THROMBOSIS RESEARCH 66; 247-258,1992 0049-3848/92 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.
PLATELET
MICROVESICLES ADHESION
ADHERE
TO SUB ENDOTHELIUM
AND
PRO$lOTE
OF PLATELETS
Michael R. Owens, Stein Holme and Sharon Cardinali American Red Cross Blood Services, Mid-Atlantic Region and the Departments of Medicine and Pathology, Medical College of Hampton Roads, Norfolk, Virginia (Received 11.ll .1992; accepted in revised form 17.3.1992 by Editor B. Hessel)
ABSTRACT A role in hemostasis has been suggested for platelet membrane microvesicles (mv). The objectives of the studies reported here include functional analysis of platelet mv in models developed for study of platelet investigation of possible adhesion, as well as interactions between mv and intact platelets in these same adhesion models. Microvesicles were prepared from washed platelet concentrates by repeated freezing and subendothelium was measured thawing. Adhesion to quantitatively by radiolabelling mv with Ill-In, and morphologically by scanning electronmicroscopy. Platelet mv adhered to subendothelium quantitatively over time. Using a modified Baumgartner chamber, we found adhesion of mv to subendothelium significantly increased with increasing shear rates. With this same model we found that prior exposure of subendothelium to mv greatly increased subsequent adhesion of platelets to the same everted vessel, compared to platelet adhesion in the absence of mv. All of these experiments were conducted with mv suspended in ACD/saline, indicating that plasma components are not essential for adhesion of mv. Our studies show that platelet mv adhere to subendothelium in much the same way as do platelets, and support the concept of a hemostatic role for mv in that they appear to increase platelet adhesion. Key words: platelets,
glycoproteins,
247
subendothelium,
shear rate
248
PLATELET MICROVESICLES
Vol. 66, Nos. 213
INTRODUCTION
In 1967, Wolf coined the term "platelet dust" to describe material derived from stored platelets(l). He characterized this material morphologically using electron microscopy, and also found that platelet dust had coagulant activity. Platelet dust, or microvesicles( were further characterized by Sandberg et al The using material from collagen-stimulated platelets(2,3). composition of mv obtained by this method was 40% protein, 42% phospholipid and 13% cholesterol(4). The coagulant activity of mv obtained from collagen-stimulated platelets has been measured using assays for platelet Factor 3. Microvesicles have been detected using monoclonal antibodies in blood bank fresh frozen plasma, cryoprecipitate and serum; the concentration in cryoprecipitate was 29 times that of plasma(5,6). Cryoprecipitate has been used to correct abnormal bleeding times in patients with uremia, and part of the hemostatic effect of this blood product may be due to the presence of mv (5). In 1987, McGill suggested that mv might play a role in hemostasis; mv labeled with Ill-Indium were found at bleeding sites in thrombocytopenic rabbits(7). Whether mv might participate directly or indirectly by potentiating platelet activity in hemostasis at bleeding sites is unknown. Quantitative measurement of platelet mv is particularly difficult. Assay of platelet factor 3 activity gives only an approximate estimate of mv. Currently the most often used method of detection of mv is flow cytometry using monoclonal antibodies to membrane glycoproteins IB or IIB/IIIA (8,9). In the studies described here we have compared the adhesion characteristics of platelet mv to that of intact platelets, using selected laboratory models which have been developed to characterize adhesion of platelets to subendothelium (10-15). In our studies, these models were used to provide both quantitative and qualitative information about mv adhesion to subendothelium, and, more importantly, to study the interaction of mv and platelets at the subendothelial surface. MATERIALS
AND
METHODS
