Platelet cGMP, but not CAMP, inhibits thrombin-induced platelet adhesion to pulmonary vascular endothelium CATHERINE Department Venturini, Catherine Kaplan. Platelet cGMP,
M. VENTURINI, LISA K. WESTON, AND JOHN E. KAPLAN of Physiology and Cell Biology, Albany Medical College, Albany, New M., Lisa K. Weston, and John E.
but not CAMP, inhibits thrombininduced platelet adhesion to pulmonary vascular endothelium. Am. J. Physiol. 263 (Heart Circ. Physiol. 32): H606-H612, 1992. -We have compared the effects of intracellular pathways initiated by nitric oxide and prostacyclin on thrombin-induced platelet adhesion to endothelial cells. Platelet aggregate adhesion was enhanced when endothelial monolayers were pretreated with NG-monomethyl-L-arginine (L-NMMA), an inhibitor of nitric oxide production. In addition, decreased platelet aggregate adhesion was seen when platelets were pretreated with 8bromoadenosine 3’,5’-cyclic monophosphate (8bromoCAMP) or 8bromoguanosine 3’,5’-cyclic monophosphate (8 bromo-cGMP). Single platelet adhesion in isolated perfused lungs under flow conditions in the presence of shear was also assessed. Pretreatment of platelets with either Iloprost, in a dose sufficient to decrease platelet aggregation, or 8-bromoCAMP did not affect platelet adhesion. However, pretreatment of platelets with 8-bromo-cGMP significantly reduced single platelet adhesion to endothelium. These studies illustrate that nitric oxide inhibits platelet adhesion to endothelium in the presence of shear. They further indicate that prostacyclin is also a regulator of this response but has effects more specifically related to the inhibition of platelet aggregation than plateletendothelium interactions. thrombosis;
nitric
oxide;
prostacyclin
ENDOTHELIAL CELLS ARE NORMALLY effective in preventing platelet adhesion. However, thrombin induces platelet adhesion to endothelium (11) and enhances release of endothelium-derived relaxing factor (EDRF) from endothelium (21). EDRF is widely believed to be nitric oxide although recent evidence suggests that other molecules such as S-nitrosocysteine may contribute to this activity (16,19). Nitric oxide downregulates platelet adhesion to endothelium in several models of thrombosis (20, 25, 27). The effect of endothelial prostacyclin on platelet adhesion to endothelium is less clear. Prostacyclin release has been correlated with endothelial resistance to platelet adhesion by some (5, 9), but this has been disputed by others (4, 12) who found no change in thrombin-induced platelet adhesion when prostacyclin generation was inhibited by aspirin pretreatment of endothelium. We have previously developed two models of platelet adhesion that examine platelet-endothelial cell interactions separately from platelet-platelet interactions. Thrombin induces adhesion of sheep platelet aggregates to cultured sheep pulmonary artery endothelial cell monolayers (13) and both nitric oxide (27) and prostacyclin (13) inhibit platelet accumulation on the endothelial surface. In a second model, thrombin pretreatment of isolated perfused rat lungs induced rat platelet adherence as single activated platelets (28). Nitric oxide inhibited platelet adhesion in this model; however, there was no change in thrombin-stimulated platelet adhesion H606
York 12208
to perfused lungs when animals were treated . with aspirin before removal of lu ngs for perfusion. The differences between the antiadhesive natures of nitric oxide and prostacyclin are potentially important in understanding the etiology of thrombosis and atherosclerosis. In the current study, we have further examined the role of nitric oxide and prostacyclin in thrombininduced platelet adhesion to endothelium by examining the platelet intracellular signaling mechanisms triggered by these factors. We hypothesize that prostacyclin does not directly inhibit platelet adhesion to endothelium. Rather, prostacyclin inhibits platelet-platelet interactions and decreases platelet adhesion in models that involve platelet aggregate adhesion, whereas nitric oxide effectively inhibits both platelet adhesion to endothelium and platelet aggregation. METHODS
Platelet isolation. Platelets were isolated from rat or sheep blood using a modification of the method of Corash et al. (3,18) utilizing isopycnic centrifugation on polyarabinogalactan gradients (Stractan II, Champion International, Tacoma, WA). 51Crlabeled rat platelets were used in the isolated perfused rat lung assay, and IllIn-labeled sheep platelets were used in the in vitro platelet adherence assay. Whole blood was drawn from the inferior vena cava of a rat or the jugular vein of a sheep through a 19gauge needle into a plastic syringe containing two parts 19% sodium citrate per 100 parts blood. Platelet-rich plasma (PRP) was prepared by adding 1 ml of a buffered saline glucosecitrate solution (BSG-citrate) containing (in M) 0.117 NaCl, 0.0136 sodium citrate, 0.0111 glucose, 0.0086 Na,HPO,, and 0.0016 KH,P04, pH 7.4, to 9 ml of blood and centrifuging at 850 g for 8 min. This procedure was performed twice for rat platelets. Rat platelets were labeled with 51Cr by adding 51Cr as sodium chromate (New England Nuclear, Boston, MA) to BSGcitrate-diluted PRP (2 &i/5 ml) and incubating for 60 min at 37°C. Sheep platelets were labeled with “‘In by adding ‘llInoxine (10 &i/ml; Amersham, Arlington Heights, IL) in the same manner. The BSG-citrate-diluted PRP was layered over a discontinuous density gradient consisting of 4 ml of 20% Stractan II and 3 ml of 10% Stractan II in a 15-ml conical tube. The tube was centrifuged for 15 min at 1,450 g, after which the plasma-BSG-citrate and 10% Stractan II layers were aspirated. The platelet layer was then removed and washed free of Stractan II by centrifugation for IO min at 600 g in excess BSGcitrate. The platelet pellet was resuspended in modified (Ca2+and Mg2+-free) Tyrode solution [containing (in M) 0.137 NaCl, 0.0030 KCl, 0.012 NaHCO,, 0.006 glucose, and 0.004 NaH,PO,, pH 7.41. This method of isolation removes plasma proteins while causing minimal damage to platelets. All procedures were carried out using plastic or siliconized glassware at room temperature. Platelets were counted using a hemacytometer (American Optical, Buffalo, NY) and Nikon phase contrast microscope (Nikon, Garden City, NY) after the method of Bjorkman (1) In some experiments, 8-bromoadenosine 3’,5’-cyclic monophosphate (8-bromo-CAMP) or 8-bromoguanosine 3’,5’-cyclic
0363-6135/92 $2.00 Copyright 0 1992 the American Physiological Society Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
cGMP
INHIBITS
PLATELET
