1. No category

From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
1994 84: 1164-1172
Low-affinity heparin stimulates the inactivation of plasminogen
activator inhibitor-1 by thrombin
PA Patston and M Schapira
Updated information and services can be found at:
http://www.bloodjournal.org/content/84/4/1164.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
Low-Affinity Heparin Stimulates the Inactivation of Plasminogen Activator
Inhibitor-l by Thrombin
By Philip A. Patston and Marc Schapira
The influence of heparin on the reaction between thrombin
and plasminogen activator inhibitor-l (PAI-1) has been examined. With a 50-fold excess of PAI-1, the rate constant
for the inhibition of thrombin was 458 mol/L"s",
which
increased to 5,000 mol/L"s" in the presence of 25 pg/mL
unfractionated heparin or heparin with lowaffinity for antithrombin. The effect of low affinity heparin was then examined by sodium dodecyl sulfate-polyacrylamide gelelectrophoresis, using close t o equimolar concentrations of
reactants. Thrombin and PAI-1formed a stablestoichiometric complex in theabsence of heparin, which did notdissociate after the addition of 25 pg/mL low-affinity heparin. In
contrast, when low-affinity heparin was added at the beginning of the reaction, there was an initial increase in PAI-1thrombin complex formation, but thiswas rapidly followed
by substantial proteolytic cleavage of unreacted PAL1 and
of the thrombin-PAL1 complex. The idea that the relative
concentrations of thrombin and PAI-1, and the presence of
low affinity heparin, could influencethe products of the reaction was examinedin detail. Quantitative zymographic analysis of tissue plasminogenactivator and PAL1activities and
chromogenic substrate assay of thrombin activity showed
that low-affinity heparin stimulated the inactivation of PAI1 by an equimolar amount of thrombin, but caused only a
minimal stimulation of thrombin inhibition. It is concluded
thatlow-affinity
heparin stimulates thrombininhibition
when PAL1 is in excess, but, unexpectedly,that low-affinity
heparin enhances PAL1inactivation when thrombin is equimolar to PAI-1.
0 1994 by The American Society of Hematology.
H
bleeding caused by the anticoagulant action of heparin on
any profibrinolytic action.ls
A possible mechanism for the profibrinolytic action of
heparin is activation of the contact system-dependent fibrinolytic pathway,lg but studies with plasma have yielded
ambiguous results.20321
The possibility that heparin might
have a fibrinolytic effect through modulating plasminogen
activator inhibitor-l (PAI-1) activity was suggested by the
in vivo study of Agnelli et al." PAI- 1, a proteinase inhibitor
of the serpin family, is the primary inhibitor of tissue plasminogen activator (t-PA),22.23and as such plays a critical
role in the regulation of fibrinolysis. The importance of maintaining the correct levels of PAL1 is demonstrated by the
association of disease states with both ele~ated'~.'~
and decreased26PAL1 levels. Further evidence for an involvement
of PAI- l in heparin-induced fibrinolysis has come from studies showing that heparin enhances the rate of thrombin inhibition by PAI-l.27.28Vitronectin also enhances the rate of
thrombin inhibition by PAI-lz8.''; however, thrombin has
also been shown to inactivate PAI-1 associated with
vitronectin in the extracellular matrix.30In the present study,
we provide evidence that heparin with low affinity for antithrombin stimulates PAI-1 neutralization by thrombin.
EPAFUN IS A HIGHLY sulfated glycosaminoglycan
consisting of alternating glucosamine and iduronic or
glucuronic residues. Within this basic structure, considerable
heterogeneity exists with regard to side-chains and extent of
sulfation. Heparin exerts its anticoagulant effect through the
activation of the serpin antithrombin lIl.'-3A specific pentasaccharide sequence: present in about 30% of the heparin
chains,* mediates the high-affinity binding to antithrombin
111. Binding to heparin induces a conformational change at
the reactive center PI arginine residue of antithrombin.6This
conformational change is responsible for the majority of the
enhanced inhibition of factor Xa. In contrast, the majority
of the activation towards thrombin results from a template
or surface approximation mechanism, whereby sequestration
of fluid phase antithrombin and thrombin onto the heparin
increases the local concentration and, hence, the rate of reacti~n.~.~
Bleeding is the major complication of heparin
In addition to its direct anticoagulant action via the antithrombin mechanism, it has been proposed by some investigators that heparin can also stimulate the fibrinolytic systern."," Indications that heparin has effects on the
fibrinolytic system have come from a number of in V ~ V O ' ~ " ~
and animal16,17studies, but such studies are sometimes difficult to interpret because of the potential superposition of
From the Division of Hematology, Department of Medicine, Vanderbilt University, Nashville, TN; and the Division of Hematology,
Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland.
Submitted January IO, 1994; accepted April 21, 1994.
Supported by an American Heart Association-Tennessee AfJiliate
New Investigator Award (P.A.P)and by National Institutes of Health
Grant No. HL-49242 (P.A.P.).
Address reprint requests to Philip A. Patston, DPhil, Division of
Hematology, C 3101 Medical Center North, Vanderbilt University,
Nashville, TN 37232-2287.
The publication costsof this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1994 by The American Society of Hematology.
0006-4971/94/8404-06$3.00/0
1164
MATERIALSANDMETHODS
Materials. The chromogenic substrates H-D-Ile-Pro-Arg-p-NA
(S-2288) and H-D-Phe-Pip-Arg-pNA (S-2238) were from Kabi Diagnostics (Franklin, OH). Phenylmethylsulfonyl fluoride (PMSF)
and chemical reagents were obtained from Sigma (St Louis, MO),
and electrophoretic reagents were from BioRad (Richmond, CA).
