© 1996 American Heart Association, Inc.
Articles |
Isolation and Characterization of Tissue-Type Plasminogen Activator–Binding Proteoglycans From Human Umbilical Vein Endothelial Cells
From the Department for Vascular Biology and Thrombosis Research, University of Vienna (Austria).
Correspondence to Prof Dr Bernd Binder, Department for Vascular Biology and Thrombosis Research, Schwarzspanierstrasse 17, A-1090 Vienna, Austria.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract We analyzed the tissue-type plasminogen activator (TPA)–binding proteoglycans (PGs) on human umbilical vein endothelial cells (HUVECs), which were metabolically labeled with [35S]Na2SO4. Cell extracts were then prepared and subjected to affinity chromatography on diisopropyl fluorophosphate (DFP)–inactivated TPA–Sepharose 4B. Approximately 6% of the incorporated 35S radioactivity bound to DFP-treated TPA–Sepharose 4B and was eluted with 2 mol/L NaCl. In addition to NaCl, heparin, arginine, and lysine but not glycine,
-amino-n-caproic acid, or aspartic acid inhibited
this binding and eluted the bound 35S radioactivity.
Urea-containing polyacrylamide gel electrophoresis of the
eluted material consistently revealed two main signals of
35S radioactivity (one with an Mr
between 600 000 and 750 000 [PGA] and the other with an
Mr between 120 000 and 180 000 [PGC]).
Occasionally a less intense signal with an Mr
between 340 000 and 440 000 (PGB) was seen. Heparitinase treatment
markedly decreased the intensities of both 35S signals (PGA
and PGB), and chondroitinases AC and ABC abolished the 35S
signal of PGC, indicating that most of the HUVEC-incorporated
radioactivity with an affinity for TPA could be attributed to heparan
sulfate– and chondroitin sulfate–like structures. Reductive
elimination, which was performed to separate the possible
glycosaminoglycan moieties from the core proteins,
confirmed the PG-like nature of this material and again revealed
heparan sulfate and chondroitin sulfate as the major
glycosaminoglycan components. We therefore conclude
that HUVECs synthesize TPA-binding, heparan sulfate– and chondroitin
sulfate–containing PGs. In vivo, similar PGs may play a role in
TPA binding to endothelial cells and thereby possibly
influence TPA activity and/or provide an intravascular storage pool of
TPA.
Key Words: glycosaminoglycans • tissue-type plasminogen activator • endothelial cells • proteoglycans
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Fibrin clots are dissolved by the serine protease plasmin, which is generated from its zymogen plasminogen by the action of plasminogen activators, eg, TPA and UPA. (For a review, see Reference 11 .) TPA is assumed to be the enzyme mainly responsible for plasmin formation during fibrinolysis,2 whereas UPA is thought to mediate plasmin-dependent extravascular tissue degradation.3
In plasma TPA circulates at low concentrations, but its plasma levels can be rapidly increased by stimuli such as venous occlusion, vasopressin analogues, platelet-activating factor, thrombin, or histamine.1 4 5 6 7 8 9 Heparin has also been shown to increase TPA plasma levels.6 10 11 The major source of TPA in plasma is the vascular endothelium. The aforementioned stimuli are thought to release TPA from ECs, although at least in the case of venous occlusion, decreased clearance of TPA by the liver could also contribute to the increases in plasma TPA levels. At present it is unclear from which cellular stores TPA is released, and the mechanisms involved in the short-term release of TPA have not been clarified. For some mediators an increase in intracellular calcium could be the signal for TPA release, as induction of TPA release has also been shown for the calcium ionophore A23187.4
An additional mechanism for the rapid increase in TPA plasma levels
could also be its release from binding sites on the EC surface. Such
binding sites for TPA have been found on cultured HUVECs by several
groups,12 13 14 15 16 17 18 19 and three different types of binding site on
HUVECs have been described. The binding site with the highest affinity
for TPA has been identified as PAI-1,14 15 16 which is
present on EC surfaces and in the extracellular
matrix.20 21 The identity of the low-affinity TPA
binding site has not been fully elucidated, although a 40-kD protein
with affinity for TPA has been identified in detergent extracts of
isolated EC membranes.17 A similar protein also has been
purified from placenta, and antibodies against this placental protein
cross-react with the endothelial
protein.18 This protein bound not only TPA but also its
substrate plasminogen and has recently been identified as
annexin II.22 23 A common binding site for TPA and
plasminogen also has been identified by another research
group19 but with binding characteristics different from
those described for annexin II. In each case, however, the
low-affinity binding sites were present on ECs in high
concentrations (
106 binding sites per cell).
The capacity of the endothelial tissue TPA storage pool has been described to be on the order of 0.5 to 2 ng/cm2 of ECs.4 If one assumes a cell density of 2 to 5x104/cm2, this would roughly correspond to 104 molecules per cell. Therefore, in vivo occupancy of even 1% of the low-affinity TPA binding sites would represent an amount of EC-associated TPA comparable to that of the calculated endothelial TPA storage pool.4 Large amounts of TPA can be recovered by perfusion of cadaver vessels with PBS,24 which also suggests that within blood vessels TPA is present on the EC surface in appreciable quantities. Therefore, release of endogenously bound TPA from low-affinity binding sites on the cell surface could also increase plasma TPA levels.