Platelet mv were prepared by modifications of the methods of McGill et al(7). In brief, 2 platelet concentrates(APC, CP2D anticoagulant) were combined and centrifuged at 3700 x g for 15 min at room temperature. The supernatant was discarded and the platelet button resuspended in 20 ml ACD/saline. This procedure was repeated, and the washed platelets were placed in a -80°C freezer overnight. The following day, the frozen platelets were thawed in a 37°C waterbath, refrozen for 3 h and thawed again. The freezethawed platelets were centrifuged at 3700 x g for 15min to remove
Vol. 66, Nos. 2i3
PLATELET MICROVESICLES
249
intact platelets, and the supernatant was collected separately. To remove large particles and remaining intact platelets the supernatant was expressed through a series of micropore filters (Acrodisc, Gelman Sciences, Ann Arbor, MI)) 1.2 u, 0.8 u, 0.45 u and 0.2 u. Aliquots from each stage of filtration were saved for bioassay. The mv preparation was stored frozen in 1.5 ml aliquots at -80" until use. BIOASSAY OF Mv Coagulant activity of mv was performed using a Stypven reagent as described by Sandberg et al (2,3). 100 ul of platelet-poor plasma, 100 ul of Russell's viper venom (American Bioproducts, Parsippany, NJ) and 100 ul of microvesicles in suspension were incubated at 37°C for 2 min, 100 ul of calcium chloride 0.025M added and clotting times recorded by a fibrometer. A curve of coagulant activity for mv was constructed by making dilutions (lOO%, 50%, 10% and 1%) of the mv in phosphate buffered saline. In addition, we sought to determine whether passage of the mv suspension through progressively smaller micropore filters resulted in a decrease in the coagulant activity. In order to measure coagulant activity of the mv filtrates, an aliquot of mv was assayed in duplicate following filtration through each of the micropore filters, 1.2 u through 0.2 u. ill-INDIUM LABELING OF MV The procedure utilized for radiolabeling mv with Ill-In was for radiolabeling adapted from procedures developed intact platelets (16). Two aliquots of mv (3 ml total) were thawed at room temperature, combined and centrifuged at 13,000 x g for lo min; the supernatant was discarded and the mv button was resuspended in 3 ml ACD/saline. 100 UCi of Ill-Indium oxine (Amersham, Arlington Heights, IL) was added and incubated at room temperature for 20 min. At the end of the incubation period, the solution was centrifuged at 13,000 x g and the supernatant separated. Both the supernatant and mv button were counted for radioactivity; over multiple labelling experiments, the average labeling efficiency in this procedure was 57+7%(SD). 51-CHROMIUM LABELING OF PLATELETS Platelet concentrates obtained on Day 1 of storage (American Red Cross, CPZD anticoagulant) were labeled with 51-Cr as described ml of platelet concentrate was earlier (17). In brief,twenty withdrawn into a syringe through a 16 guage needle and transferred to two 15 ml conical plastic tubes containing 1.5 ml ACD solution. The platelets were spun at 2100 x g for 5min to remove clumps, transferred to two new plastic tubes and spun at 3700 x g for The supernatant was removed and one platelet button 15 min. resuspended in 3 ml ACD/saline, pH 6.5. The resuspended platelet button was transferred to the second tube and the solution used to 51-Cr Approximately 75 uci second button. the resuspend (Mallinckrodt, St. Louis, MO) was added to the platelet SUSpenSiOn At the end of the and incubated for 20 min at room temperature. incubation period, 2 ml ACD/saline solution and 1 ml plasma were added to the platelet suspension, mixed gently and centrifuged at
250
PLATELET MICROVESICLES
Vol. 66, Nos. 213
2100 x g for 10 min. The supernatant was removed and the radioactivity measured and recorded. Labeled platelets were suspended in ACD/saline for adhesion experiments. PREPARATION OF RABBIT AORTAS Segments of rabbit aorta (Pel Freeze, Rogers, Arkansas) were stored in phosphate-buffered saline containing penicillin and streptomycin. Approximately 1 week before use, the endothelium was removed from each vessel segment by passage (X2) of an inflated balloon catheter. For platelet adhesion experiments, segments of aorta were everted and mounted on a rod in one of the two types of apparatus used. When Ill-indium labeled mv were used, lcm segments of aorta were rinsed in PBS and placed in 5X75 mm glass test tubes for counting of radioactivity. In experiments where EM was to be performed, aortic segments were rinsed in PBS