Agarose A was from Pharmacia (Piscataway, NJ). Heparin (sodium
salt, porcine) and low molecular weight heparin (molecular weight
[M,] 4,000 to 6,000) were from Sigma. Low- and high-affinity heparin were prepared by fractionation of full-length heparin on antithrombin-Sepharose." Heparin was assayed with Azure A.32
Proteins. Low molecular weight urokinase (human) was obtained from Sigma, and recombinant t-PA (single-chain) was from
Genentech (San Francisco, CA). Recombinant human PAL1 was a
gift from Dr T. Reilly (DuPont, Wilmington, DE)."Human athrombin was kindly supplied by Dr PaulBock (Department of
Pathology, Vanderbilt University, Nashville, TN). Bovine fibrinogen
(type l-S) was from Sigma. Human antithrombin 111 (thrombate 111)
was kindly provided by Cutter-Miles (West Haven, CT), and coupled
Blood, VOI 84, NO 4 (August 15), 1994: pp 1164-1172
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
1165
LOW-AFFINITY HEPARIN AND PAI-1 INACTIVATION
to cyanogen bromide-activated Sepharose-4B (Pharmacia) in the
presence of acetylated heparin."
Kinetic srudies. PAI-I was standardized by titration against tPA using the chromogenic substrate S-2288 at 1.47 mmoVL in 50
mmol/L Tris-HCI, pH 7.4, 127 mmol/L NaCi, at 37°C assuming a
reaction stoichiometry of 1. The recombinant PAI-1 assayed by this
method wasat least 90% active when removed from storage at
-70°C. However, once thawed, the PAI-I reverted to the latent
form.'3 Although latent PAI-I could be activated by treatment with
denaturants, the reactivated form was not used in any of the experiments described here.
For determination of the second-order rate constant of the inhibition of thrombin by PAI-I, thrombin (110 nmolk) was incubated
at23°Cwith
PAI-I (5.2 pmol/L) in the presence or absence of
heparin in a volume of 40 pL, in phosphate-buffered saline (PBS)
containing 1 mg/mL bovine serum albumin (BSA). At various times,
5-pL aliquots were assayed for residual enzymatic activity, using
330 pL of S-2238 (0.15 mmol/L) in 50 m m o K Tris-HCI, pH 7.4,
127 mmoVL NaCI, at 37°C. The absorbance change at 405 nm was
continuously recorded using a Gilford 260 spectrophotometer (CibaComing, Oberlin, OH). Second-order reaction rate constants were
derived from pseudo-first-order plots.'4
Electrophoretic studies. Sodium dodecyl sulfate-polyacrylamide
gel electrophoresis (SDS-PAGE), nonreduced, was performed using
a Bio-Rad Mini gel system by the method of Laem~nli.~'
The acrylamide concentration in stacking gels was 4% and the concentration
present in the separating gels was 12%. Gels were stained with
Coomassie blue R-250. Westem blotting was performed using a
BioRad mini-blot system. After transfer, the nitrocellulose membrane (0.45 pm; Micron Separations Inc, Westboro, MA) was
blocked in 2% BSA (fraction V; Sigma). PAI-1 was detected with
goat anti-HT l080 PAL1 antibody (American Diagnostica, Greenwich, CT) at 1/2,000 dilution, and shown with rabbit antigoat IgGhorseradish peroxidase conjugate (Fisher Biotech, Pittsburgh, PA).
The blot was developed with 0.6 mg/mL 4-chloro-l-naphthol,
0.012% HzOzin 50 mmol/L Tris-HC1, pH 7.5, 80 mmolk NaCl, 2
mmol/L CaCIz, 20% methanol.
Zymogruphicanalysis. Samples containing PAI-l or t-PA were
incubated as described in the text and in the legends to Figs 6 and
7, followed by assay on either zymograms or reverse zymograms as
described below. For the zymograms, fibrin-agarose underlays36(12
mL, poured onto a 7.2 X 10.1 cm glass plate) were prepared with
0.75% (wt/vol) Agarose A (Phmacia), 2.5 mg/mL bovine fibrinogen, 17 ng/mL human a-thrombin, and 0.05% Triton X-100, in 50
mmol/L sodium barbital, pH 7.8, 93 mmol/L sodium chloride, 1.7
mmol/L calcium chloride, and 0.7 m o l / L magnesium chloride.
Five-microliter samples containing t-PA were loaded into wells in
the fibrin-agarose gel. Wells were made with a well puncher. After
overnight incubation in a humid chamber, underlays were washed
with distilled water, dried inan oven at SOT, and stained with
Coomassie blue R-250. The area of the lysis ring was measured and
a linear standard curve of lysis area versus log of t-PA concentration
was constructed. This method could be used to quantitate t-PA from
I pg/5 pL (3.6 pmol/L) down to IO pg/5 pL (36 pmol/L).
Reverse zymograms3' were prepared in the same way as regular
zymograms except for the inclusion of 8.3 ng/mL urokinase in the
agarose underlay. Quantification of PAI-l concentration by reverse
zymography was performed by aliquoting 5 pL of the sample to test
into a hole punched in the gel, in a manner analogous to the quantitative zymography described above. The gel was incubated for 18
hours in a humid chamber, followed by washing, drying, and staining. A standard curve for PAI-I was constructed by measuring the
area of the zone protected from lysis, and plotting the area of the
protected zone against the log of the PAI-I concentration. With this
method there is a linear relationship between these parameters for
PAL1 concentrations ranging from 0.5 pg/5 pL (2.4 FmoVL) to 5
ng/5 pL (24 nmol/L). Control experiments showed that the presence
of heparin or thrombin did not interfere with the activity of t-PA or
PAI-I on the zymograms or reverse zymograms, respectively.
RESULTS
Effect of heparin fractions on the inhibition of thrombin
by PAI-l: chromogenicsubstratestudies.
Under pseudofirst-order conditions, with a 50-fold excess of PAI-l over
thrombin, the second-order rate constant for the inhibition
of thrombin by PAI-1was 458 mol/L"s-l. The effect of
various heparin species on this reaction was determined as
shown in Fig 1. A maximum rate constant of 5,000 mol/
L"s-l was obtained with unfractionated heparin concentrations between 15 and 45 pglmL (Fig 1, top left). This rate
enhancement decreased at higher concentrations of heparin.
