Cardiovascular Research 63 (2004) 130 – 138
www.elsevier.com/locate/cardiores
Effect of lipoprotein (a) on platelet activation induced by
platelet-activating factor: role of apolipoprotein (a) and
endogenous PAF-acetylhydrolase
Loukas D. Tsironis a, John V. Mitsios a, Haralampos J. Milionis b,
Moses Elisaf b, Alexandros D. Tselepis a,*
a
Laboratory of Biochemistry, Department of Chemistry, University of Ioannina, Ioannina 45110, Greece
b
Department of Internal Medicine, Medical School, University of Ioannina, Ioannina 45 110, Greece
Received 10 December 2003; received in revised form 3 February 2004; accepted 1 March 2004
Available online 26 April 2004
Time for primary review 26 days
Abstract
Objective: Lipoprotein (a) [Lp(a)] is considered an atherogenic lipoprotein, which is also implicated in thrombosis. Lp(a) binds to
platelets and modulates the effects of various platelet agonists. Platelet activating factor (PAF) is a potent platelet agonist degraded and
inactivated by PAF-acetylhydrolase (PAF-AH), which in plasma is associated with lipoproteins. Lp(a) is enriched in PAF-AH, thus a
functional characteristic of this lipoprotein is its capability to hydrolyze and inactivate PAF. In the present study, we investigated the effect of
Lp(a) on PAF-induced platelet activation. The potential roles of the apo(a) moiety and especially of the PAF-AH content of Lp(a) on the
above effect were also addressed. Methods: Lp(a) was isolated by affinity chromatography from plasma of apparently healthy fasting donors
with serum Lp(a) concentrations z 20 mg/dl. Reduced Lp(a) [Lp(a-)] was prepared by incubation of Lp(a) with dithiothreitol (DTT), whereas
inactivation of Lp(a)-associated PAF-AH was performed by incubation of Lp(a) with pefabloc [pefa-Lp(a)]. Platelet-rich plasma (PRP) or
washed platelets were prepared from peripheral venous blood samples of normolipidemic, apparently healthy fasting donors with their serum
Lp(a) levels lower than 0.8 mg/dl. The surface expression of the platelet integrin-receptor aIIbh3 and the fibrinogen binding to the activated
aIIbh3 was studied by flow cytometry. Results: Lp(a), at doses higher than 20 Ag/ml, inhibits PAF-induced platelet activation in a dosedependent manner. Pefa-Lp(a), lacking PAF-AH activity, exhibited a similar to Lp(a) inhibitory effect. Importantly, the Lp(a) inhibitory effect
was not influenced by the apo(a) isoform size, whereas Lp(a-) was a more potent inhibitor compared to Lp(a). Similarly to PAF, Lp(a) inhibits
platelet aggregation induced by ADP or Calcium ionophore A23187. Lp(a), pefa-Lp(a) or Lp(a-) effectively inhibited PAF- or ADP-induced
surface expression of aIIbh3, the Lp(a-) being more potent compared to Lp(a) or to pefa-Lp(a). Finally, Lp(a) significantly inhibited
fibrinogen binding to platelets activated with PAF. Conclusions: Lp(a) inhibits PAF-induced platelet activation in a non-specific manner. The
Lp(a)-associated PAF-AH does not play any important role in this effect, whereas the apo(a) moiety of Lp(a) significantly reduces its
inhibitory effect. The inhibition of aIIbh3 activation and fibrinogen binding to the activated platelets may represent the major mechanism by
which Lp(a) inhibits PAF-induced platelet aggregation.
D 2004 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
Keywords: Lipoprotein (a); Platelets; Platelet activating factor; PAF-acetylhydrolase
1. Introduction
Lipoprotein (a) [Lp(a)] is composed of a low density
lipoprotein (LDL)-like particle to which a large, highly
* Corresponding author. Tel.: +30-2651-098365; fax: +30-2651047832.
E-mail address: [email protected] (A.D. Tselepis).
glycosylated apolipoprotein (a) [apo(a)] is linked by a disulfide bond [1]. Several studies have demonstrated a significant
correlation between increased plasma levels of Lp(a) (>20 to
30 mg/dl) and coronary artery disease as well as stroke [2– 5].
Lp(a) is considered an atherogenic lipoprotein, which is also
implicated in thrombosis, although the underlyning mechanisms remain incompletely understood [6]. Early studies
showed that Lp(a) plays a potential role in thrombogenesis,
interfering with several steps in the fibrinolytic pathway
0008-6363/$ - see front matter D 2004 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.cardiores.2004.03.005
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
[7,8]. In this regard, more recent studies have demonstrated
that Lp(a) binds to platelets, although there are conflicting
reports as to which platelet protein or Lp(a) component plays
a major role in this interaction [9 –12]. Through its interaction
with platelets, Lp(a) may impair platelet-mediated fibrinolysis by reducing the binding of both plasminogen and t-PA to
the platelet surface and by increasing the Km values of t-PA
for plasminogen [13]. In addition, intact Lp(a) or its apo(a)
moiety may modulate the effects of various platelet agonists.
In this context, there are marked differences concerning the
results from various studies, which have examined the effects
of Lp(a) on platelet activation induced by various agonists.
Thus, it has been shown that intact Lp(a) does not influence
platelet activation induced by collagen or thrombin [14].
