From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Factors IXa and Xa Play Distinct Roles in Tissue Factor-Dependent
Initiation of Coagulation
By Maureane Hoffman, Dougald M. Monroe, Julie A. Oliver, and Harold R. Roberts
Tissue factor is the
major initiator of coagulation. Both factor
IX and factor X are activated by thecomplex of factor Vlla
and tissue factor (Vlla/TF). The goal of this study was t o
determine the specific roles of factors
IXa and Xa in initiating
coagulation. We used a model system of in vitro coagulation
initiated by Vlla/TF and that included unactivated platelets
and plasma concentrations of factors II, V, VIII, IX, and X,
tissue factor pathway inhibitor, and antithrombin111. In some
cases, factor IX and/or factor X were activated by tissue
factor-bearing monocytes, but in some experiments, picomolar concentrations of preactivated factor IX or factor X
were used t o initiate the reactions. Timed samples were
assayed for both platelet activation and thrombin activity.
Factor Xa was 10 times more potent than
factor IXa in initiating plateletactivation, but factor IXa was much more effective in promoting thrombingeneration than was factor Xa.
In the presence of Vlla/TF, factor X was required for both
platelet activation and thrombin
generation, while factor IX
was only required for thrombin generation. We conclude
that VllafTF-activatedfactors IXa and Xa have distinct physiologic roles. The main role of factor Xa that is initiallyactivated by Vlla/TF is t o activate platelets by generating an
initial, small amount of thrombin in the vicinity of platelets.
Factor IXa, on the other hand, enhances thrombin generation by providingfactor Xa on the plateletsurface, leading
to prothrombinase formation. Only tiny amounts of factors
IX and X need t o be activated by Vlla/TF t o perform these
distinct functions. Our experiments show that initiation of
coagulation ishighly dependent onactivation ofsmall
amounts of factors IXa and Xa in proximity t o platelet surfaces and thatthese factors play
distinct roles in subsequent
events, leading t o an explosion ofthrombin generation. Furthermore, the specific roles of factorsIXa and Xa generated
by Vlla/TF are not necessarily reflected by the kinetics of
factor IXa and Xa generation.
0 1995 by The American Society of Hematology.
E
The relative significance of factor IX and factor X activation by the tissue factor pathway is currently unclear. Factor
IX is clearly essential for normal hemostasis, but it can potentially be activated by at least two mechanisms: VIIa/TF
(extrinsic pathway) and factor XIa(intrinsic pathway). There
is also a putative platelet factor XI-like molecule that can
activate factorIX.’.‘j Small amountsof factor IX are activated
by a tissue factor-dependent mechanism under baseline (nonbleeding) condition^,^ but it is unknown whether factor IX
activation by tissue factor is necessary for initiation of coagulation.Although the kinetic parameters of factor X and
factor IX activation vary with the source of tissue factor,8.q
VIIa/TF consistently activates more factor X than IX under
steady-state conditions.’” However, the amounts of the two
factors activatedatsteady
state may not be an indication
of their relative physiologic importance. A
more complete
understanding of the initiation of coagulation can be gained
by considering the interactions of the activated factors with
the cells onwhich they are produced andon whichthey
act to propagate coagulation. The effect of plasma protease
inhibitors on the process is also likely to be substantial.
The goal of the current study was to examine the relative
roles of tissuefactor-generatedfactors
IXa and Xa in the
explosive burst of thrombin generation on platelets that leads
to fibrin formation. Weexamined plateletactivation and
thrombingenerationin
twotypes of experiments.Inone
typeofexperiment,we
incubatedtissue
factor-bearing
monocytes in the presence of plasma levels of purified zymogen coagulation factors (11, V, VIII, IX, and X), TFPI, antithrombin 111 (AT), and unactivated platelets with catalytic
amounts of activated factor VIIa. This complete system was
designed to mimictissue factor-initiated coagulation in vivo.
[t contained cell-associated tissue factor to initiate coagulation,platelets to provide the surface on which large-scale
thrombin generationoccurs, all the proteinsnecessary to
assemble the Xase and prothrombinase complexes, and the
relevant protease inhibitors. Platelet activation and thrombin
generation in thissystemwere also measuredwheneither
XPOSURE O F blood to tissue factor is the majorphysiologicstimulusto
initiate coagulation.’Coagulation
factor VIIa associates with tissue factor,
and the complex
activates both factor X and factor IX.’ Activated factor X
(Xa) reacts with and is inhibited by tissue factor pathway
inhibitor (TFPI). The TFPVXa complex theninhibits the
VIIa/TF complex.3s4 Thus,in order for coagulation to occur,
sufficient amounts of procoagulant factorsmust be generated
during a limited period of time when the VIIa/TF complex
is active. Theregulation of tissuefactor exposure and
activity
is crucialto
controllinginitiation
of coagulation.Even
though platelets provide the major procoagulant surface for
hemostasis, they do not express tissue factor and, therefore,
cannot initiate the process.Activated coagulation factors
produced at the site of VIJa/TF activity must find their way
to the platelet surface to effectively propagate coagulation.
Thus, induction of coagulation by tissue factor is a complex
process. Its success or failureis governed by spatial, temporal, and quantitative features
of factor IXa and factor Xa
generation (ie, where, when, and how much factor is activated).
From the Department of Pathology,DurhamVeterans
Affairs
Medical Center and Duke University, Durham; and the Depurtment
of Medicine and Centerfor Thrombosis and Hemostusis, University
of North Carolina at Chapel Hill, Chapel Hill, NC.
