To summarise, prostagiandin E2 and prostacyclin are present in inflammatory

exudates in sufficient concentrations to contribute to the erythema , oedema and

hyperalgesia associated with inflammation. The role of thromboxanes in these

responses and the elTect of cyclo-oxygenase metabolites on leucocyte function is,

however, less clear.

Intlammatory properties of lipoxygenase products

The first indication that lipoxygenase may be important in the inflammatory

response was the discovery that 12-HETE is chemotactic for PMNs (Turner, Tainer

& Lynn, 1975). A number of other mono-hydroxy acid lipoxygenase products have

now been shown to be chemokinetic and chemotactic (Goetzl & Sun, 1979).

Leucocytes also convert arachidonic acid to a 5,12 dihydroxy acid (5,12-DHETE,

leukotriene B) (Borgeat & Samuelsson, 1979) which is the most potent chemotactic

lipoxygenase product so far studied (Ford-Hutchinson, Bray, Shipley, Doig & Smith,

ANTI-INFLAM MATORY DR UGS 281

1980). 5,12-DHE T E is at least 100 fold more active than the mono-h ydroxy acids as a

chemoki netic agent for rat, rabb it or hu ma n leucocytes (Pa lrner, Step ney, Higgs &

Ea kins, 1980 ) and it has a similar potency to th e che motactic form yl-rnet hio nyl

peptides (Schiffrnan , Corcoran & Wahl , 197 5). It is possible th at gene ratio n of

lipoxygenase products such as 5,12-DHE T E may represent a local co ntrol

mechan ism for leucocyte mig ration at an infl ammat or y site.

Evidence is now appea ring wh ich suggests th at leukotrienes may aiso have a role in

vascula r respo nses . A crude pr epar ati on of SRS-A was shown to inc rease vascular

permeab ility when injected into skin (Brockleh urst, 1967 ). Purified SRS-A , whic h is

kno wn to be cysteiny l-glyciny l eicosat etrae noic acid, increases vascular perm eability

in guinea-pig skin and is a vasoconstrictor (Willia ms & Piper, 1980). Furtherrno re,

5,12- DHE TE is equipotent with PG E2 a nd prostacyclin in potentiating bradykininind uced plasma exud ation in rabbit skin (Ea kins, Higgs, Moncada, Salmon &

Spayne, 1980). It is possible, therefore, th at lipoxygenase products, as weil as cyclooxygenase products, contribute to th e development of oed ema in an ap hylactic or

inflammatory respo nses.

Anti-inflammatory drugs

Cyclo-oxygenase inhibitors

Selecti ve cyclo-oxygenas e in hibition by indomethacin or aspinn pre vents th e

synthesis of inflammatory mediators such as PG E2 and prostac yclin and th is action

accounts for their therapeuti c effects (Ferre ira & Vane, 1979). T here is a good correlation between the ra nk orde r of potency of non-steroid anti- inflamrnatory drugs in

in hibiting oedema and prostagiand in synthesis in vivo (Higgs, Ha rvey, Ferreira &

Vane, 1976). Becau se prostagland ins synergise wit h other infla mmatory mediators in

e 180

.o'"E 140 "

c_

u C 100 00

U U

~~" 60 "0

Ö

f--

20

C C

-6 2 :::-

80 C - 0

oo~

"'- O C Co v; ~ u 40

et 0

0 .02

*

--- -

* PGs

*

* * * * * *

0 . 1 0 .5 2 .0 8 .0 16 .0

Indornetbocin dose (rnq kg- ')

Figure 2 The effect of indomethacin on total leucocyte numbers (WBC) and prostagiandin

concentrations (PGs) in inflammatory exudates collected by the subcutaneous implantation of

carrageenin-irnpregnated sponges in rats. The exudates are collected at 24 hand the drug is

given orally at the time of irnplantation, 8 h later and 3 h before removal of the sponge. Each

point is the mean of 5-40 experiments and the bars represent ± s.e.mean, • indicates that

P < 0.05 compared with control values. The arrows indicate the dose requircd to inhibit

carrageenin-induced oedema by 50% in separate experiments. Data taken from Higgs, Eakins,

Mugridge, Moncada & Vane (1980), reproduced by permission.

282 G. A. HIGGS & B.J. R. WHITTLE

th e production of oedema and pain , th eir rem oval redu ces th e efTectiveness of

substances suc h as brad ykinin and histamine. This lead s to a grea ter antiinflam ma tory acti vity than would be achieved by th e removal or antagonis m of a

single direct ly-act ing med iator.

A number of cycl o-oxygenase in hibito rs ha ve a difTe rent ial efTect on leucocyte

m igrat ion in vivo (Higgs, Eak ins, Mu gridge, Moncada & Vane, 1980 ). Indometh acin ,

as pirin and flur biprofen enhance th e acc u mulation of cells in inflammatory exuda tes

at do ses which signi ficantly reduce prostagIandin product ion , bu t inhibit cell migration at higher doses (Figure 2). This obser vat ion can be explai ned if the inhibition of

cyclo-oxy genas e divert s substra te to wards th e product ion of che motactic

lipoxygenase products which th en account for increased leucocyte migration . The

subseq uent inhibition of leu cocyte migrat ion at high er do ses ma y be ex plained by a

non-spe cific in hibition of a rac hido nic acid peroxidation. Indeed, high doses of

indomethacin inhibit lipoxygen ase in vivo (Bragt & Bonta, 1980 ).

