THROMBOSIS RESEARCH 68; 145-155,1992 00493848192 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.

EFFECT OF AJOENE, THE MAJOR ANTIPLATELET COMPOUND FROM GARLIC, ON PLATELET THROMBUS FORMATION

Rafael Apitz-Castro', Juan J. Badimon' and Lina Badimon" 'Lab. Trombosis Experimental, IVIC, Ap. 21827, Caracas 1020-A, Venezuela, and 2Cardicvascular Bioloyy P.esearch,Cardiac Unit, Massachusetts General Hospital, Harvard Medical School (Received 22.6.1992; accepted in revised form 24.8.1992 by Editor S. Moncada) (Received by Executive Editorial Office 10.9.) 992)

ABSTRACT Ajoene, (E,Z)-4,5,9-trithiadodeca-1,6,11-trieneg-oxide, is a potent antiplatelet compound isolated from alcoholic extracts of garlic (Allium sativum). Ajoene rever$ibly inhibits in vitro platelet aggregation as well as release reaction induced by all known agonists. We used a well characterized cylindrical perfusion chamber to study the effect of ajoene on platelet deposition onto physiological substrates such as pig aortic subendothelium and tunica media as a model of mildly and severely damaged vessel wall respectively. Experiments were performed under flow conditions of high and low shear rate that mimic laminar blood flow in small and medium size arteries (1690 se@ and 212 se&). Our results indicate that ajoene prevents thrombus formation both at low and high shear rate in titrated whole blood. The inhibitory effect of ajoene on platelet-thrombus formation seems to be dependent on its inhibition of fibrinogen binding, since significantly higher concentrations of ajoene are needed to affect von Willebrand factor binding to GPIIb/IIIa receptors. Further, a joene does not impair Ristocetin-induced platelet agglutination, mediated by GPIb, Our results suggest that ajoene may be useful for the acute prevention of thrombus formation induced by vascular damage.

Key words: Ajoene, thrombus formation, vessel wall, platele,ts IR. Apitz-Castro is a Fellow of the John Simon Guggenheim Memorial Foundation. Corresponding author. ??

145

146

AJOENE EFFECT ON THROMBUS

Vol. 68, No. 2

INTRODUCTION It has previously been shown that ajoene, the major antiplatelet compound derived from garlic, inhibits platelet aggregation induced by all known agonists (1,2), as well as the binding of fibrinogen to activated platelets (3,4). Administered intravenously to dogs under extracorporeal circulation, ajoene prevents thrombocytopenia due to platelet activation induced by the extracorporeal artificial circuit (5). Under such conditions, the inhibitory effect of ajoene is reversible, and complete recovery of platelet function is regained after about three hours of ajoene administration. Platelet-thrombus formation at the site of vessel wall injury is modulated by the nature of the exposed substrate (related to the severity of the injury), the local rheology and the blood factors. Exposure of medial constituents (as a consequence of severe injury) induces mura; thrombosis (6-12). In contrast, mild or endothelial injury induces the recruitment of mono- or bi-layers without significant thrombus formation after more than five to ten minutes (6-11). Our aim was to study the effect of ajoene on platelet-deposition and thrombus formation on mildly and severely damaged vessel wall under controlled rheological conditions. Our results indicate that ajoene efficiently prevents platelet-thrombus formation both at low and high shear rate conditions. They also suggest that ajoene itself, or ajoene-derived compounds may be useful for the acute prevention of platelet-thrombus formation after vascular damage. MATERIAL AND METHODS The animal model for the study was the Yorkshire albino pig (body weight 22 4 3 Kg). They were obtained from a single local farm and used as blood donors. The animals were housed in the facilities of the Center for Laboratory Animal Research of the Mount Sinai Hospital (New York). Pigs were individually caged in a light, temperature (20 + 2 OC) and humidity controlled enviroment. To eliminate the stress related to transportation and change of habitat, all experiments were initiated one week after arrival of the animals. All procedures performed in this study were approved by the appropriate institutional guidelines and followed the American Heart Association Guidelines for Animal Research. Platelet Labelina with L'lIndium. After overnight fasting, blood (43 mL) was withdrawn from the femoral artery, with minimal trauma, into a syringe containing acid citrate dextrose (7 mL). Platelet rich plasma was prepared by centrifugation, and platelets were labelled with 'llIndium-tropolone (300-500 tici), as previously described (6). The efficiency of labeling in platelet-rich plasma was 48 + 4% . An average of 2.1 x lo9 labeled platelets/ml were re-injected to the same animal in a final volume of 4.5 - 5.0 mL. The entire procedure required about 2.5 hours. Labeled Whole Blood. Eighteen to 24 hours later, 45 mL of blood was withdrawn from the contra-lateral femoral artery or the carotid artery and collected in plastic tubes containing 5 mL of 90 mu

