From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
RAPID COMMUNICATION
Platelets Inhibit Fibrinolysis In Vitro
by Both Plasminogen Activator
Inhibitor-l-Dependent and -Independent Mechanisms
By William P. Fay, Daniel T. Eitzman, Amy D. Shapiro, EdwinL. Madison, and David Ginsburg
lysis, normal platelets markedly inhibited clot lysis under
Platelet-rich thrombi are resistant to lysis by tissue-type
the same conditions.However, no difference between PAIplasminogen activator(t-PA). Although platelet a-granules
l-deficient platelets and platelets with normal PAL1
contain plasminogen activator inhibitor-l (PAI-l), a fastcontent was observed when streptokinase or a PAL1-reacting inhibitor of t-PA, the contribution of PAL1 to the
antifibrinolyticeffect of platelets has remained a subject of sistant t-PA mutant were used to initiate fibrinolysis. In addition, PAl-l-resistant t-PA was significantly more efficontroversy. W e recently reported a patient with a homocient inlysingclotscontainingnormal
platelets than
zygous mutation within the PAL1gene that results in comwild-type t-PA. W e conclude that platelets inhibit t-PAplete loss of PAL1 expression. Platelets from this individmediated fibrinolysis by both PAI-1-dependent and PAIual constitute a unique
reagent with which to probe the role
1 -independent mechanisms. These results have imporof platelet PAL1 in the regulationof fibrinolysis.The effects
tant implications forthe role of PAI-1 in the resistance of
of PAI-1 -deficient platelets were compared with those of
platelet-richthrombi to lysis in vivo.
normal platelets in an in vitro clotlysis assay. Althoughthe
0 1994 by The American Societyof Hematology.
incorporation of PAI-1-deficient platelets into clots resulted in a moderate inhibition of t-PA-mediated fibrino-
T
ISSUE-TYPE PLASMINOGEN activator (t-PA) dissolves bloodclots by catalyzingthe conversion of plasminogen to plasmin, a serine protease that efficiently cleaves
fibrin. Under physiologicconditions, fibrinolysis isinitiated
by the release of t-PA from vascular endothelial cells.' In
addition, recombinant t-PA is used therapeutically to induce thrombolysis in a variety of clinical settings, most notably myocardial infarction.' However, the rate with which
t-PA induces fibrinolysis in vivo
is highly variable,
and some
thrombi are resistant to lysis, despite the administration of
pharmacologic concentrations of t-PA.3 Although several
factors determine the thrombolytic efficiency of t-PA, the
composition of a thrombus appears to be an important determinant of its sensitivity to lysis. In animal models, platelet-rich thrombi are highly resistant to lysis by t-PA," suggesting that platelets exerta potent negative effect on t-PAmediated clot lysis. In addition, human platelets inhibit clot
lysis in ~ i t r o . Because
~.~
human coronary thrombi are frequently platelet-rich,' the antifibrinolytic capacity of platelets may play
an important role inthe regulation of coronary
thrombolysis." Consistent with this hypothesis, antiplatelet
agents enhance the efficacy of thrombolytic therapy in patients with acute myocardial infarction.2
Platelets may affect clot lysis by several mechanisms. Acti1 (PAIvated platelets secrete plasminogen activator inhibitor1) and c~-antiplasmin,which inhibit plasmin formation and
activity, respectively? Plateletsalso contain factorXIII, which,
upon activation, cross-links fibrin monomer, thereby rendering it more resistant to digestion by plasmin." Receptors for
fibrin(ogen) and plasminogen are presentwithinplatelet
plasma membranes,8.1'.1' suggesting that the platelet surface
may regulate fibrinolysis within hemostatic plugs
and plateletrich thrombi. In addition, platelets mediate clot retraction, a
l3
process that inhibits clot lysis in vitro.
