Thrombosis Research 110 (2003) 255 – 258
`
`Review Article
`
`The mechanism of action of aspirin
`
`J.R. Vane*, R.M. Botting
`
`The William Harvey Research Institute, St. Bartholomew’s and the Royal London School of Medicine, Charterhouse Square, London EC1M 6BQ, UK
`
`Abstract
`
`The therapy of rheumatism began thousands of years ago with the use of decoctions or extracts of herbs or plants such as willow bark or
`leaves, most of which turned out to contain salicylates. Following the advent of synthetic salicylate, Felix Hoffman, working at the Bayer
`company in Germany, made the acetylated form of salicylic acid in 1897. This drug was named ‘‘Aspirin’’ and became the most widely used
`medicine of all time. In 1971, Vane discovered the mechanism by which aspirin exerts its anti-inflammatory, analgesic and antipyretic
`actions. He proved that aspirin and other non-steroid anti-inflammatory drugs (NSAIDs) inhibit the activity of the enzyme now called
`cyclooxygenase (COX) which leads to the formation of prostaglandins (PGs) that cause inflammation, swelling, pain and fever. However, by
`inhibiting this key enzyme in PG synthesis, the aspirin-like drugs also prevented the production of physiologically important PGs which
`protect the stomach mucosa from damage by hydrochloric acid, maintain kidney function and aggregate platelets when required. This
`conclusion provided a unifying explanation for the therapeutic actions and shared side effects of the aspirin-like drugs. Twenty years later,
`with the discovery of a second COX gene, it became clear that there are two isoforms of the COX enzyme. The constitutive isoform, COX-1,
`supports the beneficial homeostatic functions, whereas the inducible isoform, COX-2, becomes upregulated by inflammatory mediators and
`its products cause many of the symptoms of inflammatory diseases such as rheumatoid and osteoarthritis.
`D 2003 Published by Elsevier Ltd.
`
`Keywords: Aspirin; Cyclooxygenases; Bayer; Inflammation; Thromboxane; Anaphylaxis; Platelets
`
`Aspirin is the most widely used drug in the world. An
`aspirin a day doubles the chances of a long life. Studies have
`shown that a regular dose of aspirin for the over 50s can
`prolong life since aspirin reduces the risk of many diseases
`associated with aging. The history of aspirin goes back
`many thousands of years to the early uses of decoctions or
`preparations of plants that contain salicylate. Maclagan [1]
`used Salicin,
`the bitter principle of the common white
`willow, successfully in 1874 to reduce the fever, pain and
`inflammation of rheumatic fever. Also in 1874, the com-
`mercial organic synthesis of salicylic acid was formulated
`by Kolbe and his colleagues and led to the founding of the
`Heyden Chemical Company.
`The success of salicylic acid prompted the pharmaceuti-
`cal manufacturing house of Frederick Bayer to actively
`search for a derivative of comparable or better efficacy to
`salicylic acid. Arthur Eichengru¨n, head of the chemical
`research laboratories at Bayer in 1895, assigned this task
`to a young chemist named Felix Hoffman. Hoffman also
`had personal reasons for wanting a more acceptable salicylic
`acid derivative; his father had been taking salicylic acid for
`
`* Corresponding author. Tel.: +44-207-882-6179; fax: +44-207-882-
`6016.
`
`0049-3848/$ - see front matter D 2003 Published by Elsevier Ltd.
`doi:10.1016/S0049-3848(03)00379-7
`
`many years to treat his arthritis and had recently discovered
`that he could no longer take the drug without vomiting.
`Impelled then by filial affection as well as by dedication to
`his job, Hoffman searched through the scientific literature
`and found a way of acetylating the hydroxyl group on the
`benzene ring of salicylic acid to form acetylsalicylic acid.
`After initial laboratory tests, Hoffman’s father was given the
`drug; it was pronounced effective and later confirmed as
`such by a more impartial clinical trial.
`The name ‘‘Aspirin’’ was given to the new drug by
`Bayer’s chief pharmacologist, Heinrich Dreser [2], who was
`anxious to find a name that could not possibly be confused
`with salicylic acid. At least two accounts are given for
`Dreser’s choice of name; some authorities maintain that the
`drug was named after St Aspirinius, an early Neapolitan
`bishop who was the patron saint against headaches. A more
`prosaic explanation is that
`the name was derived from
`Spiraea, which is the Linnaean name for the genus of plants
`to which meadowsweet belongs. Meadowsweet contains
`salicylaldehyde, which can be oxidised to salicylic acid.