and placed in gluteraldehyde 2.5% overnight before processing for EM the following day. STUDIES WITH ROTATING PROBE DEVICE This device as described by Remuzzi et al (14) was used to 1)provide morphometric assessment of mv adhesion to subendothelium, 2)quantitative assessment of adhesion and 3)adhesion over a specified time course. Segments of rabbit aorta (2 cm length) which had endothelium removed as described above, were everted and mounted on a teflon rod (4 mm dia) which was placed in a stirring device which rotated at 250 rpm. Experiments were conducted using either unlabeled mv, for morphologic assessment, or Ill-In labeled mv, for quantitative measurement. For morphologic studies, 1 ml of mv was added to 5 ml ACD/saline and the everted aortic segments were rotated in this solution for 10 min. The segments were then rinsed in PBS and placed in gluteraldehyde overnight before processing for EM. For quantitative measurements of mv adhesion, either 50 ul, 100 ul or 200 ul of Ill-In labeled mv were added to 5 ml ACD/saline, and the everted aortic segments were rotated in this mixture for 10 min. To study the adhesion of mv to subendothelium over time, segments of aorta were rotated in 100 ul of Ill-In labeled mv in 5 ml ACD/saline for 1, 5, 10 or 20 min. EXPERIMENTS USING BAUMGARTNER APPARATUS Experiments with the Baumgartner model were designed to investigate 1) effects of shear rate on mv adhesion to subendothelium, and 2) interaction of mv and platelets at the subendothelial surface in a flowing system. Specifically we wished to learn whether mv might potentiate the subsequent attachment of platelets to subendothelium. The apparatus used was the same as previously described (18). The perfusate was composed of Ill-In labeled mv, 1 ml in 30 ml ACD/saline circulated for 10 min at shear rate of 126/see-' (n=12), or 632/set-1 (n=12). Shear rate calculations were performed as described elsewhere (19). In each case, when the 10 min perfusion period was completed, everted aortic segments were removed from the apparatus, rinsed in PBS and placed in a 5X75 mm glass tube for radioactivity counting. To study the effects of mv on subsequent platelet adhesion to subendothelium in the Baumgartner model, segments of aorta were
Vol. 66, Nos. 213
PLATELET MICROVESICLES
251
perfused with 3 ml 51-Cr labeled platelets in 30 ml ACD/saline for 1Omin at a shear rate of 126/seC-1 (control, n=12). In a second set of experiments, 2 ml of Ill-In labeled mv was first added to the perfusate of 30 ml ACD/saline and perfused through the apparatus for 5 min; 3 ml of 51-Cr labeled platelets were then added and the perfusion continued for an additional 10 min at the same shear rate (n=12). The sections of aorta were then removed from the apparatus, rinsed in PBS and counted for radioactivity using a 3" crystal LKB AutoGamma gamma counter with one window set for the 51-Cr peak and the other for the Ill-In summation peak. In each case, raw counts were corrected for background, decay based on elapsed counting time and spillover of each isotope into the other isotope channel.
TABLE I. Preparation in an Assay Using Bioactivity of Microvesicle Russell's Viper Venom. The two rows of numbers under the heading microvesicle dilutions indicate the mean clotting times for the various microvesicle dilutions tested. The rows of number shown under the heading microvesicle filtrates are mean clotting times from filtrates of microvesicles that were filtered through progressively smaller micropore filters. Coagulant activity was not significantly diminished by filtration through 0.2 u.
Microvesicle Dilution
Clotting Time (see)*
Micropore Filter size
Filtrate Clotting time
100 %
23.2
1.2 u
20.9
50 %
24.2
0.8 u
22.9
10 %
28.7
0.45 u
20.7
1%
44.1
0.2 u
21.9
* mean clotting time, n = 5
252
PLATELET MICROVESICLES
Vol. 66, Nos. 213
RESULTS
BIOASSAY OF MV: STYPVEN ACTIVITY A comparison of the stypven activity curves obtained by dilutions of cephalin or mv is shown in Table 1. The mv preparation was lo-fold more active in the assay system. The effects of filtration of the mv through progressively smaller micropore filters is also shown in the Table. There was no diminution of activity, even when filtered through a 0.2 u filter.
