Download Platelet Adhesion to Exposed Endothelial Cell Extracellular Matrixes

436
Platelet Adhesion to Exposed Endothelial Cell
Extracellular Matrixes Is Influenced by the
Method of Preparation
J. Aznar-Salatti, E. Bastida, T.A. Haas, G. Escolar, A. Ordinas,
P.H.G. de Groot, and M.R. Buchanan
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
The relative thrombogenicity of extracellular matrixes (ECMs) produced by cultured human
umbilical endothelial cells (ECs) was studied underflowconditions. ECMs were prepared using
a number of physical and chemical methods, and their reactivity toward platelets was
morphometrically evaluated, von Willebrand factor (vWF), fibronectin (FN), and 13-hydroxy9-cis,ll-<ro/is-octadecadienoic acid (13-HODE) were also determined. We found that platelet
adhesion to ECMs differed significantly, both quantitatively and qualitatively, with the method
of ECM preparation. Mechanically prepared ECM exposed a less thrombogenic surface
compared with ECM prepared by chemical methods (platelet-covered surface of 20% and 50%,
respectively). Evaluation of the ECM components vWF, FN, and 13-HODE showed significant
changes, both in their concentrations and distribution patterns, depending on the method of
ECM preparation. The decrease measured in the levels of ECM-associated vWF (from 108 to
9.2 ng/104 cells) and the minor changes observed in the distribution pattern of subendothelial
FN did not appear to be sufficient to explain the altered platelet adhesion observed in our
model. This suggests that the amount of 13-HODE probably associated to the remaining ECs
present in the mechanically exposed ECM could be one factor that specifically contributed to
the nonthrombogenic state of these preparations. We conclude that the degree of ECM
reactivity toward platelets is dependent on the method of ECM preparation and that this is
related to the removal of specific EC/ECM components that modulate their thromboresistant/
thrombogenic properties. This fact should be taken into account when ECMs produced by
cultured ECs are used in platelet adhesion studies. {Arteriosclerosis and Thrombosis
1991;ll:436-442)
H
ealthy endothelium is nonreactive to circulating blood cells, but on removal of the
endothelial cell (EC) layer, the exposed
extracellular matrix (ECM) is highly reactive to
platelets. The interaction of platelets with this exposed surface is of critical importance for primary
hemostasis.12 The nonreactivity of the endothelium
is thought to be mediated by components released by
the EC,3"5 while the reactivity of the ECM seems to
be contributed to by different types of collagen, von
From the Servicio de Hemoterapia y Hemostasia (J.A.-S.,
E.B., G.E., A.O.), Hospital Clinic i Provincial de Barcelona,
Barcelona, Spain; Department of Pathology (T.A.H., M.R.B.),
McMaster University, Hamilton, Canada; and Department of
Hematology (P.H.G.d.G.), University Hospital Utrecht, Utrecht,
The Netherlands.
Supported in part by grants from FIS #88/0163, #T1013,
#ME-9656, CAYCIT #893/84, and CIRIT (Teenies, 89).
Address for reprints: Dr. E. Bastida, Servicio de Hemoterapia y
Hemostasia, Hospital Clinic i Provincial, Villarroel 170, Barcelona
08036, Spain.
Received May 17, 1989; revision accepted November 8, 1990.
Willebrand factor (vWF), fibronectin (FN), and
other adhesive proteins.6"8
Recently, it has been reported that ECs metabolize
linoleic acid via the lipoxygenase enzyme to 13hydroxy-9-c«,ll-frarw-octadecadienoic acid (13HODE) under basal conditions, and it has been
postulated that 13-HODE facilitates the thromboresistant property of the endothelium by regulating the
expression of adhesive moieties on the surface of the
ECs.9 It has also been reported that 13-HODE
renders the underlying basement membrane nonreactive to circulating platelets.10 These latter observations are in contrast to other studies, in which the
ECM was found to be highly reactive to circulating
platelets and 13-HODE was considered not to be a
major metabolite involved in platelet-subendothelium interactions.1112
This apparent inconsistency raises the possibility
that the different subendothelial structures are not
equally reactive to platelets and that the thromboresistance of the ECM may vary depending on the
Aznar-Salatti et al Extracellular Matrix/Platelet Interactions
severity or nature of EC removal. In this study, the
reactivity to platelets of ECM prepared by different
methods was measured under flow conditions. Furthermore, the relation between the thromboreactivity
of the ECM and the amounts and distribution patterns
of vWF, FN, and 13-HODE present in the ECM were
examined. Our results indicate that platelet adhesion
to ECM is markedly influenced by the method of EC
removal and subsequent ECM exposure.