PAI- 1, the fast-acting inhibitor of t-PA and urokinase, is
an important regulator of the fibrinolytic system.14PAI-1
inhibits t-PA and urokinase by forming 1 :I protease-inhibitor complexes that are enzymatically inactive.l4 Elevated
plasma levels of this inhibitor are associated with myocardial infarction,15and PAI-1 deficiency results in abnormal
Blood. V01 83.No 2 (January 15). 1994: pp 35 1-356
Platelets are a major reservoir of PAI-1, with
approximately 90%of circulating PAL 1 contained within
platelet a-granules." Hence, it has been proposed that the
inhibitory effect of plateletson clot lysis ismediated, at least
in part, by PAL1.6 However, platelet PAL1 exists predominantly in a latent, or inactive form,20-22
suggesting that its
effect on fibrinolysis maybe limited. Prior studies addressing the role of platelet PAI-1 in fibrinolysis have yielded discordant results. Although anti-PAI-1 monoclonal antibodies enhance t-PA-mediatedlysisofplatelet-rich
clots in
vitro: studies by other investigators have suggested that
platelets inhibit clot lysis by inducing clot retraction, not by
PAL 1 release.I3
If platelet PAL 1 regulates clot lysis, then altered expression ofthis inhibitor within the platelet compartment would
be expected to affect the lysis of platelet-rich clots. We recently have identified a patient with complete PAL1 deficiencycaused by a homozygousframe-shift mutation
From the Ann Arbor Veterans Aflairs Hospital, Research Service,
Ann Arbor, MI; the University ofMichigan Medical
School, Department of Internal Medicine and Howard Hughes Medical Institute
and Department of Human Genetics, Ann Arbor, MI; the Indiana
University MedicalSchool, Department of Pediatrics, Indianapolis,
IN; and the Scripps Research Institute, Committeefor Vascular Biology, La Jolla, CA.
Submitted Octoberl I , 1993; accepted October 26, 1993.
Supported by a Department of Veterans Aflairs Medical Research
Service Advisory Group Grant (W.P.F.)and National Institutes of
Health GrantsNo. HL-02728 (W.P.F.) and HL-39137 (D.G.). D.G.
is an Associate Investigator ofthe Howard Hughes Medical Institute.
Address reprint requests to William P. Fay, MD, University of
Michigan Medical Center, 4520 MSRBI , I150 WMedical Center
Dr. Ann Arbor, MI 48109-0650.
The publication COSIS of 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 I734 solely to
indicate this fact.
0I994 by The
American Society of Hematology.
0006-4971/94/8302-0040$3.00/0
35 1
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
352
FAY ET AL
within exon4 of the PAI-1 gene.18 This mutation results in
the synthesis of an unstable PAI-I mRNA and a truncated
PAI- 1 protein that lacks the reactive center of the molecule.
Platelets from this individual contain no detectable PAI- 1,
but are otherwise normal. Hence, they constitute a unique
reagent to probe the contribution of PAL1 to platelet-mediated clot lysis resistance. We have compared the effects of
PAI-l-deficient platelets with those of normal platelets in
an in vitro clot lysis assay in which fibrinolysis was
initiated
by wild-type t-PA, streptokinase, or a mutant t-PA that is
resistant to inhibition by PAL 1. The results of thesestudies
suggest that platelets inhibit fibrinolysis by both PAI- I -dependent and PAI- 1-independent mechanisms.
MATERIALSANDMETHODS
Materials. Recombinant human t-PA (singlechain; specific activity, 580,000 IU/mg) was obtained from Genentech (South San
Francisco, CA). A recombinant human single-chain t-PA mutant
(Lys-296, Arg-298,Arg 299+ Glu) that activates plasminogennormally, but which is highly resistantto inhibition by PAL1 (inhibition rate constant 5.0 X 10' mol/L" . S" v 1.4 X IO6 mol/L" .
S-' for wild-typet-PA), was prepared as previously described.23
The
specific activityof mutant t-PA (determinedby comparison to wildtype t-PA inthe clot lysis assay described below) approximately
was
250,000IU/mg. Streptokinase (specificactivity, approximately
3,000 IU/mg) and human Glu-plasminogen were from American
Diagnostica (Greenwich, CT). Fibrinogen (prepared from human
plasma by p-alanine precipitati~n~~
and lysine-sepharose chromatograph~*~)
was the gift of M. Naski.Prostaglandin E,, apyrase, human thrombin, and fluoresceinisothiocyanatewere from Sigma (St
Louis, MO).
Platelet preparations. Blood samples were collected from
an individual with homozygous PAI-1 deficiency, an individual heterozygous forthe PAL I null allele (father of the homozygous individual), and from normal controls. The clinical characteristics and
molecular genetic analysisof this PAI-l-deficiency pedigree have
been reported previously.'8The individual with homozygous PAII deficiency (proband) was a IO-year-old female with a congenital
bleeding disorder of moderate severity. Routine indices of hemostatic function were normal. However, she had
no detectable PAI- 1
antigen or activity ineither her plasmaor platelets. Plasma levels
of
a2-antiplasminand factor XI11 were normal. Her bleeding time was
mildly prolonged (12 minutes; normal, <9), but in vitro platelet
aggregation studies werenormal. The probands father had no history of abnormal bleeding. His plasma levels
of PAL 1 activity and
antigen were in the low to normal range (activity, 0.2 U/mL [normal, I. I 0.61; antigen, 12 ng/mL [normal, 5 5 ? 261) and his platelet levels of PAL I activity and antigen were normal (activity, 4.1
U/108 platelets [normal,3.8 .+ 0.61; antigen, 272 ng/lO* platelets
[normal, 1 19 ? 271). Plasma and platelet PAL 1 levels were not determined for normal controls.