Moreover, other studies demonstrated that intact Lp(a) inhibits collagen-induced platelet aggregation [10,12,15] or ADPinduced platelet aggregation [10]. Furthermore, it has been
shown that intact Lp(a) or recombinant apo(a) promotes the
aggregation response of platelets to subaggregant doses of
arachidonic acid, whereas recombinant apo(a) does not affect
the platelet response to low doses of collagen or thrombin
[11]. Finally, intact Lp(a) or recombinant apo(a) increases
thrombin receptor-activating hexapeptide (SFLLRN,
TRAP)-induced platelet activation, whereas they do not
affect ADP- or thrombin-induced platelet activation [16].
Platelet activating factor (PAF) is a proinflammatory
phospholipid mediator and a potent platelet agonist [17].
PAF is degraded and inactivated by PAF-acetylhydrolase
(PAF-AH), an enzyme that exhibits a Ca2 +-independent
phospholipase A2 activity and in plasma, it is associated
with lipoproteins [18]. We have previously shown that Lp(a)
is enriched in PAF-AH, thus, a functional characteristic of
this lipoprotein is its capability to hydrolyze and inactivate
PAF [19]. According to our knowledge, there is a paucity of
data concerning the effect of Lp(a) on PAF-induced platelet
activation. This information could be of particular interest,
since unlike other platelet agonists, PAF-induced platelet
activation could be influenced not only through the apo(a)
moiety or the LDL-like component of Lp(a), but also via its
PAF-AH content. Thus, in the present study, we investigated
the effect of Lp(a) on PAF-induced platelet activation. The
potential roles of the apo(a) moiety and especially of the
PAF-AH content of Lp(a) on the above effects were also
addressed.
2. Methods
2.1. Isolation of Lp(a)
Lp(a) was prepared from EDTA-containing plasma
samples of apparently healthy fasting donors with serum
Lp(a) concentrations z 20 mg/dl. The investigation conforms with the principles outlined in the Declaration of
Helsinki (Cardiovascular Research 1997;35:2 – 4). Lp(a)
was isolated by affinity chromatography using a Lysine-
131
Sepharose 4B column, as previously described [20]. The
purity of isolated Lp(a) was evaluated by agarose gel
electrophoresis (Hydradel Lipo and Lp(a) kit, Sebia),
gradient SDS-PAGE in 5– 19% SDS-gradient polyacrylamide gels and by double rocket electrophoresis. Lp(a) was
extensively dialyzed against 10 mM PBS containing 0.01%
EDTA, pH 7.4, and stored at 4 jC under nitrogen for up to
2 weeks. Before use in experiments with platelets, Lp(a)
was dialyzed overnight in 10 mM PBS to remove EDTA.
No oxidation was observed in each Lp(a) preparation, as it
was revealed by agarose gel electrophoresis and the determination of the TBARS values (0.9 F 0.4 malondialdehyde
equivalents per mg of protein for all lipoprotein preparations). In some experiments, we pre-incubated Lp(a) with 1
mM pefabloc for 30 min at 37 jC, a procedure that
completely and irreversibly inactivates the endogenous
PAF-AH [21]. The pefabloc-treated Lp(a) [pefa-Lp(a)]
was stored as above and it was extensively dialyzed in
10 mM PBS before use.
2.2. Reductive cleavage of Lp(a)
Lp(a) containing 600 Ag protein/ml HEPES was submitted to reductive cleavage by incubation with 10 mM
dithiothreitol (DTT) at 37 jC for 3 h [22], a procedure that
completely dissociates apo(a) from Lp(a) [23]. Reduced
Lp(a) [Lp(a-)] was then fractionated by density gradient
ultracentifugation as previously described [19]. After ultracentrifugation, 28 fractions of 400 Al were collected and
analyzed for their content in cholesterol and PAF-AH
activity [19]. Subsequently, fractions 11 – 15 containing
Lp(a-) were pooled, and used in platelet activation studies
after an overnight dialysis in 10 mM PBS. The Lp(a-)
preparations were characterized by agarose gel electrophoresis and by 3.75% SDS-PAGE followed by immunoblotting as described previously [19]. In some experiments,
unreduced Lp(a) [(Lp(a) incubated in the absence of DTT]
was also submitted to ultracentrifugation and used as a
control.
2.3. Preparation of platelet-rich plasma
Platelet-rich plasma (PRP) was prepared from peripheral
venous blood samples of normolipidemic, apparently
healthy fasting donors with serum Lp(a) levels lower than
0.8 mg/dl, as previously described [24]. ACD solution,
consisting of 0.07 M citric acid, 0.11 M sodium citrate
and 0.11 M D(+) glucose was used as an anticoagulant. The
platelet count of PRP was adjusted to a final platelet
concentration of 2.5 108/ml with homologous plasma.
2.4. Preparation of washed platelets
Peripheral venous blood samples were collected from the
same donors using ACD as an anticoagulant. Washed
platelets were prepared over an erythrocyte pellet as previ-
132
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
ously described by Lotner et al. [25] with some modifications. Briefly, the anti-coagulated blood was centrifuged at
3900 g for 5 min at room temperature to remove plasma.
