From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
Kinetics of Factor VIII-von Willebrand Factor Association
By Andre J. Mot, Stefan J. Koppelman, Joost C.M. Meijers, Conchi Damas, H. Marijke van den Berg,
Bonno N. Bouma, Jan J. Sixma, and George M. Willems
The binding of factor Vlll t o von Willebrand factor (vWF) is
essential for the protection of factor Vlll against proteolytic
degradation in plasma. We have characterized the binding
kinetics of human factor VIII with vWF using a centrifugation
binding assay. Purifiedor plasma vWF was immobilized with
a monoclonal antibody (MoAb RU1) covalently linked t o
Sepharose (Pharmacia LKB Biotechnology, Uppsala, Sweden). Factor Vlll was incubated with vWF-RU1-Sepharose
and unbound factor Vlll was separated from bound factor
Vlll by centrifugation. The amount of bound factor Vlll was
determined from the decrease of factor Vlll activity in the
supernatant. Factor Vlll binding t o vWF-RU1-Sepharoseconformed t o the Langmuir model for independent binding sites
with a K, of 0.46 k 0.12 nmol/L, and a stoichiometry of
1.3 factor Vlll molecules per vWF monomer at saturation.
suggestingthat each vWF subunit contains a binding site for
factor VIII. Competition experimentswere performed with a
recombinant vWF (AM-rvWF), lacking residues 730 t o 910
which contain the epitope for MoAb RU1. AA2-rvWF effectively displaced previously bound factor VIII, confirming that
factor Vlll binding t o vWF-RU1-Sepharose was reversible.To
determinethe association rate constant (L.)and the dissocifactor Vlll was incubatedwith vWFation rate constant (b),
RU1-Sepharose for various time intervals. The observed association kinetics conformedt o a simple bimolecular association reaction with k, = 5.9 f 1.9 x lo6 M-' s-' and b =
1.6 f 1.2 x
s-' (mean f SD). Similar values were obtained from the dissociation kinetics measuredafter dilution
of preformed factor VIII-vWF-RUl-Sepharose complexes.
Identical rate constants ware obtained for factor Vlll binding
t o vWF from normal pooled plasma and t o vWF from plasma
of patients with hemophilia A. The kinetic parameters in
this report allow estimation of the time needed for complex
formation in vivo in healthy individuals and in patients with
hemophilia A, in which monoclonally purified or recombinant factor Vlll associates with endogenous vWF. Using the
plasma concentration of vWF (50 nmol/L in monomers) and
the obtained values for k.. and b,the time needed t o bind
50% of factor Vlll is f12 seconds.
0 1996 by The American Society of Hematology.
ance of factor VIILL3In previous studies, it was shown that
ACTOR VI11 plays an essential role in the intrinsic pathvWF protects factor VI11 from activated protein C-catalyzed
way of blood coagulation as cofactor for factor IXa
inactivation,14 and that vWF prevents activation of factor
in the activation of factor X in the presence of Ca" and
VI11 by activated factor X.I5.l6Factor VI11 binds to vWF via
phospholipids. Qualitative or quantitative deficiencies of facits light chain" with high affinity (& = 0.2 to 0.4 nmoY
tor VI11 cause bleeding tendencies characteristic of hemoL).l8." In vitro, a stoichiometry of 1:l (molecules of factor
philia A. This condition is inherited as a sex-linked recessive
VI11 per subunit of vWF) was found.20In vivo, however,
trait, affecting approximately 1 in 10,000 males.' Factor VIII
circulates in plasma at a concentration of 0.2 p g / d ( ~ 1 vWF apparently circulates with many unoccupied binding
sites because the molar concentrations of both proteins indinmol/L),' and is bound to its carrier protein von Willebrand
cate a 50-fold excess of VWin plasma. Although equilibfactor (vWF) by both hydrophobic and electrostatic interacrium binding data exist, little information is available regardt i o n ~ vWF,
.~
which is an adhesive glycoprotein that faciliing the kinetics of the factor VIII-vWF interaction. Leyte et
tates adhesion of platelets to the injured vessel
circuall8 reported a first-order dissociation rate constant of 0.9 X
lates at approximately 10 p g / d , I and comprises a series of
s-', obtained by means of solid-phase binding experidisulfide-bonded multimers ranging in size from dimers of
ments. However, no rate constant of association was reapproximately 500 kD to species of more than 20 million
ported.
daltom6
In this report, we describe a binding assay using vWF
The liver is generally believed to be the major site of
factor VI11 synthesis, although factor VIII A
"
has been
coupled to Sepharose with a monoclonal antibody (MoAb)
that allows rapid separation of bound and free factor VI11
detected in many other cells and tissues as well.' vWF is
synthesized in endothelial
and is secreted constitutively, or stored in intracellular organelles known as WeibelFrom the Department of Haematology and van Creveld Clinic,
Palade bodies.8 vWF is also synthesized in megakaryocytes
University Hospital Utrecht: and the Cardiovascular Research Instiand stored in a-granules of platelets.' Thus, both glycoprotute Maastricht, University of Limburg, Maastricht, The Netherteins are synthesized at different sites, and complex formalands.
tion probably occurs in the plasma.
Submitted May 11, 1995: accepted October 12, 1995.
