Thrombin-mediated activation of factor XI results in a thrombin-activatable fibrinolysis inhibitor-dependent inhibition of fibrinolysis.

Abstract

Recently, it has been shown that Factor XI can be activated by thrombin, and that Factor XIa significantly contributes to the generation of thrombin via the intrinsic pathway after the clot has been formed. This additional thrombin, generated inside the clot, was found to protect the clot from fibrinolysis. A plausible mechanism for this inhibitory effect of thrombin involves TAFI (thrombin-activatable fibrinolysis inhibitor, procarboxypeptidase B) which, upon activation, may inhibit fibrinolysis by removing carboxy-terminal lysines from fibrin. We studied the role of Factor XI and TAFI in fibrinolysis using a clot lysis assay. The lysis time was decreased twofold when TAFI was absent, when TAFI activation was inhibited by anti-TAFI antibodies, or when activated TAFI was inhibited by the competitive inhibitor (2-guanidinoethylmercapto)succinic acid. Inhibition of either TAFI activation or Factor XIa exhibited equivalent profibrinolytic effects. In the absence of TAFI, no additional effect of anti-Factor XI was observed on the rate of clot lysis. We conclude that the mechanism of Factor XI-dependent inhibition of fibrinolysis is through the generation of thrombin via the intrinsic pathway, and is dependent upon TAFI. This pathway may play a role in determining the fate of in vivo formed clots.

Full text

Page 1

Inhibition of Fibrinolysis by a Factor XI– and TAFI-dependent Mechanism

2323

J. Clin. Invest.
© The American Society for Clinical Investigation, Inc.
0021-9738/97/05/2323/05$2.00
Volume 99, Number 10, May 1997, 2323–2327

Rapid Publication

Thrombin-mediated Activation of Factor XI Results in a Thrombin-activatable
Fibrinolysis Inhibitor-dependent Inhibition of Fibrinolysis

Peter A. Kr. Von dem Borne,* Laszlo Bajzar,
*
Department of Haematology, University Hospital, 3508 GA Utrecht, The Netherlands;
Vermont, Burlington, Vermont 05405; and
Department of Biochemistry and Department of Medicine, Queen’s University, Kingston,
Ontario K7L 3N6, Canada

‡

Joost C.M. Meijers,* Michael E. Nesheim,

§

and Bonno N. Bouma*
Department of Biochemistry, University of



‡


§


Abstract

Recently, it has been shown that Factor XI can be activated
by thrombin, and that Factor XIa significantly contributes
to the generation of thrombin via the intrinsic pathway after
the clot has been formed. This additional thrombin, gener-
ated inside the clot, was found to protect the clot from fi-
brinolysis. A plausible mechanism for this inhibitory effect
of thrombin involves TAFI (thrombin-activatable fibrinoly-
sis inhibitor, procarboxypeptidase B) which, upon activa-
tion, may inhibit fibrinolysis by removing carboxy-terminal
lysines from fibrin.
We studied the role of Factor XI and TAFI in fibrinolysis
using a clot lysis assay. The lysis time was decreased two-
fold when TAFI was absent, when TAFI activation was in-
hibited by anti–TAFI antibodies, or when activated TAFI
was inhibited by the competitive inhibitor (2-guanidinoeth-
ylmercapto)succinic acid. Inhibition of either TAFI activa-
tion or Factor XIa exhibited equivalent profibrinolytic effects.
In the absence of TAFI, no additional effect of anti–Factor
XI was observed on the rate of clot lysis.
We conclude that the mechanism of Factor XI–depen-
dent inhibition of fibrinolysis is through the generation of
thrombin via the intrinsic pathway, and is dependent upon
TAFI. This pathway may play a role in determining the fate
of in vivo formed clots. (
J. Clin. Invest.
Key words: coagulation