Figure 1. Scanning EM photos of mv adherent to everted rabbit aorta. The alternating light-dark bars at the bottom of the photo are 10 u in length. All mv used in these studies were filtered through micropore filters of 0.2 u pore size. As is evident in the Figure, many mv are present in various sizes, some considerably larger than 0.2 u. EXPERIMENTS WITH ROTATING PROBE DEVICE Microvesicles attached to subendothelium of aortic segments mounted on a rotating probe is shown in Figure 1. The physical configuration of the mv is apparent in the Figure. Some of the mv seen are clearly greater in size than 0.2 u, suggesting that the micropore filter may not have excluded some larger sized particles, or that the mv might be deformable allowing larger particles to pass through 0.2 u pores. Quantitative studies with mv, 50 ul, 100 ul or 200 ul, in the rotating probe apparatus are shown in Figure 2. The radioactive
Vol. 66, Nos. 2l3
PLATELET MICROVESICLES
253
counts shown here were corrected for background and decay based on elapsed counting time. Quantitative increases in mv attachment is clear from the Figure. Effects of prolonged time of exposure to mv is shown in Figure 3, where aortic segments were rotated in a solution containing 50 ul of mv for times ranging from 1 min to 20 min. The progressive cumulative attachment of mv is apparent in the Figure, and the data suggest that adherent mv remain adherent rather than detaching.
MICROVESICLE
ADHESION
QUANTITATIVE 200
RADIOACTIVITY X10” ______~~~ ’
175
75
t
25
Figure 2. Ill-In labeled mv, 50 ul, 100 ul or 200 ul, prepared from washed, freeze-thawed platelets were added to 5 ml ACD/saline . Everted rabbit aorta mounted on a plexiglass probe was rotated in this mixture (250 rpm) for 10 min, rinsed in PBS and counted for radioactivity. Raw counts were corrected for background and decay based on elapsed counting time. RESULTS WITH BAUMGARTNER APPARATUS Effects of shear rate on mv adhesion to subendothelium: A perfusate containing Ill-In labeled mv in 30 ml ACD/saline was recirculated in the Baumgartner apparatus for 10 minutes at shear rate of either 126/see-' or 632/set-' (n=12 for each). As can be seen in Figure 4, quantitative adhesion of mv was almost 3 times greater at the higher shear rate than at the lower (pc.001). Interaction of mv and platelets: A perfusate containing 51-Cr labeled platelets in 30 ml ACD/saline was recirculated for 10 min over everted rabbit aorta (n=lO), after which the aortic segments were removed, rinsed and counted for To determine whether first exposing the aortic radioactivity. segments to mv might affect subsequent platelet attachment, a perfusate containing Ill-In labeled mv was first recirculate&for
PLATELET
254
Vol. 66, Nos. 213
MICROVESICLES
MICROVESICLE ADHESION TIME COURSE RADIOACTIVITY
X 10’
160 126 100 76 60 26
t MIN
20MIN
IOMIN
EXPOSUREhlINUTES)
1 ACD/saline. Figure 3. Ill-In labeled mv, 200 ul, was added to 5 In: Everted rabbit aorta mounted on a plexiglass probe was rotated in this mixture for 1 min, 5 min, 10 min or 20 min. Aortic segments were then removed from the mv solution, rinsed in PBS and counted for radioactivity.