The binding of factor VI11 to vWF is essential for the
Supported in part by the Van Creveld Foundation, the Netherlands
survival of factor VIII in the circulation. In either normal
Heart Foundation (Grant No. 90.092), and the Royal Netherlands
Academy of Arts and Sciences.
individuals or patients with hemophilia A, the half-life of
Address reprint requests to Andre' J. Vlot, MD, UniversityHospital
transfused factor VIII is approximately 12 hours, and the
Utrecht, Department of Haematology (G03.647), PO Box 85500,
clearance of factor VI11 is indistinguishable from that of
3508 GA Utrecht, The Netherlands.
either vWF or preformed factor VIII-vWF complex." HowThe publication costs of this article were defrayed in part by page
ever, in patients with severe von Willebrand disease (vWD)
charge payment. This article must therefore be hereby marked
the lack of stabilization of factor VIII by binding to vWF
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
causes a decrease of factor VIII." Similarly, a functional
indicate this fact.
defect in the factor VI11 binding domain of vWF (in vWD
0 1996 by The American Society of Hematology.
type Normandy or type 2NL2)results in an accelerated clear0006-4971/96/8705-0046$3.00/0
F
Blood, Vol87, No 5 (March l ) , 1996 pp 1809-1816
1809
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1810
VLOT ET AL
by centrifugation. Using this assay, we characterized the
kinetics of factor VI11 binding to vWF.
MATERIALS AND METHODS
All experiments were performed at 22°C in 150 mmol/L NaCI,
50 mmoVL TRIS-HCI, pH 7.4, containing 0.1%Tween-20 (E. Merck
AG, Darmstadt, Germany), 1% bovine serum albumin fraction V
(Sigma Chemical CO, St Louis, MO), and 2 mmoVL CaCI, (TBS),
unless indicated otherwise. Centrifugation steps were performed in
an Eppendorf microfuge at 10,000g for 10 seconds.
Proteins. Recombinant factor VI11 was a generous gift from Dr
R.J. Kaufman (Genetics Institute, Cambridge, MA). This preparation
was essentially free of vWF as measured with an enzyme-linked
immunosorbent assay (ELISA). Human vWF devoid of factor VI11
was purified from human cryoprecipitates (kindly provided by A.
Knevelman and H.J.C. de Wit from the Red Cross Blood Bank
Friesland, Leeuwarden, The Netherlands) using gel-filtration chromatography as described previously.16,2'Fractions were collected
and assayed for vWF:Ag, vWF:Ag multimers and factor VI11 activity. The first fractions contained high multimers, followed by intermediate and low multimeric vWF. Fractions were pooled and stored
at -70°C until use, except for the experiments with high and low
multimeric vWF, where fractions were not pooled and stored separately.
Production of a deletion variant vWF (AA2-rvWF). A detailed
description of the construction and expression of a recombinant
vWF, which lacks the A2 repeat (residues 730 to 910) in baby
hamster kidney cells will be published elsewhere (Lankhof H, et al,
submitted). Briefly, site-directed mutagenesis was used to delete the
coding region for the A2 domain, and the deletion construct was
cloned in the pNUT-expression vector." After transfection, stable
cell lines were selected with methotrexate. AA2-rvWF was purified
from serum-free medium with affinity chromatography techniques,
and showed a full range of vWF:Ag multimers on gel electrophoresis
and ristocetin-induced platelet aggregation, equivalent to normal
plasma.
Plasma vWF. Blood from 40 healthy donors was collected in
one-tenth volume of 3.8% trisodium-citrate. After centrifugation at
1,200g for 15 minutes at 4"C, the plasma was pooled and stored at
-70°C until use. In addition, plasma from three patients with severe
hemophilia A was obtained, containing less than 1% factor VI11
activity, and normal vWF:Ag and ristocetin cofactor activity levels.
Antibodies. MoAb RUI directed against the A2 repeat of human
vWF was prepared in our laboratory."
Quant$cation ofproteins. vWF concentrations were determined
with an ELISA using nonconjugated and peroxidase-conjugated
polyclonal rabbit antibodies against human vWF (Dakopatts,
Glostrup, Denmark), with normal pooled plasma as a standard. The
multimeric structure was assayed using the Pharmacia Phast Gel
System (Pharmacia LKB Biotechnology, Uppsala, Sweden) as described.*' Factor VI11 concentrations were determined by using a
chromogenic substrate assay (Factor VI11 Coatest; Chromogenix,
Molndal, Sweden).24The change in absorbance at 405 nm was measured using a V-max reader (Molecular Devices Corp, Menlo Park,
CA). Normal pooled plasma was used as a reference. The concentrations of factor VIII and vWF in normal pooled plasma were taken
from literature as 0.2 pg/mL* and 10 pg/mL,' respectively. The
molecular weights used to determine concentrations were as follows:
factor VIII, 240,000; purified and plasma vWF monomer, 270,000;
and AA2-rvWF monomer, 250,000.
Preparation of RUl-Sepharose. MoAb RUI was coupled to cyanogen bromide-activated Sepharose CL-4B (Pharmacia LKB Biotechnology, Uppsala, Sweden) to a final concentration of 1 mg MoAb
RU l/mL of Sepharose, according to the manufacturer's instructions.
After coupling and blocking of residual active sites, RU 1-Sepharose
was incubated for 30 minutes with TBS containing 4% bovine serum
albumin to prevent nonspecific binding. vWF was added to a 50%
suspension of RU1-Sepharose in a final volume of 1 mL and allowed
to bind. During incubation the sample was mixed and after 1 hour
the vWF-RUI -Sepharose was washed five times by resuspension
and repeated centrifugation in TBS.
Association experiments. Factor VI11 (150 pL) was incubated
with vWF-RUI-Sepharose (50 pL; 50% suspension in TBS). During
incubation the tubes were mixed by inversion. After IO to 600 seconds, the factor VI11 bound to vWF-RU1-Sepharose and unbound
factor VI11 were separated by centrifugation and the factor VlIl
concentration in the supernatant was measured. Initial experiments
performed at 22°C and 37°C yielded similar rate constants, thus
subsequent experiments were performed at 22°C.