•
fibrinolysis
carboxypeptidase


1997. 99:2323–2327.)
Factor XI
•




•



thrombin

•


Introduction

The activation of Factor XI by thrombin was first demon-
strated in the presence of the nonphysiological glycosami-
noglycan dextran sulfate (1, 2). Later it was shown that throm-
bin-mediated Factor XI activation could also take place in the
absence of an activating surface, although only trace amounts
of Factor XIa were formed (3). Activation of Factor XI by
thrombin is potentially important because these traces of Fac-
tor XIa contribute to the formation of large amounts of acti-
vated Factor X by the dramatic amplification in the coagula-
tion cascade (3).
The activation of Factor XI by thrombin was also demon-
strated in a plasma environment. In a tissue factor–induced
clotting system using sulfatides as a surface, Factor XI was
found to induce activation of Factor IX independent of Factor
XII (4). More interestingly, Factor XI was also shown to be ac-
tivated by thrombin in the absence of a surface (5). In a tissue
factor–induced clotting system, Factor XI increased the rate of
fibrin formation, confirming the contribution of Factor XI/XIa
to the coagulation system (5). Furthermore, the thrombin-me-
diated activation of Factor XI inhibited tissue type plasmino-
gen activator (t-PA)
-induced clot lysis (5). Trace amounts of
activated Factor XI, representing 0.01% activation, were suffi-
cient to inhibit fibrinolysis maximally. Exhaustive activation of
prothrombin takes place predominantly after the formation of
the fibrin clot, however, only in the presence of Factor XI (5).
Therefore, protection of the clot from fibrinolytic attack was
dependent on the thrombin that was generated in a Factor XI–
dependent manner via the intrinsic pathway. The inhibitory ef-
fect of Factor XI on fibrinolysis provides an explanation for
why patients with a Factor XI deficiency bleed from tissues
with high local fibrinolytic activity (6).
The mechanism behind the inhibitory effect of thrombin is
unclear. A potential candidate is TAFI (thrombin-activatable
fibrinolysis inhibitor) (7). TAFI was isolated in the search for a
mechanism behind the profibrinolytic effect of activated pro-
tein C (7–11). This protein C–dependent profibrinolytic effect
had been attributed to downregulation of thrombin activation
(12); TAFI, therefore, provides a connection between throm-
bin generation and inhibition of fibrinolysis. Amino-terminal
sequence analysis showed that TAFI is similar to plasma pro-
carboxypeptidase B (7). Plasma procarboxypeptidase B was
isolated in 1991 as a plasminogen-binding protein (13), and
was found to be very similar to the tissue form isolated some
30 yr earlier (14). In addition, a carboxypeptidase was purified
from serum, and was called carboxypeptidase U because it was
unstable (15). Carboxypeptidase U was later shown to be iden-
tical to plasma carboxypeptidase B (16). The mechanism by
which activated TAFI inhibits fibrinolysis is still unclear. It is
hypothesized that activated TAFI removes carboxy-terminal
lysines from fibrin that play a key role in plasminogen binding

1


Address correspondence to Bonno N. Bouma, Ph.D., Department of
Haematology, G03.647, University Hospital, P.O. Box 85.500, 3508
GA Utrecht, The Netherlands. Phone: 31-30-250-6513; FAX: 31-30-
2511893.
Received for publication 11 December 1996 and accepted in re-
vised form 10 March 1997.


1.
brinolysis inhibitor; t-PA, tissue type plasminogen activator.

Abbreviations used in this paper:

TAFI, thrombin-activatable fi-

Page 2

2324

Von dem Borne et al.

and activation (7, 13, 17–19). Many interactions in the fibri-
nolytic system, however, are promoted by the presence of
lysine binding sites which could be inhibited by the action of
carboxypeptidases. We investigated the inhibition of fibrinoly-
sis by thrombin (reciprocally generated via the intrinsic path-
way initiated by thrombin-mediated Factor XI activation) in
clots formed from plasma. Our results indicate that Factor XI–
dependent inhibition of fibrinolysis is TAFI dependent.