EFFECTS OF SHEAR RATE RADIOACTIVITY
COUNTS
X IO3
Ifizz
100 80 80 40 20 0 832
128
SHEAR RATE/SEC-l
Figure 4. The effects of shear rate on cumulative adhesion of mv suspended in ACD/saline and perfused over everted rabbit aorta for lo min. N=ll for each shear rate and the differences were significant (p
PLATELET
MICROVESICLES ADHESION
ADHERE
TO SUB ENDOTHELIUM
AND
PRO$lOTE
OF PLATELETS
Michael R. Owens, Stein Holme and Sharon Cardinali American Red Cross Blood Services, Mid-Atlantic Region and the Departments of Medicine and Pathology, Medical College of Hampton Roads, Norfolk, Virginia (Received 11.ll .1992; accepted in revised form 17.3.1992 by Editor B. Hessel)
ABSTRACT A role in hemostasis has been suggested for platelet membrane microvesicles (mv). The objectives of the studies reported here include functional analysis of platelet mv in models developed for study of platelet investigation of possible adhesion, as well as interactions between mv and intact platelets in these same adhesion models. Microvesicles were prepared from washed platelet concentrates by repeated freezing and subendothelium was measured thawing. Adhesion to quantitatively by radiolabelling mv with Ill-In, and morphologically by scanning electronmicroscopy. Platelet mv adhered to subendothelium quantitatively over time. Using a modified Baumgartner chamber, we found adhesion of mv to subendothelium significantly increased with increasing shear rates. With this same model we found that prior exposure of subendothelium to mv greatly increased subsequent adhesion of platelets to the same everted vessel, compared to platelet adhesion in the absence of mv. All of these experiments were conducted with mv suspended in ACD/saline, indicating that plasma components are not essential for adhesion of mv. Our studies show that platelet mv adhere to subendothelium in much the same way as do platelets, and support the concept of a hemostatic role for mv in that they appear to increase platelet adhesion. Key words: platelets,
glycoproteins,
247
subendothelium,
shear rate
248
PLATELET MICROVESICLES
Vol. 66, Nos. 213
INTRODUCTION
In 1967, Wolf coined the term "platelet dust" to describe material derived from stored platelets(l). He characterized this material morphologically using electron microscopy, and also found that platelet dust had coagulant activity. Platelet dust, or microvesicles( were further characterized by Sandberg et al The using material from collagen-stimulated platelets(2,3). composition of mv obtained by this method was 40% protein, 42% phospholipid and 13% cholesterol(4). The coagulant activity of mv obtained from collagen-stimulated platelets has been measured using assays for platelet Factor 3. Microvesicles have been detected using monoclonal antibodies in blood bank fresh frozen plasma, cryoprecipitate and serum; the concentration in cryoprecipitate was 29 times that of plasma(5,6). Cryoprecipitate has been used to correct abnormal bleeding times in patients with uremia, and part of the hemostatic effect of this blood product may be due to the presence of mv (5). In 1987, McGill suggested that mv might play a role in hemostasis; mv labeled with Ill-Indium were found at bleeding sites in thrombocytopenic rabbits(7). Whether mv might participate directly or indirectly by potentiating platelet activity in hemostasis at bleeding sites is unknown. Quantitative measurement of platelet mv is particularly difficult. Assay of platelet factor 3 activity gives only an approximate estimate of mv. Currently the most often used method of detection of mv is flow cytometry using monoclonal antibodies to membrane glycoproteins IB or IIB/IIIA (8,9). In the studies described here we have compared the adhesion characteristics of platelet mv to that of intact platelets, using selected laboratory models which have been developed to characterize adhesion of platelets to subendothelium (10-15). In our studies, these models were used to provide both quantitative and qualitative information about mv adhesion to subendothelium, and, more importantly, to study the interaction of mv and platelets at the subendothelial surface. MATERIALS
AND
METHODS