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Methods
Blood Collection and Platelet-Free
Reconstituted Perfusate
Blood obtained from healthy volunteers who had
not ingested aspirin in the previous 15 days was
collected into citric acid/citrate/dextrose-containing
sterile tubes (20 mM) for the perfusion experiments.
Other blood samples were processed to prepare the
perfusates for exposure of the ECM to shear stress.
The blood was centrifuged (150g, 10 minutes, 22°C)
to remove platelet-rich plasma. Platelet-free plasma
was prepared by centrifugation (l,300g, 20 minutes,
22°C) and stored at 4°C until use. The remaining red
blood cells were washed three times in 0.9% NaCl
containing 10 mM a-D-glucose. Platelet-free plasma
and red blood cells were reconstituted to a final
hematocrit of 47 ±2% to obtain the platelet-free
reconstituted perfusate.
Human Endothelial Cell Culture
ECs were isolated from human umbilical cell cord
veins according to a previously described method.13
Basically, ECs were harvested with collagenase treatment (2% in phosphate-buffered saline [PBS], 15
minutes, 37°C; Boehringer-Manheim, Mannheim,
F.R.G.) and were maintained and subcultured in
Eagle's minimal essential medium supplemented
with 1 mM glutamine, 2 mM N-(2-hydroxyethyl)piperazine-A^'-(2-ethanesulfonic acid) (HEPES), 100
units/ml penicillin, 50 ^glva\ streptomycin (Flow Laboratories, Irving, Scotland), and 20% pooled human
sera. The ECs were grown at 37°C in a 5% CO2
humidified incubator. ECs previously identified by
immunologic criteria14 were subcultured in their second passage on 1% gelatin-precoated glass or plastic
coverslips (Thermanox, Miles Laboratories, Naperville, 111.) and were used for adhesion assays and
vWF, FN, and 13-HODE determinations.
Preparation of Extracellular Matrix
The underlying ECM of 7-day cultured ECs was
exposed according to one of the following methods.
1) Shear stress. EC-coated coverslips inserted into
the flat chamber (described below) were perfused
with platelet-free reconstituted perfusate at a shear
rate of 1,600 sec"1 for 15 minutes to eliminate the
ECs. The coverslips were then removed from the
chamber and washed with 0.15 M PBS, pH 7.4.
2) Nitrogen drying. The chamber was perfused with
aflowof nitrogen (40 ml/min) for 10 minutes at 37°C.
437
The dehydrated ECs were then dislodged by rinsing
with PBS.
3) Cellulose stripping. Strips of cellulose acetate
were laid flat on the EC monolayers for 2 seconds
and then peeled off, thus removing the ECs. This
method has been described in detail elsewhere.15
4) Ethylene glycol-bis(fi-aminoethyl
ether)N,N,N',N'-tetraacetic acid treatment. EC-coated coverslips were incubated with 1 ml 2% ethylene glycolbis(£-aminoethyl ether)-A^Af7V',A^'-tetraacetic acid
(EGTA) (37°C, 30 minutes) causing the ECs to
detach from the surface.
5) Ammonia exposure. EC-coated coverslips were
incubated with 0.1N NH4OH (37°C, 30 minutes).
The supernatant and lysed ECs were then removed.
ECM-coated coverslips obtained by any of these
procedures were washed and stored for 60 minutes
in PBS until use.
