross-linking (lane 3) or
prior fXa binding (lane 1), reaffirming the specific
association of the high Mr band with the
TFPI·fXa complex. In the absence of cross-linking (lane
6), there appears to be a high Mr band at the top of the gel. This is likely the result of high background during
sample preparation. To confirm the identity of the high Mr species as HSPGs, PEA 13 cells were treated
with heparinase prior to binding to 125I-TFPI and fXa. The
cross-linking and immunoprecipitation with anti-TFPI antibodies were
conducted in an identical manner to that in Fig. 5B. As
shown in Fig. 5C, heparinase pretreatment diminished the
intensity of the high Mr band (lane
2) (~ 2.9-fold reduction) relative to the control (lane
1). These results demonstrate that TFPI and fXa are indeed
associated with cell surface HSPGs and that the reduced uptake and
degradation of fXa·TFPI complexes observed in Table I and Figs. 1 and
3 is likely secondary to the reduction of TFPI binding to the cells
following heparinase treatment.
Under physiological conditions,
fXa·TFPI complexes form within the vasculature. Thus, we compared the
relative ability of the bend-3 microvasculature endothelial cells,
hepatoma HepG2, and fibroblast PEA 10 cells to undergo the
fXa-stimulated TFPI degradation. All of these cell lines express
functional LRP. The cells were incubated at 4 °C with radiolabeled
TFPI to allow for cell surface binding. After washing to remove unbound
radioligand, 125I-TFPI degradation was assessed in the
absence or presence of varying concentrations of fXa. As shown in Fig.
6, in the absence of fXa the degradation of
125I-TFPI was ~6, 40, and 90 fmol/106 cells
for bend, PEA 13, and HepG2, respectively. These rates essentially
reflect the amount of 125I-TFPI degraded via LRP. With
increasing concentrations of fXa added, bend cells exhibited >10-fold
enhancement in 125I-TFPI degradation, whereas PEA 10 and
HepG2 cells displayed only ~1.5- and ~1.1-fold enhancement,
respectively. These data thus suggest that bend cells express low
levels of LRP but high levels of the endocytic receptor for the
TFPI·fXa complex. The fXa-stimulated 125I-TFPI
degradation in bend cells was also markedly reduced by treatments of
cells with heparinase, heparitinase, and chlorate (data not shown).
The possibility that bend cells may express high levels of TFPI·fXa
specific endocytic receptors prompted us to examine the receptor(s) in
this cell line. Under the conditions described in Fig. 5, TFPI was
radiolabeled, cross-linked to bend cells, and immunoprecipitated with
anti-fXa and anti-TFPI antibodies. The immunoprecipitates were run on
SDS gels under non-reducing conditions. As shown in Fig.
7A, a high Mr band was
cross-linked to 125I-TFPI (lane 4), and the band
intensity was greatly potentiated in the presence of fXa (lane
5). Densitometric analysis showed a 8.7-fold enhancement. Similar
to that seen in Fig. 5B, this high Mr
band-125I-TFPI complex was also specifically cross-linked
to fXa (lane 2). To ascertain the identity of the high
Mr band as HSPGs, heparinase treatment of bend
cells was carried out as in Fig. 5C. As shown in Fig.
7B, heparinase-treated cells displayed a 2.7-fold reduced level of surface-bound 125I-TFPI·fXa complexes. These
data demonstrate that HSPGs are also the binding species for fXa·TFPI
complexes on the endothelial cell.
Using both enzymatic and chemical approaches, we have shown that
fXa-stimulated 125I-TFPI uptake and degradation by PEA 13 cells requires cell surface HSPGs. Similarly, HSPGs are necessary for
TFPI-mediated 125I-fXa uptake and degradation. Using
chemical cross-linking in combination with immunoprecipitation
analyses, we have demonstrated that 125I-TFPI is physically
associated with HSPGs on the PEA 13 cells and that TFPI's interaction
with HSPGs is potentiated upon association with fXa. Of the various
cell lines examined for fXa-stimulated 125I-TFPI
degradation, the bend endothelial cell displays the highest level of
enhancement, suggesting high levels of the endocytic receptor(s) for
the fXa·TFPI complex are present in these cells. Cross-linking/immunoprecipitation analysis performed with bend cells
also revealed HSPGs as binding species for TFPI·fXa complexes. Thus
the direct association of the TFPI·fXa complex with HSPGs on the cell
surface of both PEA 13 and bend cells, as well as the diminution of
125I-TFPI·fXa degradation by cells following removal of
HSPGs from the cell surface, strongly suggest that HSPGs as receptors
for the uptake and degradation of the TFPI·fXa complex. Chlorate and heparinase treatments reduced the fXa-stimulated 125I-TFPI
degradation to 10-20% of control (Figs. 1 and 2), and the reduced
levels do not seem to have plateaued in the dose-response curves (Figs.
1 and 2). Thus it is likely that HSPGs are the major, if not the only,
endocytic receptors for the fXa·TFPI complex.