Platelet mv were prepared by modifications of the methods of McGill et al(7). In brief, 2 platelet concentrates(APC, CP2D anticoagulant) were combined and centrifuged at 3700 x g for 15 min at room temperature. The supernatant was discarded and the platelet button resuspended in 20 ml ACD/saline. This procedure was repeated, and the washed platelets were placed in a -80°C freezer overnight. The following day, the frozen platelets were thawed in a 37°C waterbath, refrozen for 3 h and thawed again. The freezethawed platelets were centrifuged at 3700 x g for 15min to remove
Vol. 66, Nos. 2i3
PLATELET MICROVESICLES
249
intact platelets, and the supernatant was collected separately. To remove large particles and remaining intact platelets the supernatant was expressed through a series of micropore filters (Acrodisc, Gelman Sciences, Ann Arbor, MI)) 1.2 u, 0.8 u, 0.45 u and 0.2 u. Aliquots from each stage of filtration were saved for bioassay. The mv preparation was stored frozen in 1.5 ml aliquots at -80" until use. BIOASSAY OF Mv Coagulant activity of mv was performed using a Stypven reagent as described by Sandberg et al (2,3). 100 ul of platelet-poor plasma, 100 ul of Russell's viper venom (American Bioproducts, Parsippany, NJ) and 100 ul of microvesicles in suspension were incubated at 37°C for 2 min, 100 ul of calcium chloride 0.025M added and clotting times recorded by a fibrometer. A curve of coagulant activity for mv was constructed by making dilutions (lOO%, 50%, 10% and 1%) of the mv in phosphate buffered saline. In addition, we sought to determine whether passage of the mv suspension through progressively smaller micropore filters resulted in a decrease in the coagulant activity. In order to measure coagulant activity of the mv filtrates, an aliquot of mv was assayed in duplicate following filtration through each of the micropore filters, 1.2 u through 0.2 u. ill-INDIUM LABELING OF MV The procedure utilized for radiolabeling mv with Ill-In was for radiolabeling adapted from procedures developed intact platelets (16). Two aliquots of mv (3 ml total) were thawed at room temperature, combined and centrifuged at 13,000 x g for lo min; the supernatant was discarded and the mv button was resuspended in 3 ml ACD/saline. 100 UCi of Ill-Indium oxine (Amersham, Arlington Heights, IL) was added and incubated at room temperature for 20 min. At the end of the incubation period, the solution was centrifuged at 13,000 x g and the supernatant separated. Both the supernatant and mv button were counted for radioactivity; over multiple labelling experiments, the average labeling efficiency in this procedure was 57+7%(SD). 51-CHROMIUM LABELING OF PLATELETS Platelet concentrates obtained on Day 1 of storage (American Red Cross, CPZD anticoagulant) were labeled with 51-Cr as described ml of platelet concentrate was earlier (17). In brief,twenty withdrawn into a syringe through a 16 guage needle and transferred to two 15 ml conical plastic tubes containing 1.5 ml ACD solution. The platelets were spun at 2100 x g for 5min to remove clumps, transferred to two new plastic tubes and spun at 3700 x g for The supernatant was removed and one platelet button 15 min. resuspended in 3 ml ACD/saline, pH 6.5. The resuspended platelet button was transferred to the second tube and the solution used to 51-Cr Approximately 75 uci second button. the resuspend (Mallinckrodt, St. Louis, MO) was added to the platelet SUSpenSiOn At the end of the and incubated for 20 min at room temperature. incubation period, 2 ml ACD/saline solution and 1 ml plasma were added to the platelet suspension, mixed gently and centrifuged at
250
PLATELET MICROVESICLES
Vol. 66, Nos. 213
2100 x g for 10 min. The supernatant was removed and the radioactivity measured and recorded. Labeled platelets were suspended in ACD/saline for adhesion experiments. PREPARATION OF RABBIT AORTAS Segments of rabbit aorta (Pel Freeze, Rogers, Arkansas) were stored in phosphate-buffered saline containing penicillin and streptomycin. Approximately 1 week before use, the endothelium was removed from each vessel segment by passage (X2) of an inflated balloon catheter. For platelet adhesion experiments, segments of aorta were everted and mounted on a rod in one of the two types of apparatus used. When Ill-indium labeled mv were used, lcm segments of aorta were rinsed in PBS and placed in 5X75 mm glass test tubes for counting of radioactivity. In experiments