Felez et al19 have shown that binding of TPA to ECs is inhibited by Lys, Arg, and EACA. These findings are consistent with the idea that TPA binds to negatively charged molecules, such as GAGs and PGs, that are exposed on the cell surface. The fact that TPA-EC binding is inhibited by high-Mr UPA but not low-Mr UPA19 is also consistent with this speculation, because high-Mr but not low-Mr UPA has an affinity for heparin.25 26 27 In fact, in vitro interaction between TPA and heparin (and other GAGs) has been studied extensively, and it has been shown that heparin and heparan sulfate increase the plasminogen activator activity of TPA in the absence of fibrin,26 28 29 30 whereas stimulation of TPA activity by fibrin is diminished by heparin.30 Within the TPA molecule, the finger and kringle 2 domains have been deduced to be responsible for the binding to heparin.28 If one considers both the affinity of TPA for heparin and the synthesis of PGs by ECs,31 32 33 34 one can speculate that the TPA synthesized by ECs binds to the PGs on the surfaces of these cells. Binding to cellular GAGs and release by heparin have also been described for other heparin-binding proteins, such as lipoprotein lipase35 and tissue factor pathway inhibitor.36 37 On the other hand, heparin has been shown to increase the active site–dependent binding of TPA to HUVECs.38 However, active site–dependent binding involves formation of a TPA–endothelial PAI-1 complex and therefore represents a different binding mechanism. In fact, preliminary experiments have revealed that heparin competes with DFP-inactivated TPA for binding to HUVECs (Dr S. Schönfelder, 1994, unpublished observations).
Therefore, the aim of the present study was to analyze the presence of TPA-binding PGs and GAGs on vascular ECs. Using extracts of metabolically labeled HUVECs, we have demonstrated that these cells in fact synthesize TPA-binding PGs and that the GAG component of these PGs consists mainly of heparan sulfate and chondroitin sulfate.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Materials
We obtained the following materials from the sources as indicated: Sepharose 4B (Pharmacia); DFP and heparin sodium salt (Fluka AG); heparinase III (heparitinase I) from Flavobacterium heparinum, chondroitinase AC from Arthrobacter aurescens, chondroitinase ABC from Proteus vulgaris, heparan sulfate sodium salt from bovine intestinal mucosa, chondroitin sulfate A sodium salt from bovine trachea, chondroitin sulfate C sodium salt from shark cartilage, PMSF, guanidine HCl, benzamidine HCl, Alcian Blue 8 GX, cross-linked phosphorylase, and molecular-weight markers (Sigma Chemical Co); BSA (Behring-Werke); L-Arg, L-Lys HCl, EACA, L-Asp, and CNBr (Merck); Centricon 10 ultrafiltration membranes (Amicon); H-D-IIe-Pro-Arg-pNA (S-2288, Kabi); calibration proteins for SDS-PAGE, acrylamide, bisacrylamide, SDS, low-melting agarose, and urea (Bio-Rad); and TPA (Actilyse, Boehringer Ingelheim).
Acetylation of Heparin39
Heparin (50 mg) was dissolved in 5 mL aqueous methanol (10%,
vol/vol). After addition of 25 mg Na2CO3 and
cooling to 4°C, 150 µL acetic anhydride was added in 25-µL
aliquots over 30 minutes. The final reaction mixture was dialyzed
against water, lyophilized, reconstituted in 1 mL of 0.2 mol/L HEPES
buffer, pH 7.4, containing 5% glycerol, and dialyzed against the same
buffer.
Coupling of TPA and BSA to CNBr-Activated Sepharose
4B
To protect the heparin-binding domains during the coupling
process, TPA was complexed with N-acetylated heparin
(see above) as described.39 In brief, 400 µL of a
50-mg/mL N-acetylated heparin solution in 0.2 mol/L
HEPES buffer, pH 7.4, containing 5% glycerol was added dropwise under
gentle mixing to 5 mL of the same buffer containing 1 mg/mL TPA. After
a 30-minute incubation at 22°C, TPA (250 µg/mL of Sepharose) was
coupled to CNBr-activated Sepharose 4B in the presence of
N-acetylated heparin according to the
manufacturer's instructions and rotated end over end for 2 hours at
22°C and then overnight at 4°C. BSA was coupled to
CNBr-activated Sepharose 4B according to the
manufacturer's instructions at a concentration of 250 µg/mL of
Sepharose 4B. The remaining unprotected binding sites on TPA– and
BSA–Sepharose 4B were blocked with 0.1 mol/L Tris HCl, pH 8.0, by
end-over-rotation overnight at 4°C. TPA–Sepharose 4B was
then washed with 2 mol/L NaCl to remove the bound
N-acetylated heparin. DFP treatment was performed by
incubating the TPA– and BSA–Sepharose 4B for 2 hours at 22°C with 5
mmol/L DFP in 0.01 mol/L Tris HCl, 0.1 mol/L NaCl, and 0.01% Tween 80,
pH 7.4. After incubation the unreacted DFP was removed by extensively
washing the Sepharose 4B with 0.01 mol/L Tris HCl and 0.1 mol/L NaCl,
pH 7.4. Successful inactivation of Sepharose 4B–bound TPA by the DFP
treatment was tested with S-2288. Untreated or treated (80 µL DFP)
TPA–Sepharose 4B (10 µL settled volume) in 0.01 mol/L Tris HCl, 0.1
mol/L NaCl, and 0.01% Tween 80, pH 7.4, was incubated with 20 µL
S-2288 (final concentration, 1 mmol/L) at 37°C in microtiter wells,
and the increase in absorbance at 405 nm was recorded over time.
These experiments revealed that the amidolytic activity of DFP-treated
TPA–Sepharose 4B was <1% of the untreated control.