Methods

Materials.
Sigma (St. Louis, MO). GEMSA (2-guanidinoethylmercaptosuccinic
acid) was obtained from Fluka AG (Buchs, Switzerland). Tissue type
plasminogen activator (531,000 IU/mg) was obtained from Chro-
mogenix (Mölndal, Sweden). Other chemicals obtained were the best
grade available. Factor XI was purified from fresh frozen plasma us-
ing a murine anti–Factor XI monoclonal antibody (XI-5) immobi-
lized on CNBr-activated Sepharose, as described earlier (3). The spe-
cific coagulant activity of Factor XI was 235 U/mg. Factor XIa was
prepared from Factor XI using
?
-Factor XIIa as described previously
(20). A monoclonal antibody against Factor XI capable of blocking
Factor IX activation (XI-1) has been described (21). A monoclonal
antibody against TAFI (mAbTAFI#16) was prepared as described
(22). A polyclonal antibody against TAFI was prepared in rabbits.
The antiserum was purified on protein G–Sepharose, and was able to
completely inhibit TAFI in a clot lysis assay.
TAFI was purified from fresh frozen plasma as described (7).
TAFI-deficient plasma was obtained by passing plasma from a
healthy donor over a Sepharose column to which mAbTAFI#16 (2.0
mg/ml) was coupled. The undiluted flow-through fractions of the col-
umns were pooled and used as deficient plasma. Less than 0.1% of
TAFI was present in the deficient plasma as determined by ELISA.
Factor XII–deficient plasma was obtained from a congenitally defi-
cient patient. Plasma deficient in both Factor XII and Factor XI was
prepared by passing the Factor XII–deficient plasma over a Seph-
arose column to which a murine anti–Factor XI antibody (XI-5) was
immobilized. Less than 0.01% of Factor XI was present in the Factor
XI and XII double-deficient plasma. Purified human thrombin was a
generous gift from Dr. W. Kisiel (University of New Mexico, Albu-
querque, NM) (23).
Clot lysis assay in clots formed from plasma.
induced clot lysis by thrombin-mediated Factor XI activation was
studied in clots formed from human plasma by monitoring the change
in turbidity during fibrin formation and lysis at 405 nm in a micro-
plate reader as described earlier (5). All experiments were performed
in citrated plasma recalcified with CaCl
resulting in a free calcium concentration of 2.3 mM. The plasmas con-
tained sufficient endogenous phospholipids to sustain coagulation.
Thrombin (20 nM), calcium necessary for recalcification, and t-PA
(10–30 U/ml) were added to 75
?
justed to 150
?
l with Hepes buffer (25 mM Hepes, 137 mM NaCl, 3.5
mM KCl, 3 mM CaCl
2
, 0.1% BSA, pH 7.4) resulting in a final plasma
concentration of 50%. After mixing, 100
was pipetted in a microtiter plate, and was then incubated at 37
during which turbidity was measured at different time points. Lysis
time (defined as the time to achieve a 50% reduction of the maximum
turbidity) was determined from the plot of A
independent experiments. A monoclonal anti–Factor XI antibody (XI-1)
(21), capable of blocking the activation of Factor IX by Factor XIa,
was preincubated for 30 min with the plasma to determine the effect
of Factor XI. To assess the requirement for TAFI in the effect of Fac-
tor XI or Factor XIa on clot lysis in this system, mAbTAFI#16, or a
rabbit antibody against TAFI (antibodies that both inhibit the activa-
tion of TAFI) was also preincubated for 30 min with the plasma. Al-
ternatively, the effect of XI-1 on clot lysis was also studied with
TAFI-deficient plasma in the presence and absence of a reconstitut-

Bovine serum albumin (fraction V) was purchased from




The inhibition of t-PA-

2

(final concentration of 17 mM),


l of plasma. The volume was ad-





?

l of the reaction mixture

?

C

405

versus time of three
ing amount of purified TAFI. All experiments (except those with the
TAFI-deficient, or Factor XI and XII double-deficient plasma) were
performed in pooled plasma from 40 healthy volunteers.