An essentially identical result was seen withlow-affinity
heparin (Fig 1, top right). High-affinity heparin showed the
same dose dependency as unfractionated heparin, but a
slightly higher maximum rate constant of 6,000 mol/L-Is"
was obtained (Fig 1, bottom left). In contrast, the maximal
inhibition rate constant with18 pg/mL of low molecular
weight heparin was only 1,700 mol/L-'s" at (Fig 1, bottom
right).
Effect of low-afinity heparin on the reaction of thrombin
and PAI-I: SDS-PAGE studies. The experiments shown in
Figs 2 through 5 were performed using various PA1:thrombin
molar ratios. The effect of the reactant concentration and the
presence of heparin on the outcome of a 15-minute reaction
between thrombin and PAI-1 is shown in Fig 2. The relative
molar amounts of PAI-1 to thrombin were 5:l in lanes 3
and 6, 1:l in lanes 4 and 7, and 1:5 in lanes 5 and 8 (Fig
2). With PAI-1 in excess, the reaction between PAI-1 (M,
42,000) and thrombin (M, 37,000) led to formation of a
bimolecular PAI-l-thrombin complex (Mr 73,000), with no
significant effect of low-affinity heparin on the reaction products (Fig 2, lanes 3 and 6). When thrombin was in excess,
there was significant degradation of PAI-1 and of the PAI1-thrombin complex (Fig 2, lane 5). In the presence of lowaffinity heparin, the degradation became even more profound, with minimal complex remaining undegraded (Fig 2,
lane 8). In addition, at equimolar concentrations of thrombin
and PAI-1 (Fig 2, lanes 4 and 7), significant PAI-l -thrombin
complex formation was seen in the absence of low-affinity
heparin (Fig 2, lane 4),whereas thrombin remained essentially uncomplexed in the presence of low-affinity heparin
(Fig 2, lane 7).
The time course of the reaction between PAL1 and thrombin at equimolar concentrations is shown in Fig 3. PAL1
and thrombin were incubated in the absence (lanes 3 through
6) or the presence (lanes 7 through 10) of low-affinity heparin (25 pg/mL). By comparing lanes 3 and 7 (l-minute time
point), it can be seen that low-affinity heparin enhanced the
initial rate of complex formation, as would be expected from
Fig 1. At the later time points (10 and 20 minutes; Fig 3,
lanes 9 and lo), the presence of low-affinity heparin caused
significant degradation of the complex. Under the conditions
used in this experiment, a significant amount of cleaved PAI-
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
PATSTON AND SCHAPIRA
1166
7000
6000
7000
1
F
I
v)
c
I
loo0 0
4000
I
3000
2000
Y
Y
1
‘
1
10
100 1000
[UFHI ug/ml
;;;;/,pM
1
10
100 1000
.F],=
[LAHI ug/ml
7000
c
I
v)
c
I
I
3000
2000
1000
0
2000
1000
0
Y
Y
3000
2000
0
1
10
100 1000
1
[HAHI ug/ml
10
[LMWHI ug/ml
1 (M, 41,000) was formed with a molecular weight slightly
less than the native form (M, 42,000).
The mechanism of cleavage of the PAL1 -thrombin complex was investigated in Fig 4.The complex was formed in
a 30-minute preincubation with a limiting amount of thrombin(Fig 4, lane 3), and then incubated for a further 15
minutes with either buffer (Fig4,lane 4),low-affinity heparin (Fig 4,lane S ) , thrombin (Fig 4,lane 6), or thrombin and
low-affinity heparin (Fig 4,lane 7). Addition of low-affinity
Fig 2. Effect of reactant concentration and low-affinity heparin on the reaction of PAI-l and
thrombin. Ten-microliter samples were incubated
with PBS at
23°C for l 5 minutes, followed by
mixing with sample buffer containing 1 mmol/L PMSF, run on
12% SDS-PAGE, and
stained
with Coomassie blue R-250.
Lane 1, 1.8 p g PAI-1; lane 2, 1.7
p g thrombin; lanes 3 and 6,9 p g
PAI-1 and 1.7 p g thrombin; lanes
4 and 7, 1.8 p g PAL1 and 1.7 p g
thrombin; lanes 5 and 8,1.8 p g
PAI-l and 8.5 p g thrombin. Samples were
incubated
in the
absence(lanes3,
4, and 5) or
presence(lanes 6, 7, and 8) of
low-affinity heparin at 25 pg/
mL.Right margin numbers are
M, X 10-.3.
100
1000
Fig 1. Effect of heparinfractions on the inhibition
of thrombin by PAI-l. The second-order rate constant
(mol/L”s”) for the inhibition of thrombin by PAI-1
wasdeterminedunderpseudo-first-orderconditions in the presence of different heparin species as
indicated. UFH, unfractionated heparin;LAH, low-affinity heparin; HAH, high-affinity heparin;LMWH,
low molecular weight heparin.
heparin alone did not cause spontaneous dissociation of the
PAI-I -thrombin complex, and addition of extra thrombin
caused only a minimal amount of complex degradation. Addition of both caused considerable degradation of the bimolecular PAL 1 -thrombin complex (Fig 4. lane 7). The band
seen at M, 31.000 in the SDS gels was shown by Western
blotting to containPAI-Iantigen
(data not shown). Low
molecular weight heparin did not cause cleavageof the PAIl-thrombin complex (data not shown). Thus, heparin acts
-97.4
-
66.2
45
-31
-21.5
1
2
3
4
5
6
7
8
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
LOW-AFFINITYHEPARINAND
1167
PAI-1 INACTIVATION
-97.4
Fig 3. Effect of
incubation
time and low-affinityheparin on
the reaction between PAL1 and
thrombin.Ten-microliter
samples were incubatedwith PBS at
23°C forthetimes
listed, followedbymixingwith
sample
buffer
containing
l mmol/L
PMSF, run on 12%SDS-PAGE,
and stained with Coomassie
blue R-250. Lane 1, 2 p g thrombin; lane 2,2.25 p g PAI-1; lanes
3 through 6, 2 p g thrombin and
2.25 p g PAL1 incubated for 1, 5,
10, and 20 minutes, respectively;
lanes 7 through 10,2 p g thrombin, 2.25 p g PAI-1, and 25 pg/mL
low-affinity heparin incubated
for l,5,10, and 20 minutes, respectively. Rightmargin numbers are M, x
-66.2
-
45
-31
-21.5
1
2
3
4
as a template not only for the initial encounter of thrombin
and PAI- I , but also for the interaction of thrombin with the
PAI-I-thrombin complex. To confirmthe stability of the
PAL I -thrombin complex, preformed complex was incubated with the thrombin inhibitor PPACK for up to 3 hours
in the absence (Fig S, lanes 4 through 6) or presence of 25
pg/mL low-affinity heparin (Fig S. lanes 7 through 9). No
degradation or dissociation occurred, indicating the irreversible nature of the complex and confirming its stability in the
presence of heparin.