Submitted January 17, 1995; accepted April 13, 1995.
Supported in part by Grant No. R01 HLA8320 from the National
Institutes of Health and by the United States Departmentof Veterans
Affairs.
Address reprint requests to Maureane H o f i a n , MD, PhD, LaboNC
ratoryService (113), DurhamVAMedicalCenter,Durham,
27705.
The publication costs of this article were defrayedin part by page
chargepayment. This article must thereforebeherebymarked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this facf.
0 1995 by The American Society of Hematology.
0006-4971/95/8605-0012$3.00/0
1794
Blood, Vol 86,No 5 (September l), 1995:pp 1794-1801
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
FACTORS IXaANDXaIN
TF-INITIATED COAGULATION
factor IX or factor X was omitted. In the second type of
experiment, we added purified factor IXa or factor Xa to
activated or unactivated platelets suspended in a mixture of
purified zymogen coagulation factors, TFPI, and AT. In the
latter experimental design, coagulation was initiated by the
addition of specific amounts of preactivated factors IX and/
or X, rather than allowing tissue factor to activate them.
This experimental design allowed us to isolate the effects
of factors IXa and Xa on platelet activation and thrombin
generation.
In our system, platelet activation always preceded thrombin generation. The addition of preactivated factor X efficiently promoted platelet activation but led to a relatively
small amount of subsequent thrombin generation. Preactivated factor IX was an order of magnitude less potent than
factor Xa in inducing platelet activation, but once platelet
activation had occurred, factor IXa was much more effective
than factor Xa at promoting thrombin generation. Omission
of factor X from a VIIaiTF-initiated system (containing tissue factor-bearing cells, small amounts of factor VIIa, unactivated platelets, and plasma levels of zymogen factors 11,
V, VIII, IX, and X) eliminated both platelet activation and
thrombin generation. Omission of factor IX from this system
(mimicking hemophilia B) dramatically reduced thrombin
generation, while platelet activation occurred at a nearly normal rate. These experiments support the conclusion that factor Xa and factor IXa activated by VIIaiTF play different
and distinct roles in initiating coagulation. The initial activation of factor X, which occurs before platelet activation, is
more important for generating the first small amounts of
thrombin required for platelet activation. Factor IX activation plays a major role in determining the rate of thrombin
formation by activating factor X on the platelet surface,
where it can assemble with factor Va into active prothrombinase complexes that lead to the ultimate explosion of
thrombin generation.
MATERIALS AND METHODS
Materials. Dulbecco’s modifiedEagle’smedium (DMEM) was
purchased from GIBCO (Grand Island, NY). Chromozyme TH was
purchased from Boehringer-Mannheim (Indianapolis, IN). Spectrozyme FXa and TenStop were purchased from American Diagnostica
(Greenwich, CT). Recombinant hirudin was purchased from Accurate Chemical and Scientific Corp (Westbury, NY). Antihuman factor X and antihuman factor IX polyclonal rabbit antibodies were
purchased from Dako Corporation (Carpenteria, CA). Anti-CD62
(phycoerythrin-conjugated) was purchased from Becton Dickinson
(San Jose, CA). Mono-Poly Resolving Medium was purchased from
Flow Laboratories, Inc. Lipopolysaccharide (LPS) from Escherichia
coli was purchased from Sigma Chemical CO (St Louis, MO). All
other chemicals were of high commercial grade.
Proteins. Prothrombin and factor IX were purified from human
plasma as described previously.” Factor X was purchased from
Enzyme Research Laboratories (South Bend, IN), and factor V from
Calbiochem Corporation (La Jolla, CA) or Haematologic Technologies (Essex Junction, VT). Factor VI11 was purchased from the University of North Carolina hospital pharmacy as Profilate (Alpha
Therapeutics, Los Angeles, CA). Thrombin” and ATI3were purified
from human plasma as previously described. Prothrombin, factor X,
and factor IX were treated with an inhibitor mixture (1 pmol/L tosylLysyl-chloromethyl ketone, tosyl-Phenyl-chloromethyl ketone, Glu-
1795
Gly-Arg-chloromethyl ketone, Phe-Pro-Arg-chloromethylketone,
and Phenyl-methyl-sulfonyl fluoride), followed by exhaustive dialysis to inactivate proteases in the zymogens. In addition to treatment
with low-molecular-weight inhibitors, the zymogen proteases were
incubated for 12 hours with AT at a concentration IO times that in
plasma. The proteins were diluted to plasma concentrationjust before
the experiment. Recombinant factor VIIa and recombinant TFPI
were gifts ofNOVONordisk
(Gentofte, Denmark). Factors IXa
and Xa were prepared from the respective zymogens as previously
described.14Briefly, factor IX was activated with factor XIa (Enzyme
Research Laboratories), and factor X with purified coagulant protein
from Russell’s viper venom (Haematologic Technologies). The activated factors were repurified on Q-Sepharose with calcium chloride
elution.’5
Cell isolation. Monocytes were purified on Ficoll-Hypaque density gradients from the blood of healthy volunteers.16The mononuclear cell band was collected and diluted in an equal volume of 13
mmoVL citrate (pH 7.4), 123 mmoVL NaCI, 33 mmol/L dextrose
with 10 pmol/L prostaglandin E,. The nucleated cells were sedimented at lOOg for 5 minutes, and the platelet-containing supernatant
was retained for gel filtration.” The mononuclear cell-containing
pellet was washed twice withcold calcium-free medium, andthe
resulting pellet was resuspended in DMEM with 1 mg/mL dextrose
and no serum. Two thousand monocytes per well were plated in 96well plates in DMEM containing 0.5 pg/mL bacterial LPS. The
plates were incubated at 37°C in 5% CO, for 1 hour. Nonadherent
cells were removed by washing three times with fresh DMEM. Fresh
LPS-containing DMEM was added, and the cells were cultured for
18 hours. The LPS-treated monocyte monolayers (LPS-monocytes)
were thenwashed three times with Tyrodes buffer (15 mmoVL
HEPES, pH 7.4, 3.3 mmol/L Na,P04, 138 mmol/L NaCI, 2.7 mmoU
L KCI, 1 mmol/L MgClz, 5.5 mmol/L dextrose) containing 1 mg/
mL bovine serum albumin (Tyrodeslalbumin) before being used
in experiments. The adherent population after culture and washing
consisted of greater than 95% monocytes, with less than 5% contamination with lymphocytes. Platelet contamination was from one to
two platelets per monocyte. This level of contamination was quite
low compared with thenumber of purified platelets added to reaction
wells (<0.006% of the platelets added).