All individuals from whom blood was collected for these experiments had not consumed aspirin for at least 7 days before venipuncture. After obtaining informed consent, blood samples were collected
by
peripheral venipuncture into acid-citratedextrose
anticoagulant (0.8%wt/vol citric acid,2.2% trisodium citrate,
2.4590 dextrose)at a ratio of 6 v01 ofblood to 1 v01 ofanticoagulant.
Washed platelets were preparedby repeated centrifugation as previously describedz6and then resuspended in Tyrode's buffer(1 37
mmol/L NaCI, 2.68 mmol/L KCl, 2 mmol/L CaC12, 1 mmol/L
MgCI2, 0.36mmol/L NaH2P04, 1 1.9 mmol/L NaHCO,, 5 mmol/
L HEPES, 0.1%glucose, 0.35%bovine serumalbumin, pH 7.35).
In vitro clot lysis assay. Clot lysisassayswereperformed by
*
100
75
50
25
0
0
30
60
90
120
MillUh
Fig 1. Inhibition of t-PA-mediated clot lysis by platelets. A mixture of fluorescein-labeled fibrinogen (1 mg/mL), Glu-plasminogen
(20 pg/mL), andt-PA (0.6 U/mL) was clotted in the absence (0)or
presence (m) of platelets (2.5 X 1O*/mL). andthe percentage of clot
lysis was determined at the indicated time points, as described in
Materials and Methods. Note the marked inhibitionof fibrinoiysis in
clots containingplatelets.
modification of the method of Smith et al?' in which fluoresceinlabeled fibrinogen isclotted by thrombin and plasminogen activator-mediated fibrinolysis ismonitored by the release of fluoresceinlabeled fibrin degradation products into the clot supernatant. The
concentrations of t-PA, plasminogen, fibrinogen, and platelets in
these assays weresimilar to those used in the clot lysis assay of Levi
et aL6 A mixture of Glu-plasminogen (20.8 &nL), fluoresceinlabeled fibrinogen( I .04 mg/mL), washed platelets(0 to 2.5 X lo8/
mL), and t-PA or streptokinase in Tyrode's buffer was prepared.
Aliquots of this mixture (l25 pL) were clotted in 1.5 mL polypropylene tubes by the addition of 5 pL ofthrombin (0.2 mmol/L) and
then incubated at 37°C. At timed intervals (with I O minutes after
the addition of thrombin defined as t = 0), tubes were centrifuged
(16,OOOg for 3 minutes) and 100 pL of the clot supematant was
removed and diluted into 1.4 mL of IO mmol/L Tris-HCI, 140
mmol/L NaC1, pH 7.5. The fluorescence ofthis dilution was measured (excitation wavelength, 500 nm; emission wavelength, 520
nm) in a fluorometer (SLM Instruments, Urbana, IL) and the percentage of clot lysis was calculated
as previously described.*'
RESULTS
Effect ofplatelets on t-PA-mediated clot lysis. To determine the effect of normal platelets on fibrinolysis inthis experimental system, fibrinclots were formed in the presence
or absence of plateletsand therates of wild-type t-PA-mediated clot lysis werecompared. The concentration of t-PA
in these experiments was 0.67 U/mL. In the absence of
platelets, clot lysis was greater than 90% complete within
120 minutes at 37°C (Fig l). However, in the presence of
normal platelets (2.5 X 108/mL),clot lysis wasmarkedly inhibited, with less than 1% clot lysis occurring by 120 minutes. Essentiallyidentical results ( 51% clot lysis by120 minutes in the presence of platelets) were obtained when this
experiment was repeatedwithplatelets obtained from 3
other normal individuals.