Effect of low-qflnity heparin on the thromhh-mediclted
inactivation sf PAI-I: zymographic studies. The ability of
unfractionated and low-affinity heparinto stimulate the inactivation of PAI-I by thrombin was tested usingzymographic
assays for both PAI-I and t-PA activity. PAI-I was assayed
5
6
7
8
9
1
0
by quantitative reverse zymography. The results of incubating 95 nmol/L PAI-I with 108 nmol/L thrombin (Fig 6A),
or 190 nmol/L PAI-Iwith216nmol/Lthrombin
(Fig 6B)
in the presence or absence of 25 pg/mL of unfractionated
or low-affinity heparin are illustrated. In both sets of experiments. the loss of PAI-I activity caused by thrombin was
enhanced byboth unfractionatedheparinandlow-affinity
heparin. Although an identical molar ratio of PAI-I to thrombin was used in the two sets of experiments, the incubations
at higher reactant concentrations (Fig 6B) showed greater
inactivation of PAI- I , indicating that the reaction followed
second-order kinetics. The inactivation of PAI- I by thrombin
and heparin was confirmed by assaying the ability of thrombin-treated PAI-I toinhibitt-PA (Fig 7). In Fig 7A, 190
nmol/L PAI-I was preincubated with 216 nmoVL thrombin
- 97.4
- 66.2
- 45
- 31
- 21.5
1
2
3
4
5
6
7
Fig 4. Mechanism of heparin stimulated degradation of PAI-1 by thrombin. Samples were incubated
for the times listed with PBS at 23°C followed by
mixingwith sample buffer containing l mmol/L
PMSF, run on12% SDS-PAGE,and stained with Coomassie blue R-250. Lane l,lp g thrombin; lane 2,3.8
p g PAI-l; lane 3, l p g thrombin and 3.8 p g PAL1
(10 p L volume) incubated for 30 minutes; lanes 4
through 7, 1 p g thrombin and 3.8 p g PAL1 (10 p L
volume) incubated for 30 minutes followedby a further 15 minutes of incubation after the addition of5
p L of lane 4, PBS; lane 5, low-affinity heparin t o 25
pg/mL; lane 6, 2 p g thrombin; and lane 7. low-affinity heparin t o 25 pg/mL and 2 p g thrombin. Right
margin numbers are M, x
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
1168
PATSTON AND SCHAPIRA
-
97.4
-66.2
45
31
-21.5
1
2
3
5
4
7
6
with or without 25 pg/mL unfractionated or low-affinity
heparin, followed by incubation with 29 nmol/L t-PA and
determination of the remaining t-PA activity by quantitative
zymography. Figure 7B shows a similar experiment using
105 nmol/L PAL I . Control experiments indicated that PAI1 alonecompletely inhibited t-PA (Fig 7A and B, lane l ) .
Addition of thrombin to the preincubation mixtures caused
partial inactivation of PAI-I, with about 2 0 8 of the original
8
9
lo0l
B
.-.-F
K
m
A
€
75
’.
t-PA being detectable (Fig 7A and B. lane 2). This effect
was enhanced by unfractionated or low-affinity heparin (Fig
7A and B, lanes 3 and 4). For example, the proportion o f t PA detectable by zymography after preincubation of 105
nmol/L PAI- 1 and 2 16 nmol/L thrombin increased from 14%
inthe absence of heparin (Fig 7B. lane 2) to 84% inthe
100
100
Fig 5. Stability of PAI-1thrombin complex.
Samples
were incubated as indicated in
PBS, followed by mixing with
sample
buffer
containing 1
mmollL PMSF, run on 12% SDSPAGE, andstained with Coomassie blue R-250. Lane l,2.25
p g PAI-1; lane 2, 0.76 p g thrornbin;lane 3, 2.25 p g PAI-1 and
0.76 p g thrombin incubated for
30minutes;lane
4 through 9,
2.25 p g PAL1 and 0.76 p g thrombin incubatedfor30 minutes,followed by the addition of PPACK
to 1.5 rnmollL in the absence
(lanes 4 through 6) or presence
(lanes 7 through 91 of 25 pglmL
low-affinity heparin,andincubated for a further 1 hour (lanes
4 and 71.2 hours (lanes5 and 8).
or 3 hours (lanes 6 and 9). Right
margin numbers are M, x 10
lA
75
E
a)
L
50
a
L
50
I
W
bp
T
25
25
0
1
2
3
h
1
2
3
Fig 6. Inactivation of PAI-1 by thrombin and heparin. PAL1 was
incubated as described below for 15 minutes at 23°C. and residual
activity remainingwas determinedby quantitative reverse zymography. (A) Lane 1, 95 nmollL PAI-1, 108 nmo1lL thrombin; lane 2, 95
nmol/L PAI-1.108 nmol/L thrombin, 25 pglrnL UFH; lane 3,95 nmoll
L PAI-1.108 nmollL thrombin, 25 pglmL LAH. (B1 Lane 1.190 nmoll
L PAI-1, 216 nmollL thrombin; lane 2, 190 nmollL PAI-1.216 nmol/L
thrombin, 25 pglmL UFH; lane 3, 190 nmol/L PAI-1, 216 nmollL
thrombin, 25 pglmL LAH.
