Structural requirements for preventing the aspirin- and the arachidonate-induced inactivation of platelet cyclo-oxygenase: additional evidence for distinct enzymatic sites

Thrombin-Activated Platelets Protect Vascular Endothelium against Tumor Cell Extravasation by Targeting Endothelial VCAM-1

When activated by thrombin, the platelets release their granular store of factors. These thrombin-activated platelets (TAPLT) have been shown to be capable of ameliorating pro-inflammatory processes. In this study, we tested if TAPLT could also protect the endothelium against tumor-related pro-inflammatory changes that promote angiogenesis and metastasis. Using endothelial cell (EC) models in vitro, we demonstrated that TAPLT protected EC against tumor conditioned medium (TCM)-induced increases of reactive oxygen species (ROS) production, EC permeability and angiogenesis, and inhibited transendothelial migration that was critical for cancer cell extravasation and metastasis. In vivo observations of TAPLT-mediated inhibition of angiogenesis and pulmonary colonization in a BALB/c nude mouse model were consistent with the in vitro findings. Neutralization of vascular cell adhesion molecule-1 (VCAM-1) binding significantly inhibited the ability of TAPLT to interact with EC and abrogated the TAPLT-mediated protection of EC against tumor angiogenesis and metastasis. Taken together, these findings suggest that VCAM-1-mediated linkage to EC is required for TAPLT to confer protection of EC against tumor-induced permeation and angiogenesis, thereby resisting tumor extravasation and metastasis.

Comparison of oral aspirin versus topical applied methyl salicylate for platelet inhibition

BACKGROUND

Oral acetylsalicylic acid (aspirin) is the primary antiplatelet therapy in the treatment of acute myocardial infarction and acute coronary syndrome. Methyl salicylate (MS; oil of wintergreen) is compounded into many over-the-counter antiinflammatory muscle preparations and has been shown to inhibit platelet aggregation locally and to be absorbed systemically.

OBJECTIVE

To assess the ability of topically applied MS to inhibit systemic platelet aggregation for patients who are unable to tolerate oral drug therapy.

METHODS

A randomized, prospective, blinded, crossover study was conducted in 9 healthy men, aged 30-46 years. All subjects ingested 162 mg of aspirin or applied 5 g of 30% MS preparation to their anterior thighs. There was a minimum 2-week washout period between study arms. Blood and urine were collected at baseline and at 6 hours. An aggregometer measured platelet aggregation over time against 5 standard concentrations of epinephrine, and a mean area under the curve (AUC) was calculated. Urinary metabolites of thromboxane B(2) were measured by a standard enzyme immunoassay. Differences in and between groups at baseline and 6 hours were tested by the Wilcoxon signed-rank test.

RESULTS

Baseline platelet aggregation did not differ significantly between the 2 arms of the study (median AUC [% aggregation(*)min]; binominal confidence intervals): aspirin 183; 139 to 292 versus MS 197; 118 to 445 (p = 0.51). Both aspirin and MS produced statistically significant platelet inhibition; aspirin decreased the AUC from 183; 139 to 292 to 85; 48 to 128 (p = 0.008) and MS decreased the AUC from 197; 118 to 445 to 112; 88 to 306 (p = 0.011). No significant difference was detected between baseline and 6-hour thromboxane levels for either aspirin (p = 0.779) or MS (p = 0.327).

CONCLUSIONS

Topical MS and oral aspirin both significantly decrease platelet aggregation in healthy human volunteers.

The structural basis of aspirin activity inferred from the crystal structure of inactivated prostaglandin H2 synthase

Aspirin exerts its anti-inflammatory effects through selective acetylation of serine 530 on prostaglandin H2 synthase (PGHS). Here we present the 3.4 A resolution X-ray crystal structure of PGHS isoform-1 inactivated by the potent aspirin analogue 2-bromoacetoxy-benzoic acid. Acetylation by this analogue abolishes cyclooxygenase activity by steric blockage of the active-site channel and not through a large conformational change. We observe two rotameric states of the acetyl-serine side chain which block the channel to different extents, a result which may explain the dissimilar effects of aspirin on the two PGHS isoforms. We also observe the product salicylic acid binding at a site consistent with its antagonistic effect on aspirin activity.

Salicylamide reverses the aspirin-antagonistic effect of salicylic acid on rat platelet cyclooxygenase

The antagonistic effect of salicylic acid (SA), the major metabolite of aspirin, on aspirin (ASA)-induced inhibition of cyclooxygenase has been recognized in vivo and in vitro. Salicylamide is available with aspirin in some analgesic preparations. Salicylamide shares important characteristics with salicylic acid including the lack of effect on cyclooxygenase and platelet aggregation as well as a close structural resemblance. This prompted us to study the interaction of salicylamide with aspirin and/or SA on rat platelet and cyclooxygenase. Our results showed that salicylamide has, unlike SA, no blocking effect on the anticyclooxygenase effect of aspirin in vitro. Moreover, salicylamide could dose-dependently prevent the aspirin-blocking effect of SA on platelet cyclooxygenase. These results suggest that salicylamide and SA compete for a receptor on cyclooxygenase different to that of aspirin. A functional model of cyclooxygenase enzyme is proposed.