where EM was to be performed, aortic segments were rinsed in PBS and placed in gluteraldehyde 2.5% overnight before processing for EM the following day. STUDIES WITH ROTATING PROBE DEVICE This device as described by Remuzzi et al (14) was used to 1)provide morphometric assessment of mv adhesion to subendothelium, 2)quantitative assessment of adhesion and 3)adhesion over a specified time course. Segments of rabbit aorta (2 cm length) which had endothelium removed as described above, were everted and mounted on a teflon rod (4 mm dia) which was placed in a stirring device which rotated at 250 rpm. Experiments were conducted using either unlabeled mv, for morphologic assessment, or Ill-In labeled mv, for quantitative measurement. For morphologic studies, 1 ml of mv was added to 5 ml ACD/saline and the everted aortic segments were rotated in this solution for 10 min. The segments were then rinsed in PBS and placed in gluteraldehyde overnight before processing for EM. For quantitative measurements of mv adhesion, either 50 ul, 100 ul or 200 ul of Ill-In labeled mv were added to 5 ml ACD/saline, and the everted aortic segments were rotated in this mixture for 10 min. To study the adhesion of mv to subendothelium over time, segments of aorta were rotated in 100 ul of Ill-In labeled mv in 5 ml ACD/saline for 1, 5, 10 or 20 min. EXPERIMENTS USING BAUMGARTNER APPARATUS Experiments with the Baumgartner model were designed to investigate 1) effects of shear rate on mv adhesion to subendothelium, and 2) interaction of mv and platelets at the subendothelial surface in a flowing system. Specifically we wished to learn whether mv might potentiate the subsequent attachment of platelets to subendothelium. The apparatus used was the same as previously described (18). The perfusate was composed of Ill-In labeled mv, 1 ml in 30 ml ACD/saline circulated for 10 min at shear rate of 126/see-' (n=12), or 632/set-1 (n=12). Shear rate calculations were performed as described elsewhere (19). In each case, when the 10 min perfusion period was completed, everted aortic segments were removed from the apparatus, rinsed in PBS and placed in a 5X75 mm glass tube for radioactivity counting. To study the effects of mv on subsequent platelet adhesion to subendothelium in the Baumgartner model, segments of aorta were
Vol. 66, Nos. 213
PLATELET MICROVESICLES
251
perfused with 3 ml 51-Cr labeled platelets in 30 ml ACD/saline for 1Omin at a shear rate of 126/seC-1 (control, n=12). In a second set of experiments, 2 ml of Ill-In labeled mv was first added to the perfusate of 30 ml ACD/saline and perfused through the apparatus for 5 min; 3 ml of 51-Cr labeled platelets were then added and the perfusion continued for an additional 10 min at the same shear rate (n=12). The sections of aorta were then removed from the apparatus, rinsed in PBS and counted for radioactivity using a 3" crystal LKB AutoGamma gamma counter with one window set for the 51-Cr peak and the other for the Ill-In summation peak. In each case, raw counts were corrected for background, decay based on elapsed counting time and spillover of each isotope into the other isotope channel.
TABLE I. Preparation in an Assay Using Bioactivity of Microvesicle Russell's Viper Venom. The two rows of numbers under the heading microvesicle dilutions indicate the mean clotting times for the various microvesicle dilutions tested. The rows of number shown under the heading microvesicle filtrates are mean clotting times from filtrates of microvesicles that were filtered through progressively smaller micropore filters. Coagulant activity was not significantly diminished by filtration through 0.2 u.
Microvesicle Dilution
Clotting Time (see)*
Micropore Filter size
Filtrate Clotting time
100 %
23.2
1.2 u
20.9
50 %
24.2
0.8 u
22.9
10 %
28.7
0.45 u
20.7
1%
44.1
0.2 u
21.9
* mean clotting time, n = 5
252
PLATELET MICROVESICLES
Vol. 66, Nos. 213
RESULTS
BIOASSAY OF MV: STYPVEN ACTIVITY A comparison of the stypven activity curves obtained by dilutions of cephalin or mv is shown in Table 1. The mv preparation was lo-fold more active in the assay system. The effects of filtration of the mv through progressively smaller micropore filters is also shown in the Table. There was no diminution of activity, even when filtered through a 0.2 u filter.