25
1
3
0
1
2
3
4
1
2
4
Fig 7. Inactivation of PAL1 by thrombin and heparin determined
by t-PA activity on quantitative zymography. PAL1 was preincubated
for 15 minutes at 23°C as described below, followed by the addition
of t-PA t o 29 nmollL and incubation for a further 15 minutes before
loading onto the zymogram for assay of residual t-PA activity. (A)
Lane 1.190 nmollL PAI-1; lane 2,190 nmol/L PAI-1.216 nmollL thrombin; lane 3, 190 nmol/L PAI-1.216 nmol/L thrombin, 25 pglmL UFH;
lane 4, 190 nmol/L PAI-1, 216 nmollL thrombin, 25 pglmL LAH. (B)
Lane 1,105 nmollL PAI-1; lane 2,105 nmollL PAI-1,216 nmollL thrombin; lane 3, 105 nmol/L PAI-1,216 nmol/L thrombin, 25 pglmL UFH;
lane 4, 105 nmollL PAI-1, 216 nmoll8L thrombin, 25 pg/mL LAH.
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
1169
LOW-AFFINITY HEPARIN AND PAL1 INACTIVATION
DISCUSSION
L
100 8.
.
I
E
Q)
L
75
r
+.,
.
I
>
.
I
CI
.-c
a
E
0
L
50
25
1
-I
I
0
5
10
15
20
Time (minutes)
Fig 8. Inactivation of thrombin by PAL1 and heparin determined
bythrombinchromogenicsubstrateassay.PAL1(95nmol/L)and
thrombin I108 nmol/L) were incubated at 23°C in PBS (0).or with
25 pg/mL LAH (0).or 25 pg/mL UFH (A). At the times indicated, a
5-pL aliquot was assayed for residual thrombin activity on S-2238.
Each point isthe mean of three to five separate determinations.
presence of 25 pg/mL of unfractionated heparin (Fig 7B,
lane 3).
Effect of heparin on the inhibition of thrombin amidolytic
activity by PAI-I: studies with enzyme excess. The data
reported in the preceeding paragraph indicate that unfractionated or low-affinity heparin can enhance PAI-l inactivation
by thrombin, as determined by both direct and indirect zymographic assays of PAI-1 activity (Figs 6 and 7). Additional
experiments were undertaken to assess how thrombin amidolytic activity was influenced under the reaction conditions
used for these zymographic studies. The kinetics of inhibition of thrombin (108 nmoVL) by PAI-1 (95 nmoVL) with
or without heparin are shown in Fig 8. In the absence of
heparin, thrombin activity was reduced by about 20% at
20 minutes (Fig 8, squares). In the presence of 25 pg/mL
unfractionated heparin or lowaffinity heparin, a 25% to
40% reduction of thrombin amidolytic activity was observed
within 1 to 2 minutes, followed byan additional 10% to
15% reduction at 20 minutes (Fig 8, circles and triangles).
For comparison, greater than 85% inhibition of thrombin
was observed when thrombin (108 nmol/L) was incubated
for 1 minute with antithrombin I11 (95 nmoVL) and 25 pg/
mL unfractionated heparin. Similarly, greater than 85% inhibition of t-PA was observed when t-PA (108 nmol/L) was
incubated for 1 minute with PAL1 (95 nmoVL).
PAI- 1 has affinity for heparin through lysine residues analogous to those in the heparin binding site of another serpin,
antithr~mbin,~'
and binding to heparin endows PAI-l with
thrombin inhibitory activity." Gebbink et a139showed that
stimulation of thrombin inhibition by PAL1 was specific to
heparin, with other glycosaminoglycans having much less
stimulatory effect. These studies, h o w e ~ e r , ~ 'were
* ~ ~con,~~
cerned primarily with the inhibition of thrombin by PAL1
and the stimulation of this reaction by heparin, but they did
not examine the effect of thrombin and heparin on PAI-1
activity. The objective of the study reported here was to
assess the influence of thrombin and various heparin forms
on PAI-1. Our results indicate that unfractionated heparin
and heparin with low affinity for antithrombin increase the
inhibition of thrombin by PAL1 when the serpin is in molar
excess over the serine proteinase. However, when thrombin
and PAI-1 are present at equimolar concentrations or when
there is a molar excess of thrombin, these two forms of
heparin considerably stimulate PAL1 cleavage and inactivation; importantly, thrombin is only weakly inhibited during
these reactions.
The second-order rate constant for the inhibition of thrombin by PAI-1 in the absence of heparin was 458 moVL"s"
(Fig 1). In the presence of unfractionated heparin, the maximum second-order rate constant was 5,000 moVL"s-l, occuring between 15 and 45 pg/mL of heparin. Heparin can
be fractionated into forms withlowandhighaffinity
for
antithrombin. The results with these two forms were essentially identical to the results obtained with the unfractionated
material, indicating that the pentasaccharide responsible for
the high-affinity binding to antithrombin was not necessary
for PAI-1 binding. Moreover, the smaller stimulation observed with low molecular weight heparin (maximum rate
constant, 1,700 moVL-ls"), and the shape of the dose response curves, confirmed that heparin was acting through
the template (or surface approximation) mechanism. The fact
that low-affinity heparin had the same ability to stimulate
the reaction as unfractionated heparin suggested that lowaffinity heparin could be a useful agent in distinguishing the
anticoagulant effect of heparin from its profibrinolytic action. Therefore, a detailed investigation of the action of lowaffinity heparin on the reaction between PAI-1 and thrombin
was performed.
The inhibition of thrombin by PAL1 was studied under
pseudo-first-order conditions, with a 50-fold molar excess
of PAI-l over thrombin. However, to assess its biologic
relevance, the interaction of PAI-1 and thrombin should be
examined at reactant concentrations likely to occur in vivo.