monophosphate (8bromo-cGMP; both from Sigma, St. Louis, MO) was used to increase intracellular concentrations of platelet cyclic nucleotides. Platelets were isolated and labeled as described above and then incubated with 1 or 10 mM &bromocGMP or 8bromo-CAMP for 10 min. These pretreated platelets were then used in the in vitro or isolated lung platelet adhesion assays. Platelets were incubated with 8bromo-derivatives in a small volume and then diluted before exposure to endothelium. This reduced the concentration of 8-bromo-cyclic nucleotides on the endothelium to 25 PM, minimizing the direct effects of these compounds on endothelial function. This concentration of 8-bromo-cyclic nucleotides when added to the monolayers before, or concurrently with, platelets that had not previously been incubated with cyclic nucleotides had no effect on the adherence of platelets in the presence or absence of a-thrombin (data not shown). In some cases, platelets were pretreated with 10 nM Iloprost, a stable prostacyclin analogue, for 10 min before aggregation. Iloprost was a kind gift of Dr. G. Rubanyi, Berlex Pharmaceuticals, Cedar Knolls, NJ. Platelet aggregation. In vitro aggregation of rat platelets was measured according to the method of Born (2) using a ChronoLog Aggregation Module (Chrono-Log, Havertown, PA). Aggregometry was performed at 37°C in Hanks’ balanced salt solution (HBSS; GIBCO, Grand Island, NY) with Ca2+ and Mg2+ using a stir bar rotating at 900 revolutions/min. The final volume of the assay mixture was 0.5 ml. The slope of the rate of aggregation was recorded, and a maximal scale deflection was elicited in response to 0.1 U of cr-thrombin. Thrombin preparation. Human cu-thrombin was prepared by extracting prothrombin from Cohn fraction III paste as described elsewhere (26) and was a generous gift of Dr. John W. Fenton II, Wadsworth Center for Laboratories and Research, Albany, NY. EndotheZiaZ cell cukure. Primary cultures of sheep pulmonary artery endothelial cells were isolated and grown using established techniques (6). Endothelial cells at 3 to 30 population doublings were seeded onto 24-well plates (Corning, Chicago, IL) and grown to confluence in Dulbecco’s modified Eagle’s medium (DMEM; GIBCO) containing 20% fetal bovine serum (Hyclone Labs, Logan, UT), nonessential amino acids (GIBCO), and 50 pg/ml gentamicin sulfate (M.A. Bioproducts, Walkersville, MD) in a 5% CO, incubator at 37°C. Only confluent endothelial cell monolayers were used for the platelet adherence assays. Endothelial monolayers were generously provided by Dr. Peter J. Del Vecchio, Department of Ophthalmology, Albany Medical College. Some monolayers were fixed by incubation with 500 ~1 of an osmotically balanced 1.5% paraformaldehyde solution in HBSS for 2 h at 4°C. The fixative was then removed, and the monolayers were washed three times with HBSS before use in the adherence assays. Some endothelial monolayers were pretreated with aspirin. These monolayers were washed three times with HBSS, and 1 ml of HBSS was added to each well. One hundred milligrams of aspirin (acetylsalicylic acid; Sigma) were dissolved in 450 ~1 of dimethyl sulfoxide (DMSO; Sigma). Just before use, 500 ~1 of HBSS were added, and 10 ~1 of this solution were added to each well for a final concentration of 100 PM aspirin. The vehicle (DMSO) had no effect on platelet adhesion (data not shown). Cells were incubated with aspirin 45 min at 37°C followed by the removal of aspirin and three washes with HBSS before use in the adherence assay. This method of aspirin pretreatment of endothelial cells minimizes platelet exposure to aspirin. In a series of experiments designed to study the role of nitric oxide in thrombin-induced platelet adhesion to endothelium, monolayers were pretreated with the amino acid NG-monomethyl+arginine (L-NMMA; a kind gift of Drs. S. Moncada and
ADHESION
TO ENDOTHELIUM
H607
R. M. J. Palmer, Wellcome Laboratories, Beckenham, Kent, United Kingdom) or L-arginine (Sigma). Aspirin-treated or control endothelial monolayers were incubated with either IO PM arginine or 10 PM L-NMMA for 10 min at 37°C. The amino acids were then washed from the monolayer three times with HBSS, and the platelet adhesion assay was performed. PZateZet adhesion to cuZtured endothelium. Platelet adhesion was determined as previously described (13). lllIn-labeled sheep platelets were isolated as described above on Stractan II gradients. Confluent sheep pulmonary artery endothelial monolayers grown on 24-well culture plates were washed three times with HBSS. In a final volume of 1 ml, 4 x lo7 platelets were added to the wells and incubated for 60 min at 37°C and 5% CO, in a humidified” chamber. In some experiments, thrombin (2 U/ml; 1.75 x 1O-s M) was added to the monolayers for the duration of the experiment. After 60 min of incubation, monolayers were gently washed three times with HBSS to remove nonadherent platelets. Hydrolysis of the monolayers and adherent platelets was achieved by adding 1 N NaOH to the wells and incubating for at least 1 h at 37°C. The solubilized cells were placed in counting tubes and quantified for “‘In activity in a Packard Instruments gamma counter (Downers Grove, IL). Platelet adherence is expressed throughout as percent adherence and was calculated as the level of radioactivity of the solubilized sample divided by the total number of counts in the platelets added to each well, times 100. PZateZet adhesion to isolated perfused rat lungs. As described previously (28)) male Sprague-Dawley rats weighing between 250 and 300 g were anesthetized by peritoneal injection of pentobarbital sodium and tracheotomized. The animals were then exsanguinated from the abdominal aorta, and the thorax was opened. The lungs and heart were removed and suspended in the perfusion apparatus from the trachea tube. Ventilation with room air by means of a small animal respirator (Harvard Apparatus, South Natick, MA) was begun and maintained at 70 breaths/min and 1.5 ml/breath for the duration of the experiment. The pulmonary artery and the left atrium were cannulated. Perfusion of HBSS containing 0.5% bovine serum albumin (Sigma) at pH 7.4 (HBS&) was begun within 3 min of pneumothorax with the use of a peristaltic roller pump (Harvard Apparatus) at a constant flow rate of 14 ml/min. HBSSA was maintained in a water bath at 37°C for the duration of the experiment. The lungs were regularly bathed with warm saline. After a IO-min perfusion to clear blood from the lungs, reperfusion of HBSSA with or without 2 U/ml thrombin continued for 5 min. Fluid-phase thrombin was washed from the system by circulation of HBSSA for 2 min. After the postthrombin rinse with HBSS*, 7 x IO8 isolated 51Cr-labeled rat platelets, in a total volume of 25 ml, were recirculated through the lungs for 5 min. Lungs were then perfused with HBSSA for 2 min to remove nonadherent platelets. The entire perfusion system consisted of Silastic tubing and connections, and residual binding of platelets to this tubing was found to be nominal. At the end of the perfusions, the lungs were removed from the perfusion apparatus and minced. Radioactivity retained in the lungs was assessed by counting the lung tissue in a gamma counter. Percent adherence of platelets was calculated as the radioactivity in the lungs divided by the total amount of radioactivity in the original 7 x lo8 platelets added to the lungs, times 100. All lung perfusions were confined to ~25 min. Statistical analysis was performed by analysis of variance with Scheffe’s multiple range test for comparisons within several experimental groups.
Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
H608
cGMP
INHIBITS
PLATELET
ADHESION
RESULTS
To further establish the role of nitric oxide as a modulator of platelet adhesion, endothelial cell monolayers a comwere pretreated for 10 min with 10 pM L-NMMA, petitive inhibitor of nitric oxide production (Fig. IA). Adhesion to endothelial monolayers in the absence of thrombin was not affected by pretreatment of the endothelium with L-NMMA. However, thrombin-induced platelet adhesion was significantly increased after expo(P < 0.01). To sure of the endothelium to L-NMMA insure that this was not a nonspecific effect of L-NMMA on endothelium, the experiment was repeated with 10 PM L-arginine, and no significant change in basal or thrombin-stimulated platelet adhesion was found. Nitric oxide has been implicated in the induction of prostacyclin release (7). The experiment was repeated with aspirin-treated endothelial monolayers to insure that L-NMMA inhibition of platelet adhesion was indeed via the inhibition of nitric oxide production, rather than an inhibition of prostacyclin release stimulated by nitric oxide. L-NMMA treatment of monolayers had no effect on basal platelet adhesion (Fig. IB) but significantly increased (P < 0.05) thrombin-induced adhesion to aspirin-
A 0Z 30
0
CONTROL 10
uM
ARG
m
10
uM
L-M4vlA
cn
w I
25
CONTROL
B Z
0 30 i I
THROMBIN
-
15
0
CONTROL 10
dvl
ARG
m
10
uM
L-NMMA
2
20 Iw _I 15 P 4
10
-I-
ii h
0
5
0 CONTROL
THROMBIN
Fig. 1. A: lllIn-labeled sheep platelet adherence to control sheep pulmonary artery endothelial cell monolayers pretreated with 10 pM NGmonomethyl-L-arginine (L-NMMA), an inhibitor of nitric oxide production, or L-arginine (Arg), in the presence and absence (control) of 2 U/ml of thrombin. Each bar represents 24 or more determinations completed in 6 or more experiments. B: Wn-labeled sheep platelet adherence to aspirin-treated sheep pulmonary artery endothelial monolayers pretreated with either L-NMMA or L-arginine. Both basal and thrombin (2 U/ml)-induced platelet adhesion was assessed. Each bar represents 12 or more determinations completed in 3 or more experiments. Data were expressed as means t SE.
TO ENDOTHELIUM
treated monolayers. As in control experiments, L-arginine had no effect on either basal or thrombin-induced platelet adhesion to aspirin-treated endothelium. The effects of 8-bromo-cyclic nucleotides on platelet aggregate adhesion to monolayers. We have previously dem-
onstrated that platelet adhesion to endothelial monolayers is a combination of platelet-endothelial and plateletplatelet interactions, as aggregates were observed by scanning electron microscopy to be adherent to endothelial monolayers (13). Platelets were pretreated with either 1 or 10 mM 8-bromo-CAMP at room temperature before incubation with endothelial monolayers. Pretreatment of platelets with 8-bromo-CAMP treatment did not alter basal platelet adhesion. Thrombininduced platelet adhesion (25.21 t 2.7%) was significantly decreased by the pretreatment of platelets with 1 and 10 mM 8-bromo-CAMP to 16.24 t 2.3 (P < 0.01) and 15.18 t 1.5% (P < 0.05), respectively (Fig. 2A). Endothelial monolayers were fixed and the above experiments were repeated to insure that the effect of S-bromo-CAMP was mediated through platelets and not through an alteration of endothelial cell release of active modulators of platelet adhesion. Platelet adhesion to fixed monolayers was significantly increased by incubation with 2 U/ml of thrombin. In the absence of thrombin, basal platelet adhesion to fixed monolayers was not influenced by pretreatment of platelets with 1 or 10 mM 8-bromo-CAMP (Fig. 2B). Thrombin-induced platelet adhesion was not reduced by pretreatment of platelets with 1 mM 8-bromo-CAMP. However, thrombin-induced platelet adhesion to endothelial monolayers was significantly inhibited by platelet pretreatment with 10 mM 8-bromo-CAMP (P < 0.01). To prevent prostacyclin-mediated increases in endogenous platelet adenosine 3’,5’-cyclic monophosphate (CAMP), the adhesion assay was performed with aspirinpretreated endothelial monolayers. There was no effect of 8-bromo-CAMP on basal platelet adhesion to aspirintreated monolayers (Fig. 2C). Thrombin-induced platelet adhesion was not significantly altered by pretreatment of platelets with 1 mM 8-bromo-CAMP but was significantly decreased (P < 0.01) by pretreatment with 10 mM 8-bromo-CAMP (24.36 t 1.86%) compared with adhesion of untreated platelets (33.64 t 2.80%). Similar experiments were repeated with 8-bromocGMP. When platelets were pretreated with either 1 or 10 mM 8-bromo-cGMP, no change in basal platelet adhesion was seen. When thrombin was coincubated with platelets and 2 U/ml thrombin, platelet adhesion increased significantly (Fig. 3A). Thrombin-induced platelet adhesion was decreased when platelets were preincubated with 1 mM 8-bromo-cGMP (11.62 t 0.95%; P C 0.01) or 10 mM 8-bromo-cGMP (14.77 t 1.72%; P < 0.01).
