No abstract

The synthesis and CysLT1 antagonistic activities of a new series of 2-, 3-, and 4-(2-quinolinylmethoxy)- and 3- and 4-[2-(2-quinolinyl)ethenyl]-substituted, 2'-, 3'-, 4'-, or 5'-carboxylated chalcones are described. Structure-activity relationship studies indicate a preference for the presence of a negatively charged (acidic) moiety, although in some cases nitrile or ester analogues also exhibit moderate activity. The quinoline moiety may be substituted at either the 3- or the 4-position. Replacement of this heterocycle by other aromatic groups results in compounds with comparable affinities [2-(7-chloroquinoline), 1-(1-methyl-2-benzimidazole), or 1-(2-benzothiazole)] or substantially lower activities [1-(1-ethoxyethyl)-2-benzimidazole, 2-naphthyl, or phenyl]. The quinoline and chalcone moieties may be connected by either an ethenyl or a methoxy spacer. The acidic moiety at the chalcone B ring may be attached to the 2'-, 3'-, 4'-, or 5'-position, for both the 3- and 4-substituted chalcones. There are no general patterns to specify which substitution positions gave the most potent compounds. The series contains several potent CysLT1 receptor antagonists, with K(D) values approaching the nanomolar range, as measured by the displacement of [3H]LTD4 from guinea pig lung membranes. Antagonism of LTD4-induced contraction of guinea pig ileum, the inhibition of antigen-induced contraction of guinea pig trachea in vitro, and the inhibition of LTD4-induced increase of vascular permeability in vivo are determined for chalcones with high CysLT1 receptor affinities (K(D) values below 0.1 microM). 2'-Hydroxy-4-(2-quinolinylmethoxy)-5'-(5-tetrazolyl)chalcone (14, VUF 4819) showed good activity in both in vitro and in vivo assays and has been selected for further evaluation.

The synthesis and CysLT1 receptor affinities of a new series of highly rigid 3'- and 4'-(2-quinolinylmethoxy)- or 3'- and 4'-[2-(2-quinolinyl)ethenyl]-substituted, 6-, 7-, or 8-carboxylated flavones are described. CysLT1 receptor affinities of the flavones (down to 11 nM) were determined by their ability to displace [3H]LTD4 from its receptor in guinea pig lung membranes. Structure-affinity relationship studies showed that the relative positions of the carboxylic acid and the quinoline moiety were critical for CysLT1 affinities. While the carboxyl is optimal in the 8 position but tolerated in the 6 position, only the 6- and not the 8-tetrazole has significant activity. The quinoline moiety may be connected to the flavone skeleton by an ethenyl or a methoxy linker, but the substitution position is important for high affinity, especially in the 6-carboxylated flavones. 4'-Substituted 6-carboxyflavones are essentially inactive, whereas the 3'-substituted analogues have submicromolar CysLT1 affinity. Replacement of the quinoline by other heteroaromates generally leads to decreased affinities, with the phenyl and naphthyl analogues displaying only little or no affinity, while the 7-chloroquinoline analogue is comparable in activity to the quinoline. Flavones having CysLT1 receptor affinities of 10-30 nM were selected for determination of their inhibitory effects on the LTD4-induced contraction of guinea pig ileum in vitro. The IC50 values ranged between 15 and 100 nM. Compound 5d (8-carboxy-6-chloro-3'-(2-quinolinylmethoxy)flavone, VUF 5087) was selected for further research because of its high potency in the functional assay. This series contains the most rigid CysLT1 receptor antagonists known to date, and they are useful in the development of a CysLT1 antagonist model, which is discussed in the companion paper.

This paper describes the molecular modeling of leukotriene CysLT1 (or LTD4) receptor antagonists. Several different structural classes of CysLT1 antagonists were superimposed onto the new and highly rigid CysLT1 antagonist 8-carboxy-3'-[2-(2-quinolinyl)ethenyl]flavone (1, VUF 5017) to generate a common pharmacophoric arrangement. On the basis of known structure-activity relationships of CysLT1 antagonists, the quinoline nitrogen (or a bioisosteric equivalent thereof) and an acidic function were taken as the matching points. In order to optimize the fitting of acidic moieties of all antagonists, an arginine residue from the receptor was proposed as the interaction site for the acidic moieties. Incorporation of this amino acid residue into the model revealed additional interactions between the guanidine group and the nitrogen atoms of quinoline-containing CysLT1 antagonists. In some cases, the arginine may even interact with pi-clouds of phenyl residues of CysLT1 antagonists. The alignment of Montelukast (MK-476) suggests the presence of an additional pocket in the binding site for CysLT1 antagonists. The derived model should be useful for a better understanding of the molecular recognition of the leukotriene CysLT1 receptor.

No abstract