PAL1 concentration innormalplasma is less than 20 ng/
mL (0.4 nm0l/L),4~,~'
although localized concentrations of
PAI- 1 could be much higher as a result of PAI- 1 release from
activated platelets or PAL1 association with the extracellular
m a ~ x . 4 Z . 4 3 For example, the PAL1 concentration in pathologic human thrombi can be over 1 pg/g of wet thrombus.M.
As far as the concentration of thrombin is concerned, the
following considerations have to be taken into account. Prothrombin concentration in normal plasma is 100 p g / d (1.4
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
1170
PmoVL). Thus, 1% activation of prothrombin would generate a thrombin concentration of l pg/mL, (14 nmol/L). Locally, thrombin concentration could be higher, but a significant amount will be continually sequestered by antithrombin
(and other serpins such as heparin cofactor 11), as well as
by ligands such as thromborn~dulin~~
and heparan sulfate.46
Because of these and other reasons, such as the location and
extent of vascular injury, amount of platelet activation, and
rate of flow(and therefore removal of reactants), it is difficult
to predict with certainty the in vivo concentrations of PAI1 and thrombin at the site of tissue damage and thrombus
formation. Although a large excess of PAI-l over thrombin
could be present in vivo, it is also possible that the PAI-1
concentration will be closer to that of thrombin. Therefore,
with this reasoning in mind, the effect of low-affinity heparin
on the interaction of thrombin and PAI-1 was further characterized by a series of polyacrylamide gels (Figs 2 through
5). Using approximately equimolar concentrations of PAI-l
and thrombin, it can be seen that, although low-affinity heparin stimulated the initial rate of complex formation, it also
potentiated the degradation of complex by thrombin. With
thrombin in excess over PAI-1, minimal thrombin-PAI-1
complex was formed. In this situation, neutralization of PAI1 by thrombin is likely to be the relevant process. This
process seems to occur through formation of PAI- 1-thrombin complex, proteolytic cleavage of PAI-1 at the reactive
site bond, and cleavage of the serpin at other sites. Moreover,
the observation that active serine proteinase and inactivated
serpin are detectable among the products of the reaction
between thrombin and PAI-1 is consistent with several reports indicating that serpins are suicide substrates for serine
proteinase~.~~.~~
To determine whether the reactions observed by SDSPAGE were also taking place in a nondenatured system, PAI1 activity was measured both directly by reverse zymography
(Fig 6) and indirectly by its ability to inhibit t-PA, with the
t-PA assayed by conventional zymography (Fig 7). Both
methods confirmed that thrombin neutralizes PAI-1 activity
and that unfractionated and low-affinity heparin stimulate
this reaction. In addition, when thrombin inhibition was studied with close to equimolar reactant concentrations (Fig 8),
thrombin inhibition was minimal in the absence of heparin.
Addition of unfractionated or low-affinity heparin to the incubation led to less than 40% inhibition of thrombin. Importantly, the same incubation conditions led to an 85% loss
of PAI-1 activity.
There is no evidence to suggest that inhibition of thrombin
by PAI-1 has any physiologic role per se, given the systemic
concentrations of the major thrombin inhibitors (antithrombin 111, 2.6 pmol/L; heparin cofactor 11, 0.14 pmol/L), compared with that of PAI-1 (0.4 nmol/L). Rather, in view of
the data presented here, it is more likely that the significance
of the reaction between thrombin and PAI-1 lies in the inactivation of PAI- 1. Moreover, our observation that PAI- 1 inactivation by thrombin is augmented by the presence of unfractionated or low-affinity heparin indicate that these two
heparin forms have the potential to enhance fibrinolysis. The
high rates of reocclusion seen after thrombolytic therapy are
believed to result from the release of clot bound thr~mbin,~”
PATSTON AND SCHAPIRA
de novo thrombin generati~n,~’
and platelet
with PAI-l r e l e a ~ e . ~In~ ,addition,
’~
the resistance of thrombi
to spontaneous lysis increases with increasing PAL 1 in those
thrombi.& Administration of heparinwith
thrombolytic
agents is standard p r a c t i ~ e . ~Although
~-~’
part of the potentiation of thrombolysis seen with heparin administrations8may
be caused by activation of antithrombin, the inactivation
of PAI-1 by the thrombin-dependent mechanism described
herein may also be an important factor. Evidence that reduction in PAI-1 activity is beneficial to thrombolytic therapy
is given by a study showing that a monoclonal anti-PAI1 antibody enhanced thrombolysis and reduced thrombus
extension in a rabbit
There is a growing body of
evidence that elevated PAL1 is a risk factor for thromboembolic disease, as a result of decreased fibrinolytic activity.”
The data reported here, showing thatlow-affinityheparin
can enhance the inactivation of PAI-1, suggest thatlowaffinity heparin might have a therapeutic role in such cases.
In addition, low-affinity heparin has the potential of becoming an important adjunctive agent for thrombolytic therapy,
in particular if its administration is not associated with the
bleeding complications seen with conventional heparin.
ACKNOWLEDGMENT
We thank Dr Thomas Reilly for the recombinant PAI- I , Dr Paul
Bock for the a-thrombin, and Dr PeterGettins for helpful discussions
and comments on the manuscript.
REFERENCES
1. Rosenberg RD: The heparin-antithrombin system: A natural
anticoagulant, in Colman RW, Hirsh J, Marder VJ, Salzman EW
(eds): Hemostasis and Thrombosis (ed 2). Philadelphia, PA, Lippincott, 1987, p 1373
2. Hirsh J: Heparin. N Engl J Med 324:1565, 1991
3. Bourin M-C, Lindahl U: Glycosaminoglycans and the regulation of blood coagulation. Biochem J 289:313, 1993
4. Choay J, Lormeau JC, Sinay P, Casu B, Gatti G: Structure
activity relationship in heparin: A synthetic pentasaccharide sequence with high affinity for antithrombin I11 and eliciting high antifactor Xa activity. Biochem Biophys Res Commun 116:492, 1983
5 . Lam LH, Silbert JE, Rosenberg RD: The separation of active
and inactive forms of heparin. Biochem Biophys Res Commun
69:570, 1976
6. Gettins PGW, Fan B, Crews BC, Turko IV, Olson ST, Streusand V: Transmission of conformational change from heparin binding site tothe P, residue of the reactive center of antithrombin.