To confirm that the effect of 8-bromo-cGMP was mediated through platelets and not through an alteration of release of endothelial cell modulators, endothelial monolayers were fixed and the above experiments performed. As before, platelet adhesion to fixed monolayers was significantly increased by 2 U/ml thrombin. Basal platelet
Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
cGMP
INHIBITS
PLATELET
6 u,
35
x
30
4
25
t;
20
ii +
15
$
10
$z
H609
TO ENDOTHELIUM
the presence of hemoglobin. Hemoglobin binds and inactivates nitric oxide, thereby limiting elevation of platelet guanosine 3’,5’-cyclic monophosphate (cGMP). Basal platelet adhesion in the presence of 10 PM hemoglobin was not affected by pretreatment of platelets with 1 or 10 mM 8-bromo-cGMP. Thrombin significantly increased platelet adhesion to monolayers in the presence of hemoglobin. This adhesion was not significantly decreased by
40
A
ADHESION
1Orrfblc~
5 0 CONTROL
THROMBIN
A
B
6 u35045 : 55
0
0
4o
m
Q
35
t;
30
-J
25
CONTROL
-
1KfvlcGtvP
Y
10
mM
cw
CONTROL
CONTROL
C
THROMBIN
B
0Z
50 55
i
0
THROMBIN
cobma
2
6
40
3
35
1
#
30
1olTfvlcAM>
6
25
fza1Nivlcw 45 rn40-mm’orrMcw
CONTROL l?-tvlcAh@
T
,
6
35
-
t;
30
-
d
25
-
2
20
-
’ 15
-
CONTROL
THROMBIN
0 CONTROL
THROMBIN
Fig. 2. A: effect of pretreatment of Wn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-CAMP (CAMP) on platelet adhesion to control sheep pulmonary artery endothelial cell monolayers. One and ten millimolar &bromo-CAMP-pretreatment of platelets significantly reduced adhesion to monolayers but had no effect on basal adhesion. B: sheep platelets for 10 min with 1 effect of pretreatment of lllIn-labeled or 10 mM 8bromo-CAMP on platelet adhesion to fixed sheep pulmonary artery endothelial cell monolayers. Endothelial cells were incubated with 1.5% buffered paraformaldehyde for 2 h at 4”C, and the fixative was washed off before platelet adhesion was assessed. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8bromo-CAMP. C: effect of pretreatment of Wn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromoCAMP on platelet adhesion to aspirin-treated sheep pulmonary artery endothelial cell monolayers. Endothelial cells were incubated with 1 mM aspirin for 45 min, and the aspirin was washed off before platelet adhesion was assessed. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8-bromoCAMP. In each of the above, each bar represents 24 or more determinations completed in 6 or more experiments. Data are expressed as means t SE.
adhesion to fixed monolayers was not influenced by pretreatment of platelets with I or 10 mM 8-bromo-cGMP (Fig. 3B). The decrease in 1 mM 8bromo-cGMP-treated platelet adhesion to control monolayers was not seen on fixed endothelial monolayers. However, thrombin-induced platelet adhesion to endothelial monolayers was significantly inhibited (P < 0.01) by platelet pretreatment with 10 mM 8-bromo-cGMP. Experiments were performed with 8bromo-cGMP in
CONTROL 1frtvlcw 1omfvlcw
L
30
d
25
2
20
J
15
a s
IO 5 0 CONTROL
THROMBIN
Fig. 3. A: effect of pretreatment of lllIn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-cGMP on platelet adhesion to control sheep pulmonary artery endothelial cell monolayers. Both basal adhesion and thrombin-stimulated platelet adhesion were assessed. One and ten millimolar 8-bromo-cGMP-pretreatment of platelets significantly reduced adhesion to monolayers but had no effect on basal adhesion. Each bar represents 16 or more determinations completed in 4 or more experiments. B: effect of pretreatment of l1 lIn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-cGMP on platelet adhesion to fixed sheep pulmonary artery endothelial cell monolayers. Endothelial cells were incubated with 1.5% buffered paraformaldehyde for 2 h at 4”C, and the fixative was washed off before platelet adhesion was assessed. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8-bromo-cGMP. Each bar represents 24 or more determinations completed in 6 or more experiments. C: effect of pretreatment of Wn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-cGMP on platelet adhesion to sheep pulmonary artery endothelial cell monolayers in presence of 10 PM hemoglobin. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8-bromo-cGMP. Each bar represents 8-16 determinations completed in 2-4 experiments. Data are expressed as means t SE.
Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
cGMP
THROMBIt’
d
INHIBITS
PLATELET
THROMBIN + IO mM
THROMBIN + IO mM
t-AMP
cGMP
ADHESION
Fig. 4. 51Cr-labeled platelet adhesion to thrombin-treated isolated perfused lungs was not affected by pretreatment of platelets for 10 min with 10 mM &bromo-CAMP but was significantly decreased when platelets were pretreated with 10 mM 8-bromo-cGMP. Each bar represents 4-8 perfused lungs. Data were expressed as means ,t SE.
preincubation of platelets with 1 mM 8-bromo-cGMP but was significantly decreased (P < 0.01) by 10 mM 8-bromo-cGMP (Fig. 3C). Differential effects of CAMP and cGMP on platelet adhesion to perfused lungs. Platelet adhesion to the is)olated
perfused rat lung was a measure of single -platelet adhesion to endothelium as demonstrated previously through scanning and high-voltage electron microscopy (28). Isolated lungs were perfused with platelets in the presence and absence of thrombin. The recirculation of 2 U/ml thrombin increased platelet adhesion from 4.37 t 0.90 to 10.49 t 1.72%. Th e pretreatment of platelets with 10 mM 8-bromo-CAMP had no effect (9.37 t 0.69%) on thrombin-induced pla .telet a.dhesion. Adhesion was significan .tly decreased (P < 0.05) when pl .atelets were exposed to 10 mM 8-bromo-cGMP (Fig. 4). We also determined whether platelet treatment with Iloprost, a potent stimulator of platelet CAMP, inhibited adhesion. Platelet aggregation was assessed after platelets were pretreated wi th 10 nM Iloprost, a stab1.eprostacyclin analogue. To verify the efficacy of Iloprost pretreatment, platelet aggregation was assessed (Fig. 5, inset). Platelets W 0
7
W20 LL W I cl 15 < ly IO W I
M. VENTURINI, LISA K. WESTON, AND JOHN E. KAPLAN of Physiology and Cell Biology, Albany Medical College, Albany, New M., Lisa K. Weston, and John E.