`According to this explanation, the acid derived from Spi-
`raea became ‘‘Spirsau¨re’’ in German. Acetylation of Spir-
`sau¨ re produced ‘‘Acetylspisau¨ re’’, which was soon
`shortened to Aspirin.
`
`

`

`256
`
`J.R. Vane, R.M. Botting / Thrombosis Research 110 (2003) 255–258
`
`Of course, it is quite possible that Dreser was aware of
`both possible derivations and that the ambiguous name was
`a deliberate and felicitous contrivance. Of interest,
`‘‘Euˆsparin’’ was suggested in the original document as an
`alternative name.
`
`1. Early explanations for the action of aspirin
`
`Before 1971, little was known about the real mechanism
`of action of aspirin-like drugs. They produced an anti-
`inflammatory effect that was qualitatively and quantitatively
`different from that of the anti-inflammatory steroids, and
`their analgesic action was of a different nature than that
`produced by opiates. Aspirin-like drugs are weak analgesics
`compared with ‘‘strong’’ narcotic analgesics like morphine.
`They are effective in clinical pain of low or moderate but
`not high intensity such as postoperative pain, osteoarthritis,
`rheumatoid arthritis, ankylosing spondilytis and some forms
`of headache [3]. Aspirin-like drugs are effective in experi-
`mental models involving the induction of a previous in-
`flammatory state and block the delayed stretching response
`induced with an intraperitoneal injection of phenylbenzo-
`quinone or dilute acetic acid in mice. They are not effective
`against nociception of short duration induced by pinching or
`stimulating the tail or toes of mouse, rat or guinea pig.
`Guzman et al. [4] and Lim et al. [5] provided definitive
`evidence of the peripheral analgesic activity of aspirin-like
`drugs. What then does aspirin do in the periphery to decrease
`nociception or pain? Many biochemical effects of aspirin-
`like drugs have been documented and theories based on these
`effects have been abandoned. It was observed, for example,
`that most of these drugs uncoupled oxidative phosphoryla-
`tion and that several salicylates inhibited dehydrogenase
`enzymes, particularly those dependent on pyridine nucleo-
`tides. Some aminotransferases and decarboxylases were also
`inhibited, as were several key enzymes involved in protein
`and RNA biosynthesis. All of these inhibitory actions [6]
`were at some time invoked to explain the therapeutic actions
`of aspirin. A problem with most of these theories was that the
`concentration of the drug required for enzyme inhibition was
`in excess (sometimes greatly in excess) of the concentration
`typically found in the plasma after therapy, and there was
`invariably a lack of correlation between the ability of these
`drugs to inhibit particular enzymes and their activities as
`anti-inflammatory agents. Perhaps the most serious imped-
`iment of all to acceptance of any of the above ideas was that
`their proponents could not provide a convincing reason why
`inhibition of any of these enzymes should produce the anti-
`inflammatory, analgesic and antipyretic effects of aspirin.
`
`2. Aspirin and the prostaglandin system
`
`It was against this background of knowledge that the
`investigation of aspirin’s action was taken over by prosta-
`
`glandin (PGs) researchers. Piper and Vane [7] used isolated
`lungs perfused with Krebs’ solution from sensitised guinea
`pigs. The purpose was to detect substances released during
`the anaphylactic reaction, including histamine and SRS-A,
`both of which had been known for many years as possible
`mediators of anaphylaxis.
`They used the technique of continuous bioassay with the
`cascade bioassay system developed by Vane [8] in the
`middle 1960s for use with blood or artificial salt solution.
`As expected, Piper and Vane found the release during
`anaphylaxis of histamine and SRS-A, but they also found
`some previously unreported substances: PGs, (mainly PGE2
`but some PGF2a) and another, very ephemeral substance
`that they called ‘‘rabbit aorta contracting substance’’ (RCS)
`based on the assay tissue that identified it. In the lung
`perfusate RCS had a half-life of about 2 min; even when
`cooled to a few degrees above freezing, it remained stable
`for no more than 20 min. It was identified in 1975 as
`thromboxane A2 by Hamberg et al. [9].