The pelleted cells were then resuspended in Tyrode’s pH 7.0
and washed twice using identical centrifugations at 650 g
for 10 min. The pelleted cells were then resuspended in
Tyrode’s buffer pH 7.4 and centrifuged at 120 g for 10
min at room temperature to obtain the washed platelet
suspension. The platelet count was adjusted to a final
platelet concentration of 2.5 108/ml.
under our experimental conditions. Platelets were then diluted with PBS (1:5 v/v) and immediately analyzed by flow
cytometry using a FACSCalibur flow cytometer (BectonDickinson) [28]. To quantify non-specific binding, control
samples containing FITC-conjugated IgG were also assayed. In some experiments, the effect of Lp(a) or Lp(a-)
on ADP-induced PAC-1 binding to platelets in PRP was
studied with the same method, using 100 AM ADP final
concentration [28]. Finally, in control experiments, the
peptide RGDS (1 mM final concentration), instead of
Lp(a), was added to platelets prior to activation.
2.5. Aggregation studies
2.8. Effect of Lp(a) on fibrinogen binding
Platelet aggregation studies in either PRP or washed
platelets were performed in aliquots of 0.5 ml, in a platelet
aggregometer (model 560, Chronolog) at 37 jC, with
continuous stirring at 1200 rpm. Platelets were pre-incubated at 37 jC with the various Lp(a) preparations for several
time intervals up to 30 min before the initiation of aggregation. The maximal aggregation, achieved within 3 min
after the addition of the agonist, was determined and
expressed as a percentage of 100% light transmission
calibrated for each specimen (maximal percentage of aggregation) [24]. In some experiments with PRP, platelets were
pre-incubated with aspirin (0.1 mM) for 15 min before the
addition of Lp(a). In these experiments, a combination of
creatine phosphate (CP) and creatine kinase (CPK) (4 mM
and 40 U ml 1, respectively) was added 1 min before PAFinduced aggregation [24]. In experiments with washed
platelets, fibrinogen (0.5 mg/ml) was added to the samples,
30 s before addition of the agonist.
2.6. Secretion studies
The effect of Lp(a) on PAF-induced ATP (a platelet
dense granule component) secretion was studied. ATP
secretion was determined in parallel to platelet aggregation
in aliquots of 0.45 ml PRP along with 50 Al of Luciferin/
Luciferase (to measure ATP secretion, Chrono-lume
reagent, Chronolog) [26], in a platelet aggregometer (model 560, Chronolog) at 37 jC, with continuous stirring at
1200 rpm.
2.7. Effect of Lp(a) on the expression of the platelet integrin
aIIbb3
The expression of the platelet integrin-receptor aIIbh3
was studied by flow cytometry in washed platelets or PRP
using the FITC-labeled specific monoclonal antibody PAC1 (Becton Dickinson, San Jose, CA) as previously described
[27]. Briefly, platelets (2.5 108/ml) were incubated with
30 ng/ml of PAC-1 in the absence or in the presence of
Lp(a), pefa-Lp(a) or Lp(a-) for 10 min prior to activation
with PAF (80 nM final concentration). Activation was performed for 10 min at 37 jC without stirring. This incubation
time is necessary to obtain the maximum PAC-1 binding
The effect of Lp(a) on the binding of FITC-labelled
fibrinogen (FITC-Fg) to platelets in PRP was studied by
flow cytometry, as previously described [28]. Fluorescein
labelling of fibrinogen was performed as previously described [29]. PRP with platelet number ranging from
2.5 108/ml to 4.5 108/ml was diluted 10-fold with
Walsh-albumin buffer. Diluted PRP was then mixed with
FITC-Fg (500 nM final concentration), in the presence or
absence of Lp(a). Platelet activation was performed with 80
nM PAF at room temperature for 60 min in the dark without
stirring. This incubation time is necessary to obtain the
maximum FITC-Fg binding under our experimental conditions. Then platelets were immediately analyzed by flow
cytometry, using 10,000 cell events. The mean fluorescence
intensity (MFI) values for both the non-activated and
activated platelets, in the presence or absence of Lp(a),
were calculated. The MFI values of non-activated platelets,
in the presence or absence of Lp(a) (non-specific binding),
were subtracted from those obtained after platelet activation
(total binding), respectively, thus obtaining the specific
binding of FITC-Fg [30]. To quantify non-specific binding,
control samples containing FITC-conjugated albumin were
also assayed. The effect of 1 mM RGDS on FITC-Fg
binding to activated platelets was also studied using the
same procedure. Numeric data were processed with the
Cellquest software (Becton-Dickinson).
2.9. Biochemical determinations
PAF-AH activity in Lp(a) preparations was measured by
the trichloroacetic acid precipitation procedure using 1-Ohexadecyl-2-[3H-acetyl]-sn-glycero-3-phosphocholine [3H]PAF (10 Ci/mmol; DuPont-New England Nuclear, Boston,
MA, USA) as a substrate and 5 Ag protein of Lp(a) as the
source of the enzyme [31]. Plasma Lp(a) levels were
measured by a sandwich enzyme immunoassay method
(Macra Lp(a), Terumo Medical, Elkton, MD, USA) [32].
The apo(a) isoform size of the various Lp(a) preparations
was determined in total plasma prior to the isolation of
Lp(a), by agarose gel electrophoresis and expressed as
number of K4 repeats, as previously described [4]. The
cholesterol content of Lp(a) was analyzed by commercially
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
available enzymatic reagent (BioMerieux), whereas the
Lp(a) protein content was measured by the bicinchoninic
acid (BCA) method (PIERCE).