but not CAMP, inhibits thrombininduced platelet adhesion to pulmonary vascular endothelium. Am. J. Physiol. 263 (Heart Circ. Physiol. 32): H606-H612, 1992. -We have compared the effects of intracellular pathways initiated by nitric oxide and prostacyclin on thrombin-induced platelet adhesion to endothelial cells. Platelet aggregate adhesion was enhanced when endothelial monolayers were pretreated with NG-monomethyl-L-arginine (L-NMMA), an inhibitor of nitric oxide production. In addition, decreased platelet aggregate adhesion was seen when platelets were pretreated with 8bromoadenosine 3’,5’-cyclic monophosphate (8bromoCAMP) or 8bromoguanosine 3’,5’-cyclic monophosphate (8 bromo-cGMP). Single platelet adhesion in isolated perfused lungs under flow conditions in the presence of shear was also assessed. Pretreatment of platelets with either Iloprost, in a dose sufficient to decrease platelet aggregation, or 8-bromoCAMP did not affect platelet adhesion. However, pretreatment of platelets with 8-bromo-cGMP significantly reduced single platelet adhesion to endothelium. These studies illustrate that nitric oxide inhibits platelet adhesion to endothelium in the presence of shear. They further indicate that prostacyclin is also a regulator of this response but has effects more specifically related to the inhibition of platelet aggregation than plateletendothelium interactions. thrombosis;
nitric
oxide;
prostacyclin
ENDOTHELIAL CELLS ARE NORMALLY effective in preventing platelet adhesion. However, thrombin induces platelet adhesion to endothelium (11) and enhances release of endothelium-derived relaxing factor (EDRF) from endothelium (21). EDRF is widely believed to be nitric oxide although recent evidence suggests that other molecules such as S-nitrosocysteine may contribute to this activity (16,19). Nitric oxide downregulates platelet adhesion to endothelium in several models of thrombosis (20, 25, 27). The effect of endothelial prostacyclin on platelet adhesion to endothelium is less clear. Prostacyclin release has been correlated with endothelial resistance to platelet adhesion by some (5, 9), but this has been disputed by others (4, 12) who found no change in thrombin-induced platelet adhesion when prostacyclin generation was inhibited by aspirin pretreatment of endothelium. We have previously developed two models of platelet adhesion that examine platelet-endothelial cell interactions separately from platelet-platelet interactions. Thrombin induces adhesion of sheep platelet aggregates to cultured sheep pulmonary artery endothelial cell monolayers (13) and both nitric oxide (27) and prostacyclin (13) inhibit platelet accumulation on the endothelial surface. In a second model, thrombin pretreatment of isolated perfused rat lungs induced rat platelet adherence as single activated platelets (28). Nitric oxide inhibited platelet adhesion in this model; however, there was no change in thrombin-stimulated platelet adhesion H606
York 12208
to perfused lungs when animals were treated . with aspirin before removal of lu ngs for perfusion. The differences between the antiadhesive natures of nitric oxide and prostacyclin are potentially important in understanding the etiology of thrombosis and atherosclerosis. In the current study, we have further examined the role of nitric oxide and prostacyclin in thrombininduced platelet adhesion to endothelium by examining the platelet intracellular signaling mechanisms triggered by these factors. We hypothesize that prostacyclin does not directly inhibit platelet adhesion to endothelium. Rather, prostacyclin inhibits platelet-platelet interactions and decreases platelet adhesion in models that involve platelet aggregate adhesion, whereas nitric oxide effectively inhibits both platelet adhesion to endothelium and platelet aggregation. METHODS
Platelet isolation. Platelets were isolated from rat or sheep blood using a modification of the method of Corash et al. (3,18) utilizing isopycnic centrifugation on polyarabinogalactan gradients (Stractan II, Champion International, Tacoma, WA). 51Crlabeled rat platelets were used in the isolated perfused rat lung assay, and IllIn-labeled sheep platelets were used in the in vitro platelet adherence assay. Whole blood was drawn from the inferior vena cava of a rat or the jugular vein of a sheep through a 19gauge needle into a plastic syringe containing two parts 19% sodium citrate per 100 parts blood. Platelet-rich plasma (PRP) was prepared by adding 1 ml of a buffered saline glucosecitrate solution (BSG-citrate) containing (in M) 0.117 NaCl, 0.0136 sodium citrate, 0.0111 glucose, 0.0086 Na,HPO,, and 0.0016 KH,P04, pH 7.4, to 9 ml of blood and centrifuging at 850 g for 8 min. This procedure was performed twice for rat platelets. Rat platelets were labeled with 51Cr by adding 51Cr as sodium chromate (New England Nuclear, Boston, MA) to BSGcitrate-diluted PRP (2 &i/5 ml) and incubating for 60 min at 37°C. Sheep platelets were labeled with “‘In by adding ‘llInoxine (10 &i/ml; Amersham, Arlington Heights, IL) in the same manner. The BSG-citrate-diluted PRP was layered over a discontinuous density gradient consisting of 4 ml of 20% Stractan II and 3 ml of 10% Stractan II in a 15-ml conical tube. The tube was centrifuged for 15 min at 1,450 g, after which the plasma-BSG-citrate and 10% Stractan II layers were aspirated. The platelet layer was then removed and washed free of Stractan II by centrifugation for IO min at 600 g in excess BSGcitrate. The platelet pellet was resuspended in modified (Ca2+and Mg2+-free) Tyrode solution [containing (in M) 0.137 NaCl, 0.0030 KCl, 0.012 NaHCO,, 0.006 glucose, and 0.004 NaH,PO,, pH 7.41. This method of isolation removes plasma proteins while causing minimal damage to platelets. All procedures were carried out using plastic or siliconized glassware at room temperature. Platelets were counted using a hemacytometer (American Optical, Buffalo, NY) and Nikon phase contrast microscope (Nikon, Garden City, NY) after the method of Bjorkman (1) In some experiments, 8-bromoadenosine 3’,5’-cyclic monophosphate (8-bromo-CAMP) or 8-bromoguanosine 3’,5’-cyclic
0363-6135/92 $2.00 Copyright 0 1992 the American Physiological Society Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
cGMP
INHIBITS
PLATELET