`It was RCS that provided the first clue to the relation
`between aspirin and the PGs. In the course of further
`experiments involving RCS, Piper and Vane discovered that
`in some preparations RCS was released by bradykinin. This
`suggested that aspirin’s ability to minimise some effects of
`bradykinin might be due to its blocking of RCS release. This
`idea was confirmed when Piper and Vane [10] presented
`experimental evidence that the release of RCS from isolated
`guinea pig lungs during anaphylaxis was blocked by aspirin.
`These guinea pig lung experiments also indicated that
`whenever aspirin blocked RCS release, there was a smaller
`contraction of the tissues that assayed PGs and a self-
`evident reduction in PG output after aspirin.
`The natural result of these experiments was to move the
`focus of Vane’s attention from RCS toward PGs. ‘‘While I
`was writing a review paper over the weekend’’, he recalled,
`‘‘including the results of some of these experiments, a
`thought occurred to me that perhaps should have been
`obvious earlier on. In all these experiments (and in those
`of many other workers), the ‘release’ of PGs must in fact
`amount to fresh synthesis of PGs. That is, PG output in
`these experiments, though very low, was still far higher than
`the tissues’ initial content of the hormones. Evidently, then,
`the various stimuli, mechanical and chemical, which re-
`leased PGs, were in fact ‘turning on’ the synthesis of these
`compounds. A logical corollary was that aspirin might well
`be blocking the synthesis of PGs.’’
`Vane immediately tested this exciting idea on the
`following Monday morning. In the absence of ram seminal
`vesicles, from which the synthetase enzyme was usually
`obtained, he used the supernatant of a broken cell ho-
`mogenate from guinea pig lung, the same kind of prepa-
`ration in which A¨ ngga˚rd and Samuelsson [11] had
`detected the generation of PGE2 and PGF2a in 1965.
`Aliquots of the supernatant were incubated with arachi-
`donic acid and different concentrations of aspirin, indo-
`methacin or sodium salicylate. PGF2a generation was
`
`

`

`J.R. Vane, R.M. Botting / Thrombosis Research 110 (2003) 255–258
`
`257
`
`estimated by bioassay after 30 min of incubation at 37 jC.
`There was a dose-dependent inhibition of PG formation by
`all three drugs, indomethacin being the most potent and
`sodium salicylate the least. Three other drugs, morphine (an
`opiate analgesic), hydrocortisone (a steroidal anti-inflamma-
`tory) and mepyramine (an antihistamine), had little or no
`effect.
`Vane [12] published the results of these experiments in
`Nature in 1971. Two other reports in the same issue lent
`support to his findings and extended them considerably.
`Both studies originated from the same department, and by
`coincidence, one of these stemmed from an entirely inde-
`pendent line of investigation.
`Smith and Willis [13] were investigating the effects of
`aspirin on platelet behaviour. Venous blood samples were
`obtained from three colleagues before and 1 h after taking
`600 mg of aspirin orally. Platelets were isolated, washed and
`incubated with thrombin and the supernatant was tested for
`the presence of various substances including PGs. No
`consistent changes were seen in the release of any of the
`substances except
`the PGs, which were substantially
`inhibited after aspirin. Indomethacin also blocked PG re-
`lease when taken orally or when added directly to the
`platelets in vitro.
`The importance of this study lay in its demonstration that
`these drugs were active not only in guinea pig lungs in vitro
`but also in humans, in platelets and after oral administration.
`In other words,
`the aspirin effect was not restricted by
`species, tissue or route of administration. These conclusions
`derived support from the final type of experiment reported
`in that issue of Nature. Ferreira et al. [14] demonstrated that
`the aspirin-like drugs blocked PG release from the perfused,
`isolated dog spleen. In the same year, Collier and Flower
`[15] reported in Lancet
`that administration of aspirin
`inhibited human seminal PG production.
