Treatment of inflammatory diseases by selective eicosanoid inhibition: a double-edged sword? Saul Yedgar 1 , Miron Krimsky 1 , Yuval Cohen 2 and Roderick J. Flower 3 1 Department of Biochemistry, Hebrew University-Hadassah Medical School, Jerusalem, 91120, Israel 2 Morria Biopharmaceuticals Plc, 53 Davies Street, London W1K 5JH, UK 3 Biochemical Pharmacology Centre, The William Harvey Research Institute, Queen Mary University of London, Charterhouse Square, London EC1M 6BQ, UK Eicosanoids are generally considered to be potent pro- inflammatory mediators, and their suppression has, therefore, been a desirable therapeutic goal. However, analysis of the literature reveals that inhibition of specific eicosanoids per se is a simplistic approach because it overlooks the fact that net pathophysiological effects of these lipid mediators arise from a complex balance between eicosanoids derived from different pathways, which might exhibit both pro-and anti-inflam- matory activities (depending on organs and disease stage), or which might have essential physiological roles. An alternative strategy, discussed in this review, might be to control inflammatory lipid mediators in such a way as to avoid disrupting this intricate inter-eicosa- noid balance and its physiological sequelae. Inflammation is mediated by a plethora of bioactive lipids The production of eicosanoids is initiated by the release of arachidonic acid (AA) after the hydrolysis of membrane phospholipids by phospholipase A 2 (PLA 2 ) and other enzymes. AA is further metabolized into different eicosa- noid families by several enzyme systems. The two main pathways are the cyclooxygenases COX-1 and COX-2 (and possibly COX-3 in some species), producing the prostaglan- dins (PGs) and thromboxanes (TXs), and the lipoxygenase pathways (5-LOX, 12-LOX, 15-LOX), producing the leuko- trienes (LTs), hydroperoxy fatty acids and derivatives thereof (Figure 1). The products of both pathways are involved in the induction and facilitation of numerous pathologies, particularly inflammatory diseases, and their suppression has, therefore, been a major therapeutic goal for anti-inflammatory drug development. Largely for historical reasons, the prevalent strategy has been to block selectively the enzymatic pathways leading to the pro- duction of inflammatory PGs or LTs (COX or LOX, respect- ively), or to suppress the production or action of a specific PG or LT, believed to be a key player in a specific disease. However, despite the attraction of this notion and the considerable resources invested, this approach has yielded mixed results, and high-profile drugs (e.g. some COX-2 inhibitors) have been withdrawn from the market because of safety concerns. A review of the field and analysis of the intricate path- ways of eicosanoid production and action point to several potential reasons for the unsatisfactory experience with such inhibitors. These can be mainly grouped under four headings:  Eicosanoids mediating particular functions might be produced by alternative pathways of AA metabolism.  Eicosanoids producing opposing effects on particular functions might be produced by the same pathway.  The same eicosanoids might exert opposing effects in different organs and tissues and might switch roles, from pro- to anti-inflammatory, during the course of disease.  Some ‘pro-inflammatory’ eicosanoids and AA-metabo- lizing enzymes also have essential physiological func- tions. The present review aims to highlight the factors that should be taken into account when considering the inhi- bition of eicosanoid production/action for the treatment of inflammatory diseases. Because of the enormous body of literature on this subject, the arguments discussed in this review are supported by illustrative examples and the coverage of the area is not exhaustive. Eicosanoids mediating particular functions might be produced by alternative pathways of AA metabolism It is well known that both COX and LOX pathways produce eicosanoids that mediate the inflammatory response. For example, the COX-derived TXA 2 and PGD 2 , and the LOX- produced cysteinyl leukotrienes (cys-LTs) and LTB 4 are potent inducers of smooth muscle contraction and/or con- striction of airways and blood vessels [1–4]. These eicosa- noids, as well as PGE 2 , also induce intestinal inflammation [5,6], as well as other inflammatory conditions, as illus- trated in Table 1. By blocking only one pathway, AA might be diverted into the other pathway; this would clearly not mitigate, and might even exacerbate, the pathological condition [7–9]. A well-known example is Aspirin1, the most widely used non-steroidal anti-inflammatory drug (NSAID), which by inhibiting COX can shift the AA metabolism to increased production of LTs, which might Review TRENDS in Pharmacological Sciences Vol.28 No.9 Corresponding author: Flower, R.J. (r.j.flower@qmul.ac.uk). Available online 10 August 2007. www.sciencedirect.com 0165-6147/$ – see front matter ß 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tips.2007.07.005