monophosphate (8bromo-cGMP; both from Sigma, St. Louis, MO) was used to increase intracellular concentrations of platelet cyclic nucleotides. Platelets were isolated and labeled as described above and then incubated with 1 or 10 mM &bromocGMP or 8bromo-CAMP for 10 min. These pretreated platelets were then used in the in vitro or isolated lung platelet adhesion assays. Platelets were incubated with 8bromo-derivatives in a small volume and then diluted before exposure to endothelium. This reduced the concentration of 8-bromo-cyclic nucleotides on the endothelium to 25 PM, minimizing the direct effects of these compounds on endothelial function. This concentration of 8-bromo-cyclic nucleotides when added to the monolayers before, or concurrently with, platelets that had not previously been incubated with cyclic nucleotides had no effect on the adherence of platelets in the presence or absence of a-thrombin (data not shown). In some cases, platelets were pretreated with 10 nM Iloprost, a stable prostacyclin analogue, for 10 min before aggregation. Iloprost was a kind gift of Dr. G. Rubanyi, Berlex Pharmaceuticals, Cedar Knolls, NJ. Platelet aggregation. In vitro aggregation of rat platelets was measured according to the method of Born (2) using a ChronoLog Aggregation Module (Chrono-Log, Havertown, PA). Aggregometry was performed at 37°C in Hanks’ balanced salt solution (HBSS; GIBCO, Grand Island, NY) with Ca2+ and Mg2+ using a stir bar rotating at 900 revolutions/min. The final volume of the assay mixture was 0.5 ml. The slope of the rate of aggregation was recorded, and a maximal scale deflection was elicited in response to 0.1 U of cr-thrombin. Thrombin preparation. Human cu-thrombin was prepared by extracting prothrombin from Cohn fraction III paste as described elsewhere (26) and was a generous gift of Dr. John W. Fenton II, Wadsworth Center for Laboratories and Research, Albany, NY. EndotheZiaZ cell cukure. Primary cultures of sheep pulmonary artery endothelial cells were isolated and grown using established techniques (6). Endothelial cells at 3 to 30 population doublings were seeded onto 24-well plates (Corning, Chicago, IL) and grown to confluence in Dulbecco’s modified Eagle’s medium (DMEM; GIBCO) containing 20% fetal bovine serum (Hyclone Labs, Logan, UT), nonessential amino acids (GIBCO), and 50 pg/ml gentamicin sulfate (M.A. Bioproducts, Walkersville, MD) in a 5% CO, incubator at 37°C. Only confluent endothelial cell monolayers were used for the platelet adherence assays. Endothelial monolayers were generously provided by Dr. Peter J. Del Vecchio, Department of Ophthalmology, Albany Medical College. Some monolayers were fixed by incubation with 500 ~1 of an osmotically balanced 1.5% paraformaldehyde solution in HBSS for 2 h at 4°C. The fixative was then removed, and the monolayers were washed three times with HBSS before use in the adherence assays. Some endothelial monolayers were pretreated with aspirin. These monolayers were washed three times with HBSS, and 1 ml of HBSS was added to each well. One hundred milligrams of aspirin (acetylsalicylic acid; Sigma) were dissolved in 450 ~1 of dimethyl sulfoxide (DMSO; Sigma). Just before use, 500 ~1 of HBSS were added, and 10 ~1 of this solution were added to each well for a final concentration of 100 PM aspirin. The vehicle (DMSO) had no effect on platelet adhesion (data not shown). Cells were incubated with aspirin 45 min at 37°C followed by the removal of aspirin and three washes with HBSS before use in the adherence assay. This method of aspirin pretreatment of endothelial cells minimizes platelet exposure to aspirin. In a series of experiments designed to study the role of nitric oxide in thrombin-induced platelet adhesion to endothelium, monolayers were pretreated with the amino acid NG-monomethyl+arginine (L-NMMA; a kind gift of Drs. S. Moncada and
ADHESION
TO ENDOTHELIUM
H607
R. M. J. Palmer, Wellcome Laboratories, Beckenham, Kent, United Kingdom) or L-arginine (Sigma). Aspirin-treated or control endothelial monolayers were incubated with either IO PM arginine or 10 PM L-NMMA for 10 min at 37°C. The amino acids were then washed from the monolayer three times with HBSS, and the platelet adhesion assay was performed. PZateZet adhesion to cuZtured endothelium. Platelet adhesion was determined as previously described (13). lllIn-labeled sheep platelets were isolated as described above on Stractan II gradients. Confluent sheep pulmonary artery endothelial monolayers grown on 24-well culture plates were washed three times with HBSS. In a final volume of 1 ml, 4 x lo7 platelets were added to the wells and incubated for 60 min at 37°C and 5% CO, in a humidified” chamber. In some experiments, thrombin (2 U/ml; 1.75 x 1O-s M) was added to the monolayers for the duration of the experiment. After 60 min of incubation, monolayers were gently washed three times with HBSS to remove nonadherent platelets. Hydrolysis of the monolayers and adherent platelets was achieved by adding 1 N NaOH to the wells and incubating for at least 1 h at 37°C. The solubilized cells were placed in counting tubes and quantified for “‘In activity in a Packard Instruments gamma counter (Downers Grove, IL). Platelet adherence is expressed throughout as percent adherence and was calculated as the level of radioactivity of the solubilized sample divided by the total number of counts in the platelets added to each well, times 100. PZateZet adhesion to isolated perfused rat lungs. As described previously (28)) male Sprague-Dawley rats weighing between 250 and 300 g were anesthetized by peritoneal injection of pentobarbital sodium and tracheotomized. The animals were then exsanguinated from the abdominal aorta, and the thorax was opened. The lungs and heart were removed and suspended in the perfusion apparatus from the trachea tube. Ventilation with room air by means of a small animal respirator (Harvard Apparatus, South Natick, MA) was begun and maintained at 70 breaths/min and 1.5 ml/breath for the duration of the experiment. The pulmonary artery and the left atrium were cannulated. Perfusion of HBSS containing 0.5% bovine serum albumin (Sigma) at pH 7.4 (HBS&) was begun within 3 min of pneumothorax with the use of a peristaltic roller pump (Harvard Apparatus) at a constant flow rate of 14 ml/min. HBSSA was maintained in a water bath at 37°C for the duration of the experiment. The lungs were regularly bathed with warm saline. After a IO-min perfusion to clear blood from the lungs, reperfusion of HBSSA with or without 2 U/ml thrombin continued for 5 min. Fluid-phase thrombin was washed from the system by circulation of HBSSA for 2 min. After the postthrombin rinse with HBSS*, 7 x IO8 isolated 51Cr-labeled rat platelets, in a total volume of 25 ml, were recirculated through the lungs for 5 min. Lungs were then perfused with HBSSA for 2 min to remove nonadherent platelets. The entire perfusion system consisted of Silastic tubing and connections, and residual binding of platelets to this tubing was found to be nominal. At the end of the perfusions, the lungs were removed from the perfusion apparatus and minced. Radioactivity retained in the lungs was assessed by counting the lung tissue in a gamma counter. Percent adherence of platelets was calculated as the radioactivity in the lungs divided by the total amount of radioactivity in the original 7 x lo8 platelets added to the lungs, times 100. All lung perfusions were confined to ~25 min. Statistical analysis was performed by analysis of variance with Scheffe’s multiple range test for comparisons within several experimental groups.
Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
H608
cGMP
INHIBITS
PLATELET
ADHESION
RESULTS
To further establish the role of nitric oxide as a modulator of platelet adhesion, endothelial cell monolayers a comwere pretreated for 10 min with 10 pM L-NMMA, petitive inhibitor of nitric oxide production (Fig. IA). Adhesion to endothelial monolayers in the absence of thrombin was not affected by pretreatment of the endothelium with L-NMMA. However, thrombin-induced platelet adhesion was significantly increased after expo(P < 0.01). To sure of the endothelium to L-NMMA insure that this was not a nonspecific effect of L-NMMA on endothelium, the experiment was repeated with 10 PM L-arginine, and no significant change in basal or thrombin-stimulated platelet adhesion was found. Nitric oxide has been implicated in the induction of prostacyclin release (7). The experiment was repeated with aspirin-treated endothelial monolayers to insure that L-NMMA inhibition of platelet adhesion was indeed via the inhibition of nitric oxide production, rather than an inhibition of prostacyclin release stimulated by nitric oxide. L-NMMA treatment of monolayers had no effect on basal platelet adhesion (Fig. IB) but significantly increased (P < 0.05) thrombin-induced adhesion to aspirin-
A 0Z 30
0
CONTROL 10
uM
ARG
m
10
uM
L-M4vlA
cn
w I
25
CONTROL
B Z
0 30 i I
THROMBIN
-
15
0
CONTROL 10
dvl
ARG
m
10
uM
L-NMMA
2
20 Iw _I 15 P 4
10
-I-
ii h
0
5
0 CONTROL
THROMBIN
Fig. 1. A: lllIn-labeled sheep platelet adherence to control sheep pulmonary artery endothelial cell monolayers pretreated with 10 pM NGmonomethyl-L-arginine (L-NMMA), an inhibitor of nitric oxide production, or L-arginine (Arg), in the presence and absence (control) of 2 U/ml of thrombin. Each bar represents 24 or more determinations completed in 6 or more experiments. B: Wn-labeled sheep platelet adherence to aspirin-treated sheep pulmonary artery endothelial monolayers pretreated with either L-NMMA or L-arginine. Both basal and thrombin (2 U/ml)-induced platelet adhesion was assessed. Each bar represents 12 or more determinations completed in 3 or more experiments. Data were expressed as means t SE.
TO ENDOTHELIUM
treated monolayers. As in control experiments, L-arginine had no effect on either basal or thrombin-induced platelet adhesion to aspirin-treated endothelium. The effects of 8-bromo-cyclic nucleotides on platelet aggregate adhesion to monolayers. We have previously dem-
onstrated that platelet adhesion to endothelial monolayers is a combination of platelet-endothelial and plateletplatelet interactions, as aggregates were observed by scanning electron microscopy to be adherent to endothelial monolayers (13). Platelets were pretreated with either 1 or 10 mM 8-bromo-CAMP at room temperature before incubation with endothelial monolayers. Pretreatment of platelets with 8-bromo-CAMP treatment did not alter basal platelet adhesion. Thrombininduced platelet adhesion (25.21 t 2.7%) was significantly decreased by the pretreatment of platelets with 1 and 10 mM 8-bromo-CAMP to 16.24 t 2.3 (P < 0.01) and 15.18 t 1.5% (P < 0.05), respectively (Fig. 2A). Endothelial monolayers were fixed and the above experiments were repeated to insure that the effect of S-bromo-CAMP was mediated through platelets and not through an alteration of endothelial cell release of active modulators of platelet adhesion. Platelet adhesion to fixed monolayers was significantly increased by incubation with 2 U/ml of thrombin. In the absence of thrombin, basal platelet adhesion to fixed monolayers was not influenced by pretreatment of platelets with 1 or 10 mM 8-bromo-CAMP (Fig. 2B). Thrombin-induced platelet adhesion was not reduced by pretreatment of platelets with 1 mM 8-bromo-CAMP. However, thrombin-induced platelet adhesion to endothelial monolayers was significantly inhibited by platelet pretreatment with 10 mM 8-bromo-CAMP (P < 0.01). To prevent prostacyclin-mediated increases in endogenous platelet adenosine 3’,5’-cyclic monophosphate (CAMP), the adhesion assay was performed with aspirinpretreated endothelial monolayers. There was no effect of 8-bromo-CAMP on basal platelet adhesion to aspirintreated monolayers (Fig. 2C). Thrombin-induced platelet adhesion was not significantly altered by pretreatment of platelets with 1 mM 8-bromo-CAMP but was significantly decreased (P < 0.01) by pretreatment with 10 mM 8-bromo-CAMP (24.36 t 1.86%) compared with adhesion of untreated platelets (33.64 t 2.80%). Similar experiments were repeated with 8-bromocGMP. When platelets were pretreated with either 1 or 10 mM 8-bromo-cGMP, no change in basal platelet adhesion was seen. When thrombin was coincubated with platelets and 2 U/ml thrombin, platelet adhesion increased significantly (Fig. 3A). Thrombin-induced platelet adhesion was decreased when platelets were preincubated with 1 mM 8-bromo-cGMP (11.62 t 0.95%; P C 0.01) or 10 mM 8-bromo-cGMP (14.77 t 1.72%; P < 0.01).