Results

Effect of TAFI and Factor XI on the lysis of a thrombin-induced
clot.
To study the effect of both Factor XI and TAFI on clot
lysis, thrombin (20 nM) and t-PA (30 U/ml) were added to re-
calcified plasma preincubated with antibodies that inhibit Fac-
tor XIa (XI-1), or antibodies that inhibit activation of TAFI
(mAbTAFI#16). The presence of 20 nM of thrombin pro-
duced a clot within 10 min (5). Subsequent t-PA–induced clot
lysis was followed in time by measuring turbidity of the formed
clot. In the presence of the Factor XIa blocking antibody XI-1,
an enhancement of clot lysis was observed that was caused by
inhibition of thrombin generation by the intrinsic pathway
(Fig. 1). The antibody XI-1 was found to have a maximal effect
on clot lysis at a concentration of 25
mAbTAFI#16 also enhanced lysis of the clot, an effect that
was also maximal at a concentration of 25
shown). Therefore, the presence of mAbTAFI#16 and XI-1
resulted in a rapid lysis of the clot (Fig. 1). The lysis that was
observed with XI-1 did not differ from the lysis observed with
TAFI-16 (lysis times 42
?
4 and 41
time in the absence of antibodies: 92
be enhanced further by the addition of both antibodies to-
gether (lysis time: 42
?
4 min).
The effect of both Factor XI and TAFI on the lysis of a
thrombin-induced clot was also studied in TAFI-deficient
plasma to which XI-1 and a reconstituting amount of purified
TAFI (60 nM) were added. Essentially the same effects of Fac-
tor XI and TAFI were observed (Fig. 2). Again, in the absence
of TAFI, no effect of XI-1 on clot lysis was observed.
The effect of Factor XIa and TAFI on clot lysis.
of Factor XIa and TAFI on the lysis of a thrombin-induced


?

g/ml (data not shown).

?

g/ml (data not



?

3 min, respectively; lysis
?
5 min). Lysis could not






The effect
Figure 1. Lysis of a thrombin-induced clot in the presence or absence
of antibodies against Factor XI or TAFI. Normal plasma was preincu-
bated with XI-1 (?, 160 nM), mAbTAFI#16 (?, 160 nM), XI-1 and
mAbTAFI#16 (?, 160 nM), or buffer (?). Coagulation was initiated
by adding calcium necessary for recalcification, thrombin (20 nM),
and t-PA (30 U/ml). Fibrin formation and lysis were measured in time
as the change in turbidity at 405 nm. The means?SD of three inde-
pendent experiments are given.

Page 3

Inhibition of Fibrinolysis by a Factor XI– and TAFI-dependent Mechanism

2325

clot was also studied in plasma that was deficient in both Fac-
tor XI and XII (Fig. 3). Addition of trace amounts of Factor
XIa (12.5 fM) already resulted in an increase in the lysis time
(from 47 to 66 min). The lysis time was further increased (to
110 min) by the addition of higher concentrations of Factor
XIa. Preincubation of the plasma with an anti-TAFI antibody
completely abolished the effect of Factor XIa on the rate of
clot lysis. This indicates that the effect of TAFI on clot lysis is
dependent on the amount of Factor XIa generated in a nor-
mally functioning coagulation cascade.

The effect of XI-1 and mAbTAFI#16 on clot lysis using var-
ious concentrations of t-PA.
To exclude the possibility that, in
the absence of TAFI, lysis may take place so fast that no effect
of Factor XI can be observed, we studied the effect of XI-1 and
mAbTAFI#16 on the lysis of a thrombin-induced clot using
various concentrations of t-PA (10–30 U/ml). At all t-PA con-
centrations used, no additional Factor XI effect was discern-
ible in the presence of the antibody that inhibits activation of
TAFI (Fig. 4).
Effect of GEMSA on the clot lysis assay.
petitive inhibitor of carboxypeptidase B (7, 13), was also used
to study the contribution of TAFI on the effect of Factor XI on
clot lysis. By using different concentrations of GEMSA, it
was determined that a maximum effect on clot lysis was
reached at 0.5–1 mM (data not shown). Therefore, the effect
of the Factor XIa–inhibiting antibody XI-1 on the lysis of a
thrombin-induced clot was studied in the presence and ab-
sence of 1 mM GEMSA. In the presence of XI-1, GEMSA
could not enhance the rate of clot lysis (Fig. 5), suggesting that
to inhibit fibrinolysis, TAFI needs thrombin that is generated
in a Factor XI–dependent manner via the intrinsic pathway. In
the presence of GEMSA however, the addition of XI-1 re-
sulted in a faster lysis of the clot. This result can be explained
by the fact that GEMSA is a competitive inhibitor incapable of
completely inhibiting activated TAFI.