Competition experiments. Because AA2-rvWF lacks the epitope
for MoAb RU1, it cannot compete with vWF for binding to RUlSepharose. vWF-RU1-Sepharose (100 pL; 50% suspension in TBS)
was incubated with factor VI11 (100 pL) for 30 minutes, and then
AA2-NWF (200 pL; final concentration 0 to 35 nmol/L) was added.
After 2 hours, the samples were centrifuged and the supernatants
were tested for factor VI11 activity.
Dissociation experiments. The dissociation rate constant (kc,,,)
for the factor VIII-vWF complex was determined in dilution experiments. Factor VI11 (15 pL) was added to vWF-RUI-Sepharose (250
pL, 50% suspension in TBS) and mixed. After 30 minutes, the
mixture was diluted 50-fold by adding TBS. At different time intervals, 300-pL aliquots containing buffer and Sepharose beads were
removed, centrifuged, and the supernatant was assayed for factor
VI11 activity.
Dura analysis. Equilibrium binding data of factor VI11 to vWFRU1-Sepharose were analyzed by the Langmuir model for independent binding sites:
FVIIb.
c,, = c m a xKd + FVIIIfm,
with C,, the amount of factor VI11 bound to vWF-RU1-Sepharose,
the maximal binding capacity of the vWF-RU1-Sepharose, Kd
the dissociation constant, and FVIIIf,eethe concentration of unbound
factor VIII. It should be noted that this model does not take into
account the possibility that the adjacent binding sites on the
multimeric vWF interact. Application of this model to binding data
obtained for various factor VI11 concentrations and a fixed vWF
concentration allowed establishment of the binding stoichiometry
and of the dissociation constant.
In competition experiments the nonsedimentable fraction of factor
VI11 (FVIIIns)of total added factor VI11 (FVIIb) was measured as
function of the AA2-rvWF concentration. In this situation the binding of factor VI11 to vWF-RUl-Sepharose is again described by
Equation 1, whereas the binding to AA2-rvWF is described by a
similar equation:
,C
,
where C,,,,,, is the maximal binding capacity of AA2-rvWF and
Kd,,* is the dissociation constant for factor VI11 binding to AA2rvWF. These experiments were performed in the presence of high
concentrations vWF. As a result the concentration of unbound factor
VIII, FVIIIfc, was far below the Kd and Equations ( I ) and (2) can
be simplified to:
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1811
KINETICS OF FACTOR VIII-vWF ASSOCIATION
RESULTS
and
Using the relations, C,, = FVIII,, - FVIII., and FVIII,,
+ Ccq,~AZ,
an explicit expression for FVIII,, is derived:
FVIII",
=
, c,,,
FVIIIO
I
I<d
FVIIIf,
(3)
&,AA2
I
A
=
&,AA2
+ Cmar.AA2
Equilibrium binding results were compatible (see Fig I), with one
independent factor VI11 binding site per vWF subunit. Therefore,
binding kinetics were modeled as a simple reversible bimolecular
association reaction:
k,
factor VI11 + vWF = C
(Scheme 1)
where C represents the factor VIII-vWF complex and k,, and koff
are the association and dissociation rate constants for the reaction.
These constants are related to the dissociation constant Kd = koff/
ken. The rate of change of the concentration of the factor VIII-vWF
complex with time is given by the differential equation:
(4)
with FVMo and vWFo the total added factor VI11 and vWF monomer
concentrations. The concentration of factor VIII-vWF complexes at
,
expressed in terms of FVIII,, and vWFO, is given
equilibrium, ,C
by:
1
C,, = p = -[FVIIIo
2
- &VIIIo
+ vWFO+ Kd
(5)
+ vWF, + Kd)' - 4FVIIbvWFO]
The solution of Equation 4 is given by:
C(t) = P
-
P 4 - CO1 -p
(6)
~ e * J P - d ~
4 - CO
where COrepresents the initial concentration of FVIII-vWF complexes, and
P =
c,,
and
4=
FVIIIovWFO
P
In the association experiments, CO= 0, and in the dissociation experiments, CO is calculated from Equation 5, using the dilution factor
and the concentrations of factor VI11 and vWF during the incubation.
Binding constants and reaction rates were estimated by an ordinary
least-squares fit of Equation 1, 3, or 6 to the experimental data (Fig.
P,Biosoft, Cambridge, UK) and are presented as estimate ? standard
error of the estimate (SEE), unless indicated otherwise.
Binding of vWF and factor VIII to RUI-Sepharose. The
concentration of vWF bound to RUl-Sepharose was determined by measuring the vWF concentrations in the supematant after incubation with RU1-Sepharose, and by subtracting
these values from the amount initially present in the incubation mixture. Typically, 70% to 80% of added vWF bound to
RUl-Sepharose under these conditions, and no vWF release
could be measured after subsequent washing steps. Identical
binding was found for purified and plasma vWF.
Aspecific binding of factor VI11 could result in an overestimation of factor VI11 binding. However, in control experiments where factor VI11 was added to RUl-Sepharose we
observed more than 98% recovery of factor VIII, which
indicated that aspecific binding was less than 2% and was
therefore considered to be negligible. In preliminary experiments, factor VI11 (0.5 nmol/L) was added to an excess
of vWF-RU1-Sepharose. After 30 minutes, samples were
centrifuged and supematants were assayed for factor VI11
activity. The amount of factor VI11 remaining in the supematant was 8.8% ? 2.6% (mean ? SD, n = 4) of the initial
amount. This factor VI11 also did not bind to newly added
vWF-Sepharose, and therefore most likely represents activated factor VI11 that does not bind to vWF but is normally
detected in the chromogenic assay. We corrected for this
fraction of nonbinding factor VI11 in the data analysis.