2.10. Statistical analysis
Results are expressed as mean F S.D. and were compared using analysis of variance (ANOVA) followed by the
least significant difference test (LSD). Statistically significant values were defined at a value of p < 0.05.
3. Results
3.1. Effect of Lp(a) on PAF-induced platelet aggregation in
PRP
The effect of Lp(a) on PAF-induced platelet aggregation
was first studied in PRP from normolipidemic, apparently
healthy volunteers in which Lp(a) levels were lower than 0.8
mg/dl. The PRP was initially incubated in the absence or in
the presence of various Lp(a) concentrations (10 to 80 Ag
protein/ml) for 1 min at 37 jC and then platelet aggregation
was initiated by 25 nM PAF [a concentration that induces an
133
irreversible platelet aggregation with a maximal percentage
of aggregation of about 80%, in the absence of Lp(a)]. As
shown in Fig. 1A, Lp(a) inhibits PAF-induced platelet aggregation in a dose-dependent manner. Similar results were
obtained when PRP was pre-incubated with Lp(a) for 5, 10 or
30 min before the initiation of the aggregation (data not
shown). Typical aggregation curves illustrating the dosedependent inhibitory effect of Lp(a) are illustrated in Fig.
1B. An inhibition by Lp(a) of the secondary aggregation,
which is due to the effect of other agonists secreted from
activated platelets, is also evident in this figure. Thus, in order
to determine the specific effect of Lp(a) on PAF-induced
platelet aggregation, with respect to the primary aggregation,
we pre-incubated PRP with CP/CPK (a scavenger of ADP
produced during platelet activation) as well as with aspirin,
which inhibits the agonist-induced formation of thromboxane
A2 [33]. Activation was initiated with 25 nM PAF, a concentration, which induced a maximal percentage of aggregation
of about 40% in the presence of aspirin and CP/CPK. All
aggregation curves were monophasic and reversible. Under
these conditions, Lp(a) inhibited platelet aggregation in a
dose-dependent manner (Fig. 1A). Typical aggregation
curves are illustrated in Fig. 1C. One of the platelet components secreted during PAF-induced platelet aggregation is
Fig. 1. Effect of Lp(a) on PAF-induced platelet aggregation in PRP. (A) Bar-graph showing the % inhibition of platelet aggregation in the absence (open
bars) or in the presence of aspirin and CP/CPK (closed bars). Values represent the mean F S.D., from four different platelet and Lp(a) preparations. (B)
Representative aggregation curves illustrating the dose-dependent inhibitory effect of Lp(a) on platelet aggregation induced by 25 nM PAF. (C)
Representative aggregation curves illustrating the dose-dependent inhibitory effect of Lp(a) on platelet aggregation induced by 25 nM PAF, in the presence
of aspirin and CP/CPK.
134
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
Fig. 2. Effect of Lp(a) on ADP-induced platelet aggregation in PRP.
Platelets in PRP were incubated with various concentrations of Lp(a) in 37
jC for 1 min under stirring and then platelet aggregation was initiated with
the addition of 5 AM ADP. Values represent the mean F S.D. from four
different platelet and Lp(a) preparations.
ADP and this agonist significantly contributes to the platelet
secondary aggregation. Thus, we next studied the effect of
Lp(a) on ADP-induced platelet aggregation. Platelet aggregation was initiated by 5 AM ADP, a concentration that
induced a maximum percentage of aggregation of about
70%. All aggregation curves were irreversible. As shown in
Fig. 2, Lp(a) significantly inhibits ADP-induced platelet
aggregation in a dose-dependent manner. Similar results were
obtained when PRP was pre-incubated with Lp(a) for 5, 10 or
30 min before the initiation of the aggregation (data not
shown). Finally, we studied the effect of Lp(a) on platelet
aggregation induced by 5 AM Calcium ionophore A23187, a
concentration that induces an irreversible platelet aggregation
with a maximum percentage of aggregation of about 50%.
Lp(a), at a concentration of 40 Ag protein/ml, inhibited
platelet aggregation by 61 F 12%.
washed platelets, i.e. in the absence of plasma which is
enriched in PAF-AH activity, mainly associated with LDL
[18]. In these experiments, we used Lp(a) pre-incubated
with pefabloc [pefa-Lp(a)], a treatment that completely and
irreversibly inhibits the endogenous PAF-AH activity [21].
PAF, at a concentration of 10 nM, induced a maximal
percentage of aggregation of about 50%. Under these
conditions, Lp(a) (containing PAF-AH activity of 13 F 5
nmol/mg protein/min) at a concentration higher than 20 Ag
protein/ml, significantly inhibited PAF-induced washed
platelet aggregation. Surprisingly, pefa-Lp(a), lacking
PAF-AH activity, exhibited a similar inhibitory effect to
that of Lp(a) with active PAF-AH (Fig. 3).