GEMSA, a com-

Discussion

Patients with a deficiency of Factor XI are prone to bleed from
tissues with a high local fibrinolytic activity (6). A possible ex-
planation for this clinical observation was recently provided
when it was demonstrated that the presence of Factor XI dur-
ing coagulation has an inhibitory effect on clot lysis (5). An in-
tact intrinsic pathway was found to be necessary for the inhibi-
Figure 2. Effect of TAFI on the lysis of a thrombin-induced clot in
the presence or absence of a Factor XIa–inhibiting antibody. Plasma
deficient in TAFI was preincubated with TAFI (?, 60 nM), XI-1 (?,
160 nM), TAFI and XI-1 (?, 60 and 160 nM, respectively), or buffer
(?). Coagulation was initiated by adding calcium necessary for recal-
cification, thrombin (20 nM), and t-PA (30 U/ml). Fibrin formation
and lysis were measured in time as the change in turbidity at 405 nm.
Each point represents the mean?SD of three independent experi-
ments.
Figure 3. Effect of Factor XIa on the lysis of a thrombin-induced clot.
Factor XI and XII double-deficient plasma was preincubated with
buffer (?), Factor XIa: 12.5 fM (?), 125 fM (?), 12.5 pM (?); and
Factor XIa: 12.5 pM in the presence of a polyclonal anti-TAFI anti-
body (?). Coagulation was initiated by adding calcium necessary for
recalcification, thrombin (20 nM), and t-PA (30 U/ml). Fibrin forma-
tion and lysis were measured in time as the change in turbidity at
405 nm.
Figure 4. Lysis of a thrombin-induced clot in the presence or absence
of antibodies against Factor XI or TAFI: effect of various t-PA
concentrations. Normal plasma was preincubated with 160 nM
mAbTAFI#16 in the presence (closed symbols) or absence (open
symbols) of 160 nM XI-1. Coagulation was initiated by adding cal-
cium necessary for recalcification, thrombin (20 nM), and t-PA
(10–30 U/ml). Fibrin formation and lysis were measured in time as
the change in turbidity at 405 nm. t-PA: 30 U/ml (?), 25 U/ml (?),
20 U/ml (?), 10 U/ml (?). The mean?SD of three independent ex-
periments are given.

Page 4

2326

Von dem Borne et al.