Equilibrium binding isotherm. To determine whether the
affinity of vWF for factor VI11 was affected by binding of
vWF to RU1-Sepharose, increasing amounts of factor VI11
were incubated with a constant amount of vWF-RUI-Sepharose (2.1 nmol/L) for 30 minutes and the time course of
binding was followed until equilibrium was reached. The
concentration of bound factor VI11 was obtained by subtracting the equilibrium value of unbound factor VI11 from
the total added factor VI11 concentration. Figure 1 shows
that factor VI11 binds to vWF-RUl-Sepharose in a dosedependent saturable manner. An adequate fit to these data
was obtained using the model for (identical) independent
binding sites given in Equation 1 . The dissociation constant
(Kd)and the number of binding sites (CmaX)
were estimated
by a least-squares fit of Equation 1 to the data: Kd = 0.46
? 0.12 nmol/L, and C,,, = 2.8 2 0.3 nmol/L. The obtained
value of C,,, corresponds to a binding stoichiometry of 1.3
molecules factor VI11 per vWF monomer, indicating that
every vWF monomer is accessible for factor VI11 binding.
Studies with AA2-rvWF. To compare factor VI11 binding to solution-phase vWF and to vWF immobilized to RU1Sepharose, competition experiments were performed in
which a vWF mutant was incubated with preformed factor
VIII-vWF-RU1 -Sepharose complexes. The mutant (AA2rvWF) was chosen because it lacks the A2 domain (residues
730 to 910) that contains the epitope for MoAb RU1, and
therefore does not bind to MoAb RU1. First, factor VI11
(0.16 nmol/L) was incubated with vWF-RU1-Sepharose
(vWF concentration 8.0 nmol/L) for 30 minutes. Then AA2rvWF (0 to 35 nmol/L) was added, and after 2 hours of
incubation, the samples were centrifuged and the factor VI11
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1812
VLOT ET AL
3.0
3
8
U
v
0.15
2s
0.10
2.0
P
U
1.5
P
0.05
0
0
0.5
1.0
1.5
2.0
2.5
activity in the supernatant was measured. This activity represents the sum of unbound factor VI11 and factor VI11 bound
to AA2-rvWF. Control experiments indicated that the presence of AA2-rvWF did not interfere with cofactor activity of
factor VI11 in the chromogenic assay used (data not shown).
Figure 2 shows that addition of AA2-rvWF resulted in a
dose-dependent increase of the factor VI11 concentration in
the supernatant. By a least-squares fit of Equation 3 to the
data, using the binding parameters for vWF-RUl -Sepharose
determined in Fig 1, a dissociation constant of 0.32 +- 0.04
nmol/L was found. This result provides further evidence that
factor VI11 binding to vWF-RU 1-Sepharose is reversible,
and that immobilizing vWF via an MoAb to Sepharose does
not alter the binding affinity for factor VIII.
Association experiments. The rate of factor VIII-vWF
complex formation was determined by incubating factor VI11
with vWF-RU1-Sepharose for various time intervals, and by
assaying the concentration of unbound factor VI11 in the
supernatant. Initial concentrations of factor VI11 and vWF
were chosen so that complex formation was slow enough to
be followed in time, ie, with concentrations below the plasma
concentrations for both factor VI11 and vWF. Figure 3 shows
the kinetics of binding of factor VI11 (0.49 nmol/L) to vWFRU1-Sepharose (vWF concentration 2.1 nmoVL). Adequate
fit was obtained by using the simple bimolecular association
reaction of Scheme 1 (Equation 6). The values for the association and dissociation rate constants were: k,, = 5.9 f 1.9
I
I
I
20
30
40
AA2-rVWF (nmoyL)
factor VIII free (nmoVL)
Fig 1. Binding isotherm for the interaction between factor Vlll and
vWF-RU1-Sepharose. Factor Vlll(0.5 to 5 nmol/L) was incubated with
vWF-RU1-Sepharose (vWF concentration 2.1 nmol/L) for 10 to WO
seconds. Unbound and bound factor Vlll were separated by centrifugation at 10,OOOg for 10 seconds. Supernatants were tested for the
concentrationof unbound factor VIII. Each point represents the concentration of bound factor VI11 at equilibrium, calculated according
to Equation 6. The solid line rapresents the least squares fit of the
binding isotherm of Equation 1 to these data. Estimated values of
the dissociation constant and maximal binding were: K, = 0.46 f
0.12 nmol/L and C- = 2.8 ? 0.3 nmol/L, respectively. Buffer: 150
mmol/L NaCI, 50 mmol/L TRIS-HCI, pH 7.4, 0.1% Tween-20, 1% bovine serum albumin, 2 mmol/L CaCI,; T = 22°C.
I
10
Fig 2. Dissociation of factor VIM-vWF-RUl-Sepharose by AA2rvWF. Factor Vlll (0.16 nmol/L) was added t o vWF-RU1-Sepharose
(vWF concentration8.0 nmol/L) and incubated for 30 minutes. AA2rvWF was then added and incubated for 2 hours. After centrifugation,
the supematants were testad for factor Vlll activity. Data points,
representing the mean of duplicate daterminations, were analyzed
according to Equation 3, and a value of K, = 0.32 2 0.04 nmol/L for
the disrociation constant was found.
lo6 M-' s-' and 1.6 2 1.2 X
s-', respectively (mean
+- SD, n = 3). The dissociation constant at equilibrium (KJ,
calculated as kofdkon,was 0.27 t 0.22 nmoVL.