Perfusion Adhesion Assay
Plastic coverslips coated with the ECM produced
by EC cultures processed as previously described
were inserted into flat chambers according to Sakariassen et al.16 Anticoagulated blood samples were
then recirculated for 5 minutes at 37°C. Flow was
maintained by a peristaltic pump (model 502S, Watson Marlow, Cornwall, England) and adjusted to
obtain a shear rate of 1,300 sec"1, which corresponds
to the shear rate of the microvasculature.17
Morphometric Evaluation
The exposed ECM surface and the interaction of
platelets with the ECM preparations were evaluated
morphometrically by two different methods.18
1) En face. After perfusion, the coverslips were
removed from the chamber, rinsed with PBS, and
then fixed in 0.5% glutaraldehyde in PBS (2 hours,
4°C). The coverslips were dehydrated in a graded
series of ethanols and stained with May-Griinwald
and Giemsa (Merck, Darmstadt, Germany) as previously described.16 The percentage of exposed ECM
and the degree of platelet interaction with the ECM
were evaluated by means of a semiautomated
method. The exposed ECM surface was expressed as
the percentage of the total coverslip area devoid of
ECs. Profiles of single or grouped platelets were
projected with an electromagnetic pen to a manual
optical analysis system (MOP20, Kontron, Zurich,
Switzerland) connected to a personal computer
(IBM XT, 286). This system automatically integrates
areas and expresses them as the percentage of the
total ECM screened.
2) Cross section. Once the en face evaluation was
performed, the plastic coverslips were dehydrated
in a graded series of ethanols and embedded in
JB-4 solution (Polysciences Inc., Warrington, Pa.).
Infiltration of the samples was performed in catalyzed JB-4 solution (2 hours, 4°C). Coverslips were
embedded in flat molds (Histomold, LKB, Bromma,
Sweden). The perfused side of the plastic coverslips
was laid on top of the mold filled with catalyzed
438
Arteriosclerosis and Thrombosis Vol 11, No 2 March/April 1991
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
JB-4 solution in the presence of benzoyl-ethyl
ether. After complete polymerization, plastic coverslips were carefully peeled from the JB-4 polymer, leaving the perfused side of the coverslip
firmly embedded in the JB-4 polymer. Three-micron-thick sections of the polymer film containing
the embedded material were stained with 1% Methylene Blue.
Platelet interaction with ECM was measured in
cross sections by means of a manual picture analysis
system connected to a personal computer with an
automated recognition program as previously described.19 This program allows the morphometric
evaluation of platelet-ECM interactions according
to a minor modification of the basic criteria described by Baumgartner et al.20 Platelets were classified as 1) contact platelets, those that were attached but not spread on the ECM; 2) adhesive
platelets, those that were spread on the ECM
surface, forming monolayers and; 3) aggregated
platelets, those adherent platelets forming multiple
layers that exceeded 2.5 /xm in height. The total
surface area covered with platelets was calculated
as contact plus adherent plus aggregated.
Immunofluorescence Microscopy
The glass coverslips coated with the different ECM
preparations were fixed with 1% paraformaldehyde
in PBS (4°C, 90 minutes). Fixed coverslips were
always washed in PBS (three times, 5 minutes each)
and additionally washed in three changes of 0.1%
glycine in PBS to quench the aldehyde groups. The
glass coverslips containing fixed cells were first incubated for 3 minutes with 2% Triton X-100, and a
polyclonal antibody (1:100) against the antigen being
studied (1 hour, 22°C). The antibody was removed,
and the ECM-coated coverslips were washed three
times in PBS (5 minutes for each wash) and then
incubated with a secondary antibody against rabbit
immunoglobulins (Igs) labeled with fluorescein
(1:100 dilution, Dako F23, Dakopatts, Glostrup,
Denmark). Ninety minutes later, the EC-coated coverslips were again washed in PBS and then in distilled water. Finally, the ECM-coated coverslips were
mounted in a glycerin/PBS mixture (7:3, vol/vol).
Each coverslip was then viewed with a Reichert-Jung
Microstar 110 fluorescence microscope.14 Sera and
IgGs from preimmune rabbits and antisera preabsorbed with the corresponding antigens were used as
controls. The primary or secondary antibody was
omitted to confirm the specificity of the reaction.