In Vivo Labeling of HUVECS With
[35S]Na2SO4 and Preparation of
Cell Extracts
HUVECs were isolated by mild collagenase treatment
as described.40 Cells were seeded in gelatin-coated
(Bio-Rad) 225-cm2 flasks and grown to confluence in M199
(Sigma) containing 20% supplemented calf serum (Hyclone); 50 IU/mL
penicillin, 50 µg/mL streptomycin, and 250 ng/mL amphotericin B (JHR
Biosciences); 25 µg/mL endothelial cell growth
supplement41 ; and 40 µg/mL heparin (Liquemin,
Hoffmann–La Roche). All HUVECs in this study were between passages 2
and 4. At least 3 days after reaching confluence, HUVECs were incubated
with carrier-free [35S]Na2SO4
(DuPont–New England Nuclear; final concentration, 50 µCi/mL) in
MgSO4-deficient M199 supplemented with 200 mg/L
MgCl2, 1 mg/L MgSO4, 20%
supplemented calf serum, 50 IU/mL penicillin, 50 µg/mL streptomycin,
250 ng/mL amphotericin B, 25 µg/mL endothelial cell
growth supplement, and 20 µg/mL heparin (20 mL per flask). After 48
hours the conditioned medium was removed and cells were rinsed with
Hanks' balanced salt solution containing 10 mmol/L HEPES. As judged by
light microscopy the labeling procedure did not appear to have affected
cell morphology. Cell extracts were prepared by incubating the EC
monolayers with 4 mol/L guanidine HCl, 2% Triton X-100, 100 mmol/L
EACA, 5 mmol/L PMSF, 5 mmol/L benzamidine, 10 KIU/mL aprotinin, 50
mmol/L EDTA, and 50 mmol/L sodium acetate, pH 5.7 (16 mL per flask) for
at least 2 hours at 22°C on an orbital shaker.33 The
resulting cell extracts were dialyzed extensively against 0.01 mol/L
Tris HCl buffer, pH 7.4, containing 0.01 mol/L NaCl, 0.01% Tween 80,
10 mmol/L benzamidine, and 10 KIU/mL aprotinin in a dialysis membrane
with a cutoff of 10 000.
Isolation of Metabolically Labeled TPA-Binding PGs From
HUVEC Extracts
Dialyzed cell extracts were centrifuged (20 minutes,
1100g), and aliquots (8 or 16 mL) of the resulting
supernatants were incubated and rotated end over end with 16 mL
DFP-treated TPA–Sepharose 4B or 16 mL DFP-treated BSA–Sepharose 4B (8
mL settled volume each), respectively, in 0.01 mol/L Tris HCl buffer,
pH 7.4, containing 0.1 mol/L NaCl, 0.01% Tween 80, 10 mmol/L
benzamidine, and 10 KIU/mL aprotinin (buffer A) in a total volume of 24
or 32 mL, respectively. After a 1-hour incubation the suspensions were
applied to small columns and washed extensively with buffer A at a flow
rate of 1 mL/min. Elution of bound material was performed with 2 mol/L
NaCl in 0.01 mol/L Tris HCl buffer, 0.01% Tween 80, 10 mmol/L
benzamidine, and 10 KIU/mL aprotinin, pH 7.4, at a flow rate of 1
mL/min, and fractions (0.8 mL each) were collected. From a 50-µL
aliquot of each fraction, the radioactivity (in counts per minute) due
to 35S was determined in a Beckman LS 7500 liquid
scintillation counter with ReadyGel (Beckman) as the scintillation
fluid. Fractions that contained 35S radioactivity were
eluted with 2 mol/L NaCl, pooled, and dialyzed against buffer A. This
material was called TPA-BPGs and stored in aliquots at -70°C in
either unconcentrated (800 cpm/100 µL) or concentrated (fivefold
to sevenfold with Centricon 10 ultrafiltration membranes) form.
Binding of 35S-Labeled TPA-BPGs to DFP-Treated
TPA–Sepharose 4B in the Presence of Different Agents
Aliquots (200 µL each) of 35S–TPA-BPGs (800
cpm/100 µL) were incubated and rotated end over end with DFP-treated
TPA– or BSA–Sepharose 4B (100 µL settled volume each) in the
presence of either 1 mol/L NaCl, 1 mg/mL BSA, 1 mg/mL heparin, 100
mmol/L Arg, 100 mmol/L Lys, 100 mmol/L Gly, 100 mmol/L EACA, 80 mmol/L
Asp (final concentrations) in buffer A or buffer A alone in total
volumes of 500 µL each for 1 hour at 22°C. The supernatants were
removed after centrifugation (4 minutes,
250g) and the Sepharose beads washed twice with buffer A
(400 µL each). Elution was performed twice with 400 µL each of
buffer A containing 2 mol/L NaCl instead of 0.1 mol/L NaCl. All elution
fractions and the Sepharose beads were counted for 35S
radioactivity. In this and all other binding experiments described
below, the radioactivity in the supernatants of the incubation mixture
and the radioactivity in the wash fluids were summed and designated as
unbound radioactivity. The radioactivity in the elution fluids and the
radioactivity on Sepharose 4B after the two elution steps were summed
and designated as bound radioactivity. The remaining 35S
radioactivity associated with DFP-treated TPA–Sepharose 4B after the
two elution steps was judged to represent specifically bound
radioactivity, because in controls with BSA–Sepharose 4B, the
elution-resistant bound radioactivity was always <16% of
the total applied radioactivity. Furthermore, the
elution-resistant radioactivity was lower whenever total
binding was lower (not shown).
Elution of Bound TPA-BPGs From DFP-Treated TPA–Sepharose 4B With
Different Agents
Aliquots of TPA-BPGs (200 µL, 800 cpm/100 µL) were incubated
with 300 µL DFP-treated TPA–Sepharose 4B (100 µL settled volume
each) in buffer A at 22°C. After 1 hour the supernatants were removed
and the Sepharose beads washed twice with buffer A (400 µL each) as
described above. Then 500 µL buffer A containing one of the following
agents was added to the washed Sepharose beads: NaCl (1 mol/L), BSA (1
mg/mL), heparin (1 mg/mL), Arg (100 mmol/L), Lys (100 mmol/L), Gly (100
mmol/L), EACA (100 mmol/L), or Asp (80 mmol/L). After a 3-minute
incubation the samples were centrifuged (4 minutes,
250g) and the supernatants collected and counted for
35S radioactivity. This elution step was repeated, and then
the TPA–Sepharose 4B beads were eluted twice with 400 µL of 0.01
mol/L Tris HCl buffer, pH 7.4, containing 2 mol/L NaCl, 0.01% Tween
80, 10 mmol/L benzamidine, and 10 KIU/mL aprotinin as described in the
previous section. The radioactivity in these fractions and that
remaining associated with the Sepharose beads were also determined. The
radioactivity eluted by one of the aforementioned agents and by the two
incubations with 2 mol/L NaCl and that remaining associated with the
Sepharose beads represented the total bound radioactivity
in each experiment. The 35S radioactivity eluted by a
certain agent (eg, heparin) was calculated as a percentage of total
bound radioactivity.