HSPGs are a complex and heterogeneous family of macromolecules composed
of linear sulfated polyaccharide chains, the heparan sulfate moieties,
that are covalently attached to core proteins (30, 31). HSPGs are
ubiquitously distributed on the plasma membrane. They are anchored to
the membrane either via a linkage with membrane phospholipid (32) or
via a hydrophobic transmembrane domain (33). Membrane HSPGs have been
implicated in an array of cellular functions, among which are ligand
binding and endocytosis (31, 34). Apart from serving as initial binding
sites for ligands that are eventually transferred to classical
endocytic receptors (e.g. LRP) for cellular uptake and
degradation (26-28), several lines of evidence suggest that HSPGs may
also act directly in the internalization of ligands. Lipoprotein lipase
is anchored to the cell surface via HSPGs, through which it is
internalized and recycled to intracellular compartments (35). Herpes
simplex virus does not bind to nor is internalized by cells harboring mutations in HSPG synthesis (36). While it is possible that the virus
may bind to HSPGs initially and thereafter be transferred to another
receptor that mediates infection, the observation that cell surface
proteoglycans, especially those containing heparan sulfate, are rapidly
internalized and degraded (37) argues for a direct role of the
proteoglycans.
Heparan sulfate moieties bind proteins predominantly via electrostatic
interactions between the highly charged anionic sulfate groups and
clusters of basic amino acid residues arranged in a three-dimensional
array on the protein (31). TFPI may bind to cell surface HSPGs in a
similar manner, since TFPI contains a stretch of basic residues at its
C terminus (2), deletion of this C-terminal region abolishes binding of
TFPI to the cell surface (23), and a complementary reduction of HSPG
sulfation in cells grown in chlorate-containing medium decreases the
extent of TFPI binding (data not shown) and degradation (Fig. 2).
Electrostatic interactions may render TFPI binding to a wide range of
HSPGs expressed on cell surfaces. However, not all TFPI-bound HSPGs necessarily participate in the uptake and degradation of TFPI·fXa complexes. Whether the TFPI·fXa complex receptor is a specific core
protein containing diverse GAG chains or a specific GAG on heterogeneous core proteins is a matter of future studies.
It is interesting that among the cell lines tested, the endothelial
cells (bend) exhibited >10-fold augmentation of fXa·TFPI uptake and
degradation over the basal rate. The hepatoma cells (HepG2) and the
fibroblasts (PEA 10) displayed 1.5-fold or less enhancement over the
basal rates. Interestingly, the basal rate of TFPI degradation, which
represents essentially the amount of TFPI degraded via LRP, was 15-fold
lower in bend cells than in HepG2 cells (Fig. 6). These data suggest
that bend cells express low levels of LRP, the endocytic receptor for
uncomplexed TFPI (7), but high levels of the endocytic receptor for
TFPI·fXa complexes. Physiologically this may be important since a low
rate of removal (uptake and degradation) of uncomplexed TFPI may be necessary to maintain high local concentrations of TFPI on vascular surfaces. Rapid turnover may be initiated as a means of clearance of
fXa or the TFPI·fXa complex from the cell surface.
Heparin has long been used clinically as an anticoagulant. In addition
to direct inhibition of coagulation proteases, heparin greatly enhances
the inhibitory activities of antithrombin III and TFPI. Anticoagulant
heparan sulfate moieties on the vascular endothelial cell surfaces thus
may endow them with non-thrombogenic properties (38). Such
anticoagulant HSPGs, which bind antithrombin III and greatly facilitate
the inhibition to thrombin, have been isolated from rat fat pad
microvascular endothelial cells (39). The present study provides
another potential anticoagulant role of cell surface HSPGs, one which
directly involves TFPI and the active uptake and degradation of
fXa·TFPI complexes generated on the cell surface.
Departments of Pediatrics,
Materials
-mercaptoethanol.
-galactosidase,
bovine serum albumin, and ovalbumin) (Bio-Rad) was included for
molecular weight estimation. Gels were dried and exposed to x-ray films
for visualization.
Reduction of Cell Surface Heparan Sulfate Decreases the
fXa-stimulated 125I-TFPI Degradation by PEA 13 Cells
Enzyme
fXa-stimulated
125I-TFPI degradation
1 unit/ml
%
0
100
Heparinase
47 ± 5
Heparitinase
48 ± 2
Chondroitinase
101
± 4
Fig. 1.
fXa-stimulated 125I-TFPI
degradation following heparinase digestion of PEA 13 cells. PEA 13 cells were in incubated with various concentrations of heparinase under
the conditions specified in Table I. Subsequent assays for
fXa-stimulated 125I-TFPI degradation also followed those
described in Table I. Each symbol represents the mean ± S.D. of
three independent experiments.
[View Larger Version of this Image (15K GIF file)]
Fig. 2.