Platelets were separated from plasma proteins by gel-filtering the
platelet-containing supernatant from the monocyte preparation over a
50-mL column of Sepharose CL-2B in Tyrodeslalbumin. The platelet
preparations contained less than 0.1% contamination by leukocytes.
Platelets were kept at 37°Cuntil used. Platelet activation, defined
as expression of the CD62 antigen (P-selectin, GMP-140). was measured byflow cytometry.” Reproducibly, fewer than10%ofthe
platelets were activated by the isolation procedure.
Experimental assay system. The complete cell and protein system is a modification of one we have previously used to examine
tissue factor-initiated procoagulant complex assembly on the platelet
surface.19Tissue factor-bearing monocytes were incubated with purified coagulation proteins or proteins and platelets for up to 16 minutes. The platelets, isolated from the same blood donor as the monocytes, were at a final concentration of 80,0OO/pL to 1 IO,OOO/pL. The
final concentrations of proteins were selected to approximate plasma
levels: factor X, 8 pglmL; factor IX, 4 pg/mL; factor VIII, 1 UlmL;
factor V, 7 pg/mL; TFPI, 0.1 pglmL; andAT, 150 pg/mL. The
concentration of prothrombin used, 50 pglmL, is at the lower limit
of normal plasma concentrations for this factor. The reactions were
initiated by the addition of factor VIla (0.01 pg/mL) and calcium
(3 m m o a ) . In some experiments, either factor IX or factor X was
omitted from the protein mixture incubated with monocytes and
platelets. In these experiments, 100 pg/mL of rabbit polyclonal antifactor IX or anti-factor X (both from Dako Corp, Santa Barbara,
CA) was included in the incubation mixture to inactivate any trace
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1796
of the omitted factor. Although the system is artificial, it was designed to mimic the initial in vivo coagulation reactions.
In other experiments, different amounts of activated factor IX and/
or factor X were added directly to a mixture of platelets, coagulation
proteins, and inhibitors except factor VIIa. Once the proteins had
been added to the platelets, samples were taken at timed intervals
and assayed for thrombin activity and platelet activation. The amount
of thrombin generated in anygiven set of experiments was a function
of the individual whose cells were used in the experiments. In replicate experiments, the pattern of platelet activation and the relative
amounts of thrombin generated with different treatments were consistent, even though the absolute amounts of thrombin generated
were different for cells from different donors. Each figure shows
data runin parallel from one individual andis representative of
experiments using cells from other individuals.
Assays. For the assay of thrombin activity, 10-pL samples from
the experimental wells were added to 90 pL of 0.45 m m o K Chromozyme TH, 30 pmoVL TenStop (sufficient to block the greatest
amount of factor Xa produced in these experiments), 2.2 mmol/L
EDTA, in 20 mmol/L HEPES (pH 7.4), 150 m m o K NaC1, 1 mgl
mL albumin. The assay mixture was incubated for I3 minutes, then
cleavage of the chromogenic substrate was terminated by addition
of 100 pL of 20% acetic acid. The amount of active enzyme was
related to the cleavage of the synthetic substrate (measured at 405
nm in a V,,, plate reader [Molecular Devices Corp, Menlo Park,
CA]) by a standard curve made using purified thrombin.
For assays of platelet activation and factor X(a) binding, separate
8-pL samples from the reaction mixtures were removed and added
to 30 pL of 2% paraformaldehyde in 20 mmoVL HEPES, 150 mmoll
L NaCI, 3 mmol/L CaCI2. Platelets were fixed in paraformaldehyde
for 30 minutes, then diluted in 5 v01 of Tyrodeslalbumin, and incubated for at least an additional 30 minutes. Platelet activation was
assessed by measuring expression ofthe activation-specific alpha
granule marker CD62.I8 Samples were stained with phycoerythrinconjugated anti-CD62 (final dilution, 150) for 30 minutes at 25°C.
Relative factor X(a) binding was assessed as previously described.”
Briefly, the platelet samples were incubated for 30 minutes with a
1:200 dilution of antihuman factor X (Dako), washed once, then
incubated for 30 minutes in a 1:30 dilution of fluorescein isothiocyanate(F1TC)-conjugatedgoat antirabbit IgG (Sigma). The samples
were washed once with filtered phosphate-buffered saline (PBS) and
resuspended in 0.5 mL PBS, andthemean fluorescence intensity
was determined on a FACScan flow cytometer (Becton Dickinson,
Mountain View, CA).