Binding of TPA-BPGs to DFP-Treated TPA–Sepharose 4B After
Pretreatment With GAG-Digesting Enzymes
Aliquots (200 µL) of 35S–TPA-BPGs (800 cpm/100
µL) were treated for 2 hours at 37°C with 20 µL heparitinase,
chondroitinase AC, chondroitinase ABC (5 U/mL each in buffer A), or an
equal volume of buffer A and incubated by end-over-end rotation
with 300 µL DFP-treated TPA–Sepharose 4B suspension (100 µL
settled volume) in the same buffer for 1 hour at 22°C. After
centrifugation (4 minutes, 250g) and removal
of the supernatant, each sample was washed twice with buffer A (400
µL each) and then eluted twice (400 µL each) with the same buffer
containing 2 mol/L instead of 0.1 mol/L NaCl. The remaining
TPA–Sepharose 4B beads (100 µL) were suspended in 400 µL buffer A.
All collected fractions (ie, the supernatant after incubation, wash
fluids, elution fluids, and the remaining Sepharose 4B) were counted
for 35S radioactivity. To ensure that the presence of
GAG-digesting enzymes during incubation of the
35S–TPA-BPGs with DFP-treated TPA–Sepharose 4B had not
affected its binding properties, control experiments were performed by
incubating 20 µL of either buffer or each GAG-digesting enzyme (5
U/mL each) with 200 µL buffer for 2 hours at 37°C. To each sample
300 µL DFP-treated TPA–Sepharose 4B (100 µL settled volume) was
added, incubated for 1 hour at 22°C, and washed. Binding of
35S-labeled TPA-BPGs was studied as described above. These
control experiments revealed that the binding properties of the
DFP-inactivated TPA–Sepharose 4B that had been pretreated
with heparitinase and chondroitinase AC were the same as those of
untreated TPA–Sepharose 4B. Only chondroitinase ABC treatment reduced
the subsequent binding of 35S–TPA-BPGs to DFP-treated
TPA–Sepharose 4B by 10%.
Nitrous Acid Degradation
Degradation with nitrous acid42 was performed by
incubating 200 µL 35S–TPA-BPGs (800 cpm/100 µL of
buffer A) in 40 µL of 3 mol/L NaNO2 and 40 µL acetic
acid for 80 minutes at 22°C. Excess nitrous acid was neutralized by
adding 100 µL of 3 mol/L Gly for 1 hour at 22°C. In a control
experiment 200 µL 35S–TPA-BPGs was incubated with 180
µL buffer A for the same time. Both control and acid-degraded
samples were dialyzed extensively against buffer A, and a 50-µL
aliquot of each sample was counted for 35S radioactivity.
The remaining portions of the samples were incubated with 200 µL
DFP-treated TPA–Sepharose 4B (100 µL settled volume) for 1 hour at
22°C. The supernatants were collected after
centrifugation (4 minutes, 250g) and the
Sepharose beads washed twice with 400 µL buffer A. Elution was
performed twice by adding 400 µL of 0.01 mol/L Tris HCl buffer, pH
7.4, containing 2 mol/L NaCl, 0.01% Tween 80, 10 mmol/L benzamidine,
and 10 KIU/mL aprotinin. The remaining 100 µL DFP-treated
TPA–Sepharose 4B beads were suspended in 400 µL buffer A, and bound
and unbound radioactivity values were determined as described in the
previous sections.
Statistical Analysis
Binding and elution data were subjected to one-way ANOVA
(Instat, GraphPad Software version 2.04a). Calculated means for each
binding or elution condition were compared with those of their
respective controls by Dunnett's multiple comparisons test.
PAGE
PAGE (2.75% acrylamide) was performed according to
the procedures of Weber and Osborn43 and Davies and
Stark44 as described (technical bulletin No. MWS-877X,
Sigma). The gels were stabilized by including 0.75% low-melting
agarose and then stained (with silver) for protein,45
photographed, destained, soaked in autoradiography
enhancer (Enlightning, DuPont–New England Nuclear), and dried.
Autoradiography was performed with Kodak X-Omat AR
films.
For detection of GAGs after reductive elimination, gels containing 10% acrylamide, 0.3% bisacrylamide, 0.1 mol/L Tris borate, and 1 mmol/L Na2EDTA, pH 8.3, were prepared as described46 and electrophoresed at 400 V (25 mA) until the tracking dye (0.02% bromphenol blue, 2 mol/L sucrose in Tris-borate-EDTA buffer) had migrated approximately half the length of the gel. Then the gels were fixed, stained with 0.5% Alcian Blue 8 GX in 2% acetic acid for 2 hours, destained overnight in 2% acetic acid, soaked in autoradiography enhancer (Enlightning), and dried. Autoradiography was performed as described above.
PAGE Analysis of TPA-BPGs: Effect of Treatment With
GAG-Digesting Enzymes
Aliquots (120 µL each) of concentrated
35S–TPA-BPGs (4000 cpm/100 µL) were incubated with 13
µL of either buffer A or heparitinase (5 U/mL), chondroitinase AC (5
U/mL), or chondroitinase ABC (5 U/mL) dissolved in the same buffer.
After a 3-hour incubation at 37°C, 50 mg urea, 14 µL 1% SDS, and
14 µL of 0.05% bromphenol blue were added to each sample and
incubated for an additional hour at 37°C; thereafter the samples were
subjected to 2.75% acrylamide gel electrophoresis that had
been stabilized with 0.75% low-melting agarose (see above).
PAGE Analysis of TPA-BPGs Treated With GAG-Digesting
Enzymes Followed by Reductive Elimination
Aliquots (60 µL each) of concentrated
35S–TPA-BPGs (5400 cpm/100 µL) were incubated with 7
µL of either buffer or GAG-digesting enzyme (heparitinase,
chondroitinase AC, or chondroitinase ABC; 5 U/mL each) as described
above. Then the samples were incubated with 7.4 µL of 2 mol/L KOH at
22°C. After 3 hours 13 µL of 0.1 mol/L NaOH and 200 µmol/L
NaBH4 were added; after incubation for 17 hours at 22°C
the mixture was neutralized by adding 2 µL of 50% acetic acid. Then
25 µL of 2 mol/L sucrose in Tris-borate-EDTA buffer was added and
the samples subjected to 10% PAGE as described above. One control
sample was neither treated with GAG-digesting enzymes nor subjected to
reductive elimination.