Rabbit polyclonal antibodies to human vWF (Dako,
A082), FN (Dako, A245), and 13-HODE were used
as primary antibodies. Ten micrograms of purified
13-HODE added to undiluted antisera neutralized
the detection of 13-HODE in the preparations. In
contrast, the addition of 75 fig of 5-, 12-, or 15hydroxyeicosatetraenoic acid or 100 fig of linoleic or
arachidonic acid did not. These data indicate that the
antisera were specific for the 18-carbon metabolites,
the majority of which was 13-HODE as confirmed by
high-performance liquid chromatography (HPLC).21
von Willebrand Factor Determinations
ECMs obtained by the different methods of preparation were extracted with 6 M urea and 1% Triton
X-100 in PBS for analysis of vWF.22 vWF antigen was
determined by an enzyme-linked immunosorbent assay (ELISA) as previously described.23 Pooled normal human plasma containing 10 fig/ml vWF was
used as a standard. Rabbit polyclonal antibodies,
monospecific for vWF, served as the solid phase, and
a mixture of the murine monoclonal antibodies CLBRag 35 and CLB-Rag 50 was used as an indicator, in
combination with peroxidase-labeled sheep antimouse IgG (Institute Pasteur, Marnes-la-Garenne,
France). The response of serial dilutions of ECM
samples paralleled that of purified vWF. Dissolving
the vWF in extraction buffer followed by dialysis did
not alter the results found in the ELISA.
13-Hydroxy-9-cis,ll-tTans-octadecadienoic
Acid Determinations
Other EC- or ECM-coated glass coverslips were
extracted with methanol for an estimation of their
monohydroxy fatty acid content. The amount of
fatty acid metabolites extracted into methanol was
quantified by reverse-phase HPLC according to the
method described.21 Briefly, the methanol extracts
were blown to dryness with a flow of nitrogen, and
the residues were then resuspended in 0.9 ml 40%
acetonitrile, pH 3.5. The samples were then injected into a NOVA PACK C18 cartridge (3 fiM, 5
mmx 10 cm) using a /iBondapack C18 guard column
(Water Scientific, Toronto, Canada). An acetonitrile gradient (from 40% to 80%) was used to
separate the metabolites, which were measured at
ultraviolet absorbances of 234 and 254 nm. The
sensitivity and variability in 13-HODE analysis by
HPLC was 2 ±0.1 ng.24
Statistical Analysis
Results of the experiments are expressed as
mean±SEM and were analyzed with the Wilcoxon
test for paired data. A level of/? < 0.05 was considered
statistically significant.
Results
Exposed Extracellular Matrix Surface After
Different Treatments
The percentage of the area of exposed ECM
varied according to the different methods of preparation. In untreated EC coverslips, 6.7±1.3% of
the area was exposed ECM (Figure 1), and with
NH4OH EC lysis of the EC-coated coverslips, there
was almost complete removal of the ECs, exposing
99.1±0.4% of ECM. The amount of ECM exposed
with the other methods of preparation varied between these values (Figure 1).
Aznar-Salatti et al Extracellular Matrix/Platelet Interactions
I
I H ECM
exposed
439
C.S.
100
80 •
•• ••
6O •
20 •
(I
S.S.
I
Strip.
ECM
EGTA
MH4OH
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
FIGURE 1. Bar graph of en face morphological evaluation,
with the percentage of exposed ECM after EC removal by
different methods (a). Total ECM-exposed surface area covered
with platelets after blood perfusion is also shown (•). Data are
expressed as mean±SEM, with *p<0.05, **p<0.01 in comparison with cellulose stripping; n=7. ECM, extracellular matrix;
EC, endothelial cells; S.S., shear stress, 1,600 sec'1, 5 minutes;
Nit., nitrogen drying; Strip., cellulose stripping; EGTA, 2%
ethylene gfycol-bis(fi-aminoethyl e*/ier)N,N,N',N'-te/7Haceric
acid treatment; NH40H, ammonia exposure.
Platelet Adhesion Under Flow Conditions
The interaction of platelets with the ECM was
assessed morphometrically. The total surface area
covered with platelets as measured en face (Figure 1)
on ECMs prepared by shear stress was 14.5%. The
total surface area covered with platelets onto ECMs
prepared by cellulose acetate stripping or nitrogen
flow was 22-23%. Low platelet coverages were generally found in those preparations in which ECs were
mechanically removed. In contrast, the total surface
area covered by platelets on the EGTA- and
NRtOH-prepared ECMs was approximately 50%
(Figure 1).