tion of fibrinolysis by Factor XI because an increased rate of
clot lysis was observed in Factor XI–deficient plasma as well as
in Factor II–; IX–, or X–deficient plasma, whereas in Factor
VII–deficient plasma, a normal lysis rate was observed (5, 24).
To determine which Factor of the intrinsic pathway was re-
sponsible for inhibition of fibrinolysis, clot lysis was studied in
prothrombin-deficient plasma to which Factor IXa, Xa, or
thrombin was added. Factor IXa and Xa had no effect on fi-
brinolysis, whereas a concentration-dependent antifibrinolytic
effect of thrombin was observed. We concluded that the antifi-
brinolytic effect of Factor XI was mediated by the formation of
additional thrombin in a reciprocal activation process involv-
ing the activation of Factor XI by thrombin in the intrinsic
pathway (5). These effects were observed in the absence of an
activating surface (e.g., dextran sulfate). Previous studies had
shown that activation of Factor XI by thrombin takes place in
the absence of such a surface (2, 3, 5). Small amounts of acti-
vated Factor XI are generated, but even very small amounts of
activated Factor XI, representing 0.01% activation, are capa-
ble of inhibiting clot lysis (5), showing not only the amplifica-
tion power of the coagulation cascade, but also the potential
importance of traces of Factor XIa. An activating surface on
which activation of Factor XI takes place, however, may fur-
ther potentiate this process.
Recently, a thrombin-activatable fibrinolysis inhibitor (TAFI)
has been described (7) that may provide an explanation for the
observed inhibitory effect of Factor XI on clot lysis (5). There-
fore, in this report we investigated the Factor XI–dependent
inhibition of fibrinolysis in the presence and absence of TAFI
activation.
Experiments were performed with the same clot lysis assay
that was used to show the inhibitory effect of Factor XI on clot
lysis (5). In this assay, fibrin formation is induced by the addi-
tion of thrombin to recalcified plasma, after which t-PA–medi-
ated lysis of the clot is monitored by change in turbidity. In this
way, interference of the extrinsic pathway is excluded. Fur-
thermore, we have shown previously that the contact activa-
tion does not play a role in this system, since Factor XIIa–
blocking antibodies did not affect lysis time (5). The effect of
Factor XI on lysis was determined by the addition of a Factor
XIa–inhibiting antibody (XI-1). The presence or absence of
Factor XI had no effect on clotting time in our system, nor had
it an effect on the structure of the clot (data not shown). The
plasmas contained sufficient endogenous phospholipids to sus-
tain coagulation. Addition of exogenous phospholipids gave
identical results.
First, the effect of TAFI in the lysis assay was studied with
a monoclonal antibody that inhibits activation of TAFI. Lysis
in the presence of this antibody took place at the same rate as
in the presence of Factor XIa–inhibiting antibody. An additive
or synergistic effect was not observed in the presence of both
antibodies. We conclude that the Factor XI–dependent inhibi-
tion of clot lysis is mediated through TAFI. Second, it was
shown that addition of increasing amounts of Factor XIa resulted
in an increase in the lysis time of a thrombin-induced fibrin
clot in plasma deficient in both Factor XI and XII. The effect
of Factor XIa was abolished by preincubation with anti–TAFI
antibodies, indicating that the effect of TAFI on clot lysis is de-
pendent on the amount of Factor XIa generated. Then, lysis
experiments were performed with TAFI-deficient plasma in
the presence and absence of a reconstituting amount of TAFI.
The effects of XI-1 on lysis were comparable to those observed
with the antibody that inhibits activation of TAFI. In the ab-
sence of TAFI, XI-1 had no effect on clot lysis. Finally, lysis
experiments were performed in clots formed from plasma in
the presence and absence of GEMSA, a competitive inhibitor
of the carboxypeptidase B family of enzymes, which includes
activated TAFI. In the presence of GEMSA, lysis times were
short (relative to the lysis time in the absence of GEMSA), but
XI-1 did not reduce lysis time in the presence of GEMSA.
These results further indicate that TAFI plays a central role in
the Factor XI–dependent inhibition of clot lysis.
In contrast to the small amounts of thrombin that are
needed for clot formation, high concentrations of thrombin are
needed for both the inhibition of clot lysis (5) and the activa-
tion of TAFI (7). Earlier studies in which thrombin activity
was measured during coagulation showed that relatively small
amounts of thrombin (10–20 nM) are present at the moment of
clotting, whereas high concentrations of thrombin (100–200
nM) are generated within the clot (25). We also found that
prothrombin activation continued inside the fibrin clot (5) by
measuring generation of F1
?
2. Exhaustive prothrombin acti-
vation was observed in a Factor XI–dependent way in parallel
to the antifibrinolytic effect (5). These data indicated that the
concentrations of thrombin that are generated inside the clot
in a Factor XI–dependent way are sufficient for the activation
of TAFI. Activated TAFI is then able to protect the clot from
fibrinolytic attack.
In the presence of endothelial cells, the thrombin that is
generated by the intrinsic pathway will also activate protein C
on thrombomodulin. Activated protein C can reduce thrombin
generation by inactivating Factor Va and VIIIa, resulting in a
reduced activation of TAFI and increased clot lysis. This mech-
anism may explain the reported profibrinolytic effects of acti-
vated protein C (8–11). Thrombomodulin, however, is also
capable of enhancing the activation of TAFI by thrombin by
three orders of magnitude (26). It still remains to be demon-