To confirm the specificity of the GAG-digesting enzymes, purified commercial heparan sulfate, chondroitin sulfate A, and chondroitin sulfate C were treated for 3 hours with heparitinase, chondroitinase AC, and chondroitinase ABC and then analyzed on Alcian Blue 8 GX–stained 10% PAGE gels as described above.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cell extracts were prepared from [35S]Na2SO4-labeled cultured HUVECs and subjected to affinity chromatography on DFP-inactivated TPA–Sepharose 4B or DFP-treated BSA–Sepharose 4B. Approximately 6% of the 35S-labeled, dialyzed material in the cell extracts bound to immobilized TPA and was eluted with 2 mol/L NaCl, while under the same conditions no 35S radioactivity bound to BSA–Sepharose 4B (Fig 1
). According to the results obtained by further
characterization, the 35S-labeled material that eluted from
DFP-inactivated TPA–Sepharose 4B was designated
TPA-BPGs.
|
) or BSA–Sepharose 4B (
) as described in
"Methods." Columns were washed and elution was performed with 2
mol/L NaCl. Fraction volumes were
To characterize this material, its interaction with DFP-treated
TPA–Sepharose 4B was analyzed under different conditions.
Because of the limited amount of material available, these experiments
were performed by batch adsorption and elution. To exclude nonspecific
binding, incubations were also performed with DFP-treated
BSA–Sepharose 4B and values derived therefrom were used for
nonspecific binding. Specific binding was the difference between
binding to DFP-treated TPA–Sepharose 4B and binding to DFP-treated
BSA–Sepharose4B. As shown in Table 1 binding to
DFP-treated BSA–Sepharose 4B accounted for 15% to 28% of the applied
radioactivity and was unaffected by the substances shown in Table 1.
Fifty-one percent of the applied 35S radioactivity
bound specifically to DFP-treated TPA–Sepharose 4B. The presence of
NaCl (1 mol/L) or heparin (1 mg/mL) completely suppressed specific
binding. Lys and Arg (100 mmol/L each) reduced specific binding to
25%, whereas EACA (100 mmol/L), Asp (80 mmol/L), BSA (1 mg/mL), and
Gly (100 mmol/L) did not significantly reduce binding of
35S radioactivity.
|
The substances used for the competitive-binding experiments between
TPA-BPGs and DFP-treated TPA–Sepharose 4B were also tested for their
ability to elute bound TPA-BPGs from DFP-treated TPA-Sepharose 4B
(Table 2). In addition to high concentrations of salt,
heparin, Arg, and Lys also eluted bound 35S-labeled
material from DFP-treated TPA–Sepharose 4B, although statistical
analysis revealed that only heparin (P<.01), NaCl
(P<.05), and Arg (P<.05) eluted significantly
greater amounts of radioactivity than did buffer A alone. Although only
slightly lower than the amount eluted by NaCl and Arg, the amount of
radioactivity eluted by Lys failed to reach statistical
significance.
|
Bound and unbound radioactivity values for 35S-labeled
TPA-BPGs that had been enzymatically digested and incubated with
DFP-treated TPA–Sepharose 4B were determined as described in
"Methods." As shown in Table 3, 72% of TPA-BPGs
bound to DFP-treated TPA–Sepharose 4B under control conditions.
Heparitinase treatment reduced this binding to 41%, whereas
chondroitinase AC and chondroitinase ABC were less efficient in
reducing this binding. With all GAG-digesting enzymes the decreases in
binding were statistically significant (P<.01).
|
TPA-BPGs were also treated with nitrous acid to cleave heparin and
heparan sulfate components. As shown in Table 3
nitrous acid
degradation decreased binding to 12.5% compared with 50% in
respective control experiments. This low binding percentage in control
nitrous acid degradation experiments may be due to alterations in the
GAGs caused by the time-consuming preparation procedures.
PAGE of 35S–TPA-BPGs with an acrylamide
concentration of 2.75% (Fig 2A) consistently
revealed two signals, one with an Mr of 600 000
to 750,000 (TPA-BPGA) and the other with an Mr
of 120 000 to 180 000 (TPA-BPGC). In some experiments an additional
signal was seen between 340 000 and 440 000 and was designated
TPA-BPGB. Incubation of the isolated material with heparitinase reduced
the intensities of TPA-BPGA and TPA-BPGB (not shown), indicating the
presence of heparan sulfate–like structures in these entities.
Chondroitinase ABC treatment abolished the TPA-BPGC band, suggesting
that chondroitin sulfate– as well as heparan sulfate–like GAGs
were present in the TPA-BPGs. Chondroitinase AC treatment (not
shown) had the same effect as chondroitinase ABC treatment.