Figure 1. Scanning EM photos of mv adherent to everted rabbit aorta. The alternating light-dark bars at the bottom of the photo are 10 u in length. All mv used in these studies were filtered through micropore filters of 0.2 u pore size. As is evident in the Figure, many mv are present in various sizes, some considerably larger than 0.2 u. EXPERIMENTS WITH ROTATING PROBE DEVICE Microvesicles attached to subendothelium of aortic segments mounted on a rotating probe is shown in Figure 1. The physical configuration of the mv is apparent in the Figure. Some of the mv seen are clearly greater in size than 0.2 u, suggesting that the micropore filter may not have excluded some larger sized particles, or that the mv might be deformable allowing larger particles to pass through 0.2 u pores. Quantitative studies with mv, 50 ul, 100 ul or 200 ul, in the rotating probe apparatus are shown in Figure 2. The radioactive
Vol. 66, Nos. 2l3
PLATELET MICROVESICLES
253
counts shown here were corrected for background and decay based on elapsed counting time. Quantitative increases in mv attachment is clear from the Figure. Effects of prolonged time of exposure to mv is shown in Figure 3, where aortic segments were rotated in a solution containing 50 ul of mv for times ranging from 1 min to 20 min. The progressive cumulative attachment of mv is apparent in the Figure, and the data suggest that adherent mv remain adherent rather than detaching.
MICROVESICLE
ADHESION
QUANTITATIVE 200
RADIOACTIVITY X10” ______~~~ ’
175
75
t
25
Figure 2. Ill-In labeled mv, 50 ul, 100 ul or 200 ul, prepared from washed, freeze-thawed platelets were added to 5 ml ACD/saline . Everted rabbit aorta mounted on a plexiglass probe was rotated in this mixture (250 rpm) for 10 min, rinsed in PBS and counted for radioactivity. Raw counts were corrected for background and decay based on elapsed counting time. RESULTS WITH BAUMGARTNER APPARATUS Effects of shear rate on mv adhesion to subendothelium: A perfusate containing Ill-In labeled mv in 30 ml ACD/saline was recirculated in the Baumgartner apparatus for 10 minutes at shear rate of either 126/see-' or 632/set-' (n=12 for each). As can be seen in Figure 4, quantitative adhesion of mv was almost 3 times greater at the higher shear rate than at the lower (pc.001). Interaction of mv and platelets: A perfusate containing 51-Cr labeled platelets in 30 ml ACD/saline was recirculated for 10 min over everted rabbit aorta (n=lO), after which the aortic segments were removed, rinsed and counted for To determine whether first exposing the aortic radioactivity. segments to mv might affect subsequent platelet attachment, a perfusate containing Ill-In labeled mv was first recirculate&for
PLATELET
254
Vol. 66, Nos. 213
MICROVESICLES
MICROVESICLE ADHESION TIME COURSE RADIOACTIVITY
X 10’
160 126 100 76 60 26
t MIN
20MIN
IOMIN
EXPOSUREhlINUTES)
1 ACD/saline. Figure 3. Ill-In labeled mv, 200 ul, was added to 5 In: Everted rabbit aorta mounted on a plexiglass probe was rotated in this mixture for 1 min, 5 min, 10 min or 20 min. Aortic segments were then removed from the mv solution, rinsed in PBS and counted for radioactivity.
EFFECTS OF SHEAR RATE RADIOACTIVITY
COUNTS
X IO3
Ifizz
100 80 80 40 20 0 832
128
SHEAR RATE/SEC-l
Figure 4. The effects of shear rate on cumulative adhesion of mv suspended in ACD/saline and perfused over everted rabbit aorta for lo min. N=ll for each shear rate and the differences were significant (p