Biochemistry 322385, 1993
7. Olson ST, Bjork I: Predominant contribution of surface approximation to the mechanism of heparin acceleration of the antithrombin-thrombin reaction: Elucidation from salt effects. J BiolChem
266:6353, 1991
8. Olson ST, Bjork I, Sheffer R, Craig PA, Shore JD, Choay J:
Role of the antithrombin-binding pentasaccharide in heparin acceleration of antithrombin-proteinase reactions: Resolution of the antithrombin conformational change contribution to heparin rate enhancement. J Biol Chem 267:12528, 1992
9. Levine MN, Hirsh J, Kelton JG: Hemmorhagic complications
of antithrombotic therapy, in Colman RW, Hirsh J, Marder VJ,
Salzman EW, (eds): Hemostasis and Thrombosis (ed 2). Philadelphia, PA, Lippincott, 1987, p 873
10. Marsh N: Does heparin stimulate fibrinolysis? Br J Haematol
76: 163, 1990
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
LOW-AFFINITYHEPARIN AND PAI-1 INACTIVATION
11. Fareed J, Hoppensteadt D, Bick RL, Bacher P Drug induced
alterations of hemostasis and fibrinolysis. Hematol Oncol Clin North
Am 6:1229, 1992
12. Vinazzer H, Stemberger A, Haas S, Bliimel G: Influence of
heparin; Of different heparin fractions and of a low molecular weight
heparin-like substance onthe mechanism of fibrinolysis. Thromb
Res 27:341, 1982
13. Cerek B, Lew AS, Hod H, Yano J, Reddy NKN. Ganz W:
Enhancement of thrombolysis with tissue-type plasminogen activator
by pretreatment with heparin. Circulation 74:583, 1986
14. Marini C, Di Ricco G, Rossi G, Rindi M, Palla R, Giutini
C: Fibrinolytic effect of urokinase and heparin in acute pulmonary
embolism: A randomized clinical trial. Respiration 54:162, 1988
15. Agnelli G, Levi M, Cosmi B, ten Cate JW, Nenci GG: Additive effect of dDAVP and standard heparin in increasing plasma tPA. Thromb Haemost 61:507, 1989
16. Rapold HJ, Lua HR, Wu Z, Nijs H, Collen D: Requirement
of heparin for arterial and venous thrombolysis with recombinant
tissue-type plasminogen activator. Blood 77: 1020, 1991
17. Rudd MA, George D, Johnstone MT, Moore RT, Collins L,
Rabbani LE, Loscalzo J: Effect of thrombin inhibition on the dynamics of thrombolysis and on platelet function during thrombolytic
therapy. Cin: Res 702329, 1992
18. Marder VJ, Shortell VK, Fitzpatrick PG, Kim C, Oxley D:
An animal model of fibrinolytic bleeding based on the rebleed phenomenon: Application to a study of vulnerability of hemostatic plugs
of different age. Thromb Res 67:31, 1992
19. Hauert J, Nicoloso G, Schleuning WD, Bachmann F, Schapira
M: Plasminogen activators in dextran sulfate activated euglobulin
fractions: A molecular analysis of factor XII- and prekallikreindependent fibrinolysis. Blood 73:994, 1989
20. Klauser RI: Induction of fibrinolysis by polyanions in human
plasma. Thromb Haemost 60:324, 1988
2 1. Pixley RA, Cassello A, De La Cadena RA, Kaufman N, Colman RW: Effect of heparin on the activation of factor XI1 and the
contact system in plasma. Thromb Haemost 66:540, 1991
22. Pannekoek H, Veerman H, Lambers H, Diergaade P, Verweij
CL, van Zonneveld AJ,van Mourik JA: Endothelial plasminogen
activator inhibitor (PAI): A new member of the serpin gene family.
EMBO J 5:2539, 1986
23. Loskutoff DJ, Sawdey M, Mimuro J: Type 1 plasminogen
activator inhibitor. Prog Hemost Thromb 9537, 1989
24. Bergstein JM, Riley M, Bang NU: Role of plasminogenactivator inhibitor type-l in the pathogenesis and outcome of the
hemolytic uremic syndrome. N Engl J Med 327:755, 1992
25. Schneiderman J, Sawdey MS, Keeton MR, Bordin GM, Bernstein EF, Dilley RB, Loskutoff DJ: Increased type 1 plasminogen
activator inhibitor gene expression in atherosclerotic human arteries.
Proc Natl Acad Sci USA 89:6998, 1992
26.FayWP,
Shapiro AD, Shih JL, Schleef RR, Ginsburg D:
Brief report: Complete deficiency of plasminogen-activator inhibitor
type 1 due to a frame shift mutation. N Engl J Med 327:1729, 1992
27. Ehrlich HJ, Keijer J, Preissner KT, Gebbink RK, Pannekoek
H: Functional interaction of plasminogen activator inhibitor type 1
(PAI-I) and heparin. Biochemistry 301021, 1991
28. Keijer J, Linders M, Wegman JJ, Ehrlich HJ, Mertens K,
Pannekoek H: On the target specificity of plasminogen activator
inhibitor l : The role of heparin, vitronectin, andthe active site.