To confirm that the effect of 8-bromo-cGMP was mediated through platelets and not through an alteration of release of endothelial cell modulators, endothelial monolayers were fixed and the above experiments performed. As before, platelet adhesion to fixed monolayers was significantly increased by 2 U/ml thrombin. Basal platelet
Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
cGMP
INHIBITS
PLATELET
6 u,
35
x
30
4
25
t;
20
ii +
15
$
10
$z
H609
TO ENDOTHELIUM
the presence of hemoglobin. Hemoglobin binds and inactivates nitric oxide, thereby limiting elevation of platelet guanosine 3’,5’-cyclic monophosphate (cGMP). Basal platelet adhesion in the presence of 10 PM hemoglobin was not affected by pretreatment of platelets with 1 or 10 mM 8-bromo-cGMP. Thrombin significantly increased platelet adhesion to monolayers in the presence of hemoglobin. This adhesion was not significantly decreased by
40
A
ADHESION
1Orrfblc~
5 0 CONTROL
THROMBIN
A
B
6 u35045 : 55
0
0
4o
m
Q
35
t;
30
-J
25
CONTROL
-
1KfvlcGtvP
Y
10
mM
cw
CONTROL
CONTROL
C
THROMBIN
B
0Z
50 55
i
0
THROMBIN
cobma
2
6
40
3
35
1
#
30
1olTfvlcAM>
6
25
fza1Nivlcw 45 rn40-mm’orrMcw
CONTROL l?-tvlcAh@
T
,
6
35
-
t;
30
-
d
25
-
2
20
-
’ 15
-
CONTROL
THROMBIN
0 CONTROL
THROMBIN
Fig. 2. A: effect of pretreatment of Wn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-CAMP (CAMP) on platelet adhesion to control sheep pulmonary artery endothelial cell monolayers. One and ten millimolar &bromo-CAMP-pretreatment of platelets significantly reduced adhesion to monolayers but had no effect on basal adhesion. B: sheep platelets for 10 min with 1 effect of pretreatment of lllIn-labeled or 10 mM 8bromo-CAMP on platelet adhesion to fixed sheep pulmonary artery endothelial cell monolayers. Endothelial cells were incubated with 1.5% buffered paraformaldehyde for 2 h at 4”C, and the fixative was washed off before platelet adhesion was assessed. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8bromo-CAMP. C: effect of pretreatment of Wn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromoCAMP on platelet adhesion to aspirin-treated sheep pulmonary artery endothelial cell monolayers. Endothelial cells were incubated with 1 mM aspirin for 45 min, and the aspirin was washed off before platelet adhesion was assessed. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8-bromoCAMP. In each of the above, each bar represents 24 or more determinations completed in 6 or more experiments. Data are expressed as means t SE.
adhesion to fixed monolayers was not influenced by pretreatment of platelets with I or 10 mM 8-bromo-cGMP (Fig. 3B). The decrease in 1 mM 8bromo-cGMP-treated platelet adhesion to control monolayers was not seen on fixed endothelial monolayers. However, thrombin-induced platelet adhesion to endothelial monolayers was significantly inhibited (P < 0.01) by platelet pretreatment with 10 mM 8-bromo-cGMP. Experiments were performed with 8bromo-cGMP in
CONTROL 1frtvlcw 1omfvlcw
L
30
d
25
2
20
J
15
a s
IO 5 0 CONTROL
THROMBIN
Fig. 3. A: effect of pretreatment of lllIn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-cGMP on platelet adhesion to control sheep pulmonary artery endothelial cell monolayers. Both basal adhesion and thrombin-stimulated platelet adhesion were assessed. One and ten millimolar 8-bromo-cGMP-pretreatment of platelets significantly reduced adhesion to monolayers but had no effect on basal adhesion. Each bar represents 16 or more determinations completed in 4 or more experiments. B: effect of pretreatment of l1 lIn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-cGMP on platelet adhesion to fixed sheep pulmonary artery endothelial cell monolayers. Endothelial cells were incubated with 1.5% buffered paraformaldehyde for 2 h at 4”C, and the fixative was washed off before platelet adhesion was assessed. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8-bromo-cGMP. Each bar represents 24 or more determinations completed in 6 or more experiments. C: effect of pretreatment of Wn-labeled sheep platelets for 10 min with 1 or 10 mM 8-bromo-cGMP on platelet adhesion to sheep pulmonary artery endothelial cell monolayers in presence of 10 PM hemoglobin. Thrombin-induced platelet adhesion was significantly decreased by pretreatment of platelets with 10 mM 8-bromo-cGMP. Each bar represents 8-16 determinations completed in 2-4 experiments. Data are expressed as means t SE.
Downloaded from www.physiology.org/journal/ajpheart at SUNY Buffalo (128.205.114.091) on March 17, 2019.
cGMP
THROMBIt’
d
INHIBITS
PLATELET
THROMBIN + IO mM
THROMBIN + IO mM
t-AMP
cGMP
ADHESION
Fig. 4. 51Cr-labeled platelet adhesion to thrombin-treated isolated perfused lungs was not affected by pretreatment of platelets for 10 min with 10 mM &bromo-CAMP but was significantly decreased when platelets were pretreated with 10 mM 8-bromo-cGMP. Each bar represents 4-8 perfused lungs. Data were expressed as means ,t SE.
preincubation of platelets with 1 mM 8-bromo-cGMP but was significantly decreased (P < 0.01) by 10 mM 8-bromo-cGMP (Fig. 3C). Differential effects of CAMP and cGMP on platelet adhesion to perfused lungs. Platelet adhesion to the is)olated
perfused rat lung was a measure of single -platelet adhesion to endothelium as demonstrated previously through scanning and high-voltage electron microscopy (28). Isolated lungs were perfused with platelets in the presence and absence of thrombin. The recirculation of 2 U/ml thrombin increased platelet adhesion from 4.37 t 0.90 to 10.49 t 1.72%. Th e pretreatment of platelets with 10 mM 8-bromo-CAMP had no effect (9.37 t 0.69%) on thrombin-induced pla .telet a.dhesion. Adhesion was significan .tly decreased (P < 0.05) when pl .atelets were exposed to 10 mM 8-bromo-cGMP (Fig. 4). We also determined whether platelet treatment with Iloprost, a potent stimulator of platelet CAMP, inhibited adhesion. Platelet aggregation was assessed after platelets were pretreated wi th 10 nM Iloprost, a stab1.eprostacyclin analogue. To verify the efficacy of Iloprost pretreatment, platelet aggregation was assessed (Fig. 5, inset). Platelets W 0
7
W20 LL W I cl 15 < ly IO W I