Vol. 68, No. 2

sodium citrate. cell count (5.28 fibrinogen level the start of the

AJOENE EFFECT ON THROMBUS

147

Platelet count (3.85 + 0.5 x 108/mL), red blood f 0.48 x 109/mL), hematocrit (25.3 + 3.8%) and (296 + 20 mg/dl) were routinely determined before experiment.

We used a previously described Perfusion Chamber and Substrates. chamber that mimics the cylindrical shape of blood In this chamber, the substrate is placed in a vessels (6,7). lateral position, in such a way that the surface to be tested constitutes an integral part of the channel wall and is directly Two chambers of different internal exposed to the flowing blood. diameters (1.0 and 2.0 mm) are sufficient to allow a broad range of wall shear rates on the substrate, with only moderate changes' in average blood flow rate. In the experiments to be described, blood was perfused at local shear rates of 212 see-l and 1689 see-'. These shear rates cover the range from large to medium size arteries (1.3).

perfusion

Pig aortic subendothelium and tunica media were used as model substrates of mildly and severely damaged vessel wall respectively. were obtained frozen from Pel-Freez Aortas from normal pigs Biologicals (Rogers, AR) and kept at -70 OC until used. The day of the experiment, aortas thawed in cold phosphate-buffer saline (PBS, and cut into PH = 7.2) were freed from excess adventitia rectangular pieces (3 x 0.5 cm). The thawed aortas (devoid of the endothelial layer) were used as such, without further manipulation, substrate (mild vessel wall injury). as subendothelium For experiments using tunica media as substrate (deep vessel wall damage), the intima with a thin portion of sub-adjacent media was lifted and removed by peeling. The exposed media was used as substrate in the chamber. -. The day of the experiment, blood was drawn from a previously labeled animal and collected in 90 mM sodium citrate (9:l v/v). Blood was distributed in 30 mL aliquots in plastic tubes, and kept at room temperature until perfusion begun. Ajoene, dissolved in Intralipid as previously described (4) was added to pre-warmed blood (37 'C) under continuous stirring. Ajoene from a 1 M stock solution in ethanol was appropriately diluted in Intralipid to a concentration such that 0.6 mL of this solution into 30 mL of whole blood gave a final concentration of 300 and 600 PM respectively. Control experiments were performed using the same amount of the vehicle (ethanol) dissolved in Intralipid. After addition of ajoene or vehicle! blood was maintained at 37 OC for 10 minutes under occasional stirring, and then perfusion was initiated and maintained for a period of 5 minutes at the preselected shear rate (peristaltic pump, Master-flex, Model 7013). Substrates were mounted in the chamber pre-filled with PBS. Prior to the start of blood circulation, Vassar-saline solution (37 OC) was circulated through the chamber for one minute. After the end of blood perfusion (5 min), Vassar-saline was again perfused for 30 seconds under identical flow conditions in order to clear away remnent blood. The changes from buffer to blood and v?ce versa were done manually with a three-way valve, without introduction of stasis or air bubbles in the chamber. The chamber was immed$ately disassembled and the perfused segments were fixed in a mixture of

148

AJOENE EFFECT ON THROMBUS

Vol. 68, No. 2

3% glutaraldehyde in 0.2 M cacodylate buffer (pi 7.4). 'l'Indium radioactivity was measured in a gamma-counter and used for quantitation of deposited platelets in the substrate as previously described (6). Platelet Count and hematocrit, determined at the end of each experiment showed no significant statistical difference with the pre-perfusion values. .