In cross sections, the qualitative evaluation of
platelets adherent to the ECM displayed a varying
degree of platelet activation, depending on the
method of ECM preparation used (Figure 2). Thus,
the majority of platelets associated with the shear
stress preparation were mainly in contact (>80%),
and only a few platelets had been stimulated
enough to spread or to form aggregates (Figure 3),
indicating that this preparation rendered a less
adhesive surface. A similar pattern of interaction
was generally observed in those coverslips in which
ECs had been removed by mechanical methods
(nitrogen drying and cellulose stripping). In contrast, the majority of the platelets associated with
the EGTA and NH4OH preparations were spread
and/or aggregated (>50%), suggesting that the last
methods expose a more reactive surface.
Statistical comparisons performed between preparations that resulted in the exposure of a considerable surface of ECM (cellulose stripping, EGTA,
and NH,OH) showed significant differences in their
reactivity toward platelets. As shown in Figure 3, a
FIGURE 2. Photomicrographs showing platelet interaction
with ECM, both en face (upper photomicrographs) and cross
sectionalfy (lower photomicrographs). ECMs were exposed by
A: Shear stress (note some remaining ECs). B: Cellulose
acetate stripping. Image obtained is similar to one observed for
nitrogen-dried ECM. In both cases, there is a low-thrombogenic surface. C: Chemical methods, which produced highfy
thrombogenic surfaces, inducing circulating platelets to adhere
and to aggregate. ECM, extracellular matrix; EC, endothelial
cells. x760
statistical increase in the presence of spread platelets was observed in EGTA- and NI-ttOH-treated
ECs (p<0.01 vs. cellulose stripping). A similar
increase in the presence of aggregates was also
observed (p<0.05 vs. cellulose stripping).
440
Arteriosclerosis and Thrombosis
I
I COTvJTACT
SPREAD
Vol 11, No 2 March/April 1991
AGGREGATE
EGTA
(-H4OH
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
FIGURE 3. Bar graph of cross-sectional morphometric evaluation of platelet interaction with ECMs (a, contact; @, spread;
M, aggregate) under flow conditions. Evaluation of platelet
interaction is expressed as the percentage of total covered surface
(mean±SEM). *p<0.05, **p<0.01 in comparison with cellulose stripping; n=7. 5.5., shearstress, l,600sec~', 5minutes;Nit,
nitrogen drying; Strip., cellulose stripping; EGTA, 2% ethylene
gfycol-bis(f}-aminoethyl ether)NJ$,N' Ji'-tetmacetic acid treatment; NH4OH, ammonia exposure.
Histoimmunolocalization of von Willebrand Factor,
Fibronectin, and 13-Hydraxy-9-ck,ll-transoctudecadienoic Acid
Paraformaldehyde-fixed coverslips containing
ECMs from which ECs had been removed were
labeled for vWF and FN localization. In control
preparations, the pattern of staining for vWF followed the regular distribution pattern as previously
described.14 In the mechanically exposed ECMs,
remaining ECs showed a punctate pattern of staining
corresponding to the vWF located in Weibel-Palade
bodies.25 ECM-vWF showed a punctate pattern of
staining following fibrilJar structures. In the ECMs
prepared with NH4OH and EGTA, the fluorescence
was located on the fibrilJar structures of the matrix.
Control samples incubated with FN antibodies
showed fluorescent labeling following fibrillar arrangements defining the limits of the cells (Figure 4).
Mechanically exposed ECM showed homogeneous
distribution of FN with some areas devoid of fluorescent labeling, corresponding to the damaged zones.
Chemical methods exposed a very homogeneously
distributed staining for FN fibrils.
13-HODE staining was detected in ECM preparations in which there were remaining ECs and also in
FIGURE 4. Photomicrographs of histoimmunofluorescent
images of different ECM preparations. A: Untreated ECs
in which 1) vWF was detected in the ECM and localized in
cell bodies; 2) FN was localized only in the ECM; and 3)
intracellular amounts of 13-HODE were detected. x250;
B: ECM exposed by mechanical methods, in which 1) vWF
fluorescence appeared inside some remaining ECs and was
also observed over the ECM, following fibrillar arrangements; 2) FN staining showed a homogeneous distribution
and some damaged areas devoid of fluorescent labeling;
and 3) variable amounts of 13-HODE were detectable in
remaining ECs. x650; C: Chemically exposed ECM, in
which 1) only extracellular vWF was stained, confirming
that there were no ECs in those preparations; 2) a very
homogeneous mesh of FN staining was observed; and 3)
lack of fluorescent staining indicates absence of 13-HODE;
x650. ECM, extracellular matrix; EC, endothelial cell;
vWF, von Willebrand factor; FN, fibronectin; 13-HODE,
13-hydroxy-9-ds, 11-trans-octadecadienoic acid.