The demonstration th at some non-steroid anti-inflammatory drugs ha ve a differenti al efTect on leucocyte migration ma y explain some anoma lies arisi ng from

their use in the c1inic. The asp irin-l ike drugs give good sympto ma tic rel ief in chronic

inflammation but it is often sta ted that they do little to ameliorate th e underlying

progress of the disease. Symptomatic relief such as analgesia and a reduction in

swelling would result from th e removal of cyclo-ox ygenase deri ved mediators, but

failure to inhibit lipoxygenase at com pa rable do ses ma y allow the more chronic,

leu coc yte-mediated components ofinflammation to proceed un ch ecked.

Corticosteroids

The anti-inflammatory corticostero ids do not directl y inhibit prostagIandin

synthes is in vitro (Va ne, 1971; Flower , G ryg lewski, Herbaczynska-Cedro & Vane,

1972) although steroids suppress prostagIandin production in rat skin (G reaves &

McDonald-Gibson , 1972) and they inhibit prostagiandin release from th e ra bbit fat

pad (Lewis & Piper, 1975). Steroids inh ibit the release of prostaglandins from

perfused lungs of the guinea-pig (G ryglews ki, Pan czenko, Korbut, Grodzinska &

Ocetkiewicz, 1975) and these aut hors concl uded that steroids preve nt th e release of

a rachidonic acid from phospholipids. This work was supported by th e demonstration

that corticosteroids inhibit the seru m-stim ulated release of a rachido nic aci d from

transformed mouse fibroblasts (Hong & Levine, 1976). It is no w thought that steroids

induce th e release or form ation of a substa nce which inhibits phospholipase A2

(Flower & Blackwell, 1979) and thi s may explain th eir ' rnembra ne sta blising' efTects.

The inhibition of arac hidonic aci d release from phospholipids wou ld result in

reduced formation of cyclo-oxygenase and lipoxygenase products, thus decreasing

the generation of inflammatory medi ators. The indirect inhibition of both pathways

of arachidonic acid ox ygenation may explain wh y dexamethasone is equi-active in

reducing prostagiandin concentra tions and totalleucocyte numbers in inflammatory

ex udates (Higgs, Flower & Vane, 1979).

Inhibitors 0/cyclo-oxygenase and /ipoxygenase

The experimental anti-infla mma tory compound BW755C (3-amino-I-[m-

(trifluoromethyl)-phenyl]-l-pyrazoline) is equi-active in pr eventing cyclo-oxygena se

a nd lipoxygenase acti vit y in vitro (Higgs, Flower & Van e, 1979). U nlike

indo methac in and the sa licylates, BW755C does not en hance leucoc yte migrat ion at

a ny do se and red uces oedema, prostagiandin concentrat ions and total leucocyte

numbers in inflam matory ex uda tes at the same dose . Benoxaprofen a lso inhibits

lipoxygen ase (Walker & Da wson , 1980) and this ma y account for its efTects on leucocyte function in inflam mation (Meacoc k & Kitchen, 1979).

ANTI-INFLAMMATORY DRUGS

Gastrointestinal toxicity of anti-inflammatory drugs

283

A major c1inical problem encountered in anti-inflarnmatory therapy has been the

high incidence of gastrointestinal irritation, particularly with aspirin-like drugs.

Gastric mucosaldarnage may result from disruption ofthe normal resistance ofthe

gastric mucosa to the back-diffusion of gastric acid from the lumen into the mucosal

tissue. Topical administration of aspirin, as weil as a number of unrelated irritants,

has been shown to produce this effect (Davenport, 1964). Such a mechanism cannot, however, explain why many anti-inflamrnatory drugs cause gastrointestinal damage

when administered parenterally (Grossrnan, Matsumoto & Lichter. 1961: Whittle.

1977). Such actions may result from the ability of these compounds to inhibit

prostagiandin biosynthesis (Vane, 1971). Prostacyclin is the predominant cyclooxygenase product in the gastric mucosa and, in common with other endogenous

prostaglandins such as PGE2, prostacyclin can protect the gastric mucosa from

10

..

-e

.:

c

o

"8

w

20 20 44 22 30

o PO's in sponge

o PGI2 trom galtricmucosa

ASPIRIN SALICYLATE INDO

200 200 4

FLURBI

0.5

BW755c

100

Oral dose (mg kg-')

Figure 3 Effect of oral administration of non-steroid anti-inflamrnatory agents on prostaglandin E2 - like activity in the sponge inflammatory exudate, the formation of prostacyclin

(PGh) by the gastric mucosa ex vivo and the induction of gastric erosions in rats. Drugs were

administered orally in the dose shown 3 times over 24 h. The results are expressed as mean ±

s.e.mean of(n) values. The changes in prostaglandin levelsare given as % change compared to

control levels, where * indicates significance (P < 0.01). The degree of gastric damage is

expressed in terms of an erosion index. Data taken from Whitt1e, Higgs, Moncada & Vane

(1980), reproduced by permission.

284 G . A. HIGGS & B.J. R. WHITTLE

damage (Whittle, Boughton-Smith, Moncada & Vane, 1978). Being a potent mucosal

vasodilator, prostacyclin may be involved in the local regulation of the gastric

microcirculation under physiological or pathophysiological conditions. Removal of

such vasodilator tone following the inhibition of endogenous prostagiandin biosynthesis with aspirin-like compounds could lead to areas of local ischaemia within

the gastric mucosa which would be susceptible to attack by acid or pepsins and wh ich

could ultimately lead to tissue necrosis and erosion formation.