The effects on clot lysis of washed platelets obtained from
the individual withhomozygous PAI-l deficiencywere
compared with those of plateletsfrom a normal control (Fig
2). Whereas essentiallyno lysis occurred by 120 minutes in
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
353
PLATELET PAI-1 AND FIBRINOLYSIS
Y
8
ag
0
30
60
120
Minutes
Fig 2. Comparison of effects of normal versus PAl-l-defmient
platelets on clot lysis. Fibrin clots
were formed in the presence oftPA (0.67 U/mL)and washed platelets (2.5 X 10*/mL) prepared
from an individual with homozygous PAL1 deficiency (m) or a normal control (A). At theindicated time points, the percentage of clot
lysis was determined.
the control reaction, 42% clot lysis was observed
in thepresence of platelets from the PAI- 1-deficient individual. This
experiment was repeated at a later date with a second sample
offreshly prepared platelets, and similar, although less
marked, results wereobtained ( 17%clot lysis at 120 minutes
in thepresence of PAL 1-deficient platelets v 1% clot lysis in
the presence of normal platelets). During the latter experiment, the effects on clot lysis of washed platelets prepared
from the proband's father were also studied. These platelets
markedly inhibited clot lysis (< 1% lysis by 120 minutes),
producing an effect indistinguishable from that of normal
platelets. These resultsare consistent with the normal levels
of platelet PAI1 activity and antigen previously observed in
carriers of this defective PAI-1 allele.18
The effect of platelet concentration on t-PA-mediated
clot lysis wasalso examined. Clot lysis was allowedto occur
for 60 minutes in the presence of t-PA (0.67 U/mL) and
platelets (0 to 2.4 X 108/mL).For both normal and homozygous PAL1-deficient platelets,the extent of clot lysis was
inversely proportional to platelet concentration, although
more extensive lysis consistently was observed in the presence of PAI-1-deficient platelets (Fig3). Hence, PAI- 1-deficient platelets inhibit clot lysis less than normal platelets.
However, platelets lacking PAI-1 retain the capacity to inhibit fibrinolysis, and do so in a concentration-dependent
manner. Of note, clot retraction was observed in this experimental system, and the extent of clotretraction (estimated
by visual inspection) appeared proportional to platelet concentration.
Eflect ofplatelets on mutant t-PA- and streptokinase-mediated clot lysis. To further assess the role of platelet PAI1 in regulating clot lysis, studies similar to those described
above were performed, except that PAI-I-insensitive plasminogen activators (ie, t - P A ~ , ~ 2 9 6 , ~ ~ ~ 2 9 8 , ~ r g 2 9or
9 - strep
~,"
tokinase) were substituted for wild-type t-PA. In these experiments, the effects of platelets from the PAI- I -deficient
individual were compared with those of her father's platelets. Platelet-rich clots were formed in the presence of mutant t-PA (1.5 U/mL) or streptokinase (8 mU/mL) and the
rates of clot lysis were determined. In contrast to experiments withwild-typet-PA, clots containing PAI-l-deficient platelets (obtained from the proband) and those containing plateletswith normal PAI-I content (obtained
from the proband's father) lysed at similar rates when
t - P A ~ y s - 2 9 6 , ~ r g - 2 9 8 , ~ r g 2 9 9(Fig
- ~ 1 ~ 4A)or streptokinase (Fig 4B)
were used to initiate plasmin formation.
Comparison of wild-type and mutant t-PA in the lysis of
platelet-rich clots. If platelet PAL1 contributes to plateletmediated clot lysis resistance, then a PAI- 1-insensitive t-PA
should lyse platelet-rich clots more efficientlythan wild-type
t-PA. To test this hypothesis, either wild-type or mutant tPA was usedto lyse clots formed in the presence or absence
of normal platelets. In preparation for these experiments,
standard curves of t-PA concentration (both for wild-type
and mutant enzymes) versusthe percentage ofclot lysis at l
hour were constructed (data not shown) to allow selection
of enzyme concentrations that resulted in essentially equal
amounts of clot lysis in the absence of platelets (ie, 0.6U/
mL wild-type t-PA and 2.0 ng/mL PAI-l-resistant t-PA).
As shown inFig 5, using theseconcentrations of t-PA, similar rates of clot lysis were observedin the absence of platelets. However, in the presence of platelets (2.5 X 108/mL),
the same concentrations of wild-type or mutant t-PA resulted in significantly different rates of clot lysis. Whereas
normal platelets markedly inhibited the fibrinolytic capacity of wild-typet-PA, the fibrinolytic activity of PAI- l-resistant t-PA was minimally inhibited under these same conditions.