We next asked whether the apo(a) moiety of Lp(a) could
play any role in the inhibitory effect of Lp(a). Initially, we
investigated whether the length of the apo(a) influences the
Lp(a) inhibitory effect. In these experiments, we used Lp(a)
preparations containing low V 21K4 or high z 27K4 apo(a)
size. No significant differences were observed between Lp(a)
preparations containing low or high apo(a) isoform size (data
not shown). Subsequently, we treated Lp(a) with DTT, a
procedure that completely removes apo(a) from Lp(a), and
then purified reduced Lp(a) by ultracentrifugation [19]. This
treatment did not significantly influence the endogenous
PAF-AH activity. The effect of such Lp(a) preparations
[Lp(a-)] was studied on PAF-induced washed platelet aggregation in comparison to that of unreduced Lp(a). As shown in
Fig. 3, both Lp(a) preparations significantly inhibited platelet
aggregation in a dose-dependent manner; however, Lp(a-)
was a more potent inhibitor compared to Lp(a).
3.2. Effect of Lp(a) on PAF-induced ATP secretion
To further investigate the inhibitory effect of Lp(a) on the
PAF-induced secondary wave of aggregation, we studied the
effect of Lp(a) on ATP secretion (a marker of dense granule
secretion) from activated platelets. PAF at a concentration of
25 nM induced a secretion of 1.9 F 0.4 nmol ATP. Lp(a) at
doses higher than 20 Ag/ml significantly inhibited ATP
secretion, at a dose-dependent manner. This inhibition
reached the 100% at an Lp(a) concentration of 80 Ag
protein/ml.
3.3. Effect of Lp(a) on PAF-induced washed platelet
aggregation
In order to investigate the role of the Lp(a)-associated
PAF-AH on the inhibitory effect of this lipoprotein on PAFinduced platelet activation, we performed studies with
Fig. 3. Role of endogenous PAF-AH and apo(a) on Lp(a)-mediated
inhibition of PAF-induced platelet aggregation. Washed platelets (2.5 108/
ml) were incubated with Lp(a) (expressing PAF-AH activity of 13 F 5
nmol/mg protein/min) (open bars) or pefa-Lp(a) (inactive PAF-AH) (gray
bars) or Lp(a-) (closed bars) at 37 jC for 1 min under stirring. Fibrinogen,
0.5 mg/ml was added 30 s prior to aggregation with 10 nM PAF. Values
represent the mean F S.D. from three different experiments. *p < 0.03
compared to either Lp(a) or pefa-Lp(a).
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
135
as well as the role of the PAF-AH content and the apo(a)
moiety, we studied the effect of Lp(a), pefa-Lp(a) and
Lp(a-) on the binding of the specific monoclonal antibody PAC-1, which recognizes the activated form of the
platelet integrin receptor aIIbh3. In these experiments, the
above Lp(a) preparations were used at concentrations of
40 and 80 Ag protein/ml. As shown in Fig. 4A, all forms
of Lp(a) studied, at both concentrations, effectively
Fig. 4. Role of endogenous PAF-AH and apo(a) on Lp(a)-mediated
inhibition of PAF-induced activation of aIIbh3. (A) Bar-graph showing the
%inhibition of PAC-1 binding in the presence of Lp(a) (expressing PAF-AH
activity of 13 F 5 nmol/mg protein/min) (open bars) or pefa-Lp(a) (inactive
PAF-AH) (gray bars) or Lp(a-) (closed bars). Values represent the
mean F S.D. from three different experiments. *p < 0.04 compared to
either Lp(a) or pefa-Lp(a). (B, C) Representative histograms, obtained by
FACS analysis, illustrating the effect of 80 Ag protein/ml of either Lp(a) (B)
or Lp(a-) (C) on PAC-1 binding on activated platelets.
3.4. Effect of Lp(a) on PAF-induced PAC-1 binding
To further investigate the mechanism of Lp(a)-induced
inhibition of platelet aggregation in the presence of PAF,
Fig. 5. Lp(a) mediated inhibition of PAF-induced FITC-Fg binding. (A)
Bar-graph showing the %inhibition of FITC-Fg binding in the presence of
Lp(a) (open bars or 1 mM RGDS (black bars)). Values represent the
mean F S.D. from three different experiments. (B, C) Representative
histograms, obtained by FACS analysis, illustrating the effect of Lp(a) (40
Ag protein/ml) (B), or RGDS (C) on FITC-Fg binding to platelets activated
with 80 nM PAF.
136
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
inhibited the PAC-1 binding to activated platelets in the
presence of 80 nM PAF, the Lp(a-) being more potent
compared either to Lp(a) or to pefa-Lp(a), a finding
which accords to the results from the aggregation studies.
No difference in their inhibitory potency was observed
between Lp(a) and pefa-Lp(a) (Fig. 4A). Representative
flow cytometry curves illustrating the effect of 80 Ag
protein/ml Lp(a) or Lp(a-) on PAC-1 binding are shown
in Fig. 4B,C. Similarly to PAF, Lp(a) and Lp(a-), at
concentrations of 40 and 80 Ag protein/ml, significantly
inhibited ADP-induced PAC-1 binding, the Lp(a-) being
more potent than Lp(a) (24 F 9% versus 15 F 6% inhibition at 40 Ag protein/ml and 68 F 12% versus 43 F 14%
inhibition at 80 Ag protein/ml, p < 0.03 for both comparisons). Finally, as expected, the RGDS tetrapeptide at a
concentration of 1 mM completely inhibited PAC-1
binding induced by either PAF or ADP.
3.5. Effect of Lp(a) on PAF-induced fibrinogen binding
To investigate whether the inhibition of PAF-induced
aIIbh3 expression and activation (PAC-1 binding) by
Lp(a) influences the association of fibrinogen with the
receptor, we performed binding studies using FITC-Fg.