Action of some salicylate derivatives on in vitro platelet aggregation. Inhibitory and inhibition antagonistic effects

Twenty salicylate derivatives were tested for their antagonistic activity on the inhibitory effect of aspirin on platelet aggregation. The blocking effect was not limited to the salicylate but also characterised some of its substituted compounds. The substituant influence did not seem to be related to electronic or size parameters. This antagonistic activity of these derivatives decreased as concentrations increased, owing to the emergence of their own inhibitory activity: several salicylate derivatives showed dual inhibitory and inhibition antagonistic activity, with both properties present at the same concentration. A mechanism involving dissociated activities on the two enzymatic sites of cyclooxygenase is proposed.

Inhibitory effect of aryl thienyl-ketones and -thioketones on arachidonic acid-induced malondialdehyde formation in human platelets: biological data and molecular modelling

A series of anti-thrombotic aryl thienyl-ketones and -thioketones was assayed in vitro for their inhibitory effect on malondialdehyde (MDA) production induced by arachidonic acid in human platelets. For several compounds MDA formation was strongly inhibited indicating that the anti-platelet target was situated on the cyclooxygenase pathway. A comparison between the inhibition constant Ki and the IC50 values revealed competitive inhibition kinetics. The molecular structure of one active compound was analysed by X-ray diffraction and theoretical calculations to provide information on its electronic and lipophilic properties.

Inhibition of prostaglandin generation in the rabbit brain in-vivo by AD-1590, a non-steroidal anti-inflammatory agent with potent antipyretic activity

The inhibition of prostaglandin generation by AD-1590 was investigated in the rabbit brain in-vivo. AD-1590 (0.4 mg kg-1 i.v.) markedly prevented both the increases in body temperature and PGE2 level in cerebrospinal fluid (CSF) caused by i.v. injection of lipopolysaccharide. On the other hand, 2,4-dinitrophenol (20 mg kg-1 i.v.)-induced hyperthermia, which was not affected by AD-1590, was not accompanied by an increase in PGE2 level in CSF. When injected intracerebroventricularly, AD-1590 dose-dependently inhibited the hyperthermia caused by arachidonic acid given by the same route; its ED50 was 1.6 micrograms compared with about 35 micrograms for indomethacin. From these results, it is suggested that AD-1590 is more active than indomethacin in suppressing prostaglandin synthetase in rabbit brain.

Pharmacology of platelet inhibition in humans: implications of the salicylate-aspirin interaction

The current dispute over the effects of "low" vs "high" doses of aspirin should take into consideration the pharmacokinetics of this drug. In fact, different pharmaceutical formulations of aspirin may deliver little or no aspirin to the systemic blood. This was the case, for instance, in healthy volunteers taking 320 mg of compressed aspirin or 800 mg of enteric-coated aspirin. In all instances thromboxane B2 generation in serum was fully inhibited. Platelet cyclooxygenase might therefore be effectively acetylated by exposure to aspirin in the portal circulation, whereas vascular cyclooxygenase could be spared. Thus aspirin formulations ensuring complete first-pass deacetylation should be sought rather than "low" or "high" doses of unspecified aspirin formulations. Regardless of the type and dose of aspirin administered, salicylate is formed and accumulates in the circulation. It may antagonize the effects of aspirin on cyclooxygenase, at least in acute conditions. As an example, after administration of 1 g of salicylate to healthy volunteers, when plasma levels of the drug were about 75 micrograms/ml, the effect of 40 mg iv aspirin (given 40 min later) on platelet cyclooxygenase and aggregation was significantly diminished. In contrast, in patients undergoing saphenectomy, the same dose of salicylate (1 g) gave plasma drug levels of about 25 micrograms/ml; salicylate was unable to prevent the inhibitory effect on platelets of 40 mg iv aspirin (given 1 hr later) but did act on vascular prostacyclin. Thus the combination of salicylate with aspirin at an appropriate dose and blood level ratio may result in almost complete dissociation of the drug's effect on platelets and vessels in man.(ABSTRACT TRUNCATED AT 250 WORDS)

Platelet desensitization induced by arachidonic acid is not due to cyclo-oxygenase inactivation and involves the endoperoxide receptor

