Lipidic mimetics as inhibitors of pancreatic phospholipase A2

Abstract

The regulation of phospholipase A2 (plA2) is a topic of pharmacological and medicinal interest due to the direct implication of the enzyme in the pathogenesis of inflammation. The hydrolysis of phospholipids by plA2, liberates fatty acids from the sn2 position. Arachidonic acid, produced mainly through this pathway from phosphatidylcholine and phosphatidylethanolamine, is the precursor of eicosanoids that are considered causative factors in inflammation. Numerous compounds have been synthesised as potent and specific inhibitors of plA2 including analogues of natural products and transition state phospholipid analogues, etc. ( I ) . Lipidic mimetics are synthetic molecules that resemble natural lipids either structurally or in hnction. In this study a series of a) amides b) esters c) alcohols and d) dipeptide analogues of lipidic amino acids were tested as inhibitors of porcine pancreatic plA2. Their inhibitory activities, expressed as an apparent IC50 value, are summarised in Table 1. The synthesis of these compounds was done by standard procedures (2) and their ihibitory activity was based on a bulk radiometric assay with 1% phosphatidylcholine as substrate (3). The reaction products were separated by TLC and quantitated on a Berthold Scanner. The dipeptide analogues cl , bl , b2 and b3 (Table 1) were designed to explore the specificity of the groups X and Y. By comparison of the activities of the dipeptides c l and bl. the free H2N group was shown to be necessary for inhibitory activity. The other lipidic mimetics included in this study, were designed to carry a free H2N group. A methylester (bl) or an alcohol (b3) in position Y were not distinguished by the enzyme (equally good inhibitory activity). The free carboxylic group was not favoured (b2) in position Y. The series of amides al , ll ,5 and a2,1-a2,5 were designed to explore the specificity of the R1 and R2 groups. When a Rl was a short side chain (4 C) the corresponding compounds exhibited significant inhibition (28pM for a1.l and 13pM for a1.5). The inhibitory activity was found to depend on the number of double bonds of the long unsaturated chain R2 (17-19 C): the active compounds carried one (al.1) or three (a1.3) double bonds. None of the compounds that carried a long saturated R1 side chain (a2,la2,5) exhibited inhibitory activity. The amides a1.5 and a2.5 carried a saturated long R2 chain, and differed in the length of the R1 chain. Their activity was in aggrement with the previous findings: a1.5 (4 C Rl chain) exhibited good inhibitory activity ( I Q o 13pM) but a2,5 (14 C R1 chain) was inactive. The pair a2,5 and c2 confirmed the requirement for a free H2N group at position X. The esters dl and d2 and did not exhibit any inhibitory activity, although they were designed with respect to the specifications for R1 and R2 as these resulted from the previous data. The compounds el and e2 were designed with one more spacer and a reversed esteric bond but proved also inactive, The new findings suggest that for the design of active compounds the amide bond is favoured. The most potent compounds of the presented series of lipidic mimetics exhibited an inhibitory activity with apparent ICso values of 11-19pM (Table I). Based on the results of this study, more potent and specific inhibitors of plA2 are under development. These findings suggested that this type of lipidic mimetic could contribute not only to the development of potent inhibitors of plA2 but other lipolytic enzymes as well, due to their resemblance to the natual lipid substrates.