|
To confirm the PG nature of the 35S-labeled material that
had eluted from DFP-treated TPA–Sepharose 4B, reductive elimination
was performed to separate the possible carbohydrate components from the
core proteins. These experiments were done with untreated and
enzymatically digested material. The resulting cleavage products
were analyzed by PAGE according to the method of Cowman et
al,46 followed by autoradiography. The
results (Fig 2B) show that without enzyme pretreatment and reductive
elimination (lane 0), most of the 35S radioactivity did not
migrate into the gel but remained at the top. After reductive
elimination (lane C) the radioactive signal on top of the gel
disappeared and two new broad radioactive bands appeared in the upper
part of the gel, corresponding to positions HS and CS. After
heparitinase pretreatment (lane H) the radioactive band at position HS
disappeared while the intensity of the band at position CS was
unchanged, indicating that the band at position HS was mainly heparan
sulfate. Chondroitinase ABC pretreatment had no effect on the signal at
position HS but led to the disappearance of the signal at position CS,
indicating that this band contained mainly chondroitin sulfate. Control
experiments (Fig 2C) revealed that heparitinase (lane H) degraded
heparan sulfate exclusively, whereas chondroitinase AC (lane AC) and
chondroitinase ABC (lane ABC) digested chondroitin sulfate A and
chondroitin sulfate C but not heparan sulfate. These data suggest that
the HUVEC-derived, metabolically labeled material that had
eluted from DFP-treated TPA–Sepharose 4B was in fact composed of PGs
and that the predominant GAGs were heparan sulfate and chondroitin
sulfate.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In this study TPA-BPGs on HUVECs were analyzed by [35S]Na2SO4 in vivo labeling of sulfate-containing components, and in vivo–labeled, dialyzed HUVEC extracts were subjected to affinity chromatography on DFP-inactivated TPA–Sepharose 4B. DFP-inactivated TPA–Sepharose 4B was chosen because it excludes the binding of cellular components (eg, PAI-1) to the active site of TPA. Furthermore, coupling of TPA to CNBr-activated Sepharose 4B was performed in such a way that the heparin-binding domains of TPA remained accessible to the immobilized protein.39 Two high-Mr 35S-labeled entities (TPA-BPGA with an Mr of 600 000 to 750 000 and TPA-BPGC with an Mr of 120 000 to 180 000) bound consistently to DFP-treated TPA–Sepharose 4B and were eluted with 2 mol/L NaCl. In some experiments an additional faint band (Mr of 340 000 to 440 000 [TPA-BPGB])was also observed. Heparin, Arg, Lys, and solutions of high salt concentration competed with 35S-labeled EC-derived material for binding sites on DFP-treated TPA–Sepharose 4B and were able to elute the latter, whereas Gly, Asp, and EACA had no effect. These data support the hypothesis that cellular GAGs are involved in binding. Heparin was the most efficient agent for eluting bound 35S radioactivity from DFP-treated TPA–Sepharose 4B and for competing with 35S radioactivity for binding sites on DFP-treated TPA–Sepharose 4B, suggesting that heparin and EC extract isolates share related binding domains on the TPA molecule, which for heparin are on the finger and kringle 2 domains.28
Treatment of TPA-BPGs with GAG-digesting enzymes revealed that
TPA-BPG (A and B) were sensitive to heparitinase treatment, whereas
TPA-BPGC was digested by chondroitinases AC and ABC. The data in Fig 2B
confirm the PG nature of the 35S-labeled HUVEC
material eluted from DFP-treated TPA–Sepharose 4B, because reductive
elimination resulted in separation of 35S radioactivity
from (a) high-Mr component(s) and formation of
new, lower-Mr radioactive bands that were
sensitive to heparitinase and chondroitinase treatment.
Isolated TPA-BPGs treated with GAG-digesting enzymes were also
analyzed for their binding behavior to DFP-treated
TPA–Sepharose 4B. After treatment with heparitinase, chondroitinase
AC, and chondroitinase ABC, binding was markedly reduced. Nitrous acid
degradation, a procedure that specifically destroys heparin and heparan
sulfate, almost completely abolished binding of TPA-BPGs to DFP-treated
TPA–Sepharose 4B. After nitrous acid degradation only 12.5% of the
added radioactivity bound to DFP-treated TPA–Sepharose 4B, an amount
similar to that in control experiments with DFP-treated BSA–Sepharose
4B. This value was lower than expected, because the data in Fig 2
suggested that TPA-BPGC contained chondroitin sulfate GAGs, which is
not cleaved by nitrous acid (not shown). However, this does not exclude
the possibility that nitrous acid might have changed the structure in
such a way that the isolated PG lost its affinity for TPA.
From our data we cannot conclude that specific GAG sequences are
responsible for the interaction between these BPGs and TPA. It is also
possible that the binding of cellular GAGs to TPA is caused simply by
nonspecific ionic interactions. As shown in Fig 1 only 6% of the
high-Mr 35S radioactivity in HUVEC
extracts bound to DFP-treated TPA–Sepharose 4B. Such a low percentage
of binding might suggest a requirement for specific GAG sequences.
However, it is also possible that under our experimental conditions,
heparin competes with metabolically labeled PGs for binding
to TPA, and heparin is present in the culture media and also binds
to HUVECs. Additional experiments to analyze the carbohydrate
composition of the TPA-BPGs might answer the question about GAG
specificity for TPA binding to endothelial PGs.
In this report we have shown that ECs synthesize specific heparan sulfate– and chondroitin sulfate–containing PGs to which TPA can bind, most likely via its heparin-binding sites. Although our data do not allow us to localize these PGs to a specific cellular compartment, heparan sulfate–containing PGs are known to be present on the EC surface,47 making it likely that such TPA-binding PGs may also be found on the luminal surfaces of vascular ECs. This large TPA-binding capacity might in fact represent a TPA storage pool, from which TPA is released and made locally available under certain circumstances. TPA might then be transferred to locally formed fibrin along its affinity gradient and thus provide the profibrinolytic capacity of ECs.48 On the other hand, ECs could also shed their GAG coat after cell activation,49 which would result in the release of GAGs and bound TPA. The latter mechanism might be operative in situations involving EC injury, such as hypoxia during venous occlusion4 or receptor-induced EC activation via, eg, histamine or thrombin.8 9 In fact, such a mechanism might be responsible for the increases in plasma TPA that are often observed after venous occlusion. Furthermore, as heparin-bound TPA has been shown to elaborate higher plasminogen-activating activity,26 28 heparan sulfate–bound TPA released from ECs would provide an additional mechanism for locally increased fibrinolytic activity and render ECs resistant to fibrin deposition.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported in part by grants P9478-M and P10435-M from the Austrian Science Foundation.
Received May 4, 1995; accepted February 1, 1996.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Bachmann F. Molecular aspects of plasminogen, plasminogen activators and plasmin. In: Bloom AL, Forbes CD, Thomas DP, Tuddenham EGD, eds. Haemostasis and Thrombosis, Volume 1. Edinburgh, Scotland: Churchill Livingstone; 1994:575-613.