Association experiments were also performed in which
varying amounts of vWF (1.5 or 8 nmol/L) were coupled to
X
o.6
1
0.4
0.2
0
I
I
1
200
400
600
time(sec)
Fig 3. Association of factor Vlll with purified vWF coupled to RUlSepharose. Recombinant factor VIU was inahtad with vWF-RU1Sapharosefor various time i n t w a h Concsntrations were 0.49 nmoll
L and 2.1 nmollL for factor Vlll and vWF, respucthfely. At the indiited
times, unbound and bound factor VIN were separatd by centrifugation, and the concentration of unbound factor Wll was measured in
the supomatant. The solid line rapresents the best fit of Equation 6
to thase data, with estimated values k,,= 7.9 f 0.4 x 10' M-' s-'
and kW
= 2.5 f 0.4 x lO-'s-'.
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1813
KINETICS OF FACTOR VIII-vWF ASSOCIATION
Table 1. Kinetics of Factor VI11 Binding to Purified vWF
2o
~
0.27 t 0.22
1.6 2 1.2
1.5 2 0.9
1.1 t 0.5
5.9 t 1.9
6.5 t 2.5
3.4 2 1.4
All multimers
High multimers
Low multimers
~
1
0.24 2 0.17
0.32 2 0.20
~
~
Factor VI11 was incubated with purified vWF coupled to RU1-Sepharose for 10 to 600 seconds. Concentrations varied from 0.34 to 0.52
nmol/L and 1.5 to 6.0 nmol/L for factor VI11 and vWF, respectively.
Unbound and bound factor VI11 were separated by centrifugation, and
the concentration of unbound factor VI11 in the supernatant was measured with a chromogenic assay. Kinetic rate constants were derived
by a least squares fit of Equation 6 to the data. The values are indicated as mean t SD of three experiments performed in duplicate.
0
a constant amount of RU1-Sepharose and incubated with a
low (0.02 nmol/L) or high (0.34 nmol/L) factor VI11 concentration. As expected, the rate of complex formation increased
with increasing factor VIII and vWF concentrations, whereas
the estimated rate constants were similar (knn = 5.2 and 4.3
s-', respecx lo6 M-' s-' and knff = 3.9 and 2.2 X
tively). This indicates that the model of bimolecular association (Scheme 1) provides adequate description and that complex formation is not influenced by a transport rate limitation.
The purified v W F used in the preceding experiments contained a mixture of high and low vWF multimers, as present
in plasma. To investigate the influence of multimeric size of
vWF on the rate of complex formation, v W F subfractions
containing only high or low multimers were used to immobilize vWF to RUl-Sepharose, and association experiments
were performed. Binding of high and low multimeric v W F
to RU1-Sepharose was similar (data not shown). Table 1
shows that similar values were found for the rate constants
of association and dissociation, indicating that factor VIIIv W F interaction is independent of the degree of vWF multimerization.
Because the coupling of vWF to RU1-Sepharose is followed by a washing procedure, it also permits selective binding of vWF from plasma. We performed binding experiments
with vWF from normal pooled plasma and from plasma from
patients with severe hemophilia A. As shown in Table 2,
Table 2. Kinetics of Factor Vlll Binding to Plasma vWF
Normal pooled plasma
Hemophilic plasma
Patient 1
Patient 2
Patient 3
L"
Ln
Kd
xlOs M-'s-'
~ 1 0 - 3S-1
~ 1 0 - 9M
6.6 t 0.8
1.5 t 0.5
0.23 t 0.08
6.7 t 1.4
6.9 2 4.5
4.9 c 4.2
1.1 t 0.1
1.7 2 0.9
1.2 t 0.4
0.16 t 0.04
0.25 t 0.14
0.24 t 0.08
vWF from normal pooled plasma or from plasma from patients with
severe hemophilia A was coupled to RU1-Sepharose and incubated
with factor Vlll for 10 to 600 seconds. Concentrations varied from 0.36
to 0.56 nmol/L and 1.3 to 3.1 nmol/L for factor VI11 and vWF, respectively. Assays were performed and data were analyzed as described
in the legend to Table 1. The values are indicated as mean t SD of
three experiments performed in duplicate.
I
I
I
I
30
60
90
120
time (min)
Fig 4. Dissociationof factor Vlll from vWF bound to RU1-Sepharose. Factor VI11 (1.6 nmol/L) was incubated with vWF-RU1-Sepharose (vWF concentration7.7 nmol/L) for 30 minutea. The mixture was
diluted 50 times and mixed by inversion. At the indicated times,
samples of 300 p L were taken and centrifuged. The supematants
wera assayed for factor Vlll activity with a chromogenic assay. The
solid line represents the best fit of Equation 6 to these data, with
estimated values hn= 4.2 f 0.5 x 10' M-' s-' and IC,,, = 0.72 f 0.06
x 10-3
vWF from normal pooled plasma and from plasma of patients with severe hemophilia A exhibits kinetic rate constants comparable to purified vWF.