`
`3. Correlation of anti-enzyme activity of aspirin with its
`therapeutic activity
`
`The major importance of these findings was that they
`provided a simple explanation of the manner in which
`aspirin-like drugs exerted their therapeutic actions. When
`the reports were published in 1971,
`there was already
`evidence suggesting that PGE1 was an extremely potent
`pyretic agent in several species [16] and that PGE1 or PGE2
`mimicked the inflammatory response when injected intra-
`dermally. PGs had also been detected in inflammatory
`exudates [17], so there were grounds for speculating that
`PGs might be responsible, at least in part, for the genesis of
`fever or inflammation and that the aspirin-like drugs might
`owe their therapeutic activity to their ability to prevent PG
`biosynthesis. Certainly, as Flower et al. [18] pointed out, the
`concentrations of these drugs required to inhibit synthesis
`were within the plasma levels found during therapy, even
`when protein binding was taken into account.
`
`4. Inhibition of cyclooxygenase
`
`A homogeneous, enzymatically active cyclooxygenase
`(COX) or prostaglandin endoperoxide synthase (PGHS) was
`isolated in 1976 [19]. This membrane-bound hemoprotein
`and glycoprotein with a molecular weight of 72 kDa is
`found in greatest amounts in the endoplasmic reticulum of
`prostanoid-forming cells [20]. It exhibits COX activity that
`both cyclizes arachidonic acid and adds the 15-hydroperoxy
`group to form PGG2. The hydroperoxy group of PGG2 is
`reduced to the hydroxy group of PGH2 by a peroxidase that
`uses a wide variety of compounds to provide the requisite
`pair of electrons. Both COX and hydroperoxidase activities
`are contained in the same dimeric protein molecule.
`Aspirin selectively acetylates the hydroxyl group of one
`serine residue (Ser 530) located 70 amino acids from the C
`terminus of the enzyme [21]. Acetylation leads to irrevers-
`ible COX inhibition; thus, a new enzyme must be synthe-
`sized before more prostanoids are produced. When the
`purified enzyme is acetylated, only the COX, not the hydro-
`peroxidase, activity is inhibited. The stoichiometry of this
`reaction is 1:1, with one acetyl group transferred per enzyme
`monomer of this dimeric protein. At low concentrations,
`aspirin acetylates PGHS rapidly (within minutes) and selec-
`tively. At high concentrations, over longer time periods,
`aspirin will also non-specifically acetylate a variety of
`proteins and nucleic acids [22]. Acetylation of the enzyme
`by aspirin places a bulky substituent on the Ser 530 oxygen
`that inhibits binding of arachidonic acid [23].
`
`5. Discovery of COX-2 and COX-3
`
`the
`By the late 1980s, several reports appeared that
`synthesis of PGHS could be stimulated by growth factors,
`tumour promoters,
`interleukin-1 [24],
`lipopolysaccharide
`and tumour necrosis factor. Interleukin-1 exerted its effect
`during the transcriptional rather than during the translational
`phase of induced synthesis of PGHS [25]. Induction of
`PGHS gene expression by serum factors occurred after
`approximately 2 h in mouse 3T3 cells, in which PGs are
`essential for cell division. These reports culminated with the
`discovery by Dan Simmons [26] of a second, distinct COX
`gene which could be induced with mitogens, growth factors,
`tumour promoters and lipopolysaccharide, and the induction
`of which could be inhibited with glucocorticoids. This gene
`expresses COX-2 which elaborates PGs, mostly PGE2,
`during inflammatory reactions in contrast to COX-1, which
`produces PGs involved in physiological processes such as
`protection of the stomach mucosa, platelet aggregation and
`kidney function.
`There is 60% homology between the amino acid struc-
`tures of COX-1 and COX-2 and aspirin binds to Ser 516 in
`the active site of COX-2 in the same way as it binds to Ser
`530 in the active site of COX-1. However, the active site of
`COX-2 is slightly larger than the active site of COX-1, so
`
`

`

`258
`
`J.R. Vane, R.M. Botting / Thrombosis Research 110 (2003) 255–258
`
`that arachidonic acid can still ‘squeeze past’ the aspirin
`molecule inactivating COX-2 and become converted to 15-
`R-HETE [27].