As shown in Fig. 5A, Lp(a) at a concentration of 40 and
80 Ag protein/ml, significantly inhibited FITC-Fg binding
to platelets activated with 80 nM PAF. Representative
flow cytometry curves illustrating the effect of Lp(a) at
40 Ag protein/ml, on PAF-induced FITC-Fg binding are
shown in Fig. 5B. Finally, as expected, the PAF-induced
FITC-Fg binding was completely inhibited by 1 mM
RGDS (Fig. 5A,C).
4. Discussion
The results of the present study show for the first time
that Lp(a) inhibits PAF-induced platelet activation in a dosedependent manner. Importantly, even in the presence of all
plasma components including lipoproteins (i.e. in PRP),
Lp(a) at a concentration higher than 20 Ag protein/ml, which
corresponds to plasma Lp(a) concentration >8 mg/dl, significantly inhibits PAF-induced platelet activation. A similar
inhibitory effect of Lp(a) was observed in washed platelets,
an indication that the plasma components do not significantly influence the Lp(a)-mediated inhibition of platelet
aggregation.
An important observation of the present study is that
the PAF-AH component of Lp(a) did not significantly
contribute to the inhibitory effect of Lp(a) on PAFinduced platelet activation, thus suggesting that in plasma, which is enriched in PAF-AH activity mainly associated with LDL [31], the Lp(a)-PAF-AH may not play
any significant role in the Lp(a)-mediated inhibition of
platelet aggregation induced by PAF. However, we cannot
exclude the possibility that this enzyme may significantly
influence other Lp(a) functions, which implicate this
lipoprotein in atherothrombosis. Thus, the pathophysiological role of the Lp(a)-associated PAF-AH remains to
be established.
According to our results, Lp(a) inhibits both the
primary and the secondary aggregation as well as the
secretion of dense granules (measured by determining
their ATP content) induced by PAF. A content of the
platelet dense granules and one of the endogenous
platelet agonists that significantly contributes to the
secondary aggregation induced by PAF is ADP [35].
Lp(a) inhibited ADP-induced platelet aggregation; consequently, the inhibition of the PAF-induced secondary
aggregation by Lp(a) can be at least partially attributed
both to the inhibition of ADP secretion and ADP-induced
platelet aggregation.
The inhibition of platelet aggregation induced by
different agonists (i.e. PAF, ADP and Calcium ionophore
A23187) by Lp(a) suggests that it may not be a specific
effect of this lipoprotein towards a certain agonist but
rather that Lp(a) influences a common step through
which all agonists lead to platelet aggregation. This
common step concerns the surface expression and conformational change of the platelet integrin receptor aIIbh3,
which is observed during platelet activation induced by
any agonist [34]. The aIIbh3, in its activation state,
expresses several binding sites for fibrinogen, which
binds to the receptor, thus leading to platelet aggregation
[34]. Our results showed that Lp(a) inhibits both the
PAF- and the ADP-induced activation of aIIbh3, measured
by the capacity of the receptor to bind PAC-1, which
recognizes only the activated form of this integrin [36].
In addition, Lp(a) significantly inhibited the binding of
fibrinogen to activated platelets. These results suggest
that Lp(a) inhibits platelet aggregation by influencing
the expression of aIIbh3 and fibrinogen binding to the
activated receptor. Several studies over the last years have
shown that Lp(a) interacts with human platelets at their
resting state in a specific, saturable and reversible manner, although a specific Lp(a) receptor on platelets has
not been described yet [9,10,12,13]. In this context, it has
been suggested that Lp(a) binds to the aIIbh3 subunit of
the aIIbh3 in its resting state. Thus, this subunit may
represent the major Lp(a) binding protein on platelets [9],
although other studies showed that aIIbh3 in its resting
state may not be implicated in Lp(a) binding [10,13].
Taking into account the above results, we may suggest
that binding of Lp(a) to resting platelets results in the
inhibition (through a yet unidentified mechanism) of
aIIbh3 surface expression and activation induced by any
agonist, thus leading to the inhibition of both fibrinogen
binding and platelet aggregation. Alternatively, in addition
to its association with the resting platelets, Lp(a) may
bind to the activated form of aIIbh3, thus inhibiting the
binding of PAC-1 and fibrinogen. This hypothesis is
currently under investigation in our laboratory.