The activation state of the platelet-bound factor X was determined
by Western blot analysis. Samples were removed from the incubation
mixtures, and the platelets were centrifuged through a layer of 10%
sucrose with 3 mmol/L CaCI,, to separate free from bound ligand.
The platelet pellets were resuspended in 20 pL of IO mmol/L EDTA
to dissociate membrane-bound factors X and Xa. The platelets were
sedimented in a microfuge, and the supernatant was added toan
equal volume of sodium dodecyl sulfate (SDS)-containing sample
buffer and boiled for 1 minute. The samples were separated by
polyacrylamide gel electrophoresis (PAGE) and transferred to nitrocellulose membranes on a Phastgel System (Pharmacia, Uppsala,
Sweden). Factor X antigen was detected using the same polyclonal
rabbit immunoglobulin used for the flow cytometric assays and a
chemiluminescent horseradish peroxidase substrate (LumiGLO; Kirkegaard & Perry Laboratories, Gaithersburg, MD).
Activation of factor X and factor IXby tissue factor-bearing
monocytes was assayed as follows. Factor X (8 p g / m L ) , factor VIIa
(0.01 pglmL). and 100,000 monocytes per well were incubated together in 200 pL of 20 mmoVL HEPES (pH 7.4), 150 mmol/L NaCl
(HBS) with 5 mmol/L CaClz. At timed intervals, 50-pL samples
were removed from the supernatant and incubated with 1 mmolL
HOFFMAN ET AL
Spectrozyme FXain 20 mmoUL HEPES (pH 7.4). 150 mmol/L
NaCI, 5 mmoVL EDTA. The rate of cleavage of Spectrozyme was
compared with the rate measured for a standard curve using factor
Xa of known concentration. The background was determined from
a sample that did not include factor X with the monocytes. Similarly.
plasma concentrations of factor IX (4 pglmL) were incubated with
0.01 pglrnL factor VIIa and 100,000 monocytes per well in 200 pL
HBS with 5 mmol/L CaCI,. At timed intervals, 50-pL samples were
removed from the supematant and incubated with 50 pLof 60 prnol/
L phospholipid vesicles (30% phosphatidylserine, 70% phosphatidylcholine), 5 nmol/L factor VlIIa (activated with thrombin that was
neutralized with hirudin), 300 nmol/L factor X , and 3 mmoilL CaCI?
(final concentrations). Cleavage of Spectrozyme wasmeasured at
405 nm, and the rate was determined as the second order value in
a polynomial fit to the data. This rate was compared with the rate
measured for a standard curve using factor IXa of known concentration. The background was determined from a sample thatdidnot
contain factor IX with themonocytes. The standard curves for factors
IXaandXawere
calibrated by active-site titration withAT. The
concentration ofAThad been determined by titration with active
site-titrated thrombin.I4”’
RESULTS
Platelet activation and thrombin generation in the tissue
factor-initiated system. Platelets (IOO,OOO/pL final concentration) were incubated with tissue factor-bearing monocytes
and plasma levels of purified prothrombin, factor X, factor
IX, factor VIII, AT, TFPI, and a catalytic amount of factor
VIIa.I9 Platelet activation, as measured by exposure of the
alpha granule membrane protein CD62, began within about
2 minutes after the addition of calcium (Fig 1A). However,
in the absence of monocytes, very little platelet activation
occurred. We have previously shown that platelet activation
is dependent on the generation of small amounts of thrombin,
because no platelet activation occurs in the absence of prothrornbin.lg
The time course of thrombin generation in the presence
of platelets and monocytes was compared with thrombin
generation in wells containing platelets alone or monocytes
alone. As shown in Fig IB, very little thrombin generation
occurred in wells containing the coagulation proteins and
platelets only. However, in the presence of both monocytes
and platelets, thrombin generation began after a 2- or 3minute lag period followed by a rapid burst of thrombin
generation. Monocytes alone did not support generation of
significant amounts of thrombin (data not shown). No thrombin generation was observed in the absence of calcium or
when monocytes and platelets were both omitted. In addition, when monocytes were preincubated with an anti-tissue
factor antibody at concentrations that completely inhibited
monocyte tissue factor activity,” platelet activation and
thrombin generation were nearly identical to that seen in the
absence of monocytes.
Effect of factor IXa and factor Xa on platelet activation
and thrombin generation. Once wehad ascertained that
platelet activation and thrombin generation were dependent
on the presence of tissue factor-bearing monocytes in a complete system that included platelets and coagulation zymogens, we conducted experiments to determine whether the
addition of low concentrations of purified factor IXa, factor
Xa, or combinations of the two factors, could reproduce the
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1797
FACTORS IXa AND Xa IN TF-INITIATED COAGULATION
thrombin in amounts too small to be detected in the supernatant. By contrast, l 0 pmol/L factor IXa (0.015% of the
plasma IX concentration) had nodetectable effect on platelet
activation (Fig 2, lower panel). Considerably higher concentrations of factor IXa wererequired to induce platelet activation. For example, addition of 100 pmoVL factor IXa resulted
in 50% of the platelets being activated after about 10 minutes. The addition of 100 pmol/L factor Xa resulted in 50%
of the platelets being activated after only 4 minutes, Neither
factor IXa nor factor Xa directly activated platelets (data not
shown).