DISCUSSION
Several results from these in vitro experiments suggest
that PAI-I is an important determinant of the antifibrinolytic effect ofplatelets. First, PAIl -deficient platelets inhibited t-PA-mediated clot lysisto a substantially lesser extent
than normal platelets. Second, the relative inhibition of fi-
100
75
la
U
8
25
k
0
50
100
150250 200
PlateletshL (XO
,l O
O
O
,O
O
)
Fig 3. Effect of platelet concentration on t-PA-mediated fibrinolysis. Fibrin clotswere formed in the presence oft-PA (0.67 U/
mL) and eithernormal platelets (m)or homozygousPAI-l-deficient
platelets (0) at the indicated concentrations.Clots were incubated
for 60 minutesat 37'C. The percentage of clot lysiswas then measured. Both normal and
PAl-l-deficient platelets inhibitfibrindysis
in a concentration-dependent manner, although normal platelets
exert a morepotent antitibrinolytic effect.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
354
FAY ET AL
that PAI- 1 is particularly abundant on fibrin fibersthat are
closely associated withthe platelet surface. These investigators hypothesize that the binding of PAI-1 to platelet-associated fibrin renders it more resistant to lysis, because fibrinbound PAL1 retains the capacity to inhibit t-PA.29,30 In
contrast, in studies using pharmacologic concentrations of
t-PA, Kunitada et all3 concluded that PAI-1 is not a determinant of platelet-mediated clot lysisinhibition. These in0
30
60
90
vestigators observedthat a PAI- 1-resistant t-PA mutant at
a
concentration of 800 U/mL did not lyse platelet-rich clots
Minutes
faster than wild-type enzyme. In addition, no differences
were observed in the t-PA activity of supernatants of clots
prepared from platelet-poor versusplatelet-richplasma.
The
enhanced efficiency of a PAI-1-resistant t-PA (com40
6o
pared with wild-type enzyme) in our experimental system,
but lack thereof in the experiments of Kunitada et a1,I3 is
most likely explained bythe substantial differences in t-PA
concentrations
used in the two studies (approximately
2o
00
30
60
90
120
1,000-fold).Platelets contain approximately 4,000 molecules of PAL1 percell.19Therefore, in clot lysis experiments
using a platelet concentration of 3 X 108/mL and a t-PA
Minutes
(specific activity, 580,000 U/mg) concentration of 800 U/
Fig 4. Effects of platelets on PAl-1-resistant t-PA- and strepmL (ie, the conditions of Kunitada et all3),the molar ratio
d it-PA:PAItokinase-mediated clot lysis. (A) t - P A L , . 2 P e , r ~ . - a s s ~ ~ s s ~ ~ ~ - m eof
1 is 10:1. In addition, because as much as 95%
ated clot lysis. Fibrin clotswere formed in the presence of PAL1l is reported to exist in a latent, or inactive,
of
platelet
PAIresistant t-PA (1.5 U/mL) and platelets (2.5 X 108/mL)and the perthe ratio of t-PAto functional inhibitor may have
centage of clot lysis was determined at theindicated time points.
exceeded 100-fold in theseexperiments. Under these condi(A) Homozygous PAl-l-deficient platelets (no detectable platelet
PAI-1); (A)platelets from the proband's father (normal levels of
tions, it is probably not possible to observe a PAI-l effect
platelet PAL1). (B) Streptokinase-mediated clot lysis. Experimental and, hence, to discriminate between the lytic efficienciesof
conditions were the same as for (A), except that streptokinase (8
wild-type and PAI- l -resistant enzymes. Taken together,
mU/mL) was substituted formutant t-PA. Labels are as
in (A).
these studies suggest that clot-associated PAL1may inhibit
endogenously mediated fibrinolysis,inwhichlocalt-PA
brinolysis by plateletswith normal PAI-l content (comconcentrations are probably in the nanogram per milliter
paredwithPAI-l-deficient
platelets) was not observed
range,31but not pharmacologically mediated fibrinolysis, in
when plasminformation was initiated by PAI- l -insensitive
which plasma t-PA concentrations can exceed 1 pg/mLS3*
plasminogen activators (ie, the PAI- 1-resistant t-PA mutant
or streptokinase).And finally, PAI1-resistant t-PAwas significantly more effective than wild-type t-PA in lysingclots
containing normal platelets. However, these experiments
'OOt
also suggest that platelets inhibit fibrinolysis by a mechanism(s) that does not involve PAI-l. Althoughreduced
compared with normal platelets, platelets lacking PAI1 retained the capacity to inhibit fibrinolysis. These findings
are
consistent with a contribution of platelet-mediated clot retraction to lysis resistance,as suggested by Kunitada et al.I3
Indeed, visible clot retraction was observed during our experiments and appeared to be proportional to platelet con0
30
60
90150 120
centration for both normal and PAI- 1-deficient platelets.