.

riflcation of o rclne vWF. van Willebrand Factor (VWF) was purified from norial pig plasma, obtained frozen from Pel-Freez Biologicals, (Rogers, AR). Blood for this purpose was collected in 3.8% sodium citrate (1:9) containing phenylmethyl sulfonyl fluoride (PMSF) at a final concentration of 2 mM. The frozen plasma was partially thawed overnight in the cold room at 4' C, until soft. To this partially thawed plasma, cold ethanol (95% in 0.02 M Tris pH 7.0, kept at -5O C) was added dropwise to a final concentration of 3%. After standing at -2O C, the ethanol-plasma suspension was centrifuged at 5ijOZ, xg for i0 minutes. The precipitate, washed with cold 8% ethanol in 0.02 M Tris pH 7.0, was extracted by dissolving it in Tris 0.02 M, pH 7.0, containing 0.02 M Eaminocaproic acid (EACA). From this step on, the procedure was carried out at room temp. This solution was sequentially adsorbed with a) aluminum hydroxide and b) Bentonite powder, in order to remove factors II, IX, X, and most of the contaminant fibrinogen. To the resulting solution, 0.5 M sodium citrate was added at a final concentration of 0.02 M and the pH adjusted at 6.15 with 0.02 Most of the remaining fibrinogen was removed by M citric acid. precipitation with polyethylene glycol (PEG, Carbowax 3350, Fisher P-146) at a final Concentration of 5%. The supernatant was further precipitated with PEG to a final concentration of 12%, and the resulting precipitate, dissolved in a minimal volume of 0.05 M Tris-0.15 M NaCl, pH 7.1 was fractionated by gel filtration through Agarose A-15m (BioRad). Material eluting from the column with the void volume was pooled and kept in aliquots at -70' C until use. Polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate, under reducing conditions, resulted in a single apparent molecular mass of 250 kD. protein band with Immunodiffusion of the purified vWF against two different antifibrinogen antibodies showed no detectable reactivity, indicating that the preparation was free of contaminant fibrinogen. The purified porcine vWF induced platelet aggregation of human PRP in the absence of ristocetin. It supported ristocetin-induced Assay of agglutination of paraformaldehyde-fixed platelets. Ristocetin-Willebrand cofactor activity, as described by Olson et al., (13) gave an average of 3.8 _+0.07 pg/unit (n = 3), equivalent to a specific activity of 263 units/mg. Labelina of oorcine von Willebrand Factor. Purified porcine vWF was labeled with 12Y using the Iodogen method (14). The specific radioactivity obtained was on average 1.5 x 10' cpm/mg, >95% was precipitable with 10% tricloroacetic acid. Bindina of oorcine vWF to ADP-stimulated washed olatelets. Washed porcine platelets were prepared as described (15), but after the second centrifugation, the platelet pellet was gently resuspended in the washing buffer (one tenth the original volume of PRP) and gel filtered through a column of Sepharose 4B equilibrated with HEPES-Tyrode buffer of pH 7.35 Containing 3.5% albumin. Binding

Vol. 68, No. 2

AJOENE EFFECT ON THROMBUS

149

was assayed essentially as described by Timmons et al., (16). All experiments were done in 0.6 mL volume containing 1.4 - 1.6 x 10' ADP, for 3 min, in platelets. Platelets were stimulated with 13 I.~M the presence of ajoene (50 PM) or the corresponding amount of was added to vehicle (control) and then 1251-porcine vWF (14 I.cg/ml) Incubation without stirring was continued for 30 each sample. After incubation, aliquots of 550 ~1 minutes at room temperature. were loaded on top of a 1:4 (v/v) of a mixture of silicon oils D200 and D550 (Dow Corning) and the platelets separated from unbound Specific '?-vWF by centrifugation for 3 minutes at 7000 xg. binding was calculated by subtracting nonspecific binding (obtained in the absence of ADP), from total binding. The effect of aioene on the interaction of nlasma vWF with the pfl. Washed, paraformaldehyde-fixed porcine platelets were prepared as described (17), except for an extra step consisting 0S rlitracion OS' tilewashed platelets through sepharose CL-4B. The filtered platelets were divided into two aliquots. One of them was incubated with ajoene (100 PM or 600 pM) for 3 minutes. It has been previously shown that activation of washed platelets is completely inhibited after 90 seconds of incubation with less than 50 PM ajoene (3,4). Inmediately after the incubation period, both aliquots were fixed by adding an equal volume of 1.8 % buffered paraformaldehyde. After 24 hours at 4' C, the platelets were washed three times with imidazole-saline and resuspended in citrate-saline (17). Agglutination of both control and ajoenetreated platelets was assayed in the presence of 1.5 mg/ml of ristocetin and fresh porcine plasma as the source of vWF (17). S atis t was performed on a Macintosh II computer (Apple) using the statistical tool provided by Stat-View 512+ (Brain Power Inc.). Variance about the mean is given as C 1 SEM (standard error of the mean). Materials. Ajoene (MW = 234), synthesized as described previously, (5,18) was a gift of Dr. Mahendra K. Jain (Dept. of Chemistry, LT. of Delaware, Newark, DE). It was used as a 1:l mixture of cis and trans isomers. Intralipid, a fat emulsion for intravenous use, was obtained as 10 % solution from Kabi-Vitrum (Sweden). As indicated by the manufacturers, Intralipid 10 % contains: 100 gr soybean oil, 12 gr fractionated egg phospholipids, 22.5 gr glycerol, and water (bd), qsp 1000 ml. All other reagents were of the best commercially available grade. RESULTS Platelet Deposition on Severely Damaaed Vessel Wall. To study the effect of ajoene on platelet deposition on severely damaged vessel wall, pig aortic tunica media was used as substrate. a) Hi 1). sh a The effect of ajoene on platelet deposition on tunica media from pig aorta, exposed to whole blood perfused in vitro under high shear rate conditions, is shown in Fig. la. Five minutes of perfusion of titrated whole blood, incubated with Intralipid (vehicle) for 10 minutes, produced an average platelet deposition of 20.13 2 .52 x 10' platelets/cm2