2. Collen D, Lijnen HR. Fibrin-specific fibrinolysis. In: Brakman P, Kluft C, eds. Plasminogen Activation in Fibrinolysis, in Tissue Remodeling, and in Development: Annals of the New York Academy of Sciences, Volume 667. New York, NY: New York Academy of Sciences; 1992:259-271.
3. Murphy G, Atkinson S, Ward R, Gavrilovic J, Reynolds JJ. The role of plasminogen activators in the regulation of connective tissue metalloproteinases. In: Brakman P, Kluft C, eds. Plasminogen Activation in Fibrinolysis, in Tissue Remodeling, and in Development: Annals of the New York Academy of Sciences, Volume 667. New York, NY: New York Academy of Sciences; 1992:1-12.
4. Emeis JJ. Regulation of the acute release of tissue-type plasminogen activator from the endothelium by coagulation activation products. In: Brakman P, Kluft C, eds. Plasminogen Activation in Fibrinolysis, in Tissue Remodeling, and in Development: Annals of the New York Academy of Sciences, Volume 667. New York, NY: New York Academy of Sciences; 1992:249-258.
5. Stadler M, Hauert J, Kruithof EK, Bachmann F. Release of vascular plasminogen activator (v-PA) after venous stasis: electrophoretic-zymographic analysis of free and complexed v-PA. Br J Haematol. 1985;61:169-176. [Medline] [Order article via Infotrieve]
6. Agnelli G, Levi M, Cosmi B, ten Cate JW, Nenci GG. Additive effect of dDAVP and standard heparin in increasing plasma t-PA. Thromb Haemost. 1989;61:507-510. [Medline] [Order article via Infotrieve]
7. Nilson IM, Holmberg L, Aberg M, Vilhardt H. The release of plasminogen activator and factor VIII after injection of dDAVP in healthy volunteers and in patients with von Willebrand's disease. Scand J Haematol. 1980;24:351-359.
8. Levin EG, Marzec U, Anderson J, Harker LA. Thrombin stimulates tissue plasminogen activator release from cultured human endothelial cells. J Clin Invest. 1984;74:1988-1995.
9. Hans M, Collen D. Secretion of tissue type plasminogen activator and plasminogen activator inhibitor by cultured human endothelial cells: modulation by thrombin, endotoxin, and histamine. J Lab Clin Med. 1987;109:97-104. [Medline] [Order article via Infotrieve]
10. Agnelli G, Borm J, Cosmi B, Levi M, ten Cate JW. Effect of standard heparin and low molecular weight heparin (Kabi 2165) on fibrinolysis. Thromb Haemost. 1988;60:311-313. [Medline] [Order article via Infotrieve]
11. Huber K, Resch I, Rosc D, Probst P, Kaindl F, Binder BR. Heparin induced increase of t-PA antigen plasma levels in patients with unstable angina: no evidence for clinical benefit of heparinization during the initial phase of treatment. Thromb Res. 1989;55:779-784. [Medline] [Order article via Infotrieve]
12. Beebe DP. Binding of tissue plasminogen activator to human umbilical vein endothelial cells. Thromb Res. 1987;46:241-254. [Medline] [Order article via Infotrieve]
13. Hajjar KA, Hamel NM, Harpel PC, Nachman RL. Binding of tissue plasminogen activator to human endothelial cells. J Clin Invest. 1987;80:1712-1719.
14.
Barnathan ES, Kuo A, Van der Keyl H, McCrae R, Larsen
GR, Cines DB. Tissue-type plasminogen
activator binding to human endothelial
cells: evidence for two distinct binding sites. J
Biol Chem. 1988;263:7792-7799.
15.
Russell ME, Quertermous T, Declerk PJ, Collen D, Haber
E, Homcy CJ. Binding of tissue-type plasminogen
activator with human endothelial cell
monolayers: characterization of the high affinity interaction with
plasminogen activator
inhibitor-1. J Biol Chem. 1990;265:2569-2575.
16.
Ramakrishnan V, Sincropi DV, Dere R, Darbonne WC,
Bechtol KB, Baker JB. Interaction of wild-type and
catalytically inactive mutant forms of tissue plasminogen
activator with human umbilical vein
endothelial cell monolayers. J
Biol Chem. 1990;265:2755-2762.
17.
Hajjar KA, Hamel NM. Identification and
characterization of the human endothelial cell binding
sites for tissue plasminogen activator and
urokinase. J Biol Chem. 1990;265:2908-2916.
18.
Hajjar KA. The endothelial cell
tissue plasminogen activator receptor: specific
interaction with plasminogen. J
Biol Chem. 1991;266:21962-21970.
19.
Felez J, Chanquia CA, Levin EG, Miles LA, Plow
EF. Binding of tissue plasminogen
activator to human monocytes and monocytoid cells.
Blood. 1991;78:2318-2327.
20.
Sakata Y, Okada M, Noro A, Matsuda M.
Interaction of tissue-type plasminogen
activator and plasminogen activator
inhibitor 1 on the surface of endothelial
cells. J Biol Chem. 1988;263:1960-1969.
21.
Schleef RR, Podor TJ, Dunne E, Mimuro J, Loskutoff
DJ. The majority of type 1 plasminogen
activator inhibitor associated with cultured
human endothelial cells is located under the cells and
is accessible to solution-phase tissue-plasminogen
activator. J Cell Biol. 1990;110:155-163.
22.
Hajjar KA, Jacovina AT, Chacko J. An
endothelial cell receptor for
plasminogen/tissue plasminogen
activator, I: identity with annexin II.
J Biol Chem. 1994;269:21191-21197.
23.
Cesarman GM, Guevara CA, Hajjar KA. An
endothelial cell receptor for
plasminogen/tissue plasminogen
activator (t-PA), II: annexin II mediated enhancement of
t-PA-dependent plasminogen activation.
J Biol Chem. 1994;269:21198-21203.
24.