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
An important observation of the present study is that
removal of apo(a) from Lp(a) significantly enhances the
antiaggregatory effect of Lp(a) as well as its inhibitory
effect on PAC-1 binding to activated platelets. Previous
studies have demonstrated that apo(a) is not necessary for
the interaction of Lp(a) with resting platelets (9). Furthermore, Lp(a-) can also bind to platelets although with
lower affinity [9,10]. However, it has been shown that
recombinant apo(a) is also capable of binding to resting
platelets, in an aIIbh3-independent manner, implying that
another unidentified yet platelet protein may be involved
in this process [11]. The interaction of apo(a) with
platelets enhances their response to arachidonic acid or
SFLLRN [11,16] thus, in contrast to Lp(a), its apo(a)
moiety may play a proaggregant role. This suggests that
Lp(a) and apo(a) may modulate agonist-induced platelet
activation, interacting with distinct binding sites on the
platelet surface. In accordance to that, it has been shown
that lysine residues on apo(a) may play an important role
in the apo(a)-platelet interaction [11], whereas it was
recently suggested that the lysine residues of apo(a)
may not mediate the Lp(a)-induced inhibition of platelet
aggregation [12]. Based on our results as well as on the
results of previous studies, we may suggest that the
inhibitory effect of Lp(a) on PAF-induced platelet aggregation is primarily due to its Lp(a-) moiety, since it
exhibits a more potent inhibitory effect compared to
Lp(a). Overall, we may speculate that Lp(a) interacts
with distinct binding sites on resting platelets, i.e. through
its apo(a) moiety and through its LDL-like particle [Lp(a)]. The binding of Lp(a) through its Lp(a-) moiety leads
to the inhibition of platelet activation, whereas its binding
through the apo(a) moiety may enhance agonist-induced
platelet activation. Thus, the effect of Lp(a) on PAF- or
ADP-induced platelet aggregation, described in the present study, could be the result of the Lp(a-) inhibitory
effect and the apo(a) stimulatory effect. This hypothesis
may also explain the controversial results published in
the literature concerning the effects of Lp(a) or apo(a) on
platelet aggregation.
In conclusion, the present study demonstrates that
Lp(a) inhibits PAF-induced platelet activation in a nonspecific manner. The Lp(a)-associated PAF-AH does not
play any important role in this effect, whereas the apo(a)
moiety of Lp(a) significantly reduces its inhibitory effect
in an apo(a) isoform size-independent manner. We suggest
that the inhibition of aIIbh3, activation and fibrinogen
binding to the activated platelets may represent the major
mechanism by which Lp(a) inhibits PAF-induced platelet
aggregation.
References
[1] Utermann G, Weber W. Protein composition of Lp(a) lipoprotein from
human plasma. FEBS Lett 1983;154:357 – 61.
137
[2] Danesh J, Collins R, Peto R. Lipoprotein (a) and coronary heart
disease. Meta-analysis of prospective studies. Circulation 2000;102:
1082 – 5.
[3] Seed M, Ayres KL, Humphries SE, Miller GJ. Lipoprotein (a) as a
predictor of myocardial infarction in middle-aged men. Am J Med
2001;110:22 – 7.
[4] Katsouras CS, Karabina SA, Tambaki AP, et al. Serum lipoprotein(a)
concentrations and apolipoprotein(a) isoforms: association with the
severity of clinical presentation in patients with coronary heart disease. J Cardiovasc Risk 2001;8:311 – 7.
[5] Zenker G, Koltringer P, Bone G, Niedorkorn K, Pfeiffer K, Jurgens G.
Lipoprotein (a) as a strong indicator for cerebrovascular disease.
Stroke 1986;17:942 – 5.
[6] Chapman MJ, Huby T, Nigon F, Thillet J. Lipoprotein (a): implication
in atherothrombosis. Atherosclerosis 1994;110:S69 – 75.
[7] Hajjar KA, Gavish D, Breslow JL, Nachman RL. Lipoprotein(a)
modulation of endothelial cell surface fibrinolysis and its potential
role in atherosclerosis. Nature 1989;339:303 – 5.
[8] Harpel PC, Gordon BR, Parker TS. Plasmin catalyzes binding of
lipoprotein (a) to immobilized fibrinogen and fibrin. Proc Natl Acad
Sci U S A 1989;86:3847 – 51.
[9] Malle E, Ibovnik A, Steinmetz A, Kostner GM, Sattler W. Identification of glycoprotein IIb as the lipoprotein (a)-binding protein on
platelets. Lipoprotein (a) binding is independent of an arginyl-glycylaspartate tripeptide located in apolipoprotein (a). Arterioscler Thromb
1994;14:345 – 52.
[10] Pedreno J, Fernandez R, Cullare C, Barcelo A, Elorza MA, de Castellarnau C. Platelet integrin alpha IIb beta 3 (GPIIb – IIIa) is not
implicated in the binding of LDL to intact resting platelets. Arterioscler Thromb Vasc Biol 1997;17:156 – 63.
[11] Martinez C, Rivera J, Loyau S, et al. Binding of recombinant apolipoprotein (a) to human platelets and effect on platelet aggregation.
Thromb Haemost 2001;85:686 – 93.
[12] Barre D. Human lipoprotein (a)-induced reduction of platelet aggregation is not mediated by apolipoprotein (a)’s lysine-binding regions.
Front Biosci 2003;8:1226 – 8.
[13] Ezratty A, Simon DI, Loscalzo J. Lipoprotein(a) binds to human
platelets and attenuates plasminogen binding and activation. Biochemistry 1993;32:4628 – 33.
[14] Malle E, Ibovnik A, Leis HJ, Kostner GM, Verhallen PFJ, Sattler W.
Lysine modification of LDL or lipoprotein (a) by 4-hydroxynonenal
of malondialdehyde decreases platelet serotonin secretion without
affecting platelet aggregability and eicosanoid formation. Arterioscler
Thromb Vasc Biol 1995;15:377 – 84.
[15] Gries A, Gries M, Wurm H, et al. Lipoprotein(a) inhibits collageninduced aggregation of thrombocytes. Arterioscler Thromb Vasc Biol
1996;16:648 – 55.
[16] Rand ML, Sangrar W, Hancock MA, et al. Apolipoprotein(a) enhances platelet responses to the thrombin receptor-activating peptide
SFLLRN. Arterioscler Thromb Vasc Biol 1998;18:1393 – 9.