`The small difference in size between the active sites of
`COX-1 and COX-2 has been exploited by pharmaceutical
`companies to develop selective COX-2 inhibitors, such as
`celecoxib [28], rofecoxib [29] and meloxicam [30], which
`reduce inflammation without damaging the stomach muco-
`sa. One company has also produced nitroaspirin [31], which
`combines aspirin with a nitric oxide-releasing moiety. The
`nitric oxide liberated in the stomach protects the stomach
`mucosa from damage by gastric hydrochloric acid. As new
`anti-inflammatory drugs with fewer severe side effects than
`non-selective non-steroid anti-inflammatory drugs
`(NSAIDs) are developed, the use of aspirin for osteoarthritis
`and rheumatoid arthritis will decline. However, its use as a
`potent anti-thrombotic agent for the prevention of second
`heart attacks is likely to increase.
`A recent report from Chandrasekharan et al. [32]
`describes a third cyclooxygenase (COX-3) selectively
`inhibited not only by paracetamol but also by low concen-
`trations of some non-steroid anti-inflammatory drugs in-
`cluding aspirin. COX-3 is a variant of COX-1 which has
`retained intron-1 during translation and which is found in
`human tissues in a polyadenylated form. Selective inhibition
`of COX-3 will discover potent and valuable new drugs for
`controlling pain and fever.
`
`References
`
`[1] Maclagan TJ. The treatment of acute rheumatism by salicin. Lancet
`1876;i:342 – 83.
`[2] Dreser H. Pharmacologisches u¨ ber Aspirin (Acetylsalicyl-sau¨ re).
`Pflugers Arch 1899;76:306 – 18.
`[3] Beaver WT. Mild analgesics: a review of their clinical pharmacology.
`Am J Med Sci 1965;250:577 – 604.
`[4] Guzman F, Braun C, Lim RKS, Potter GD, Rodgers DW. Narcotic
`and non-narcotic analgesics which block visceral pain evoked by
`intra-arterial injections of bradykinin and other algesic agents. Arch
`Int Pharmacodyn 1964;149:571 – 88.
`[5] Lim RKS, Guzman F, Rodgers DW, Goto K, Braun C, Dickerson GD,
`et al. Site of action of narcotic and non-narcotic analgesics determined
`by blocking bradykinin-evoked visceral pain. Arch Int Pharmacodyn
`1964;152:25 – 58.
`[6] Whitehouse MW. Some biochemical and pharmacological properties
`of anti-inflammatory drugs. Prog Drug Res 1965;8:321 – 429.
`[7] Piper PJ, Vane JR. Release of additional factors in anaphylaxis and its
`antagonism by anti-inflammatory drugs. Nature 1969;223:29 – 35.
`[8] Vane JR. The use of isolated organs for detecting active substances in
`the circulating blood. Br J Pharmacol Chemother 1964;23:360 – 73.
`[9] Hamberg M, Svensson J, Samuelsson B. Thromboxanes: a new group
`of biologically active compounds derived from prostaglandin endo-
`peroxides. Proc Natl Acad Sci U S A 1975;72:2994 – 8.
`[10] Piper PJ, Vane JR. The release of prostaglandins during anaphylaxis
`in guinea-pig isolated lungs. In: Mantegazza P, Horton EW, editors.
`Prostaglandins, peptides and amines. London: Academic Press; 1969.
`p. 15 – 9.
`[11] A¨ ngga˚rd E, Samuelsson B. Biosynthesis of prostaglandins from
`arachidonic acid in guinea-pig lung. J Biol Chem 1965;240:
`3518 – 21.
`
`[12] Vane JR. Inhibition of prostaglandin synthesis as a mechanism of
`action for Aspirin-like drugs. Nat New Biol 1971;231:232 – 5.
`[13] Smith JB, Willis AL. Aspirin selectively inhibits prostaglandin pro-
`duction in human platelets. Nature 1971;231:235 – 7.
`[14] Ferreira SH, Moncada S, Vane JR. Indomethacin and Aspirin abolish
`prostaglandin release from spleen. Nature 1971;231:237 – 9.
`[15] Collier JC, Flower RJ. Effect of Aspirin on human seminal prosta-
`glandins. Lancet 1971;ii:852 – 3.
`[16] Milton AS, Wendlandt S. A possible role for prostaglandin E1 as a
`modulator of temperature regulation in the central nervous system of
`the cat. J Physiol 1970;207:76 – 7.