Minutes
The results ofour studies are also consistent with the reFig 5. Comparison ofwild-type and PAl-l-resistant t-PA in the
in which an anti-PAI- 1 monoclonal anport of Levi et
clots were
lysisof platelet-rich andplatelet-poorclots.Fibrin
tibody markedly accelerated the capacity of t-PA to lyse
inthe absence or presence of normalplatelets (2.5 X 10')
platelet-rich clots in vitro. These investigators also showed formed
andeither0.6U/mLofwild-typet-PA(WT-t-PA)or2ng/mLofPAIthat this antibody, which inhibits the interaction of PAI-l
l-resistant t-PA (Mut-t-PA). The percentage of dot lysis was dewitht-PA, enhances spontaneous clotlysis and inhibits
termined at the indicated time points. ( 0 ) W-t-PA/ - platelets; (W
(WT-t-PA/ + platelets; (A) Mut-t-PA/ - platelets; (A)Mut-t-PA/
thrombus extension in vivo.Braaten et a12*also showedthat
+ platelets. Data are from a singleexperimentthat was performed
an anti-PAI- 1 monoclonal antibody accelerates clot lysisin
in triplicate with similar results. Note the similar rates of lysis of
vitro. In addition, this group used immuno-electron microsplatelet-poor clots for either wild-type or PAl-l-resistant t-PA.
copy to demonstrate that PAI- 1 in platelet-rich thrombi is
However, the PAl-l-resistant t-PA is significantly more efficient in
dissolvingplatelet-richclots than wild-type t-PA.
localized on fibrin, as opposed to the platelet surface, and
loot
1
l
/
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
PLATELET PAL1AND
FIBRINOLYSIS
355
3. Collen D Coronary thrombolysis: Streptokinaseor recombinant tissue-type plasminogenactivator? Ann Intern Med 112:529,
1991
4. Jang I-K, Gold HK, Ziskind AA, Fallon JT, Holt RE, k i n bach RC,May JW, Collen D Differential sensitivityoferythrocyterich and platelet-rich arterial thrombi to lysis withrecombinant tissue-type plasminogenactivator. Circulation 79:920, 1989
5. Booth NA, Robbie LA, Croll AM, Bennett
B Lysis ofplateletrich thrombi: The role of PAI-I . Ann NY Acad Sci 667:70, 1992
6. Levi M, Biemond BJ, van Zonneveld A-J, Wouter Ten Cate
J, Pannekoek H: Inhibition of plasminogen activator inhibitor-l
activity resultsin promotion of endogenousthrombolysis and inhibition of thrombus extension in models of experimental thrombosis. Circulation 85:305, 1992
7. Warnes CA, Roberts W C Sudden coronary death Comparison of patients with to those without coronary thrombus at necropsy. Am J Cardiol54: 1206, 1984
8. Coller BS: Platelets and thrombolytic therapy. N Engl J Med
322:33, 1991
9. Erickson LA, Ginsberg MH, Loskutoff DJ:Detection and partial characterization of an inhibitor of plasminogen activator in human platelets. J Clin Invest74: 1465, 1984
10. Francis CW, Marder VJ: Increased resistanceto plasmic degradation of fibrin with highly crosslinked alpha-polymer
chains
formed at high factor XI11 concentrations. Blood 7 1:I36 1, 1988
1 1. Miles LA, Plow EF: Bindingand activation of plasminogen
on the platelet surface.J Biol Chem 260:4303, 1985
12. Miles LA, Ginsberg MH, White JG, Plow E F Plasminogen
interacts with human platelets through two distinct mechanisms. J
Clin Invest 77:2001, 1986
13. Kunitada S, Fitzgerald GA, Fitzgerald DJ:
Inhibition of clot
lysis and decreased binding of tissue-type plasminogen
activator as
a consequence of clot retraction. Blood 79: 1420, 1992
14. Loskutoff DJ, Sawdey M, Mimuro J: Type 1 plasminogen
activator inhibitor. Prog HemostThromb 9:87, 1989
15. Hamsten A, Wiman B, de Faire U, BlombackM: Increased
plasma levelsof a rapid inhibitor of tissue plasminogenactivator in
young survivorsof myocardial infarction. N Engl J Med 3 13: 1557,
1985