[17] Zimmerman GA, McIntyre TM, Prescott SM, Stafforini DM. The
platelet-activating factor signaling system and its regulators in syndromes of inflammation and thrombosis. Crit Care Med 2002;
30(Suppl. 5):S294 – 301.
[18] Tselepis AD, Chapman MJ. Inflammation, bioactive lipids and atherosclerosis: potential roles of a lipoprotein-associated phospholipase
A2, platelet activating factor-acetylhydrolase. Atherosclerosis 2002;
30(Suppl. 3):57 – 68.
[19] Karabina SA, Elisaf MC, Goudevenos J, Siamopoulos KC, Sideris D,
Tselepis AD. PAF-acetylhydrolase activity of Lp(a) before and during
Cu(2+)-induced oxidative modification in vitro. Atherosclerosis
1996;125:121 – 34.
[20] Doucet C, Huby T, Ruiz J, Chapman MJ, Thillet J. Non-enzymatic
glycation of lipoprotein(a) in vitro and in vivo. Atherosclerosis
1995;118:135 – 43.
[21] Dentan C, Tselepis AD, Chapman MJ, Ninio E. Pefabloc, 4-[2-aminoethyl]benzenesulfonyl fluoride, is a new, potent nontoxic and irrevers-
138
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
L.D. Tsironis et al. / Cardiovascular Research 63 (2004) 130–138
ible inhibitor of PAF-degrading acetylhydrolase. Biochim Biophys
Acta 1996;1299:353 – 7.
Armstrong VW, Walli AK, Seidel D. Isolation, characterization, and
uptake in human fibroblasts of an apo(a)-free lipoprotein obtained on
reduction of lipoprotein(a). J Lipid Res 1985;26:1314 – 23.
Scanu AM, Pfaffinger D, Fless GM, Makino K, Eisenbart J, Hinman
J. Attenuation of immunologic reactivity of lipoprotein(a) by thiols
and cysteine-containing compounds. Structural implications. Arterioscler Thromb 1992;12:424 – 9.
Goudevenos J, Tselepis AD, Tsoukatos D, Grekas G, Kritikakos J,
Sideris D. Platelet aggregability to platelet activating factor at rest and
after exercise in patients with coronary artery disease. Eur Heart J
1995;16:1036 – 43.
Lotner GZ, Lynch JM, Betz SJ, Henson PM. Human neutrophil-derived platelet activating factor. J Immunol 1980;124:676 – 84.
Taylor ML, Misso NL, Stewart GA, Thompson PJ. The effects of
varying doses of aspirin on human platelet activation induced by
PAF, collagen and arachidonic acid. Br J Clin Pharmacol 1992;33:
25 – 31.
Golden A, Brugge JS, Shattil SJ. Role of platelet membrane glycoprotein IIb – IIIa in agonist-induced tyrosine phosphorylation of platelet proteins. J Cell Biol 1990;111:3117 – 27.
Biris N, Abatzis M, Mitsios JV, et al. Mapping the binding domains of
the alpha(IIb) subunit. A study performed on the activated form of the
platelet integrin alpha(IIb)beta(3). Eur J Biochem 2003;270:3760 – 7.
Xia Z, Wong T, Liu Q, Kasirer-Friede A, Brown E, Frojmovic MM.
Optimally functional fluorescein isothiocyanate-labelled fibrinogen
[30]
[31]
[32]
[33]
[34]
[35]
[36]
for quantitative studies of binding to activated platelets and platelet
aggregation. Br J Haematol 1996;93:204 – 14.
Frojmovic MM, Mooney RF, Wong T. Dynamics of platelet glycoprotein IIb – IIIa receptor expression and fibrinogen binding: I. Quantal activation of platelet subpopulations varies with adenosine
diphosphate concentration. Biophys J 1994;67:2060 – 8.
Tselepis AD, Dentan C, Karabina SA, Chapman MJ, Ninio E. PAFdegrading acetylhydrolase is preferentially associated with dense LDL
and VHDL-1 in human plasma. Catalytic characteristics and relation
to the monocyte-derived enzyme. Arterioscler Thromb Vasc Biol
1995;15:1764 – 73.
Labeur C, Michiels G, Bury J, Usher DC, Rosseneu M. Lipoprotein(a) quantified by an enzyme-linked immunosorbent assay with
monoclonal antibodies. Clin Chem 1989;35:1380 – 4.
Cazenave JP, Benveniste J, Mustard JF. Aggregation of rabbit platelets by platelet-activating factor is independent of the release reaction
and the arachidonate pathway and inhibited by membrane-active
drugs. Lab Invest 1979;41:275 – 85.
Phillips DR, Charo IF, Parise LV, Fitzgerald LA. The platelet membrane glycoprotein IIb – IIIa complex. Blood 1988;71:831 – 43.
Kloprogge E, De Haas GH, Gorter G, Akkerman JWN. Stimulus –
response coupling in human platelets. Evidence against a role of PAFacether in the third pathway. Thromb Res 1983;30:107 – 12.
Taub R, Gould RJ, Garsky VM, et al. A monoclonal antibody against
the platelet fibrinogen receptor contains a sequence that mimics a
receptor recognition domain in fibrinogen. J Biol Chem 1989;264:
259 – 65.