Aznar-Salatti et al
TABLE 1. von WUlebrand Factor and 13-Hydroxy-9-cii,ll-/nuisoctadecadienoic Acid
13-HODE
vWF
(ng/lCrt cells)
(ng/106 cells)
Condition
12.60±1.0
108.0±12.6
ECs
70.4±8.5
Shear stress
Cellulose acetate
ND'
41.6±4.5«
stripping
ND'
28.7±3.5«
2% EGTA, 30 minutes
ND*
0.1N NH.OH
9.2±1.1*
vWF, von Willebrand factor, 13-HODE, 13-hydroxy-9-cis,llfrunj-octadecadienoic acid; EC, endothelial cell; EGTA, ethylene
glycol-bis(/3-aminoethyl ether)Ar,Af,A'',A/'-tetraacetic acid;
NH4OH, ammonia treatment; ND, not detectable.
Values are mean±SEM, n=l.
V<0.01.
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
those that were mechanically exposed. In any of the
ECMs obtained with Nri,OH and EGTA, 13-HODE
was detected (Figure 4).
Analysis of von Willebrand Factor and
13-Hydroxy-9-c\s,l 1-tr ans-octadecadienoic Acid
The amount of vWF bound to the intact ECs was
108±12.6 ng/104 ECs. There was 76.2±5.4 ng/equivalent area in ECMs prepared by shear stress and
41.6±4.5 ng/equivalent area in ECMs exposed by
cellulose acetate stripping. The amount of vWF in
ECMs prepared by chemical methods ranged from
28.7±3.5 (EGTA) to 9.2±1.1 (NH,0H) ng/equivalent area as shown in Table 1.
The amounts of 13-HODE associated with the ECM
preparations are shown in Table 1. There was 12.6±1.0
ng 13-HODE/104 ECs associated with intact ECs.
There was no 13-HODE detected in the ECMs exposed by cellulose acetate, EGTA, and NH,OH.
Discussion
A number of studies have demonstrated that a
variety of components of the ECM underlying the
endothelium modulate platelet-ECM interactions.
Among others, these components include several
types of collagen and the noncollagenous proteins
vWF7 and FN.8 Recently, it has been suggested that
the EC-derived linoleic acid lipoxygenase metabolite
13-HODE is also important in the regulation of
platelet-ECM interactions,910 although the role of
this metabolite in platelet subendothelium is still not
well elucidated and remains a matter of controversy.12
In this study, we have explored the importance of
ECM preparation with regard to its thrombogenic
properties under flow conditions. We have focused
on the changes in the ECM concentration and distribution of two adhesive proteins, vWF and FN, and of
13-HODE, the lipoxygenase metabolite allegedly involved in the thromboresistant properties of ECM.
The results of our study indicate that the thrombogenic or thromboresistant properties of the ECM
underlying the endothelium, examined under flow
conditions, are influenced by the method of EC
removal and subsequent ECM exposure. Morpho-
Extracellular Matrix/Platelet Interactions
441
metric analysis of treated ECMs showed significant
differences in the extent of exposed ECM after the
different procedures of EC removal. Nonchemical
procedures were less effective in exposing the matrix
than those that required chemical treatment of the
ECs. When the ECMs were placed in the chamber
and perfused with anticoagulated whole blood under
the same conditions, the percentage of exposed ECM
covered with platelets was not proportional to the
extent of exposed subendothelial material, indicating
that the reactivity of the ECM is dependent on the
method of its preparation. Histoimmunolocalization
studies showed that vWF and FN were present in the
differently prepared ECMs. Conversely, immunofluorescent localization of 13-HODE in the ECM exhibited marked differences in the amounts of the
metabolite present. Quantification of the levels of
vWF by an ELISA and of 13-HODE by HPLC
confirmed these observations. The levels of vWF
were significantly reduced in the preparations in
which most of the cells were eliminated. It is important to point out that in all cases studied, the levels of
this adhesive protein exceeded the minimum amount
that, according to Sixma et al,26 is necessary to
support platelet adhesion. It is also interesting to
consider that although vWF and FN in the subendothelium are important in supporting platelet deposition, in both cases plasma provides an alternative
source of these proteins, and this fact may minimize
the effects of their removal from the ECM. On the
contrary, according to our current knowledge, 13HODE is present only inside ECs14; thus, the remnant 13-HODE found inside the remaining ECs may
cause the low thrombogenicity found in mechanically
exposed ECM.