Chlorate inhibits fXa-stimulated
125I-TFPI degradation. PEA 13 cells were cultured for
2 days in medium containing various concentrations of sodium chlorate
with or without additional 10 mM sodium sulfate. Following
~4 doublings the cells reached a density of ~70% on the day of
assay. Degradation studies were conducted as described in Table I. The
data were generated from at least three independent experiments.
[View Larger Version of this Image (21K GIF file)]
Fig. 3.
Heparinase and chlorate reduce the cellular
uptake and degradation of 125I-fXa mediated by TFPI.
A, PEA 13 cells were treated with or without 1 unit/ml
heparinase in a manner identical to that described in Table I. The
cells were then incubated at 4 °C for 30 min in the presence of 50 nM TFPI, following which unbound TFPI was removed. After
incubation at 4 °C for 30 min in the presence of increasing
concentrations of 125I-fXa, the cells were placed at
37 °C for 3 h to assay degradation. B, PEA 13 cells
were grown in medium containing 10 mM chlorate, or 10 mM chlorate plus 5 mM sulfate, or no additional
chemicals in a manner similar to that described in Fig. 2. TFPI
prebinding to and subsequent 125I-fXa degradation by the
cells were carried out as described in A. Numbers
represent duplicate determinations.
[View Larger Version of this Image (19K GIF file)]
Fig. 4.
Heparinase and chlorate have no effect on the
cellular degradation of uncomplexed 125I-TFPI. PEA 10 cells were incubated without or with 2 units/ml heparinase in
conditions described in Table I (A) or grown in the absence
or presence of 10 mM sodium chlorate for 2 days
(B). Cells from A and B were then
incubated with increasing concentrations (2-12 nM) of
125I-TFPI at 37 °C for 3 h to assess degradation.
Numbers represent duplicate determinations.
[View Larger Version of this Image (18K GIF file)]
Fig. 5.
Binding of fXa increases the interaction of
125I-TFPI with HSPGs on the PEA 13 cell surface.
A, PEA 13 cells (grown in 100-mm dishes) were incubated at
4 °C with or without 10 nM TFPI for 1.5 h. After
washing to remove unbound TFPI, the cells were incubated at 4 °C
with 10 nM 125I-fXa for another 1.5 h.
Following washing to remove unbound radioligand, the cells were
incubated with or without 0.5 mM DTSSP at 4 °C for 30 min. Following cell lysis, immunoprecipitations were performed with
anti-TFPI, anti-fX, or normal rabbit IgG (N.R.) as described under "Experimental Procedures." The immunoprecipitates were run on
7.5% SDS gels under nonreducing conditions. Equal amounts of total
cellular proteins were loaded in each lane. B, PEA 13 cells were incubated at 4 °C with 10 nM 125I-TFPI
for 1.5 h. After washing to remove unbound radioligand, the cells
were incubated at 4 °C with or without 10 nM fXa for an
additional 1.5 h. Subsequent cross-linking, immunoprecipitation, and gel analysis were identical to those specified in A.
C, PEA 13 cells were incubated without (lane 1)
or with (lane 2) 2 units/ml heparinase at 37 °C for
1.5 h. The enzyme-treated cells were then incubated sequentially
with 125I-TFPI and fXa as in B. After incubation
at 4 °C for 30 min with DTSSP, the cell lysates were
immunoprecipitated with anti-TFPI antibodies. Gel analysis was
identical to that described in A.
[View Larger Version of this Image (25K GIF file)]
Fig. 6.
fXa-stimulated 125I-TFPI
degradation by various cell lines. HepG2, bend, and PEA 10 cells
were incubated with 12 nM 125I-TFPI at 4 °C
for 30 min to allow for binding. After washing to remove unbound
radioligand, the cells were incubated at 4 °C for 30 min with
increasing concentrations (0-12 nM) of fXa and subsequently incubated at 37 °C for 3 h to assess
125I-TFPI degradation. Means ± S.D. are derived from
two or three independent experiments.
[View Larger Version of this Image (22K GIF file)]
Fig. 7.
TFPI and fXa associate with HSPGs on bend
cells. A, binding of 125I-TFPI and fXa to bend
cells, and subsequent cross-linking/immunoprecipitation were performed
as described in Fig. 5B. B, heparinase treatment of bend cells and the subsequent experimental scheme followed that
described in Fig. 5C.
[View Larger Version of this Image (24K GIF file)]
*
This work was supported in part by the National Institutes
of Health and Monsanto.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.
§
To whom correspondence should be addressed: Dept. of Pediatrics,
Box 8116, Washington University School of Medicine, St. Louis, MO
63110. Tel.: 314-454-6286; Fax: 314-454-2685.
1
The abbreviations used are: fXa, factor Xa;
TFPI, tissue factor pathway inhibitor; GAG, glycosaminoglycan; HSPG,
heparan sulfate proteoglycan; LRP, low density lipoprotein-related
protein; DTSSP, dithiobis(sulfosuccinimidyl propionate); PBS,
phosphate-buffered saline.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.