\ool
Factor Xa
0 1 2 3 4 5 6 7 8 9 1 0
I
0
I
I
I
I
2
4
6
8
1
I
0
I
2
4
6
8 1 ~ 0
10
12
tlme (min)
0
°
Fig l. (A) Effect of tissue factor-bearing monocytes on platelet
activation. Wells contained prothrombin, TFPI, AT, and factorsIX, X,
VIII, V, and Vlla incubated with platelets, 100,0001pL. with 12.0001
well; 0)
or without monocytes
l + 1. Platelet activation wasmeasured
at the times indicated
as described in Materials and Methods and is
shown as the percent of plateletsexpressing CD62. The figure shows
data from one representative experiment10
ofperformed. (B) Effect
of tissue factor-bearing monocytes on thrombin
generation. Thrombin generation wasmeasured as described in Materials and Methods.
The data are from thesame experiment as in panel A and are representative of 10 experiments performed. (W), monocytes plus platelets; (V), platelets only.
effects of tissue factor on platelet activation and thrombin
generation. Different concentrations of factors IXa and Xa
were added to wells containing platelets, all the zymogen
coagulation factors, inhibitors, and calcium. The effect of
factors IXa and Xa on platelet activation is shown in Fig 2.
We found that as little as 10 pmoUL factor Xa (0.008%of the
plasma concentration of factor X) was sufficient to initiate
platelet activation in the presence of the other zymogen coagulation factors (Fig 2, upper panel).However, platelet activation did not occur when prothrombin was omitted from the
mixtures. This suggests that small amounts of factor Xa
promote platelet activation by activating prothrombin to
0
2
4
6
8 1 - 0
time (min)
Fig 2. Effect of factors IXa and
Xa on platelet activation. Platelets
(100,0001pL~were incubated in wells containing prothrombin, TFPI,
AT, and factors V, IX, X, and VIII, with the
addaion of different
concentrations of factor Xa (upper panel) or factor IXa (lower panel). Samples were removed for fixationat theindicated times. Expression of
CD62 was measured by flow cytometry as described in Materials
and Methods. The data are shown from one experiment and are
representative of four performed. IO), 0 pmol1L; ( 01, 1 pmol/L; (0).
10 pmol1L; (A). 100 pmol1L.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1798
HOFFMAN ET AL
Factor Xa
l
2
1
4
I
I
I
I
I
I
I
0
2
4
6
8
1 0
12
’
Factor IXa
12
8-
be more effective in promoting thrombingeneration than
factor IXa, because it is more effective at inducing platelet
activation. To separate an effect on platelet activation from
an effect specifically on thrombin generation, we measured
the amount of thrombin generated when the platelets were
activated before the addition of coagulation factors. Platelets
were activated by incubation for 10 minutes with 50 pg/mL
of a thrombin receptor agonist peptide (SFLLRN),” which
has no proteolytic activity ofits own. Thisconcentration was
more than sufficient to induce maximal CD62 expression.
Zymogen coagulation factors I1 and V, inhibitors, and calcium were added to activatedplatelets. In addition, either
factor Xaor factors IXa, VIII, and X were added. The zymogenswerealwaysadded
at concentrations approximating
those in plasma as detailed in Materials and Methods, while
the activated factors were added at varying, but miniscule,
concentrations. As shown in Fig 3, the amount of thrombin
generated on activated platelets to which 10 or 100 pmol/L
factor IXa had been added (lower panel) was significantly
greater than platelets to which 10 or 100 pmol/L factor Xa
had beenadded (upper panel). This findingsuggeststhat
factor IXa, in the presence of zymogen coagulation factors,
is more effective than the same amount of exogenous factor
Xa at promoting thrombin generation. We hypothesized that
this is because factor IXa, as part of the Xase complex, can
continue to activate factor X and provide a source of platelet-
I
304
0
2
4
6
8
10
1 2
time
(minutes)
Fig 3. Effect of factors IXa and Xa on thrombingeneration. Platelets (100,OOO/pL) were preactivated with SFLLRN (50 Fg/mL). The
in wells containing prothrombin,TFPI,
platelets were then incubated
AT, and factors Xe and V (upper panel), or prothrombin, TFPI.AT,
and factors IXa, VIII, V, and X (lower panel), as described in Materials
and Methods. Samples were removed for assay at the indicated
times. The results are representative of fiveexperiments performed.
(U), 0 pmol/L; (0).10 pmol/L; (A) 100 pmollL.