16. Schleef RR, Higins DL, Pillemer E, Levitt LJ: Bleeding diathesis due to decreased functional activity of type 1 plasminogen
activator inhibitor. J Clin Invest 83: 1747, 1989
17. Diival J, Nguyen G, Gross S, Delobel J, Kruithof EKO: A
lifelong bleedingdisorder associated witha deficiency of plasminogen activator inhibitor type 1. Blood 77528, 1991
18.FayWP, Shapiro AD, Shih JL, Schleef RR, Ginsburg D
Complete deficiency of plasminogen-activator inhibitor type 1 due
to a frame-shift mutation. N Engl J Med 327: 1729, 1992
19. Kruithof EKO, NicolosaG, Bachmann F Plasminogen activator inhibitor 1: Development of a radioimmunoassay and observations on its plasma concentration during venous occlusion and
after platelet aggregation.Blood 70: 1645, 1987
ACKNOWLEDGMENT
20. Hekman CM, Loskutoff DJ: Endothelial cells produce a laThe authors thank Dr David Williams (Howard Hughes Medical tent inhibitor of plasminogen activators that can be activated by
Institute, Indiana University MedicalCenter), in whose laboratory
denaturants. J Biol Chem 260:1 1581, 1985
experiments withPAI-l-deficientplateletswere
performed; Dr
2 l. Schleef RR, Sinha M, LoskutoffDJ: Immunoradiometric asDaniel Lawrence, who provided helpful suggestions
during the desay to measure the binding of a specific inhibitor to tissue-type plassign and analysis of these experiments; and Suzann Labun, who
minogen activator. J Lab Clin Med 106:408, 1985
assisted inthe preparation of the manuscript.
22. Booth NA, Simpson AJ,Croll A, Bennett B, MacGregor IR:
Plasminogen activator inhibitor (PAI-I ) in plasmaand platelets. Br
REFERENCES
J Haematol70:327, 1988
I . Erickson LA, Schleef RR, Ny T, Loskutoff DJ: The fibrino23. Madison EL, Goldsmith EJ, Gerard RD, Gething MH, Samlytic systemof the vascular wall. Clin Haematol 145 13, 1985
brook JF, Bassel-Duby RS: Amino acid residuesthat affect interac2. Collen D, Lijnen HR: Basic and clinical aspectsof fibrinolysis
tion of tissue-type plasminogenactivator with plasminogen activaand thrombolysis. Blood 78:3114, 1991
tor inhibitor 1. Proc NatlAcad Sci USA 87:3530, 1990
However, the concentration of platelets typically used in in
vitro clot lysis assays( lo8 to 109/mL5*6)
is probably considerably less than the concentration of platelets within a hemostatic plug or platelet-rich thrombus. For example, in our
experimental system, platelets constituted less than 1% of
the total volume of the clot lysis mixture. However, microscopic evaluation of coronary thrombi retrieved from patients with acute myocardial infarction or sudden cardiac
death shows that they consist, to a considerable extent, of
essentially solid masses platelets
of
with interspersedislands
of fibrin.’Hence, the concentration of PAI-1 in platelet-rich
clots in vivo may be considerably higher than is present in
in vitro assays. Consistent with this hypothesis, Potter van
Loon et aP3showed that the average amount of PAI-1 antigen in human arterial thrombi, which are frequently platelet-rich, is l50 times that detected in an equivalent volume
of platelet-poor plasma. Similarly, the mean concentration
ofPAI-1in experimentally induced porcine platelet-rich
thrombi (36 pg/mL) exceeds that detected in plasma by
greater than two orders of magnitude.34 Whether active
PAI-I accumulates in human platelet-rich thrombi in
sufficient concentrations to inhibit pharmacologically-induced fibrinolysis is unknown. However, platelets contain
vitronectin, which stabilizes PAI-1 in the active conformation,35*36
and phospholipid vesicles can convert latent PAIl to the active form,37suggesting that reactivation of latent
PAI- 1 may occur on cell surfacesin vivo.