`[17] Di Rosa M, Giroud JP, Willoughby DA. Studies of the mediators of
`the acute inflammatory response induced in rats in different sites by
`carrageenin and turpentine. J Physiol 1971;104:15 – 29.
`[18] Flower R, Gryglewski R, Herbaczynska-Cedro K, Vane JR. The ef-
`fects of anti-inflammatory drugs on a cell-free prostaglandin synthe-
`tase system from dog spleen. Nature 1972;238:104 – 6.
`[19] Hemler M, Lands WEM, Smith WL. Purification of the cyclooxyge-
`nase that forms prostaglandins: demonstration of two forms of iron in
`the holoenzyme. J Biol Chem 1976;251:5575 – 9.
`[20] Smith WL. Prostaglandin biosynthesis and its compartmentation in
`vascular smooth muscle and endothelial cells. Annu Rev Physiol
`1986;48:251 – 62.
`[21] Roth GJ, Majerus PW. The mechanism of the effect of Aspirin on
`human platelets: 1 Acetylation of a particulate fraction protein. J Clin
`Invest 1975;56:624 – 32.
`[22] Smith WL. The eicosanoids and their biochemical mechanisms of
`action. Biochem J 1989;259:315 – 24.
`[23] De Witt DL, El-Harith EA, Kraemer SA, Andrews MJ, Yao EF,
`Armstrong RL, et al. The Aspirin and heme-binding sites of ovine
`and murine prostaglandin endoperoxide synthases. J Biol Chem
`1990;265:5192 – 8.
`[24] Raz A, Wyche A, Siegel N, Needleman P. Regulation of fibroblast
`cyclooxygenase synthesis by interleukin-1. J Biol Chem 1988;263:
`3022 – 8.
`[25] Raz A, Wyche A, Needleman P. Temporal and pharmacological divi-
`sion of fibroblast cyclooxygenase expression into transcriptional and
`translational phases. Proc Natl Acad Sci U S A 1989;86:1657 – 61.
`[26] Xie W, Chipman JG, Robertson DL, Erikson RL, Simmons DL. Ex-
`pression of a mitogen-responsive gene encoding prostaglandin syn-
`thase is regulated by mRNA splicing. Proc Natl Acad Sci U S A
`1991;88:2692 – 6.
`[27] Vane JR, Bakhle YS, Botting RM. Cyclooxygenases 1 and 2. Annu
`Rev Pharmacol Toxicol 1998;38:97 – 120.
`[28] Silverstein FE, Faich G, Goldstein JL, Simon LS, Pincus T, Whelton
`A, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-
`inflammatory drugs for osteoarthritis and rheumatoid arthritis: the
`CLASS study: a randomised controlled trial. Celecoxib Long-term
`Arthritis Safety Study. J Am Med Assoc 2000;284:1247 – 55.
`[29] Bombardier C, Laine L, Reicin A, Shapiro D, Burgos-Vargas R, Davis
`B, et al. Comparison of upper gastrointestinal toxicity of rofecoxib
`and naproxen in patients with rheumatoid arthritis. N Engl J Med
`2000;343:1520 – 8.
`[30] Dequeker J, Hawkey C, Kahan A, Steinbru¨ck K, Alegre C, Baumelou
`E, et al. Improvement in gastrointestinal tolerability of the selective
`cyclooxygenase (COX)-2 inhibitor, meloxicam, compared with pirox-
`icam: results of the safety and efficacy large-scale evaluation of COX-
`inhibiting therapies (SELECT) trial in osteoarthritis. Br J Rheumatol
`1998;37:946 – 51.
`[31] Fiorucci S, Santucci L, Gresele P, Faccino RM, Del Soldato P, Morelli
`A. Gastrointestinal safety of NO-aspirin (NCX-4016) in healthy hu-
`man volunteers: a proof of concept endoscopic study. Gastroenterol-
`ogy 2003;124:600 – 7.
`[32] Chandrasekharan NV, Dai H, Roos KL, Evanson NK, Tomsik J, Elton
`TS, et al COX-3, a cyclooxygenase-1 variant inhibited by acetamino-
`phen and other analgesic/antipyretic drugs: cloning, structure and
`expression Proc Natl Acad Sci U S A 99:13926 – 31.
`
`