Not only were there quantitative differences in the
interactions of platelets with the ECM but also there
were qualitative differences in the nature of the interactions of platelets with the different ECM preparations. The platelet interactions with the ECMs exposed
by mechanical methods were only contact interactions,
indicating lower thrombogenicity. Conversely, the
platelets analyzed in the sections prepared from ECMs
exposed by chemical methods were mostly spread on
the surface or aggregated, suggesting that these methods markedly affect the composition of the resting
endothelium/subendothelium.
Our microscopic examination shows that a variable
number of ECs remain attached to the ECM after
the different methods of exposure. It is clear that this
fact may also influence ECM reactivity; thus, the
possibility that the remaining ECs are still metabolically active, thereby synthetizing prostacyclin, endothelium-derived relaxing factor, or even 13-HODE
itself, all inhibitors of platelet clumping, cannot be
ruled out.
In summary, the data we report here indicate that
under dynamic conditions, the reactivity of the ECM
toward platelets depends on the procedure of EC
removal, which directly influences the levels of different ECM components. ECM is a widely used
442
Arteriosclerosis and Thrombosis
Vol 11, No 2 March/April 1991
substratum for platelet adhesion experiments; the
results of our investigations suggest that it is important to consider the method of preparation to avoid
spurious results in the studies concerning plateletECM interactions.
Acknowledgments
We would like to thank F. Royo and J. Aznar from
Boehringer Ingelheim S.A.E., Spain, for their statistical analysis of the data and to I. Calopa, M.
Garrido, P. Ant6n, L. Eltringham, and R. Burley for
their technical assistance.
References
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
1. Gordon JL, Pearson JD: Biology of the vascular endothelium,
in Bloom AL, Thomas DP (eds): Hemostasis and Thrombosis,
ed 2. Edinburgh/London/Melbourne/New York, Churchill Livingstone, 1987, pp 303-311
2. Harker LA, Schwartz SM, Ross R: Endothelium and arteriosclerosis. Clin Haematol 1981;10:283-296
3. Gamse G, Fromme HG, Kresse H: Metabolism of sulfated
glycosaminoglycans in cultured endothelial cells and smooth
muscle cells from bovine aorta. Biochim Biophys Ada 1978;
554:514-528
4. Loskutoff DS, van Mourik JA, Erikson LA, Lawrence D:
Detection of an unusually stable flbrinolytic inhibitor produced by bovine endothelial cells. Proc Nad Acad Sci USA
1984;80:2956-2960
5. Moncada S, Vane JR: Arachidonic acid metabolites and the
interactions between platelets and blood-vessel walls. N EnglJ
Med 1979;3O0:1142-1147
6. Madri JA, Dreyer B, Pitlick FA, Furthmayr H: The collagenous components of subendothelium, correlation structure
and functions. Lab Invest 1980;43:303-315
7. Jaffe EA, Hoyer LW, Nachman RL: Synthesis of von Willebrand factor by cultured human endothelial cells. Proc Nail
Acad Sci USA 1974;71:19O6-19O9
8. Houdijk WPM, Sixma JJ: Fibronectin in artery subendothelium is important for platelet adhesion. Blood 1985;65:
598-604
9. Buchanan MR, Bastida E: The role of 13-HODE and HETES
in vessel wall/circulating blood cells interactions. Agents
Actions 1987;22:l-3
10. Buchanan MR, Richardson M, Haas TA, Hirsh J, Madri JA:
The basement membrane underlying the vascular endothelium
is not thrombogenic: In vivo and in vitro studies with rabbit
and human tissues. Thromb Haemost 1987^8:698-704
11. Ross R: The pathogenesis of atherosclerosis: An update. N
EnglJ Med 1986;314:488-500
12. De Graaf JC, Bult H, Meyer GRY, Sixrna JJ, de Groot PG:
Platelet adhesion to subendothelium structures under flow
conditions: No effect of the lipoxygenase product 13-HODE.