Blood 78:1254, 1991
29. Naski MC, Lawrence DA, Mosher DF, Podor TJ, Ginsburg
D: Kinetics of inactivation of a-thrombin by plasminogen activator
inhibitor-l: Comparison of the effects of native andurea treated
forms of vitronectin. J Biol Chem 268:12367, 1993
30. Ehrlich H J , Gebbink RK, Preissner KT, Keijer J, Esmon NL,
Mertens K, Pannekoek H: Thrombin neutralizes plasminogen activa-
1171
tor inhibitor 1 (PAI-l) that is complexed with vitronectin in the
endothelial cell matrix. J Cell Biol 115:1773, 1991
3 1. Hijijk M, Bjork I, Hopwood J, Lindahl U: Anticoagulant activity of heparin: Separation of high-activity and low-activity species
by affinity chromatography on immobilized antithrombin. FEBS Lett
66:90, 1976
32. Andrade-Gordon P, Strickland S: Fractionation of heparin by
chromatography on a tissue plasminogen activator-Sepharose column. Proc Natl Acad Sci USA 87:1865, 1990
33. Reilly TM, Seetharam R, Duke JL, Davis GL, Pierce SK,
Walton HL, Kingsley D, Sisk WP: Purification and characterization
of recombinant plasminogen activator inhibitor- 1 from Eschericia
Coli. J Biol Chem 265:9570, 1990
34. Kitz R, Wilson IB: Esters of methanesulfonic acid as irreversible inhibitors of acetylcholinesterase. J Biol Chem 237:3245, 1962
35. Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680, 1970
36. Granelli-Piperno A, Reich E: A study of proteases and protease-inhibitor complexes in biological fluids. J ExpMed 148:223,
1978
37. Erickson LA, Lawrence DA, Loskutoff DJ: Reverse fibrin
autography: A method to detect and partially characterize protease
inhibitors after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Anal Biochem 137:454, 1984
38. Ehrlich W,Gebbink RK, Keijer J, Pannekoek H: Elucidation
of structural requirements on plasminogen activator inhibitor 1 for
binding to heparin. J Biol Chem 267:11606, 1992
39. Gebbink RK, Reynolds CH, Tollefsen DM, Moertens K,
Pannekoek H: Specific glycosaminoglycans support the inhibition
of thrombin by plasminogen activator inhibitor 1. Biochemistry
32:1675, 1993
40. Booth NA, Simpson AJ, Croll A, Bennett B, MacGregor IR:
Plasminogen activator inhibitor (PAI-I) in plasma and platelets. Br
J Haematol 70:327, 1988
41. Declerck PJ, Alessi MC, Verstreken I, Kruithof EKO, JuhanVague I, Collen D: Measurement of plasminogen activator inhibitor
1 in biologic fluids with a murine monoclonal antibody-based enzyme-linked immunosorbent assay. Blood 71 :220, 1988
42. Knudsen BS, Nachman RL: Matrix plasminogen activator
inhibitor: Modulation of the extracellular proteolytic environment.
J Biol Chem 263:9476, 1988
43. Mimuro J, Loskutoff DJ: Purification of a protein from bovine
plasma that binds to type 1 plasminogen activator inhibitor and
prevents its interaction with extracellular matrix: Evidence that the
protein is vitronectin. J Biol Chem 264:936, 1989
44. Potter van Loon BJ, Rijken DC, Brommer EJP, van der Maas
APC: The amount of plasminogen, tissue-type plasminogen activator
and plasminogen activator inhibitor type 1 in human thrombi and
the relation to ex-vivo lysis. Thromb Haemost 67:101, 1992
45. Dittman WA, Majerus PW: Structure and function of thrombomodulin: A natural anticoagulant. Blood 75:329, 1990
46. de Agostini AI, Watkins SC, Slayter HS, Youssoufian H,
Rosenberg RD: Localization of anticoagulantly active heparan sulfate proteoglycans in vascular endothelium: Antithrombin binding
on cultured endothelial cells and perfused rat aorta. J Cell Biol
111:1293, 1990
47. Patston PA, Gettins P, Beechem J, Schapira M: Mechanism
of serpin action: Evidence that Cl-inhibitor functions as a suicide
substrate. Biochemistry 30:8876, 1991
48. Gettins PGW, Patston PA, Schapira M: The role of conformational change in serpin structure and function. Bioessays 15:461,
1993
49. Cooperman BS, Stavridi E, Nickbarg E, Rescorla E, Schechter
NM,Rubin H: Antichymotrypsin interaction with chymotrypsin:
Partitioning of the complex. J Biol Chem 268:23616, 1993
From www.bloodjournal.org by guest on December 3, 2014. For personal use only.
1172
50. Owen J, Friedman KD, Grossman BA, Wilkins C, Berke AD,
Powers ER: Thrombolytic therapy with tissue plasminogen activator
or streptokinase induces transient thrombin activity. Blood 72:616,
1988
5 1. Naski MC, Shafer JA: a-Thrombin within fibrin clots: Inactivation of thrombin by antithrombin-111. Thromb Res 69:453, 1993
52. Fitzgerald DJ, Catella F, Roy L, FitzGerald GA: Marked
platelet activation following intravenous streptokinase in acute myocardial infarction. Circulation 77: 142, 1988
53. Serizawa K, Urano T, Kozima Y, Takeda Y, Takada A: The
potential role of platelet PAI-I in t-PA mediated clot lysis of platelet
rich plasma. Thromb Res 71:289, 1993
54. Torr-Brown SR, Sobel BE: Attenuation of thrombolysis by
release of plasminogen activator inhibitor type-l from platelets.
Thromb Res 72:413, 1993
PATSTON AND SCHAPIRA
55. Eisenberg PR: Role of heparin in coronary thrombolysis.
Chest 101:131S, 1992
56. Fuster V: Coronary thrombolysis: A perspective for the practicing physician. N Engl J Med 329:723, 1993
57. Rapaport E: Thrombolysis, anticoagulation, and reocclusion.
Am J Cardiol 67: 17E, 1991
58. Burke SE, Lubbers NL, Nelson RA, Henkin J : Recombinant
pro-urokinase requires heparin for optimal clot lysis and restoration
of blood flow in a canine femoral artery thrombosis model. Thromb
Haemost 69:375, 1993
59. Levi M, Biemond BJ, van Zonnenveld AJ,ten Cate JW,
Pannekoek H: Inhibition of plasminogen activator inhibitor-l activity results in promotion of endogenous thrombolysis and inhibition
of thrombus extension in models of experimental thrombosis. Circulation 85:305, 1992