150

AJOENE EFFECT ON THROMBUS

Vol. 68, No. 2

at high shear rate (1690 set-I).Under the same conditions, ajoene dramatically inhibits platelet deposition on the tunica media, in a dose-dependent manner (Fig. la). At high shear rate, the lowest dose of ajoene used (300 PM), produces about 50% inhibition of platelet deposition (9.57 + .76 x lo6 platelets/cm*), while about 80% inhibition of platelet deposition is obtained with 600 PM ajoene (4.09 _+ .43 x lo6 platelets/cm2).

A

6

* L *

T

ii6

mm

AJOEM CWCENTRATION

600 pN

* IILIIL *

l-

morn

mm AJOENE CONCENTRATION

J

1. Effect of ajoene concentration on platelet deposition on pig tunica media in vitro. A) High shear rate Each bar (1690 set-1); B) Low shear rate (212 set-1). corresponds to the mean + SEM of six individual experiments (* = p < 0.001).

b) Low shear rate conditions (212 set-'). As shown in Fig. lb, at low shear rate platelet deposition on the tunica media is clearly decreased, reaching in the control experiments an average value of 8.44 + .38 x lo6 platelets/cm2. This observation, already described (5,6), reflects the shear-dependence of platelet activation. At low shear rate the effect of ajoene is slightly more marked than at high shear rate. Platelet deposition at 300 PM only amounts to 2.95 + .43 x lo6 platelets/cm' (65% inhibition), while at 600 I.IM it averages 1.29 rt.13 x lo6 platelets/cm" (85% inhibition). Platelet deposition in the range of 1 x 10' to 4 x lo6 platelets/cm'has been previously shown to account for a monolayer of single platelets deposited on the substrate surface (6). Plat g. 1 To study the effect of ajoene on platelet deposition on mildly damaged vessel wall, pig aortic subendothelium was used as substrate. a) Hiah shear rate conditions. As shown in Fig. 2a, platelet deposition on subendothelium, at high shear rate, reaches an average value of 6.03 _+ .95 x lo6 platelet/cm2 in control perfusions. Under the same conditions, ajoene (300 PM) produces an 80% decrease on platelet deposition, reaching an average value of

1.15

.3 x

-I

PM) did

not

A higher ajoene concentration (600 10' platelets/cm'. in further decrease of platelet deposition.

result

8

0-

2

A

B 21

6-

O-

151

AJOENE EFFECT ON THROMBUS

Vol. 68, No. 2

ZOIJTRCL

300 FM

600 et4

AJOENE CONCENTRATION

6.

OL

CONTROL

300 !A4

600 pM

AJOENE CCNXNTRATION

Fig. 2. Effect of ajoene concentration on platelet on pig subendothelium in vitro. A) High shear set-1) ; B) Low shear rate (212 set-1). Each bar mean It SEM of five individual experiments (* = p = p < 0.005).

deposition rate (1690 represents < .OOl, **

b)

Effect of ajoene, the major antiplatelet compound from garlic, on platelet thrombus formation.

Ajoene, (E,Z)-4,5,9-trithiadodeca-1,6,11-triene 9-oxide, is a potent antiplatelet compound isolated from alcoholic extracts of garlic (Allium sativum)...
919KB Sizes 0 Downloads 0 Views