Dissociation experiments. The dissociation rate constant
of factor VIII binding to vWF-RU1-Sepharose was determined independently by dilution of preformed factor VIIIvWF-RU1-Sepharose complexes, prepared by incubation of
factor VI11 (1.6 nmol/L) with vWF-RU1-Sepharose (vWF
concentration 7.7 nmoyL). The dilution (50-fold) was chosen
so that the concentration of factor VI11 decreased to below
the dissociation constant of the factor VIII-vWF complex
(Kd 0.4 nmol/L, see Fig 1). Figure 4 shows that the simple
bimolecular association model (Scheme 1, Equation 6) provides an excellent fit to these data. The fit resulted in a
dissociation rate constant, kOn= 0.91 ? 0.24 X
s-'
(mean ? SD, n = 4), which is in good agreement with
s-' obtained in the association
the value k,, = 1.6 X
experiments. Similar experiments with vWF from normal
pooled plasma or from patients with severe hemophilia A
yielded equivalent results (data not shown).
-
DISCUSSION
In the present study we developed a method that allowed
the measurement of the association and dissociation kinetics
of the factor VIII-vWF complex with a time resolution of
15 seconds. vWF was immobilized by binding to an MoAb
(MoAb RU1) coupled to Sepharose beads. Free factor VI11
and factor VI11 associated with vWF bound to the Sepharose
could be separated by centrifugation. The data presented
here show that factor VIII-vWF complex formation is a rapid
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1814
process with an association rate constant of 5.9 t 1.9 x 10'
M-' s-' and a dissociation rate constant of 1.6 t 1.2 x lo-?
s-'. In addition, we found identical rate constants for vWF
from normal pooled plasma and from plasma of patients
with hemophilia A.
The dissociation constant, Kd = 0.46 t 0.12 nmoVL,
found in equilibrium experiments, was in good agreement
with the Kd values for the factor VIII-vWF interaction previously reported by us2' and other^.'^.'^^^^ The stoichiometry
found in this study is consistent with the value of 1: 1 (molecules of factor VI11 per subunit of vWF) reported by Lollar et
a1," who analyzed the interaction between the corresponding
porcine proteins. It is also consistent with the value that we
found previously in experiments with vWF conjugated to 15nm colloidal gold." In experiments with vWF immobilized
directly or indirectly via an antibody on the wells of a microtiter plate, a different stoichiometry of 150 to 1:lOO was
found.'8.2',26We previously observed that in solution, vWF
binding to MoAbs with epitopes distinct from the factor
VI11 binding region per se does not interfere with the 1:l
stoichiometry of factor VI11 binding." Therefore, it appears
that binding of vWF directly to a plastic surface or indirectly
via an MoAb causes steric restrictions, which then results
in a loss of factor VI11 binding capacity of the majority (98%
to 99%) of the vWF subunits. However, in this study all
vWF subunits retained their binding capacity when bound
to Sepharose with an antibody, and the value of Kd found in
competition experiments (see Fig 2 ) indicates that binding
of vWF to RU1-Sepharose does not significantly affect the
affinity of vWF for factor VI11 binding.
The observed association and dissociation kinetics were
adequately described by assuming one single class of binding
sites for factor VI11 and vWF. The value k,, = 5.9 t 1.9 X
10' M-' s-' found for the association rate constant of the
factor VIII-vWF complex is well below the diffusion limit
of lo9 M-' s-', and typical for protein-protein reactions in
solution. The value kOff = 1.6 t 1.2 X lo-' s-' for the
first-order rate constant of dissociation and the half-life of
dissociation of 430 seconds found in this study agrees well
with the results from Leyte et a1,I8 who reported a first-order
dissociation rate constant of 0.9 X lo-' s-' and a half-life
of dissociation of 780 seconds. The dissociation of bound
factor VI11 from vWF-RU1-Sepharose in exchange experiments with AA2-rvWF supports the reversibility of the factor VIII-vWF interaction. It also indicates that factor VI11
can dissociate from vWF in the presence of an excess of
vWF, and in the absence of thrombin. This is consistent
with our previous observation in vivo, in which radiolabeled
factor VI11 in complex with lower multimeric vWF was
redistributed over all multimeric forms of vWF after infusion
into patients with hemophilia A."
The assay used in this study showed no difference in factor
VIII-vWF binding kinetics between low and high multimeric
vWF. Likewise, the calculated values for the dissociation
constants were similar. This extends our earlier observation
that both forms of vWF have the same affinity toward factor
VI11 (Koppelman S, et al, Blood, in press). It is important
to note that vWF was bound to RUl-Sepharose beads with
VLOT ET AL
a diameter of 100 pm, which could introduce a rate limit
for the mass transfer of factor VI11 from solution to the
Sepharose surface and thereby influence the rate of association. To evaluate this possibility, we performed association
experiments with different densities of vWF bound to RU 1Sepharose. Identical values were found for the rate constant
of association, indicating that the observed rate constants
reflect intrinsic kinetics and are unaffected by transport limitations.
The porous structure of the Sepharose beads, with a 20
X 10' Dalton exclusion size of the pores, also could present
a point of concern with respect to the accessibility for factor
VI11 binding. Our equilibrium binding data show that all
vWF monomers are accessible for factor VI11 binding. In
addition, no apparent deviations from the simple bimolecular
association model were found in measurements of the association and dissociation kinetics. More convincingly, restricted
accessibility could only be anticipated for low molecular
vWF multimers, with molecular dimensions similar to factor
VIII, whereas our data in Table 1 show that the kinetics
observed for low and high multimers are not significantly
different.
The site of complex formation between factor VI11 and
vWF in the human body is not precisely known. The liver
is the major site of factor VI11 synthesis, as shown by immunochemical localization studies,2Rwhereas vWF is synthesized by vascular endothelial cells' and megakaryocytes.'