Thromb Haemost 1989;62:802-806
13. Jaffe EA, Nachman RL, Becker CG, Minick CR: Culture of
human endothelial cells derived from umbilical veins. J Clin
Invest 1973^2:2745-2756
14. Aznar-Salatti J, Bastida E, Buchanan MR, Castillo R, Ordinas
A, Escolar G: Differential localization of von Willebrand
factor, fibronectin and 13-HODE in human endothelial cell
cultures. Histochemistry 1990;93:507-511
15. Buchanan MR, Dejana E, Cazenave JP, Richardson M, Mustard JF, Hirsh J: Differences in inhibition of PGI2 production
by aspirin in rabbit artery and vein segments. Thromb Res
1980;20:447-460
16. Sakariassen KS, Aarts PAMM, de Groot PG, Houdijk WPM,
Sixma JJ: A perfusion chamber developed to investigate
platelet interaction in flowing blood with human vessel wall
cells, their extracellular matrix, and purified components. J
Lab Chn Med 1983;102:522-535
17. Sakariassen KS, Muggli R, Baumgartner HR: Measurement of
platelet interaction with components of the vessel wall in
flowing blood, in Hawiger JJ (ed): Platelets: Methods in Enzymology, Volume 169. New York, Academic Press, Inc, 1989,
pp 37-70
18. Aznar-Salatti J, Escolar G, Ant6n P, Bastida E, Ordinas A: A
rapid embedding procedure for the study of platelet interactions with extracellular matrices in a flowing system: Effect of
aspirin on platelet activity. Methods Find Exp Clin Pharmacol
1990;12:149-154
19. Escolar G, Bastida E, Castillo R, Ordinas A: Development of
a computer program to analyze the parameters of platelet-vessel wall interactions. Haemostasis 1986;16:8-14
20. Baumgartner HR, Muggli R: Adhesion and aggregation: Morphological demonstration and quantitation in vivo and in vitro,
in Gordon JL (ed): Platelets in Biology and Pathology. Amsterdam, North Holland Biomedical Press, 1976, pp 23-60
21. Buchanan MR, Haas TA, Lagarde M, Guichardant M: 13-Hydroxy-octadecadienoic acid is the vessel wall chemorepellant,
LOX. / Biol Chem 1985;260:16056-16059
22. Mosher DF, Vaheri A: Thrombin stimulates the production
and the release of a major surface-associated glycoprotein
(fibronectin) in cultures of human fibroblast. Exp Cell Res
1978;112:323-334
23. De Groot PG, Reinders JH, Sixma JJ: Perturbation of human
endothelial cells by thrombin or PMA changes the reactivity of
their extracellular matrix towards platelets. / Cell Biol 1987;
104:697-704
24. Haas TA, Buchanan MR: An automated extraction and
quantitation procedure for lipoxygenase metabolites. / Chromatogr 1988;410:l-10
25. Wagner DD, Olmsted JB, Marder VJ: Immunolocalization of
von Willebrand protein in Weibel-Palade bodies of human
endothelial cells. / Cell Biol 1982;95:355-360
26. Sixma JJ, Nievelstein PFEM, Zwaginga JJ, de Groot PG:
Adhesion of blood platelets to the extracellular matrix of
cultured endothelial cells. Ann N Y Acad Sci 1987^16:39-52
KEY WORDS •
platelet adhesion
endothelial cells
extracellular matrix
•
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Platelet adhesion to exposed endothelial cell extracellular matrixes is influenced by the
method of preparation.
J Aznar-Salatti, E Bastida, T A Haas, G Escolar, A Ordinas, P H de Groot and M R Buchanan
Arterioscler Thromb Vasc Biol. 1991;11:436-442
doi: 10.1161/01.ATV.11.2.436
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1991 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/11/2/436
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which permission
is being requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/