Human platelets pre-exposed to arachidonic acid (AA) (0.1-1 mM) or to the endoperoxide analogue U46619 (1-3 microM) and then washed and resuspended, failed to respond with aggregation or secretion to a second challenge by either agonist. The response to thrombin at low (0.04-0.1 u ml-1) but not at high (2.5 u ml-1) concentrations was also inhibited by pre-exposure to AA and U46619. The ability of platelets to synthesize thromboxane (Tx) B2 from AA or upon challenge with thrombin persisted despite platelet desensitization. In the presence of the reversible cyclo-oxygenase (CO) inhibitors methyl salicylate (MS) or L8027, pre-exposure to AA had no effect on subsequent challenge by the same agonist or by U46619, whereas platelet desensitization by pre-exposure to U46619 persisted. However, platelet activation by, and desensitization to AA and U46619, was prevented by trimetoquinol and compound L636499, two thromboxane/endoperoxide receptor antagonists. In contrast to the CO inhibitors, the thromboxane synthetase inhibitor dazoxiben, which in 3 'responders' out of 5 subjects suppressed aggregation, secretion, and Tx formation induced by AA, failed to prevent AA-induced desensitization. Compared to quiescent cells the distances between platelets desensitized after re-exposure to AA were reduced in electron microscopy, but the tight connections associated with aggregated cells were not observed. Degranulation was also not observed and cell morphology resembled that of normal quiescent platelets. In conclusion, (a) AA and U46619 desensitize human platelets at a similar site sensitive to prostaglandin/thromboxane receptor antagonists, and show cross-desensitization; (b) desensitization by AA appears to be mediated by a CO-dependent metabolite, as CO inhibitors prevent desensitization by AA but not to U46619; (c) the failure of dazoxiben to prevent desensitization by AA suggests that a metabolite other than TxA2, possibly the endoperoxides, mediates the phenomenon; (d) desensitization does not involve inactivation of CO or thromboxane synthetase enzymes.

Comparative pharmacology of cyclooxygenase inhibitors on platelet function

Previous studies from our laboratory have demonstrated a critical role for ferrous heme in prostaglandin synthesis. Based upon these studies, we proposed a model for heme-arachidonic acid interaction and demonstrated that compounds which interfere with this interaction inhibit arachidonic oxidation by ferrous heme. In this study, we have examined the effect of four different inhibitors for their effect on platelet arachidonic acid metabolism and function. The compounds studied were an iron chelator, 2,2'-dipyridyl, the cyclooxygenase inhibitors, Ibuprofen and aspirin, and a polyenoic acid, docosahexaenoic acid. All four compounds at approximately 100 microM concentration blocked the second wave of platelet aggregation in response to epinephrine or adenosine diphosphate. They were equally potent in inhibiting 14C-arachidonic conversion by platelets to thromboxane. However, inhibition of platelet thromboxane production and function by dipyridyl and DHA was reversible. Removal of these compounds from the medium restored platelets ability to respond to agonists and generate products through the cyclooxygenase pathway. The inhibitory effect of Ibuprofen and aspirin on cyclooxygenase activity could not be reversed by washing the platelets. However, Ibuprofen treated platelets aggregated when stirred with arachidonate in a normal way. No such response could be elicited from aspirin treated platelets. All compounds (except DHA) interfered with heme-arachidonic acid interaction in a cell-free system and prevented arachidonic acid oxidation. Results of our studies suggest a common mechanism of action for these different classes of compounds. In spite of the common mechanism, each class of drug seems to have a relatively different effect upon platelet cyclooxygenase and function.

Interactions of inhibitors of the lipoxygenase and cyclooxygenase pathways with a supplementary binding site on soybean lipoxygenase

The oxygenation of [1-14C]-arachidonic acid by a soluble soybean lipoxygenase (E.C.1.13.11.12) preparation was determined in the presence of various cyclo-oxygenase and lipoxygenase inhibitors. The results showed that several non-inhibitory compounds drastically blunted the inhibitory potency of potent lipoxygenase inhibitors. Studies on the combined effects of a variety of structurally unrelated inhibitors of lipoxygenase, cyclo-oxygenase or both oxygenation pathways provided strong evidence for the existence of a supplementary binding site on soybean lipoxygenase which reduces the effective interactions of inhibitors with the catalytic site. Thus several cyclo-oxygenase inhibitors (which do not inhibit at the lipoxygenase catalytic site), as well as low concentrations of lipoxygenase inhibitors, interact with this putative supplementary site and blunt the inhibitory efficacy of potent lipoxygenase inhibitors. Although the degree of interaction with the catalytic site determines the absolute potency of inhibitors, the additional interaction at the putative supplementary binding site is also obligatory for inhibitory potency. In this new multiple-site model the potent lipoxygenase inhibitors (e.g. acetone phenylhydrazone, phenidone) possess high affinities for both sites, whereas weak inhibitors and certain cyclo-oxygenase inhibitors (e.g. benoxaprofen, phenylbutazone, indomethacin) interact predominantly with the supplementary site on the lipoxygenase but lack affinity for the catalytic site.