In summary, our experiments with PAI-l-deficient platelets and plasminogen activators with differential sensitivity
to inhibition by PAI-l suggest that PAI-I is an important
determinant of platelet-dependent clot lysis inhibition in vitro. Our studies also suggestthat platelets inhibit fibrinolysis
in a PAI-l-independent manner, consistent with prior observations that platelet-mediated clot retraction inhibits fibrin01ysis.l~Although these results suggest
that platelet PAI1 may playan important role inthe regulation of fibrinolysis
within the microenvironment of the platelet-rich clot, additional studies are necessary to adequately test this hypothesis. For example, animal studies comparing the lysisof
platelet-rich thrombi in responseto wild-type versus PAL
1resistant t-PAs should help to define the role of PAL1 in
platelet-mediated clot lysis inhibition in vivo. In addition
to defining mechanisms underlying the inhibitory effect of
platelets on fibrinolysis, such studies may also suggest alternative strategies for enhancing thrombolytic therapy in
patients with acute thrombotic disease.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
356
24. Lewis SD, Shafer JA: A thrombin assay based upon the release of fibrinopeptide A fromfibrinogen:Definition of a new
thrombin unit. Thromb Res 35: 1 I I , 1984
25. Deutsch D G , Mertz ET: Plasminogen: Purificationfrom human plasma by affinity chromatography.Science 1 7 0 1095, 1970
26. Legrand C,Dubemard V, Nurden A T Characteristicsof collagen-induced fibrinogenbinding to human platelets. Biochim Biophys Acta8 I2:802, 1985
27. Smith DS, Harmon J, Owen WG: A sensitive and specific
assay for plasminogen activators.
Thromb Res 37533, 1985
28. Braaten JV, Handt S, Jerome WG, Kirkpatrick J, Lewis JC,
Hantgan RR: Regulation of fibrinolysis by platelet-released plasminogen activator inhibitor 1: Light scattering and ultrastructural
examination of lysis of a model platelet-fibrinthrombus. Blood 8 1:
1290,1993
29. Wagner OF, de, VriesC,
Hohmann C, Veerman H,
Pannekoek H: Interaction between plasminogenactivator inhibitor
type 1 (PAI-l) bound to fibrin and either tissue-type plasminogen
activator (t-PA) or urokinase-type plasminogenactivator (U-PA).J
Clin Invest 84:647, 1989
30. Reilly CF,Hutzelmann JE: Plasminogenactivator inhibitor1 binds to fibrin and inhibits tissue-type plasminogenactivator-mediated fibrin dissolution.J Biol Chem 267: 17 128, 1992
3 1. Francis CW, Marder VJ: Physiologic regulation and pathologic disorders of fibrinolysis,in Colman RW, HirshJ, Marder VJ,
FAY ET AL
Salzman EW (eds): Hemostasis and Thrombosis. Basic Principles
and Clinical Practice. Philadelphia, PA,
Lippincott, 1987, p 358
32. Collen D, Bounameaux H, De Cock F, Lijnen HR, Verstraete
M: Analysis of coagulation and fibrinolysis during intravenous infuin pasion of recombinant human tissue-type plasminogen activator
tients with acute myocardial infarction. Circulation
735 l l, 1986
33. Potter van Loon BJ, Rijken DC, Brommer UP, Van der Maas
APC: The amount of plasminogen, tissue-type plasminogen activator
and plasminogen activator inhibitor
type l in human thrombi and the
relation to ex-vivo lysibility.Thromb Haemost 67: 101,1992
34. Fay WP, Schwartz RS, Holmes DR, Owen WG: High concentrations of plasminogen activator inhibitor-l within coronary
artery thrombi. Circulation 82:2388, 1990 (abstr, suppl 11)
35. Declerck PJ, De MolM,AlessiMC,Baudner
S, Piques
E-P, Preissner KT, Muller-BerghausG , Collen D Purification and
characterizationof a plasminogen activator inhibitor 1 binding protein fromhuman plasma. J Biol Chem 263:15454, 1988
36. Mimuro J, Loskutoff DJ: Purification of a protein from bovine plasma that binds to type 1 plasminogen activator inhibitor
and prevents itsinteraction with extracellular matrix. J Biol Chem
264:936, I989
37. Lambers JW, Cammenga M, Konig BW, Mertens K,
Pannekoek H, van Mourik JA: Activation of human endothelial
cell-typeplasminogen activator inhibitor (PAI-l) by negatively
charged phospholipids.J Biol Chem 262:17492, 1987
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1994 83: 351-356
Platelets inhibit fibrinolysis in vitro by both plasminogen activator
inhibitor-1-dependent and -independent mechanisms
WP Fay, DT Eitzman, AD Shapiro, EL Madison and D Ginsburg
Updated information and services can be found at:
http://www.bloodjournal.org/content/83/2/351.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.