Interestingly, hepatic sinusoidal endothelial cells do not contain vWF:Ag' or Weibel-Palade bodies," which suggests
that these cells are different from their vascular counterparts
in that they do not synthesize vWF. Directly after synthesis
and secretion, factor VI11 enters the space of Disse, the narrow space between the hepatocyte and the endothelium lining the sinusoid." The endothelium of the hepatic sinusoids
contains fenestrations with openings up to 1 pm'" that permit
plasma vWF to gain access to the surface of the hepatic
cells. Thus, complex formation most likely occurs in the
liver in the space of Disse, where factor VI11 and vWF first
meet. The rate of complex formation, expressed as the halftime for complex formation, can be calculated as tlir= ln2/
(kc>"x vWF concentration + kc,ff).Based on the plasma concentration of vWF (50 nmol/L. in monomers),' together with
the estimates for k,, and k,,, obtained in this study, the time
needed to bind 50% of factor VlII is approximately 2 seconds.
The results obtained in the present work also provide L\
theoretical basis for the successful treatment of patients with
hemophilia A with factor VI11 concentrates presently available. Since the 1970s, therapy in patients with severe hemophilia A depended on cryoprecipitates, a product that retains
the normal amount and distribution of vWF multimers observed in plasma. However, during the period from 1980 to
1990, monoclonally purified and recombinant factor VI11
concentrates became available, and those products have become the standard in the treatment of patients with hemophilia A.''." However, it should be noted that these concentrates contain little or no vWF, and therefore factor VI11 is
dependent on endogenous vWF with which it has to form a
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
KINETICS OF FACTOR VIII-vWF ASSOCIATION
complex to be protected against proteolytic degradation in
plasma. Although clinical studies have proven that the
plasma half-life of monoclonally purified and recombinant
factor VI11 is not substantially different from that of less
pure concentrate^,".^^ the time needed for complex formation
between infused factor VI11 and endogenous vWF was heretofore unknown. Our results show that vWF from patients
with hemophilia A resembles normal vWF with respect to
factor VI11 binding kinetics. Moreover, infused factor VI11
is bound rapidly to the plasma vWF, with more than 95%
binding in 12 seconds, which is extremely rapid in comparison to the half-life of 1 hour observed for factor VI11 in the
absence of vWF in plasma.33
Rapid complex formation may also serve to promote stable assembly of factor VI11 subunits at the site of secretion,34
or to prevent factor VI11 from interacting with components
that bind with lower affinity, such as factor IXa (Kd = 15
n m ~ l / L ) 'and
~ phospholipids (Kd = 2 to 4 n m ~ l / L )Previ.~~
ous studies have established that factor VI11 can be activated
by factor Xa during the initial phase of factor X activation.
Because factor VI11 bound to vWF is protected from activation by activated factor X,I5 rapid complex formation between factor VI11 and vWF may be important in regulating
factor VI11 activity by inhibiting this feedback activation of
factor VIII.
In summary, the kinetic experiments in this study show
that factor VI11 forms a complex with vWF in plasma, with
a half-life of complex formation of 2 seconds. This rapid
complex formation between vWF and factor VIII, either in
the liver or after infusion of factor VI11 in patients with
hemophilia A, may be important in the stabilization and
regulation of factor VI11 activity in plasma.
ACKNOWLEDGMENT
We thank Dr R.J. Kaufman for the recombinant factor VIII, and
A. Knevelman and H.J.C. de Wit for the vWF concentrates.
REFERENCES
1. Hoyer LW: The factor VI11 complex: Structure and function.
Blood 58:1, 1981
2. Wion KL, Kelly D, Summerfield JA, Tuddenham EG, Lawn
RM: Distribution of factor VI11 mRNA and antigen in human liver
and other tissues. Nature 317:726, 1985
3. Fay PJ: Factor VI11 structure and function. Thromb Haemost
70:63, 1993
4. Weiss HJ, Baumgartner HR, Tschopp TB, Turitto VT, Cohen
D: Correction by factor VI11 of the impaired platelet adhesion to
subendothelium in von Willebrand disease. Blood 5 1:267, 1978
5. Sakariassen KS, Bolhuis PA, Sixma JJ: Human blood platelet
adhesion to artery subendothelium is mediated by factor VIE-von
Willebrand factor bound to the subendothelium. Nature 279:636,
1979
6. Sadler JE: von Willebrand factor. J Biol Chem 266:22777,
1991
7. Jaffe EA, Hoyer LW, Nachman R: Synthesis of antihemophilic
factor antigen by human endothelial cells. J Clin Invest 52:2757,
1973
8. Wagner DD, Marder VJ: Biosynthesis of von Willebrand protein by human endothelial cells. J Biol Chem 258:2065, 1983
9. Nachman R, Levine R, Jaffe EA: Synthesis of factor VI11
1815
antigen by cultured guinea pig megakaryocytes. J Clin Invest 60914,
1977
10. Over J, Sixma JJ, Bruine MH, Trieschnigg MC, Vlooswijk
RA, Beeser-Visser NH, Bouma BN: Survival of 125iodine-labeled
factor VI11 in normals and patients with classic hemophilia. Observations on the heterogeneity of human factor VIII. J Clin Invest 62:223,