A number of different events occur during the process of
platelet activation. The platelet activation marker we used,
CD62, reflectssecretion of alpha granule contents. Alpha
granulerelease
is relativelyslow
compared with other
events, such as shape change,activation of glycoprotein IIbl
IIIa,andsurfaceexposure
of phosphatidyl serine. Thus,
while we did notdirectly measure exposure of procoagulant
lipids, CD62 expression is a good indicator that theplatelet
surface has already undergone thetransformation to a procoagulant state. In fact, we found that
CD62 expression always
preceded large-scale thrombin generation in our model system. Thus, we might expect that exogenous factor would
Xa
I
I
T
I
I
0
500
1000
pM added
Fig 4. Flow cytometricanalysis of theeffect of factor IXa on platelet-associated factor X antigen. Platelets (lW,OOO/pL) were preactiin wells convated with SFLLRN. The platelets were then incubated
taining the indicated amounts of factor IXa (D) or factor Xa ( 0 ),
prothrombin, TFPI,AT, and factors Vlll, V, and X, as described in
Materials andMethods. Samples were removed and fixed
in 2% paraformaldehyde after 8 minutes incubation. The relative amount of
platelet-bound factor X antigen was assessed by indirect immunofluorescence and flow cytometry, as detailed in Materials endMethods. The results shown are from one representative experiment of
five performed.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
FACTORS IXa AND Xa IN TF-INITIATED COAGULATION
1799
Fig 5. Western blot analysis of the effect of 100 pmol/L factor IXa on platelet-associated factors X and Xa. Platelets (1OO,OOO/pL) were
incubated in wells containing plasma levels of prothrombin, TFPI, AT, and factors V, X, and V111; 3 mrnollL CaCI,; and 0, 10, or 100 pmol/L
factor IXa. The gel shown is from platelets incubated
with l00 pmol/Lfactor IXa. Samples were taken at the indicated times
(S, seconds; m,
minutes), and the platelets were centrifuged through sucrose. Bound factors were eluted from the platelets by the addition of EDTA. The
eluates were subjected t o SDS-PAGE and Western blotting using anti-factor X antibody, as detailed in Materials and Methods. The major
band of higher molecular weight has the same electrophoretic mobility as authentic factor X, while the band of slightly lower molecular
weight has the same mobility as factor Xa. The supernatant sample (Super) was taken after a 16-minute incubation from the supernatant
after the platelets had been sedimented. Samples in the absence of factor IXa were similar t o those shown at 30 seconds incubation and
showed no change in the proportionsof factors X and Xa over time. Samples from platelets incubated with10 pmollL factor IXa showed a
smaller increase in platelet-associated factor Xa than those incubatedwith 100 pmol/L factorIXa. The 30-second through 12-minutelanes and
supernatant lane were run on the same gel, while the 16-minute sample was on a separate gel. The gels were transferredt o nitrocellulose,
reacted with antibodies and substrate at the same time, and exposed t o the same piece of x-ray film. The 16-minute lane was cut out and
pasted nextt o the remaining lanes for this photograph. These data are representative of three experimentsperformed.
associated factor Xa to form prothrombinase complexes. To
test this hypothesis, we also measured the amount of plateletassociated factor X(a) resulting from the addition of different
amounts of exogenous factor IXa or Xa in the experiments
detailed above. As shown in Fig 4, the addition of increasing
amounts of factor IXa, in the presence of factors VI11 and
X, resulted in increasing amounts of factor X antigen associated with the platelet surface, up to twice the amount of
factor X bound in the absence of added factor IXa. The
polyclonal antibody used in these studies does not distinguish between zymogen and activated factor X. However,
the increase in platelet-associated factor X correlated with
the rate of thrombin generation, suggesting that addition of
factor IXa results in increased functional factor Xa on the
platelets. The addition of increasing amounts of factor Xa
was not reflectedin a significant increase in thrombin generation or platelet-associated factor X, suggesting that much of
the exogenous factor Xa was unable to reach the platelet
surface, perhaps being bound by inhibitors.
We next examined the activation state of the plateletbound factor X(a) antigen. Platelets were activated with
SFLLRN and then added to a mixture of zymogen coagulation factors 11, V, VIII, and X, as well as AT and TFPI.
Different amounts of factor IXa were added, and timed samples were removed as in the experiments described above.
These samples were immediately centrifuged through sucrose to separate free from platelet-bound factors, and then
EDTA was added to the platelet pellets to dissociate bound
factor X(a). The eluates were subjected to Western blot analysis. Initially, the great majority of platelet-associated factor
X antigen was in the zymogen form. In the presence of added
factor IXa, there was a progressive increase in the amount
of activated factor X (Fig 5). In the absence of factor IXa,
there was no increase in the proportion of platelet-bound
factor X in the activated form. Very little of the factor X in
the supernatant was activated, and very little antigen was
detected at a molecular weight consistent with factor XdAT
complexes. Thus, both factor X and Xa are bound to activated platelets, but factor IXa promotes the progressive activation of factor X on platelet surfaces. The platelet-associated factor Xacouldthenbe
directly incorporatedinto
prothrombinase complexes.
Factor IX and factor X activation on tissuefactor-bearing
monocytes. To measure how much factor X and factor IX
was activated on the tissue factor-bearing monocytes,
100,OOO monocytes per well were incubated with factor VIIa,
calcium, and plasma concentrations of either factor IX or X
for 1 hour. The amount of each factor that had beenactivated
was measured by activity assays.We calculated that the
amount of the factors activated in 1 minute by the number
of monocytes used in our experiments (2,000 per well) would
result in no more than 10 pmoVL of each activated factor
being transferred to the platelet suspensions.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1800
HOFFMAN ET AL
Roles of factors IXa and Xu in a tissue factor-initiated
model system. Finally, we returned to a tissue factor-initiated system to determine whether factor IXand factor X
activatedon a cell surface functioned similarly to exogenously added factors IXa and Xa. Tissue factor-bearing
monocytes and platelets were incubated with the complete
complement of coagulation proteins (AT, TFPI, factors 11,
V, VIIa, VIII, IX, and X) and compared with incubation
mixtures lacking either factor IX or factor X. Mixtures lacking factor X demonstrated neither platelet activation nor
thrombin generation (Fig 6). Mixtures lacking factor IX
showed a nearlynormal rate of platelet activation (upper
panel) but only a very small amount of thrombin generation
(lower panel). Thus, in a tissue factor-initiated system, factor
X activation by the extrinsic pathway promotes platelet activation, while factor IX activation by VIIa/TF is required for
optimal thrombin generation on the platelet surface.