Figure 5. Effect of GEMSA on the lysis of a thrombin-induced clot in
the presence or absence of a Factor XIa–inhibiting antibody. Normal
plasma was preincubated with GEMSA (?, 1 mM), XI-1 (?, 160
nM), GEMSA and XI-1 (?, 1 mM and 160 nM), or buffer (?). Coag-
ulation was initiated by adding calcium necessary for recalcification,
thrombin (20 nM), and t-PA (30 U/ml). Fibrin formation and lysis
were measured in time as the change in turbidity at 405 nm. Each
point represents the mean?SD of three independent experiments.

Page 5

Inhibition of Fibrinolysis by a Factor XI– and TAFI-dependent Mechanism

2327

strated whether the protein C– or TAFI–stimulating effect of
thrombomodulin dominates in vivo.
We conclude that TAFI plays a crucial role in the inhibition
of fibrinolysis by thrombin generated via the intrinsic pathway
through thrombin-mediated Factor XI activation. This path-
way may thereby play an important role in determining the
fate of clots that are formed in vivo. Inhibition of this pathway
may result in clots that are more susceptible to fibrinolysis, and
may therefore prove to be a valuable addition to fibrinolytic
therapy. This therapy can be done directly by inhibition of
TAFI with antibodies or peptides, or indirectly by inhibiting
thrombin generation via the intrinsic pathway by interfering
with thrombin-mediated Factor XI activation.
Future research will have to show which approach will ren-
der the best results with the fewest (bleeding) complications.

Acknowledgments

This work was supported in part by a fellowship of the Royal Nether-
lands Academy of Arts and Sciences to J.C.M. Meijers, and by a post-
doctoral fellowship from the Medical Research Council of Canada to
L. Bajzar. J.C.M. Meijers is an Established Investigator of the Neth-
erlands Heart Foundation.

References

1. Naito, K., and K. Fujikawa. 1991. Activation of human blood coagulation
factor XI independent of factor XII. Factor XI is activated by thrombin and
factor XIa in the presence of negatively charged surfaces.
7353–7358.
2. Gailani, D., and G.J. Broze. 1991. Factor XI activation in a revised model
of blood coagulation.

Science (Wash. DC).
3. von dem Borne, P.A.K., S.J. Koppelman, B.N. Bouma, and J.C.M.
Meijers. 1994. Surface independent factor XI activation by thrombin in the
presence of high molecular weight kininogen.
4. Gailani, D., and G.J. Broze. 1993. Factor XII-independent activation of
factor XI in plasma: effects of sulfatides on tissue Factor-induced coagulation.
Blood.
82:813–819.
5. von dem Borne, P.A.K., J.C.M. Meijers, and B.N. Bouma. 1995. Feed-
back activation of factor XI by thrombin in plasma results in additional forma-
tion of thrombin that protects fibrin clots from fibrinolysis.
3042.
6. Asakai, R., D.W. Chung, E.W. Davie, and U. Seligsohn. 1991. Factor XI
deficiency in Ashkenazi Jews in Israel.
N. Engl. J. Med.
7. Bajzar, L., R. Manuel, and M.E. Nesheim. 1995. Purification and charac-
terization of TAFI, a thrombin activatable fibrinolysis inhibitor.
270:14477–14484.

J. Biol. Chem.


266:


253:909–912.


Thromb. Haemostasis.


72:397–402.



Blood.

86:3035–


325:153–158.