Binder BR, Spragg J, Austen KF. Purification and
characterization of human vascular plasminogen
activator derived from blood vessel
perfusates. J Biol Chem. 1979;254:1998-2003.
25. Ecke S, Geiger M, Binder BR. Glycosaminoglycans specifically stimulate the inhibition of 55kD but not of 33kD urokinase by protein C inhibitor. Ann Hematol. 1992;64(suppl):A64. Abstract.
26. Pâques EP, Stöhr HA, Heimburger N. Study on the mechanism of action of heparin and related substances on the fibrinolytic system: relationship between plasminogen activators and heparin. Thromb Res. 1986;42:797-807. [Medline] [Order article via Infotrieve]
27. Dosne AM, Bendetowicz AV, Kher A, Samama M. Marked potentiation of the plasminogenolytic activity of pro-urokinase by unfractionated heparin and a low molecular weight heparin. Thromb Res. 1988;52:627-630.
28.
Stein PL, van Zonneveld A-J, Pannekoek H, Strickland
S. Structural domains of human tissue-type
plasminogen activator that confer stimulation
by heparin. J Biol Chem. 1989;264:15441-15444.
29. Edelberg JM, Pizzo SV. Kinetic analysis of the effects of heparin and lipoproteins on tissue plasminogen activator mediated plasminogen activation. Biochemistry. 1990;29:5906-5911. [Medline] [Order article via Infotrieve]
30. Andrade-Gordon P, Strickland S. Interaction of heparin with plasminogen activators and plasminogen: effects on the activation of plasminogen. Biochemistry. 1986;25:4033-4040. [Medline] [Order article via Infotrieve]
31.
Oohira A, Wight TN, Bornstein P. Sulfated
proteoglycans synthesized by vascular endothelial cells
in culture. J Biol Chem. 1983;258:2014-2021.
32. Kjellen L, Lindahl U. Proteoglycans: structures and interactions. Annu Rev Biochem. 1991;60:443-475. [Medline] [Order article via Infotrieve]
33.
Marcum JA, Atha DH, Fritze LMS, Nawroth P, Stern D,
Rosenberg RD. Cloned bovine aortic endothelial
cells synthesize anticoagulantly active heparan sulfate
proteoglycan. J Biol Chem. 1986;261:7507-7517.
34.
Nader HB, Dietrich CP, Buonassisi V, Colburn P.
Heparin sequences in the heparan sulfate chains of an
endothelial cell proteoglycan. Proc Natl
Acad Sci U S A. 1987;84:3565-3569.
35. Olivecrona T, Bengtsson G, Marklund SE, Lindahl U, Höök M. Heparin-lipoprotein lipase interactions. Fed Proc. 1977;36:60-69. [Medline] [Order article via Infotrieve]
36. Rapaport SI. The extrinsic pathway inhibitor: a regulator of tissue factor-dependent blood coagulation. Thromb Haemost. 1991;66:6-15. [Medline] [Order article via Infotrieve]
37. Lindahl AK, Sandset PM, Abildgaard U. The present status of tissue factor pathway inhibitor. Blood Coagul Fibrinolysis. 1992;3:439-449. [Medline] [Order article via Infotrieve]
38.
Rosenfeld L, Kuo A, Hirsh J, Klugherz B, Gardell SJ,
Cines DB, Barnathan ES. Heparin enhances active
site-dependent binding of tissue-type plasminogen
activator to endothelial cells.
Blood. 1992;80:1486-1495.
39.
Andrade-Gordon P, Strickland S. Fractionation of
heparin by chromatography on a tissue
plasminogen activator-sepharose
column. Proc Natl Acad Sci U S A. 1990;87:1865-1869.
40. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical cord veins: identification by morphologic and immunologic criteria. J Clin Invest. 1973;52:2745-2756.
41.
Gospodarowicz J. Purification of a fibroblast
growth factor from bovine pituitary. J Biol
Chem. 1975;250:2515-2520.
42. Shively JE, Conrad HE. Formation of anhydrosugars in the chemical depolymerization of heparin. Biochemistry. 1976;15:3932-3942. [Medline] [Order article via Infotrieve]
43.
Weber K, Osborn M. The reliability of molecular
weight determination by dodecyl
sulfate-polyacrylamide gel electrophoresis.
J Biol Chem. 1969;244:4406-4412.
44.
Davies G, Stark G. Use of suberimidate, a
cross-linking reagent, in studying the subunit structure of
oligomeric proteins. Proc Natl Acad Sci U S A. 1970;66:651-659.
45. Morrissey JH. Silver stain for proteins in polyacrylamide gels: a modified procedure with enhanced uniform sensitivity. Anal Biochem. 1981;117:307-310. [Medline] [Order article via Infotrieve]
46. Cowman MK, Slahetka MF, Hittner DM, Kim J, Forino M, Gadelrab G. Polyacrylamide-gel electrophoresis and alcian blue staining of sulfated glycosaminoglycan oligosaccharides. Biochem J. 1984;221:707-716. [Medline] [Order article via Infotrieve]
47. Lowe-Krentz LJ, Joyce JG. Venous and aortic porcine endothelial cells cultured under standardized conditions synthesize heparan sulfate chains which differ in charge. Anal Biochem. 1991;193:155-163. [Medline] [Order article via Infotrieve]
48. Speiser W, Anders E, Binder BR, Müller-Berghaus G. Clot lysis mediated by cultured human microvascular endothelial cells. Thromb Haemost. 1988;60:463-467. [Medline] [Order article via Infotrieve]
49. Silvestro L, Ruikun C, Sommer F, Minh Duc T, Biancone L, Montrucchio G, Camussi G. Platelet-activating factor-induced endothelial cell expression of adhesion molecules and modulation of surface glycocalix, evaluated by electron spectroscopy chemical analysis. Semin Thromb Hemost. 1994;20:214-222.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  B. P. Schick, J. F. Gradowski, and J. D. S. Antonio Synthesis, secretion, and subcellular localization of serglycin proteoglycan in human endothelial cells Blood, January 15, 2001; 97(2): 449 - 458. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||