0
2
4
6
8
10
DISCUSSION
The work described in this report was undertaken to investigate the relative roles of factor IX and factor X activation
65.0by tissue factor in the process of initiating coagulation. One
of the unanswered questions in hemostasis iswhy a defi4.0ciency of factor IX or factor VIII can cause such a severe
bleeding diathesis. These factors work together to form the
W
platelet (intrinsic) Xase complex. However, the VIIdTF
c 3.0complex should be able, at least partially, to offset a defia
E
ciency of factor IX or VI11 by activating factor X itself. As
2 2.0factor IX is a competitive inhibitor of factor X activation by
c
VIIdTF,'" one might expect tissue factor-dependent activaltion of factor X to be increased in a compensatory manner
1 .owhen factor IX is lacking. A number of investigators have
addressed this problem by studying the kinetic features of
0.0factor IXand X activation by tissue factor.x,'".21-2'Others
have suggested that VIIa/TF cannot compensate for a lack
0
2
4
6
8
1'0
of factor IX because TFPI rapidly inactivates it once some
factor Xa has been generated.26However, kinetic studies of
time (minutes)
the activities of tissue factor and its inhibitor do not necessarily address questions related to the physiology of the system.
Fig 6. (A) Effect of omitting factor IX or factor X on platelet activaOur studies were designed to look at the effects of the prodtion. Wells contained prothrombin, TFPI, AT, and factors V, VIII, IX,
ucts of VIIa/TFactivityon
platelets and platelet surface
X, and Vlla incubated with platelets (100,00O/pL) and 2,000 monocoagulation complexes as a step toward determining whether
cytes per well (0).
Factor IX was omitted from some wells (W, and
factor X from others ( 0 l. Platelet activation was measured at the
factors IXa and Xa, activated by VIIa/TF, play distinct physitimes indicated as describedin Materials and Methods and is shown
ologic roles in the clotting process.
as the percent of platelets expressing C D M . The figure shows data
Our data suggest that factor IX and factor X activated by
from one representative experiment of four performed. (B) Effect of
VIIdTF play distinct roles in coagulation. Exogenously
omitting factor IX or factor X on thrombin generation. Thrombin
generation was measured as described in Materials and Medhods.
added factor Xa efficiently promoted platelet activation, and,
The data are from the same experiment as in panel A and are reprein a tissue factor-initiated system, factor X was required for
sentative of four experiments. Wellscontainedthe complete system
platelet activation and thrombin generation. However, once
(01, lacked factor IX (01,or lacked factor X ( 0 1.
platelets were activated, the amount of thrombin generated
wasnot a function of the amount of exogenously added
factor Xa. We conclude that factor X activation on a tissue
of factor IX allowed nearly normal platelet activation but only
factor-bearing cell leads to the generation of small amounts
minimal thrombin generation. We conclude that factor
IX actiof thrombin, which then leads to platelet activation.
vation is not required for platelet activation but is required for
In contrast to the effects of factor Xa, exogenously added
factor IXa had little ability to promote platelet activation. How-optimal thrombin generation. Our results suggest that this is
because factor E a , in complex with factor VIIIa, provides an
ever, added factor IXa led to increasing amounts of factor Xa
ongoing supply of factor Xa on the platelet surface for formaon the platelet surface, ultimately leading atoburst of thrombin
generation. Inour tissue factor-initiated model system, omission tion of the prothrombinase complex. Factor Xa added into the
h
5
.-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
FACTORS IXa AND Xa IN TF-INITIATEDCOAGULATION
1801
Factor IX is activated in vivo by the tissue factor mechanism. Blood
fluid phase at the beginning ofthe reaction (as provided by
76:731, 1990
VIIa/TF) is not as effective in generating thrombinas is factor
8. Rao LV, Robinson T, Hoang AD: Factor VIIdtissue factorXa generated on the platelet surface by the factor
IXaNma
catalyzed activation of factors IX and X on a cell surface and in
complex. Thereare probably at leasttwo reasons for this. First,
suspension: A kinetic study. Thromb Haemost 67:654, 1992
factor Va constitutes part of the platelet surface-binding site
9. Bom VJ, van Hinbergh V W , Reinalda-Poot JH, Mohanld RW,
for factor Xa. Factor Xa that is formed
by V I I m before
Bertina RM: Extrinsic activation of human coagulation factors IX
platelets and factor Vaare activated has no meansof localizing
and X on the endothelial surface. Thromb Haemost 66:283, 1991
10. Bom VJ, Reinalda-Poot JH, Cupers R, Bertina RM: Extrinsic
on the platelet surface. Instead, this factor Xa remains in soluactivation of human blood coagulation factors IX and X. Thromb
tion,whereitissusceptibletoinhibitionby
AT and TFPI.
Haemost 63:224, 1990
Second,factor Xathatisactivatedbythefactor
IXaNma
11. Jenny R, Church WR, Odegaard B, Litwiller R, Mann KG:
complex on the activated platelet may remain localized on the
Purification of six human vitamin K-dependent proteins in a single
platelet surface and move directly into complex with factor Va.
chromatographic step using immunoaffinity columns. Prep Biochem
This factor Xa formed by the platelet Xase remains protected
16:227, 1986
from inhibitors and can most efficiently participate in formation 12. Church FC, Whinna HC: Rapid sulfopropyl-disk chromatoof prothrombinase complexes. Thus, factor X activated
by W d
graphic purification of bovine and human thrombin. Anal Biochem
TF on monocytes, endothelial cells,or stroma is a poor substi- 157:77, 1986
13. Church FC, Meade JB, Treanor RE, Whinna HC: Antithrombin
tute for factorXactivated by theplateletfactor IXaNma
activity of fucoidan. The interaction of fucoidan with heparin cofactor
complex. This interpretation is consistent withthe factthat
II, antithrombin m,and thrombin. J Biol Chem 264:3618, 1989
factor X activated byWa/TFdoes not substitute for a lackof
14. Monroe DM, Sherril GB, Roberts HR: Use of p-aminobenplatelet Xase activity in patients with hemophilia.
zamidine to monitor activation of trypsin-like serine proteases. Anal
Kinetic studies have provided detailed information about
Biochem 172:427, 1988
the rates of factor IX and factor X activation by VIIa/TF.