J. Biol. Chem.

8. Comp, P.C., and C.T. Esmon. 1981. Generation of fibrinolytic activity by
infusion of activated protein C into dogs.
9. Sakata, Y., D.J. Loskutoff, C.L. Gladson, C.M. Hekman, and J.H. Grif-
fin. 1986. Mechanism of protein-C dependent clot lysis: role of plasminogen ac-
tivator inhibitor.
Blood.
68:1218–1223.
10. de Fouw, N.J., F. Haverkate, and R.M. Bertina. 1990. Protein C and fi-
brinolysis: a link between coagulation and fibrinolysis.
281:235–243.
11. Bajzar, L., J.C. Fredenburgh, and M.E. Nesheim. 1990. The activated
protein C-mediated enhancement of tissue-type plasminogen activator-induced
fibrinolysis in a cell-free system.
J. Biol. Chem.
12. Bajzar, L., and M.E. Nesheim. 1993. The effect of activated protein C on
fibrinolysis in cell free plasma can be attributed specifically to attenuation of
prothrombin activation.
J. Biol. Chem.
268:8608–8616.
13. Eaton, D.L., B.E. Malloy, S.P. Tsai, W. Henzel, and D. Drayna. 1991.
Isolation, molecular cloning and partial characterization of a novel carboxypep-
tidase B from human plasma.
J. Biol. Chem.
14. Folk, J.E., and J.A. Gladner. 1960. Carboxypeptidase B. IV. Purification
and characterization of the porcine enzyme.
15. Hendriks, D., W. Wang, S. Scharpé, M.P. Lommaert, and M. van Sande.
1990. Purification and characterization of a new arginine carboxypeptidase in
human serum.
Biochim. Biophys. Acta.
1034:86–92.
16. Wang, W., D.F. Hendriks, and S.S. Scharpé. 1994. Carboxypeptidase U,
a plasma carboxypeptidase with high affinity for plasminogen.
269:15937–15944.
17. Pannell, R., J. Black, and V. Gurewich. 1988. Complementary modes of
action of tissue-type plasminogen activator and pro-urokinase by which their
synergistic effect on clot lysis may be explained.
18. Miles, L.A., C.M. Dahlberg, J. Plescia, J. Felez, K. Kato, and E.F. Plow.
1991. Role of cell-surface lysines in plasminogen binding to cells: identification
of
?
-enolase as a candidate plasminogen receptor.
19. Redlitz, A., A.K. Tan, D.L. Eaton, and E.F. Plow. 1995. Plasma carboxy-
peptidases as regulators of the plasminogen system.
2538.
20. van der Graaf, F., J.S. Greengard, B.N. Bouma, D.M. Kerbiriou, and
J.H. Griffin. 1983. Isolation and functional characterization of the active light
chain of activated human blood coagulation factor XI.
9675.
21. Meijers, J.C.M. 1988. Regulation of the blood coagulation mechanism
by plasma proteinase inhibitors. Ph.D. thesis. State University of Utrecht, The
Netherlands. 93–108.
22. Bajzar, L., M.E. Nesheim, and P.B. Tracy. 1996. The pro-fibrinolytic ef-
fect of activated protein C in clots formed from plasma is TAFI-dependent.
Blood.
88:2093–2100.
23. Kisiel, W., K.J. Smith, and B.A. McMullen. 1985. Proteolytic inactiva-
tion of blood coagulation factor IX by thrombin.
24. Broze, G.J., Jr., and D.A. Higuchi. 1996. Coagulation-dependent inhibi-
tion of fibrinolysis: role of carboxypeptidase-U and the premature lysis of clots
from hemophilic plasma.
Blood.
88:3815–3823.
25. Kessels, H., S. Béguin, H. Andree, and H.C. Hemker. 1994. Measure-
ment of thrombin generation in whole blood. The effect of heparin and aspirin.
Thromb. Haemostasis.
72:78–83.
26. Bajzar, L., J. Morser, and M. Nesheim. 1995. TAFI, or plasma procar-
boxypeptidase B, couples the coagulation and fibrinolytic cascades through the
thrombin-thrombomodulin complex.
J. Biol. Chem.

J. Clin. Invest.

68:1221–1228.



Adv. Exp. Med. Biol.



265:16948–16954.




266:21833–21838.

J. Biol. Chem.

235:2272–2278.



J. Biol. Chem.


J. Clin. Invest.

81:853–859.



Biochemistry.

30:1682–1691.

J. Clin. Invest.

96:2534–

J. Biol. Chem.

258:9669–



Blood.

66:1302–1308.






271:16603–16608.