1978
11. Weiss HJ, Sussman 11, Hoyer LW: Stabilization of factor VI11
in plasma by the von Willebrand factor. Studies on posttransfusion
and dissociated factor VI11 and in patients with von Willebrand's
disease. J Clin Invest 60:390, 1977
12. Sadler JE: A revised classification of von Willebrand disease.
Thromb Haemost 71:520, 1994
13. Nishino M, Girma JP, Rothschild C, Fressinaud E, Meyer D:
New variant of von Willebrand disease with defective binding to
factor VIII. Blood 74:1591, 1989
14. Koedam JA, Meijers JC, Sixma JJ, Bouma BN: Inactivation
of human factor VI11 by activated protein C. Cofactor activity of
protein S and protective effect of von Willebrand factor. J Clin
Invest 82:1236, 1988
15. Koedam JA, Hamer RJ, Beeser-Visser NH, Bouma BN,
Sixma JJ: The effect of von Willebrand factor on activation of factor
VI11 by factor Xa. Eur J Biochem 189:229, 1990
16. Koppelman SJ, Koedam JA, van Wijnen M, Stem D, Nawroth
P, Sixma JJ, Bouma BN: Von Willebrand factor as a regulator of
intrinsic factor X activation. J Lab Clin Med 123:585, 1994
17. Hamer RJ, Koedam JA, Beeser-Visser NH, Bertina RM, van
Mourik JA, Sixma JJ: Factor VI11 binds to von Willebrand factor
via its Mr-80,000 light chain. Eur J Biochem 166:37, 1987
18. Leyte A, Verbeet MP, Brodniewicz-Proba T, van Mourik JA,
Mertens K: The interaction between human blood-coagulation factor
VI11 and von Willebrand factor. Characterization of a high-affinity
binding site on factor VIII. Biochem J 257:679, 1989
19. Nesheim M, Pittman DD, Giles AR, Fass DN, Wang JH,
Slonosky D, Kaufman RJ: The effect of plasma von Willebrand
factor on the binding of human factor VI11 to thrombin-activated
human platelets. J Biol Chem 266:17815, 1991
20. Lollar P, Parker CG: Stoichiometry of the porcine factor VIIIvon Willebrand factor association. J Biol Chem 262:17572, 1987
21. Vlot AJ, Koppelman SJ, van den Berg HM, Bouma BN,
Sixma JJ: The affinity and stoichiometry of binding of human factor
VI11 to von Willebrand factor. Blood 85:3150, 1995
22. Palmiter RD, Behringer RR, Quaife CJ, Maxwell F, Maxwell
IH, Brinster RL: Cell lineage ablation in transgenic mice by cellspecific expression of a toxin gene. Cell 50:435, 1987
23. Lawrie AS, Hoser MJ, Savidge GF: Phast assessment of
vWfAg multimeric distribution. Thromb Res 59:369, 1990
24. Tripodi A, Mannucci PM: Factor VI11 activity as measured
by an amidolytic assay compared with a one-stage clotting assay.
Am J Clin Pathol 86:341, 1986
25. Lopez-Femandez MF, Blanco-Lopez MJ, Castineira MP, BatIle J: Further evidence for recessive inheritance of von Willebrand
disease with abnormal binding of von Willebrand factor to factor
VIII. Am J Hematol 40:20, 1992
26. Layet S, Girma JP, Obert B, Peynaud-Debayle E, Bihoreau
N, Meyer D: Evidence that a secondary binding and protecting site
for factor VI11 on von Willebrand factor is highly unlikely. Biochem
J 282:129, 1992
27. Over J, Bouma BN, Sixma JJ, Bolhuis PA, Vlooswijk RA:
Heterogeneity of human factor VIII. 111. Transitions between forms
of factor VI11 present in cryoprecipitate and in cryosupematant
plasma. J Lab Clin Med 95:323, 1980
28. Zelechowska MG, van Mourik JA, Brodniewicz-Proba T: U1-
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1816
trastructural localization of factor VIII procoagulant antigen in human liver hepatocytes. Nature 317:729, 1985
29. Horn T, Lyon H, Christoffersen P: The blood hepatocytic
barrier: A light microscopical, transmission- and scanning electron
microscopic study. Liver 6:233, 1986
30. Wisse E: An electron microscopic study of the fenestrated
endothelial lining of rat liver sinusoids. J Ultrastruct Res 31:125,
1970
31. Kasper CK, Kim HC, Gomperts ED, Smith KJ, Salzman PM,
Tipping D, Miller R, Montgomery RM: In vivo recovery and survival
of monoclonal-antibody-purified factor VI11 concentrates. Thromb
Haemost 66:730, 1991
32. Lusher JM, Arkin S, Abildgaard CF, Schwartz RS, Kogenate
Previously Untreated Patient Study Group: Recombinant factor VI11
for the treatment of previously untreated patients with hemophilia
VLOT ET AL
A. Safety, efficacy, and development of inhibitors. N Engl J Med
328:453, 1993
33. Brinkhous KM, Sandberg H, Garris JB, Mattsson C, Palm M,
Griggs T, Read MS: Purified human factor VI11 procoagulant protein: Comparative hemostatic response after infusions into hemophilic and von Willebrand disease dogs. Proc Natl Acad Sci USA
82:8752, 1985
34. Wise RJ, Domer AJ, Krane M, Pittman DD, Kaufman RJ: The
role of von Willebrand factor multimen and propeptide cleavage in
binding and stabilization of factor VIII.J Biol Chem 266:21948, 1991
35. Lenting PJ, Donath MJ, van Mourik JA, Mertens K: Identification of a binding site for blood coagulation factor IXa on the light
chain of human factor VIII. J Biol Chem 269:7150, 1994
36. Gilbert GE, Furie BC, Furie B: Binding of human factor VIII
to phospholipid vesicles. J Biol Chem 265:815, 1990
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1996 87: 1809-1816
Kinetics of factor VIII-von Willebrand factor association
AJ Vlot, SJ Koppelman, JC Meijers, C Dama, HM van den Berg, BN Bouma, JJ Sixma and GM
Willems
Updated information and services can be found at:
http://www.bloodjournal.org/content/87/5/1809.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.