15. Yan SCB, Razzano P, Chao YB, Walls JD, Berg DT, McClure
However, kinetic studies may not address the physiologic
DB, Grinnell BW: Characterization and novel purification of recomroles of factor IX and factor X activation in tissue factorbinant human protein C from three mammalian cell lines. Biotechnology 8:655, 1990
initiated hemostasis. Our current data, based on an in vitro
16. Ferrante A, Thong YH: Optimal conditions for the simultanemodel that reflects in vivo conditions, suggest that factors
ous purification of mononuclear and polymorphonuclear leucocytes
IXa and Xa generated by the tissue factor pathwayplay
from human peripheral blood by the Ficoll-Hypaque method. J Imdistinct and complementary roles, and that both are required
munol Methods 36:109, 1980
for effective initiation of coagulation. Optimal thrombin gen17. Hoffman M, Monroe DM, Roberts HR: A rapid method for
eration is achieved only when both factor X and factor IX
the isolation of platelets from human blood. Am J Clin Pathol98:531,
are activated by the VIIa/TF complex. Factor Xa promotes
1992
platelet activation, probably by activating small amounts of
18. Stenberg PE, McEver RP, Shuman MA, Jacques YV, Bainton
prothrombin. Factor IXa promotes rapid thrombin generation
DF: A platelet alpha-granule membrane protein (GMP-140) is expressed on the plasma membrane after activation. J Cell Biol
on the activated platelet surface by activating factor Xa,
101:880, 1985
which can then be incorporated into the platelet surface pro19. Monroe DM, Roberts HR, Hoffman M: Platelet procoagulant
thrombinase complex.
ACKNOWLEDGMENT
We acknowledge the excellent technical assisstance of Suzie Wortham and Missy Martin.
REFERENCES
1. Broze GJ Jr:Why do hemophiliacs bleed? Hosp Pract (Off
Ed) 27:71, 1992
2. Bsterud B, Rapaport SI: Activation of factor M by the reaction
product of tissue factor and factor VII: Additional pathway for initiating blood coagulation. Proc Natl Acad Sci USA 74:5260, 1977
3. Broze GJ Jr, Warren LA, Novotny W, Higuchi DA, Girard
JJ, Militich PJ: The lipoprotein-associated coagulation inhibitor that
inhibits the factor VD-tissue factor complex also inhibits factor Xa:
Insight into its possible mechanism of action. Blood 71:335, 1988
4. Rapaport SI: The extrinsic pathway inhibitor: A regulator of
tissue factor-dependent blood coagulation. Thromb Haemost 66:6,
1991
5. Schiffman S, YehCH: Purification and characterization of
platelet factor XI. Thromb Res 60:87, 1990
6. Walsh PN: Platelet-mediated trigger mechanisms in the contact
phase of blood coagulation. Semin Thromb Hemost 13:86, 1987
7. Bauer KA, Kass BL, ten Cate H, Hawiger JJ, Rosenberg RD:
complex assembly in a tissue factor-initiated system. Br J Haematol
88:364, 1994
20. McCord DM, Monroe DM, Smith KJ, Roberts HR: Characterization of the functional defect in factor IX Alabama. J Biol Chem
265:10250, 1990
21. Hoffman M, Monroe DM, Roberts HR: Human monocytes
support factor X activation by factor VIIa independent of tissue
factor: Implications for the therapeutic mechanism ofhigh dose
factor VIIa in hemophilia. Blood 83:38, 1994
22. Vassallo RRJ, Kieber ET, Cichowski K, Brass L F Structurefunction relationships in the activation of platelet thrombin receptors
by receptor-derived peptides. J Biol Chem 2676081, 1992
23. Komiyama Y, Pedersen AH, Kisiel W: Proteolytic activation
of human factors IX and X by recombinant human factor VIIa:
Effects of calcium, phospholipids, and tissue factor. Biochemistry
29:9418, 1990
24. Morrison SA, Jesty J: Tissue factor-dependent activation of
tritium-labeled factor IX and factor X in human plasma. Blood
63:1338, 1984
25. Wan-Cramer BJ, Bajaj S P Intrinsic versus extrinsic coagulation. Kinetic considerations. Biochem J 239:757, 1986
26. Repke D, Gemmell CH, Guha A, Turitto VT, Broze GJJ,
Nemerson Y: Hemophilia as a defect of the tissue factor pathway
of blood coagulation: Effect of factors VI11 and IX on factor X
activation in a continuous-flow reactor. Proc Natl Acad Sci USA
87:7623, 1990
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1995 86: 1794-1801
Factors IXa and Xa play distinct roles in tissue factor-dependent
initiation of coagulation
M Hoffman, DM Monroe, JA Oliver and HR Roberts
Updated information and services can be found at:
http://www.bloodjournal.org/content/86/5/1794.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.